Why Early Confucianism Cannot Generate Democracy

Transcription

1 Dao (2010) 9: DOI /s Why Early Confucianism Cannot Generate Democracy David Elstein Published online: 30 September 2010 # Springer Science+Business Media B.V Abstract A central issue in Chinese philosophy today is the relationship between Confucianism and democracy. While some political figures have argued that Confucian values justify non-democratic forms of government, many scholars have argued that Confucianism can provide justification for democracy, though this Confucian democracy will differ substantially from liberal democracy. These scholars believe it is important for Chinese culture to develop its own conception of democracy using Confucian values, drawn mainly from Kongzi (Confucius) and Mengzi (Mencius), as the basis. This essay describes some obstacles to this form of Confucian democracy. It argues that considering the political philosophies of Kongzi and Mengzi in the context of their views on personal cultivation reveals that they oppose some of the central assumptions of democracy. They do not trust the public to make good decisions, and advocate government for the people, but not by the people. These philosophies alone cannot generate democracy. Keywords Confucianism. Democracy. Kongzi. Mengzi 1 Introduction Several years before I entered graduate school, I attended a lecture by Wm. Theodore de Bary. The topic, as I recall, was human rights and Confucianism. Prof. de Bary talked about various expressions of rights in early Chinese thought, including the right of the people to replace a bad government. 1 I remember coming away with a new appreciation of early Confucian political thought and its support for democratic institutions. However, I am a little older now and, if not wiser, at least more wary of reinterpretations of ancient philosophies that make them fit modern conceptions. Certainly, if any tradition is to remain relevant, it must apply to life as we live it today and not just life two thousand years ago. 1 As this was probably close to fifteen years ago, I may have some of the details of Prof. de Bary s talk wrong. David Elstein (*) Department of Philosophy, SUNY New Paltz, 1 Hawk Drive, New Paltz, NY 12561, USA elsteind@newpaltz.edu

2 428 David Elstein Much has not changed about the fundamental questions we as individuals and as society face, and I believe Confucian thinkers have many insights into these questions, as do other traditions. Confucianism may well have a distinctive contribution to make to contemporary political theory, but perhaps we need to put aside the question of whether Confucian thought is democratic before that contribution can be appreciated. Confucian thought surely need not fit modern conceptions of democracy to be relevant. A glance at recent scholarship illustrates that many people do not share my view, as a number of conceptions of Confucian democracy have been put forth in recent years. Why has this question attracted so much attention? Obviously, a principal factor is the fact that China, the birthplace of Confucianism, is the world s most populous country, will soon have the largest economy in the world, and is not democratic. 2 Many thinkers find both Confucianism and democracy valuable, and try to find some way they can coexist. My goal for this essay is to clarify the terms of the debate around Confucian democracy and show some fundamental difficulties for one type of democratic interpretation of Confucian thought, the view that Confucianism can itself provide the theoretic foundation for democratic governance and generate democratic values. I will thus begin by describing some of the ambiguity in the debate so far. I will present significant and so far unresolved (to my knowledge) obstacles for this democratic interpretation of Confucian political thought, and discuss some previous views on this question and where I think they have gone wrong. This article is thus largely negative, though I hope it can help refocus the debate around Confucian philosophy and contemporary political theory into more constructive avenues. 2 Problems with the Debate In this section I will concentrate on three problems: how to define both Confucianism and democracy, and what Confucian democracy should be understood to mean. I will describe what I see as some of the problems, found in the work of many who have taken a position on whether Confucianism supports democracy, that make it difficult to achieve any kind of consensus on the question. Although reasonable people can definitely disagree on questions of interpretation (within certain limits), there is a lack of agreement on some basic concepts that leads to one wondering whether people on different sides are actually talking about the same thing. If they are not (as I will suggest is often the case), meaningful dialogue will be difficult. A first step to clarifying the question is thus understanding how democracy and Confucianism have been understood by various scholars writing on these issues. It is probably impossible to give a definition of democracy that will be universally accepted, as the numerous accounts show. ZHANG Shiqiang 張世強 and ZHANG Shize 張世澤 believe it is impossible to come to any consensus on what democracy or the Confucian tradition really is, so the question of whether they are compatible can never be resolved (Zhang and Zhang 2006: 10). This may be true, when neither term is defined clearly. However, even if (and because) total consensus on what these terms mean is unattainable, it is incumbent on anyone writing on the question of whether Confucianism can support democracy to define what 2 There have been limited experiments with democratic elections at the local level, but opinions on the fairness of these elections are mixed, and so far elections have been limited to rural villages (O Brien and Li 2000; Kennedy 2002; Zweig and Fung 2007).

