Just Hierarchy: Why Social Hierarchies Matter in China and the Rest of the World 9780691200880

Citation preview

j ust h i e r a rch y

Just Hierarchy w h y soci a l h i e r a rch i e s ­m at t e r i n ch i na a n d t h e r e st of t h e wor l d

da n i e l a . be l l wa ng pe i

pr i nce­t on u n i v e r sit y pr e ss pr i nce­t on & ox for d

Copyright © 2020 by Prince­ton University Press Requests for permission to reproduce material from this work should be sent to permissions@press​.­princeton​.­edu Published by Prince­ton University Press 41 William Street, Prince­ton, New Jersey 08540 6 Oxford Street, Woodstock, Oxfordshire OX20 1TR press​.­princeton​.­edu All Rights Reserved Library of Congress Control Number 2019954572 ISBN 978-0-691-20089-7 ISBN (e-­book) 978-0-691-20088-0 British Library Cataloging-­in-­Publication Data is available Editorial: Rob Tempio and Matt Rohal Production Editorial: Jill Harris Jacket Design: Carmina Alvarez Production: Brigid Ackerman Publicity: Alyssa Sanford and Kate Farquhar-­Thomson Copyeditor: Jay Boggis This book has been composed in Arno Pro Printed on acid-­free paper. ∞ Printed in the United States of Amer­i­ca 10 ​9 ​8 ​7 ​6 ​5 ​4 ​3 ​2 ​1

c on t e n t s

Acknowl­edgments  ​vii Introduction​1 1.  What’s Wrong with Hierarchy?

8

2.  In Defense of Hierarchy

13

3.  From China to the World

21

1 Just Hierarchy between Intimates: On the Importance of Shifting Roles

29

1.  Relations with Friends

31

2.  Relations with Lovers

36

3.  Relations with ­Family Members

45

4.  Relations with House­keepers

55

2 Just Hierarchy between Citizens: On the Importance of Ser­vice

66

1. Justifying Hierarchical Po­liti­cal Rule in the Chinese Context

68

2. Is It Pos­si­ble to Limit Po­liti­cal Power without Competitive Elections?

78

3. Po­liti­cal Meritocracy as the Prob­lem, Po­liti­cal Meritocracy as the Solution

84

v

vi  c o n t e n t s

4. Justifying a Hierarchical Po­liti­cal System to ­Those Outside the System

93

3 Just Hierarchy between States: On the Need for Reciprocity

106

1. Hierarchical Ideals of Global Order in Ancient India

111

2. Hierarchical Ideals of Global Order in Ancient China

117

3. One World, Two Hierarchical Systems?

129

4 Just Hierarchy between ­Humans and Animals: Subordination without Cruelty

143

1. Are Animals Our Equals?

144

2. Domesticated Animals: Subordination with Care

154

3. Eating Animals: Subordination with Humane Exploitation

166

5 Just Hierarchy between ­Humans and Machines: On the Need for a Master-­Slave Relation

177

1. Marx on Machines

179

2. The Role of Confucian Role Ethics

187

3. Silicon Valley vs. the Chinese Communist Party

198

Notes  ​207 Selected Bibliography  ​249 Index  ​263

ac k now l­e d g m e n t s

first but not least, we would like to thank each other. The ideas in this book evolved via prolonged conversations and arguments over the past few years, to the point that we forgot who said what. Daniel wrote most of  the En­glish version of this book, and Pei wrote most of the Chinese version, but we are jointly responsible for its ideas, ­whether good or bad. We are most grateful to Rob Tempio, our insightful and supportive editor at Prince­ton University Press, along with two anonymous referees who allowed us to further refine our ideas. We are also grateful to our research assistant, Sun Qiming, and Daniel would like to thank his assistants at Shandong University—­Liu Yuhan, Huang Ping, Wang Fuxiang, and Wang Chengchao—­for help. We would also like to thank Cheng Jiaolong for writing the beautiful calligraphy on the book’s back cover. Daniel owes special thanks to f­ amily members for emotional and intellectual support. He would also like to thank Kong Lingdong, Cao Xianqiang, Kong Xinfeng, Liu Lin, and all his other colleagues and leaders at Shandong University’s School of ­Po­liti­cal Science and Public Administration for providing an intellectually stimulating setting that allowed him to write this book. Daniel is also grateful to Tsing­hua University president Qiu Yong for continued support at Tsing­hua as well as his co-­ teacher at Schwarzman College, Wang Hui, for fascinating discussions on Chinese intellectual history, Bai Tongdong at Fudan vii