3 Why Early Confucianism Cannot Generate Democracy 429 democracy is. Leaving the concept democracy vague and unspecified means it could be compatible with nearly anything (see also Hu 2000: 55). Yet some writers have failed to do that much (for example, Murthy 2000; Xu2006; Wang and Titunik 2000). Of the writers that do give some definition, these vary considerably. Daniel Bell defines democracy in terms of political institutions: A democratic system at minimum must include regular elections based on universal franchise (Bell 1999: 452). HU Shaohua s definition is a little broader but still emphasizes political and legal aspects. He boils democracy down to the right to participate in government on the one hand and freedom from undue government interference on the other; as he puts it, democracy is popular sovereignty and individual liberty (Hu 2000: 57). LI Chenyang argues that what is important are democratic values, because without these, elections and other democratic institutions become meaningless. He defines democracy as a value structure that encompasses liberty, equality (which itself needs some definition), and pluralism (LI Chenyang 1997: 184). Presumably, democratic institutions would be important as well, as these values demand no particular political structure. Many scholars take issue in particular with liberal democracy and look to Confucianism for an alternative. What Brooke Ackerly finds most problematic in Western liberal democracy is the assumption of an autonomous individual rightsbearing citizen. Like many other writers who treat the Western liberal tradition as unitary, what Ackerly seems to mean is actually the American version of liberalism: Capitalism and consumerism have allowed autonomy to manifest itself as a pursuit of self-interest that threatens social cohesion (Ackerly 2005: 548). This seems a more accurate characterization of American (or Anglo-American) rather than continental European values, which put more emphasis on social welfare. In her view, democracy is governance consistent with political equality that is functional, not merely formal. She emphasizes functional equality since formal, legal equality can be undermined by unequal social and economic arrangements. Despite her challenge to liberalism as the only foundation for democracy, her list of essential democratic institutions is fairly conventional: elected representative government with mechanisms for accountability, a free press, independent judiciary, and a civil society (Ackerly 2005: ). Those scholars who argue for a pragmatist, Deweyan understanding of democracy would likely find Ackerly s approach congenial in some respects, as they also find a challenge to liberalism in Confucian and Deweyan democracy. David Hall and Roger Ames did much to bring the question of the possibility of Confucian democracy to the prominence it enjoys today through their important work, The Democracy of the Dead. TAN Sor-Hoon has presented her own Dewey-inspired pragmatist understanding of Confucian democracy in several works, notably Confucian Democracy: A Deweyan Reconstruction. These pragmatists offer the most radical understanding of democracy of all, since as Hall and Ames describe it, Deweyan democracy is not really about governmental institutions or political practices at all. Democracy is about having an informed, cooperative community working together to resolve common problems, a communicating community (Hall and Ames 1999: ). 3 In this respect, these thinkers diverge from Ackerly, who, while challenging liberalism, is still committed to an essentially political understanding of democracy. 3 Since my topic in this article is Confucian democracy rather than Deweyan, I will not consider whether this is an accurate interpretation of Dewey s views. This is the Dewey-inspired understanding of Confucian democracy that Hall and Ames want to argue for, which is the important thing for my purposes. For some suggestions of areas in which Hall and Ames s account of Dewey is problematic, see Cline 2003.

4 430 David Elstein So here is our problem. Some writers understand democracy as elected representative government, and some as certain rights or values that people should be committed to, which are best protected by democratic political institutions. Some writers reject liberalism as the only style of democracy while wanting to preserve democratic institutions, while others explicitly reject voting based on universal suffrage while wanting to uphold classic liberal values such as individual freedom from government interference (for certain people at least) and rule of law (Bai 2007: 202, 204). For most scholars, democracy is essentially a philosophy of government, while the pragmatists define it in terms of a certain kind of community, and this kind of democratic community seems compatible with almost any form of government. The wide variety of definitions proposed means that people may not even be talking about the same thing when they talk about democracy. If they are not, it is no surprise that there is little consensus in the debate. To know if Confucianism can support democracy, we first have to know what democracy means. We also have to know what Confucianism means. As HU Shaohua has pointed out, both sides of the debate pick and choose those parts of Confucianism that support their arguments. To generalize, philosophers and historians generally focus on one of two areas: the founding fathers Kongzi (Confucius, ca BCE) and Mengzi (Mencius, ca BCE), 4 and what is commonly known as Neo-Confucianism, the resurgence of Confucian thought in the Song, Ming, and Qing dynasties (HUANG Zongxi 黃宗羲, , is a particular favorite). Political theorists and sociologists focus on current cultural values that they associate with Confucianism, often without trying to trace them to a specific thinker (for example, Rozman 1991). 5 As HU Shaohua has described it, Confucianism can refer to three fairly distinct, though overlapping, things: (1) the thought of Kongzi, Mengzi, and Xunzi; (2) a state ideology that dominated China for more than two thousand years ; and (3) a metonym for Chinese culture, particularly pre-19th century Chinese culture (Hu 2000: 56 57). This unfortunately ignores Neo-Confucianism but remains useful. I agree with Hu that this last use of Confucianism to refer to Chinese culture is mistaken and biased, as it ignores the considerable impact of Daoism, Buddhism, and other influences on Chinese culture while simultaneously treating Chinese culture as something static and unchanging (for example, see Hall and Ames 1999: 31). Like democracy, Confucianism is a term that means different things to different people. Even if, as I hope, people will stop using Confucianism when they mean Chinese culture, that leaves us with a very broad range of ideas and writers who have been called Confucian, from Kongzi himself in the 5 th century BCE to contemporary New Confucians (xin rujia 新儒家 ). 6 Distinctions like Neo-Confucianism and New Confucianism help indicate that these philosophies are not exactly the same as the Confucianism of the Warring States, but these distinctions are often lost and Confucianism is presented as an unchanging monolith. This is far from the case. 7 Just like any tradition, Confucianism has changed over its history, and the early philosophy is often quite different from later thinkers views. Thus, I do not think we can talk about 4 A third early figure, Xunzi, is generally ignored, partly due to the fact that he had less influence on later Chinese thought, and partly because his thought is even less conducive to democracy. For more on Xunzi, see Hutton Obviously, this generalization does not cover everyone who has written on the subject. 6 On New Confucianism, see Liu 2003; Makeham The Neo-Confucians of the Song and Ming dynasties may have seen themselves as faithful followers of Kongzi and particularly Mengzi, but scholars such as Philip Ivanhoe have made a convincing case that they actually transformed their thought considerably (Ivanhoe 2000 and 2002).

5 Why Early Confucianism Cannot Generate Democracy 431 whether Confucianism can support democracy without asking which democracy and which Confucianism we are talking about. The nature of the Confucian tradition (and tradition itself) also raises the question of how much revision and reinterpretation are possible while still remaining part of that tradition. No one, as far as I know, suggests that any pre-modern Chinese thinker advocated a form of democracy, nor did any consciously reject it. So the question is, rather, can the thought of any Confucian philosopher be interpreted in such a way that it supports democracy? Much will hinge on how much latitude one can take with such interpretation. An interpretation that either ignores or implausibly revises key beliefs and values of the philosophers under discussion should be questioned, and one goal of this essay to show how some supporters of Confucian democracy have ignored or implausibly revised key beliefs and values represented in Confucian thought. TAN Sor-hoon often demonstrates this, such as when she concedes that Confucianism advocates government for the people rather than by the people, but then adds, It can and must be reconstructed to advocate government by the people (Tan 2004: 145). The question becomes how to do that. Tan s main argument comes from Dewey that government by the people is necessary to ensure government for the people. 8 If the basic idea comes from outside the tradition but can still provide a coherent and consistent interpretation of Confucian thought I would have no objection; however, I will consider this interpretation in detail later and show that it does not seem consistent with Kongzi and Mengzi s beliefs. It seems that if a persuasive philosophical argument is to be made that Confucianism can generate democracy, it must show that some Confucian values can provide grounds for democratic views and that such values are fundamental to at least certain parts of the Confucian tradition and other main values do not conflict with these views while those that do conflict, such as support for hereditary rule, are tangential. Not conflicting with democracy alone should not be the standard, because one way of not conflicting would be by being wholly unrelated. If core Confucian values simply have no bearing on democracy at all, I do not think it can be said that Confucianism supports democracy, nor would it be opposed to democracy. 9 We will need to find some core values that imply democratic political structures, while explaining how other values either do not conflict with democracy or can be reasonably thought to be less important than those that support democracy. I think scholars such as Tan and XU Keqian try to make such an argument, and I think it ultimately fails, as I will try to illustrate. To put it another way, I understand Confucian democracy as (1) a form of democratic governance based on Confucian values, and not (2) a democratic polity that implements Confucian values but is not structured upon them. I do this for three reasons. Foremost, the authors whose constructions of Confucian democracy I will be objecting to in this article employ a similar understanding. TAN Sor-hoon, for example, in one article sees her project as developing a philosophical basis for Confucian democracy that provides a positive answer to the question, To what extent could democracy and other related ideas of liberty, equality, or rights be generated from Chinese culture? (Tan 2007: 141, 147). Similarly, XU Keqian asks, Should China find from its own traditional culture some resource that will be able to legitimize the modern democratization? and subsequently makes clear that his answer to this question is it should (Xu 2006: 136). 10 Second, we must recognize that the 8 CAI Renhou makes a similar argument, though he defines democracy as a political system and not in pragmatist terms (Cai 1987: 13 14). 9 This, I think, would be an accurate description of HU Shaohua s position (Hu 2000: 61). 10 The subtitle of this article, The Potential Theoretic Foundations of Democracy in Modern China, is also telling.