viii 

a c k n o w l­e d g m e n t s

University for inspiring conversations in po­liti­cal theory, and Eric X. Li for friendship that goes beyond the ordinary meaning of friendship. He would also like to thank his students at Shandong University and Tsing­hua University for constructive and often well-­deserved criticism of the teacher’s ideas. Daniel also owes special thanks to Nicolas Berggruen. We cannot construct a better world without serious engagement with the world’s ideas—­including ideas from previously marginalized parts of the world—­and Nicolas has both the vision and the means to realize this aspiration. The idea for this book emerged when Daniel was director of the Berggruen Institute’s Center for Philosophy and Culture. Daniel helped to or­ga­nize a conference on “Hierarchy and Equality” at Stanford University’s Center for the Advanced Study in the Behavioral Sciences in March 2016. The papers w ­ ere excellent—­and we have made use of several of them in this book—­and Daniel realized that ­there’s the need for a full-­length book on the topic of “Just Hierarchy.” Daniel would like to thank the workshop’s participants as well as co-­organizers Margaret Conley and Jennifer Bourne for help and inspiration. He would also like to thank Berggruen Institute leadership in Los Angeles: Nathan Gardels, Dawn Nakagawa, Nils Gilman, and (former) president Craig Calhoun for help over the years. Daniel owes special thanks to the Berggruen Institute’s China Center at Peking University which supported several workshops in New Delhi, Qingdao, Beijing, and Bangkok, comparing Chinese and Indian thought, and Daniel is grateful to workshop participants as well as to co-­organizers Song Bing, Roger Ames, Yan Xuetong, Amitav Acharya, Rajeev Bhargava, Shelley Hu, Li Xiaojiao, and Li He. Pei would like to thank her friends and colleagues at Fudan University’s China Institute. She is particularly grateful to Zhang Weiwei, Eric X. Li, Fan Yongpeng, Chen Ping, Li Bo, Yu Liang,

a c k n o w l­e d g m e n t s 

ix

Lin Ling, Meng Weizhan, and Feng Zhun. She would also like to thank visitors to the China Institute, especially Alexander Dugin, Yukon Huang, Martin Jacques, Kishore Mahbubani, and Dominique de Villepin. She is grateful to the institute for providing time and support and an intellectually stimulating environment for research. Pei owes much to Wang Hui for constant concern and intellectual inspiration. Wang Hui supervised Pei’s postdoctoral research at Tsing­hua University and showed the importance of relating philosophy to real politics and social life. Pei would also like to thank her former colleagues at the Tsing­hua Institute for Advanced Study in the Humanities and Social Sciences (TIAS): Kong Yuan, Fu Zheng, Yuan Xianxin, Li Peiyan, and Yang Tao, as well as visitors to TIAS, Michael Dutton, Viren Murthy, and Yu Zhizhong. Pei and Daniel are also grateful to Geir Helgesen, Liu Chunrong, and Unn Irene Aasdalen for facilitating and modeling the best form of cross-­cultural dialogue at Nansen Acad­emy in Lillehammer, Norway. Pei’s friends Cheng Jiaolong, Li Shuzhi, and Wang Hairong have provided unconditional support and have inspired her with new perspectives on real­ity. Pei is deeply indebted to her f­ amily members and especially to her m ­ other, who always surprises with her direct and sharp comments, and spoils Pei with all her tenderness. ­Earlier versions of chapter 2 w ­ ere published in Philosophy and Public Issues (by Pei) and the Journal of Chinese Humanities / 文史哲 (by Daniel), and we are grateful to the publications for permission to draw on t­ hose articles, as well as to Aeon for permission to publish the online appendix. E ­ arlier versions of chapter 2 ­were presented (by Daniel and Pei) at the University of Malaya’s Institute of China Studies, the Penang Institute, the Beijing Thinkers’ Forum, and the annual Reset conference in