6 432 David Elstein question of whether Confucianism supports democracy is more than an abstract philosophical question. It arose largely in response to actual socio-political conditions in countries where Confucianism is important, and while I cannot detail these reasons here, it would be a mistake to ignore them. 11 Preserving the connection with the Confucian tradition is one important motive in making an argument for Confucian democracy (e.g., Tan 2004: 133; Tan 2007; Xu2004; 2006), and if the democratic values themselves have other sources, this connection would be weakened. Third, since by itself it is not clear whether Confucian democracy refers to (1) or (2) above, I think distinguishing Confucian democracy for (1) and Confucian-inspired democracy for (2) is a useful disambiguation. My argument in this article will focus on the difficulties of Confucian democracy so understood, but a Confucian-inspired democracy may still be a viable option (e.g., Ackerly 2005, Yang 2004). To sum up then, I think an effective discussion of whether Confucianism can support democracy first has to define both democracy and Confucianism, taking into account the dynamic nature of Confucian tradition. When I say support, I mean more than the possibility of coexistence. Confucianism and democracy could coexist in the same society by being entirely separate value systems that have little effect on each other; Confucianism would be largely relegated to personal ethical values and have little political impact in such a case. As I understand it, this is not what is meant by a Confucian democracy. By support, I mean it is possible to make an argument for democracy using Confucian values, in a way that is consistent with how those values are applied in Confucian philosophy. 12 When I speak of Confucian democracy, I mean an understanding of democracy supported by Confucianism in this way. The challenge for those arguing that Confucianism can be democratic is to explain why the undemocratic elements of Confucian thought do not make this understanding of Confucian democracy impossible. In section four I will examine some of those undemocratic elements and possible responses to them. 3 Defining Democracy and Confucianism I will begin by giving the definitions of Confucianism and democracy that I will use in this section. I will begin with Confucianism, since that will be the easier of the two. Confucianism in this study will refer to the thought of the Analects, a text associated with Kongzi, and the Mengzi, the book that is the source for Mengzi s thought. 13 When I refer to Kongzi or Mengzi, it should be understood as indicating the philosophies in those books, not to any specific historical figures. I focus on the thought of Kongzi and Mengzi for two reasons. First, they have been the most influential Confucian philosophers in Chinese history, such that later Confucian thinkers saw themselves as commenting on and explaining their thought even when they were doing something original. Second, most 11 I plan to examine these in greater detail in future work, but let me give just one example for the moment: the Asian values position initially claimed by LEE Kuan Yew of Singapore. He argued that democracy conflicts with traditional Asian values, and this was a reason for rejecting it (Zakaria 1994). In response, some scholars have tried to show a Confucian basis for democratic values to show that traditional Asian values do in fact support democracy. 12 This understanding of support for democracy can also be found in Xu 2006: 137. In his Chinese article, he talks about compatibility (jianrong 兼容 ),but also states that Confucian thought can serve as the theoretical premises of democratic government (Xu 2004: 5), so the meaning seems essentially the same. 13 I refer to the books rather than the thinkers themselves because neither of these books was written by Kongzi or Mengzi.

7 Why Early Confucianism Cannot Generate Democracy 433 previous writers who argue for a Confucian democracy use the ideas of Kongzi and Mengzi. Mengzi in particular is often considered to be the most democratic Confucian philosopher until at least the Ming dynasty, so if I can show that the best case for Confucian democracy faces substantial challenges, it is unnecessary to examine thinkers who are generally thought to be less democratic. This means that Neo-Confucian thought is outside the scope of this article. Democracy is, as I have acknowledged, a more difficult problem. One could begin by going back to the Greek roots of the term and show it means rule by the people, but understandings of who the people are have changed considerably over the course of history. In early Greek democracy a minority of citizens ruled over a larger populace who had no input into politics, a situation that today might be described as oligarchy rather than democracy. The history of the United States is characterized by an expansion of the people, from property owners to white men generally, women, people of color, and finally anyone over the age of eighteen. How wide does the franchise have to be to qualify as democracy? There is no agreement on this question. Further confusion comes from treating democracy and republic as synonymous. The founding fathers of the United States wanted a republic, which sets limits on the government s powers (Everdell 1983: 2 7). The essence of republican philosophy is sometimes described as freedom from arbitrary power, including arbitrary action by a majority (Laborde and Maynor 2008: 3). A democracy, on the other hand, is simply government by majority will. The majority or its representatives can make any laws they like. However, as William Everdell wrote, The term democracy has ballooned so alarmingly that it has absorbed almost all that remained of the alternative meanings of the word republic (Everdell 1983: 7). Many advocate democracy in China because they want to limit the government s powers, a common feature of republicanism. This terminological confusion may be regrettable but is probably here to stay. It seems to me that a common feature of most understandings of democracy is effective popular sovereignty: the people have real power in making decisions concerning the operation of their government, whether directly or through elected representatives. Effective power likely involves certain presumptions about freedom, equality, and rights but these need not be examined here. This broad definition would include republics, which differ in the establishment of limits on what the majority can do. I propose this as a minimal definition of democracy that captures a commonsense understanding of the term, and that can be related to most of the other definitions that I mentioned above. Note that this will exclude pragmatist understandings of democracy which do not see it as a political philosophy, so I do not intend my arguments here as an objection to a pragmatist conception of Confucian democracy. Pragmatist democracy is rather idiosyncratic and arguably does not capture much of what most people think of as democracy, such that even a pragmatist Confucian democracy might well not be considered very democratic by non-pragmatists. Connections between pragmatist democracy and Confucianism will thus largely be outside the scope of this article, which will concentrate on democracy as a political philosophy. However, I will still engage with pragmatists (notably TAN Sor-hoon) insofar as they do focus on questions of governing. I emphasize decision-making power as critical to democracy because I think it is crucial that the people do not simply express opinions; they have power to realize their ideas in government. What many writers argue is that early Confucian political theory is government for the people (for example, Hall and Ames 1999: 219). I think this is absolutely true. But a benevolent dictator can rule for the people (and I would argue that this is essentially what Kongzi and Mengzi had in mind). Ruling in the people s interestis

8 434 David Elstein not enough for democracy (LI Chenyang 1997: 186; LI Zehou 1998: 205). The people have to be involved in making decisions concerning how to govern the state, not just giving advice to the rulers who decide on their own. Furthermore, that decision-making power must be protected, and not simply granted or taken away at the whim of the ruler. This may be the appropriate time to talk about Confucianism and rights before I go on to give my reading of Confucian political thought. Obviously, there is considerable overlap between discussions of human rights and democracy, though in principle many rights are distinct from any particular form of government. LI Chenyang suggests that democracy requires rights, but the converse is not necessarily true (Li 1997: 187). A non-democratic regime could not guarantee a right to vote, but it could guarantee a right to free speech. China, for example, does say citizens have freedom of speech in its constitution, though actually protecting this right is another matter. Since my topic is democracy rather than human rights, I will not consider much of the considerable scholarship on Confucianism and human rights, except insofar as it directly relates to the question of democracy (on rights see de Bary and Tu 1998; Angle 2002; Svensson 2002; Bauer and Bell 2002). This means that I will focus on questions related to the possibility of political participation and not protection of rights that do not relate to this. 4 My Reading of Confucian Thought With that said, I can now present my understanding of Confucian political thought and its relationship to democracy. My position is closest to LI Chenyang and HU Shaohua. Li has argued that essential conflicts between Confucian and democratic values make them impossible to integrate, though they could coexist as distinct value structures (Li 1997: 190). Hu concludes that Confucianism is not democratic but not necessarily opposed to democracy (Hu 2000: 61). My position is a bit more negative: I agree that Confucianism is not democratic, but in terms of political philosophy, it is opposed to democracy. There are certainly elements in Confucianism that are compatible with democracy, but this does not mean that Confucianism is democratic, without further qualification. To say that Confucianism is compatible with democracy means discounting its own political theory. What can be sacrificed while still remaining Confucian is a contentious question, which likely will receive very different responses. My own view is that sacrificing enough to allow for democracy results in a political philosophy that has substantial conflicts with the thought of Kongzi and Mengzi. I suggest that such a philosophy may be Confucian-inspired but is certainly a modification of early Confucianism. The essence of early Confucian political theory is, to turn John Adams s phrase around, a government of men and not of laws. Rather than advocating institutions to limit what bad rulers could do, Kongzi and Mengzi put their efforts into trying to convince rulers to rule well. In the remainder of this section I will first examine elements of Kongzi s political views that present difficulties for a democratic interpretation: distrust of the rule of law, and lack of confidence in the ability of people to make good decisions. Following that, I will examine some of the best cases for providing a foundation for democracy in Mengzi s thought that proponents of Confucian democracy often point to. I will show that reading these in context leads to quite different conclusions. For Kongzi, the determining factor for whether a state will be governed well or poorly is the character of the ruler. The consequence of this is that the Analects has little confidence in laws and institutions as means of governing. Governing well requires morally transforming the people, and laws simply cannot do that: If you guide the people with

9 Why Early Confucianism Cannot Generate Democracy 435 laws and reform them with punishments, they will evade them 14 and have no sense of shame. If you guide them with virtue and reform them with the rites, they will have a sense of shame and correct themselves (Analects 2.3). 15 The Analects has very little to say about the substance of government: what laws should be enacted, how government should be structured, how to prevent corruption. There is not much interest in these things because it all comes down to the moral character of the ruler. Under a good ruler, officials will not be corrupt and the people will be good. Under a bad ruler, institutions are not going to be much help. JI Kangzi once suggested executing those who did not follow the proper Way. Kongzi responded, Why must you use executions in your governing? If you yourself want goodness, the people will be good. The virtue of the gentleman is like the wind; that of the people is like the grass. When the wind blows over it, the grass is sure to bend (Analects 12.19). There is no sign that any government institutions can prevent the ruler affecting the people, positively or negatively. The importance placed on the ruler s virtue causes problems for seeing Kongzi s thought as compatible with democracy. The Analects has a very top-down model of governing. It is true that Kongzi talks about caring for the people and winning their trust (Hu 2000: 58). Indeed, he says that having the people s confidence is more important than a state having sufficient food or weapons (Analects 12.17). But again, caring for the people and having their trust are very different from giving them a say in government. I see no evidence that Kongzi thinks this would ever be a good idea in the Analects. Democracy might be one way to try to ensure that the government does care for the people and gains their trust, but while these might be necessary conditions for a successful democracy, they are certainly not sufficient. A government need not be democratic to care for the people and have their confidence. The Analects instead displays very little confidence in the ability of most people to understand enough to be trusted with active participation in government. In my opinion, one of the strongest anti-democratic passages in the whole book is 8.9, which is ignored by most who believe Confucianism can support democracy: The people can be made to follow, but they cannot be made to understand. 16 Shaun O Dwyer is one of the few that mentions this passage and its implications against any possibility of effective government through popular sovereignty (O Dwyer 2003: 44). However, he still seems to ignore what it suggests about the impossibility of having an articulate public direct their own associations, as in a more pragmatist understanding of democracy. It must be said, though, that O Dwyer is not attempting to give a faithful interpretation of Kongzi s thought but instead is freely adapting him (O Dwyer 2003: 47). Perhaps there might be room for democratic administration of non-governmental associations according to the pragmatist understanding of democracy, but the impossibility of the public understanding government certainly weighs against democratic participation. The Analects also shows a distrust of majority opinion that indicates opposition to democracy, especially direct democratic methods such as referenda. When a disciple asks what Kongzi thinks of someone whom everyone in a village likes, or someone whom 14 This phrase could grammatically refer to evading either the laws or the punishments. 15 On the importance of the rites, see Ames There are various interpretations concerning exactly what they can follow but not understand, but none seem to me very conclusive. Another way to understand the passage would be, One can let the people follow, but not let the people understand, making it prescriptive rather than descriptive. Either understanding causes serious problems for a democratic reading of Kongzi. This passage caused trouble for many later Confucian philosophers, who often tried to interpret it in such a way that it did not give the impression that the people are necessarily ignorant. For one such interpretation, see Huang 2008.

10 436 David Elstein everyone hates, Kongzi rejects this as a standard for determining who is good. Better, he says, is someone whom the good like and the bad hate (Analects 13.24). Similarly, he says when the masses like someone or hate someone, one must still investigate the person oneself. 17 This suggests that Kongzi would disagree that majority vote is a good method for selecting rulers, since the fact that the majority chooses someone is no guarantee that he is good. Mengzi s thought is generally offered as the best case for a Confucian democracy. For example, XU Keqian bases his argument that classical Confucianism supports democracy largely on Mengzi s ideas, particularly the importance of the people s will and the doctrine that the people are more important than the ruler (Xu 2006: ). Both of these points are often considered as part of a more general political theory that the people are the foundation of the state (minben 民本 ), commonly mentioned in discussions of Confucian democracy (such as Murthy 2000; Wang and Titunik 2000; LI Jinquan 2006). Some consider this as evidence of political equality in Mengzi that could be developed into a democratic theory. Others point to Mengzi s defense of a right of revolution, justifying overthrowing tyrants (Tu 1993: 6; Bai 2007: 206). However, I think when we look more closely at these, we will again find that Mengzi absolutely advocates rule for the people (in their interests), but not rule by the people. In one frequently cited passage, Mengzi does say that the people s will is important in determining who the ruler will be. In a passage about the legendary sage-kings, Mengzi explains that Yao recommended Shun, and it was actually heaven that gave the empire to him, not Yao. What Mengzi means by this is that the people accepted him: Yao recommended Shun to heaven and heaven accepted him. He presented him to the people and the people accepted him. However, I think this should not be understood as a kind of election. The main indication that the people accepted him is not any kind of vote, but that he was put in charge of government affairs and they were well ordered and the people were content. This is being accepted by the people (Mengzi 5A5). This kind of passive acceptance falls short of true exercise of substantive popular sovereignty. If this is the standard, any country where the people are content with their government is a democracy. The passage says that the public preferred Shun to Yao s son, but this is hardly surprising considering that he was Yao s chosen successor and the public was familiar with him through his twenty-eight years of assisting Yao. There is another dimension to this passage that is not often mentioned. According to Mengzi s understanding of history, Yao picked Shun to succeed him and Shun picked Yu. Yu then established the principle of hereditary succession and founded the first dynasty. At that point, the people s acceptance no longer mattered (Wang and Titunik 2000: 79). Being accepted by the people only seems to matter when a dynasty has not been established, or when the dynasty has become so corrupt that it is replaced. Neither Mengzi nor any other early Confucian challenged the principle of hereditary succession (LI Chenyang 1997: 185). 18 There is another expression of the people s will that may seem democratic, but it too is often exaggerated. This is the need for the ruler to solicit public opinion in hiring and removing officials, and in carrying out executions. In both of these instances, Mengzi cautions the ruler not to listen just to those around him. Only when the citizens (guoren 國人 ) all say a person is 17 This is probably addressed to the ruler, who should not rely on majority opinion when selecting officials. We will see a similar idea in Mengzi. 18 Confucian thinkers did not challenge hereditary succession, but other authors in the Warring States did, though their writings were lost for most of Chinese history (Pines ). Even these authors did not consider anything like a democracy, but rather argued for rulers abdicating after choosing a worthy successor.

11 Why Early Confucianism Cannot Generate Democracy 437 worthy should the ruler hire him, and only when the citizens all say a person is unworthy should the ruler remove him. 19 The same is true for carrying out executions. But we must be careful. CHENG Chung-ying says that, in this passage, It is the people s judgment that should be decisive (Cheng 1998: 151). A careful reading does not bear this out. Mengzi does not say to hire whomever the citizens say is worthy; he says that when they say someone is worthy, the ruler should investigate, and only hire him if he really is worthy (Mengzi 1B7). The same is true for removing officials and executing criminals. After hearing public opinion, the ruler must still investigate on his own, and judge for himself whether public opinion is right before acting. The ruler should listen to public opinion, but he still makes the final decision (Murthy 2000: 37;Bai 2007: 208). It is instructive to pay attention to the metaphors Mengzi uses, and a common metaphor when talking about governing is that the ruler is parent to the people (Mengzi 1A4, 1B7, 3A3). I do not suggest this means that rulers have unconditional power over the people. I think the point of this metaphor is again to encourage rulers to act in the people s interests. As parents act in the best interests of their children, so should the ruler in the interest of his people (Chang 1998: ). At the same time, it establishes a clear hierarchy that is incompatible with true political equality. 20 It is impossible for parents and children to change roles, and I think Mengzi and other Chinese thinkers feel it would be equally unnatural for the people to decide who the ruler is. In the case of parents, this may be mitigated by the fact that children grow up and can become parents themselves, so that progression through life means that one is not always in the subordinate position (Hall and Ames 2003: 140). While this is true for parents and children, it is not true in Kongzi s and Mengzi s understandings of government. There is no progression from subject to ruler. A ruler should absolutely govern for the benefit of the people, and even solicit their opinions in certain instances, but a ruler who can be replaced by popular decision does not fit well with the image of a parent to the people. Mengzi further argues for hierarchy when discussing rulership with a follower of another philosopher. Mengzi argues that the demands of governing make it incompatible with pursuing other occupations at the same time, and that is why division of labor is necessary. Citing what may have been an adage of the time, he says, Some work with their minds, and some work with their strength. Those who work with their minds govern others; those who work with their strength are governed by others. This is the universal principle of the world (Mengzi 3A4). The point could not be much clearer: some rule and some are ruled, and these are distinct groups. Mengzi comes close to suggesting that these classes are determined at birth (certainly the ruler is), though I do not think he necessarily has to be read that way. It may be more a matter of social environment and opportunities. Someone born into a family of subsistence farmers is probably going to find it impossible to spend the years in study that Mengzi thinks are necessary to qualify to work as a government official, but that does not mean the person is incapable of learning (Mengzi 1B9; de Bary 1998: 8). A number of writers have found these ideas in the Mengzi very difficult to square with his more egalitarian sentiments, which are presumed to take precedence (for example, Xu 2006: ; Bai 2007: 206). The difference is more than simply that different jobs have 19 Some say that this means everyone in the state, which I think is unlikely. First, there would be a tremendous practical problem in gathering the opinions of the entire population, or even a majority. Second, while in modern Chinese guo does indeed mean country, its original meaning, still used in Mengzi s time, was a walled city, particularly the capital of the state. A more likely reading is that Mengzi means listen to the people in the capital. 20 This is often considered crucial to democracy; see the definitions by Ackerly 2005 and LI Chenyang above.

12 438 David Elstein different rewards, as CHANG Wejen says (Chang 1998: 124). Mengzi suggests that there is a difference in kind when he says some rule and some are ruled. So what can we make of such statements as, The sage and I are the same in kind, or that it is not only the worthy who have moral principles but all people have them (Mengzi 6A7, 6A10)? It is not so difficult to explain, I think. It is the difference between having equal potential and actually being equal. Mengzi and later Xunzi combine a profound optimism on the theoretical perfectibility of all people with a deep pessimism about whether people will actually do what is necessary to develop their potential. Xunzi put it well: The man on the street could become a Yu, but is not necessarily able to become a Yu (Xunzi ). What both Mengzi and Xunzi are talking about is the difference between having a potential and realizing it. Everyone may have the potential to be a sage, but most people will never realize it. 21 Universal education does not solve this problem, because it cannot ensure that people will make the necessary effort. In another passage Mengzi states that government should only be entrusted to those with the specialized training to know how to administer a state (Mengzi 1B9). He considers the masses essentially incapable of understanding the Way (Mengzi 7A5). This implies that he would not say that equal potential is grounds for universal political participation, any more than arguing that I could potentially run a marathon, which would get me a spot in the Olympics. Developing the potential is what determines participation. One might grant this but then argue that this is still consistent with a representative democracy, where the majority s role is to vote for those who actually run the government. 22 If the public simply chooses among candidates who are qualified to hold office (perhaps enforced by relevant restrictions on who can be a candidate), this could ensure that only the qualified hold office while still allowing democratic elections. This proposal has some merits, but faces a number of problems. First, since Mengzi argues that the masses are incapable of understanding the Way, could they determine which candidate best embodies it? It is difficult to see how. Alternatively, if participation in elections is restricted to candidates who do understand the way of good governance (however that is to be determined), do the voters still have a substantive role in the process? It looks as if there will be no real choice of candidates, since anyone who runs will essentially have to meet Confucian standards for qualifications. The real power seems to be held by those who control the nomination process. 23 When voters choices are so restricted, the system begins to look quite undemocratic. Mengzi s supposed right of revolution is the final point that needs examination. If Mengzi defends such a right, or something analogous to it (it being noted by many that classical Chinese has no term for rights 24 ), it would be an obvious parallel to John Locke, whose thought was particularly influential in the development of British and American democracy. I agree with Wm. Theodore de Bary and Justin Tiwald that what Mengzi had in mind is significantly different from Locke (de Bary 1998: 8; Tiwald 2008). Let us compare them. Locke wrote, Whenever the legislators endeavour to take away, and destroy the property of the people, or to reduce them to slavery under arbitrary power, they put themselves in a 21 Hutton further illustrates this point in Xunzi s philosophy (Hutton 2008: ). 22 This was suggested by an anonymous reviewer. 23 In fact, such a situation is quite similar to officials attempts to control the nominations in some village elections in China. See citations in note Perhaps the most complete account of the development of Chinese rights language is Angle 2002:

13 Why Early Confucianism Cannot Generate Democracy 439 state of war with the people, who are thereupon absolved from any farther obedience.whensoever therefore the legislative shall transgress this fundamental rule of society; and endeavor to grasp themselves, or put into the hands of any other, an absolute power over the lives, liberties, and estates of the people; by this breach of trust they forfeit the power the people had put into their hands for quite contrary ends, and it devolves to the people, who have a right to resume their original liberty, and, by the establishment of a new legislative, (such as they shall think fit) provide for their own safety and security. 25 (Locke 1980: sec. 222) It is important to note here that Locke explicitly talks about the people as both the source of original governmental authority and those with the right to establish a new government if the old infringes its mandate. There are problems with understanding this as a right at all according to some concepts of rights in which they are interests or entitlements which either are or ought to be legally protected. As Randall Peerenboom puts it, One of the defining features of rights for most people is that they are legally enforceable (Peerenboom 1998: 259; see also Twiss 1998: 28; Ching 1998: 68). This is certainly not the only way to understand rights, but it is a common understanding. 26 But how could a right of revolution be legally protected? Revolution is the response to absence of democracy, where there is no other way for the public to make their will effective (Wang and Titunik 2000: 88; Tan 2004: 140). But perhaps we should not make too much of semantic issues and exactly what defines rights. The more important issue is that the way Mengzi talks about this is significantly different from the way Locke does. King Xuan of Qi asked, Is it true that Tang banished Jie and King Wu attacked Zhou? Mengzi answered, According to what has been transmitted, it is so. The king asked, Is it permissible for a subject to commit regicide? Mengzi answered, Someone who harms benevolence is called a bandit and someone who harms appropriateness is a savage. A bandit and savage is just a man. I have heard of the punishment of the man Zhou, but I have not heard of regicide. (Mengzi 1B8) This is a commonly cited passage that purports to show that the people can replace a tyrant without it being considered regicide (for example, Hu 2000: 59;WangandTitunik2000: 82, 85; Yang 2004: 90). However, we need to look closely at what Mengzi says and take into account the history he is talking about. Mengzi does not say anything about the people. Tang and King Wu acted correctly, but they were both already nobles, not commoners, and happened to be sages. This passage must be read in light of Mengzi 5B9, which most writers ignore. 27 Here, Mengzi distinguishes between two kinds of ministers: ministers who are related to the royal house, and those who are not. In the case of a tyrannical ruler, their responses are very different. Those who are not related to the royal house first remonstrate, and if not heeded they leave to seek employment elsewhere. Ministers of the royal house first remonstrate with the ruler, and if not heeded they replace him. So it is not the public who decides to replace a bad ruler; it is his relatives. Ordinary people as well as officials have the option to leave, and Mengzi thinks they will (Mengzi 1A6). This is what enables benevolent rulers to attract people to their states. There is no general right to overthrow a bad ruler (see Tiwald 2008). 25 Italics in original. 26 Among contemporary rights theorists, this view is strongly implied by Alan Gewirth 1978: 72; 1996: Angle and de Bary are the only ones I have seen that take it into account (see 2002: 125; and de Bary 1998: 8).