x 

a c k n o w l­e d g m e n t s

Venice; and (by Daniel) as a keynote speech at the 2019 IPP International Conference on Civilization and National Governance in Guangzhou, at the Po­liti­cal Meritocracy in Comparative Historical Perspective conference and the China India Meritocracy conference, both at the Harvard Center in Shanghai, and at the Nansen Acad­emy in Lillehammer, Norway. ­Earlier versions of chapter 3 ­were presented (by Daniel) at “From a Westcentric to Post-­Westcentric World” in Taipei, at the Peking University’s Berggruen China Center, and at CKGSB in Beijing. ­Earlier versions of chapter 5 ­were presented (by Daniel and Pei) at Santa Clara University and (by Daniel) at Pembroke College, Cambridge University (as the annual China Goes Global lecture), at Sun Yat-­Sen University (as a Global Justice Lecture), at Xi’an Jiaotong–­Liverpool University (as a keynote speech at the Second Annual Conference of the Jiangnan Research Group on China Studies), at Peking University’s Berggruen China Center, and at the Nansen Acad­emy in Lillehammer, Norway. Formal settings that allow us to refine our arguments are impor­tant, but informal settings that allow for the expression of half-­baked ideas even more so. We are grateful to friends who share our love of crazy conversation fueled by fine wine and good food.

j ust h i e r a rch y

Introduction

the seating arr angements for formal meals in Shandong province—­the home of Confucian culture, with a population of nearly 100 million p­ eople—­are rigidly hierarchical. The host with the highest social status sits at the “top” of a round t­ able with a view of the door, the host with the second-­highest social status sits at the other end of the ­table; the guest with the highest social status sits on the right-­hand side of the host with the highest social status and next to the host with the third-­highest social status; the guest with the second-­highest social status sits on left side of the host with the highest social status and next to the host with the fourth-­highest social status; the guest with the third-­highest social status sits on the right side of the host with the second-­highest social status and next to host with the fifth-­ highest social status; the guest with the fourth-­highest social status sits on the left side of the host with the second-­highest social status and next to the host with the sixth-­highest social status. The other seats are randomly distributed among t­ hose with the least social status, with the number of randomly assigned seats depending on the number of hosts and guests. Sounds complex? The pictorial depiction of the social hierarchy in figure 1 might be helpful.1 What’s wrong with Shandong’s seating arrangements for formal meals? Nothing at all! The only t­ hing wrong is the expectation that all social relations are supposed to be equal. As dean 1

2  I n t r o du c t i o n Principal Host

Guest of Honor

Second Guest of Honor

Host 3

Host 4

Random Seat

Random Seat

Host 5

Host 6

Guest 4

Guest 3

Assistant Host

Door

figure 1. Seating hierarchy for dinner party.

at Shandong University’s school of po­liti­cal science and public administration, Daniel has hosted countless meals with such seating arrangements, including hosting of foreign guests, and he has not once received any complaints. Perhaps his guests are too polite to complain. But we’d like to think that such seating arrangements are tolerated b­ ecause they do not express and reinforce unjust hierarchies that rank ­people according to race or gender. P ­ eople who are not ethnically Chinese—­such as Daniel—­are seated in the same position as ethnically Chinese ­people with the same social status (i.e., with a title of dean). Men

I n t r o du c t i o n  

3

and ­women occupy their seats according to their social roles regardless of gender: For example, President Fan of Shandong University is female, and she occupies the principal host seat at formal meals, in the same position as previous (male) presidents at Shandong University. This is not to deny that patriarchal ways still inform the seating arrangements in rural parts of Shandong province—­women often sit at dif­fer­ent and less comfortable ­tables—­but such norms are rejected in university settings.2 The seating arrangements at Shandong University are also tolerated ­because the hierarchies are nearly invisible to the untrained eye. The t­ ables are round, with the appearance of perfectly equal symmetry,3 and the visiting guests ­won’t know about the social hierarchies u­ nless they are informed of the norms by local hosts. In contrast, the rectangular “high ­tables” at traditional Oxford and Cambridge universities are literally higher than ­tables for students, and students are not allowed to start eating ­until the teachers formally get the proceedings under way. What­ever the case for special treatment (better food and wine) for teachers and guests in university settings, Oxbridge-­style “in your face” social hierarchies often generate a vague sense of unease even for beneficiaries of ­these arrangements. But we’d like to defend a stronger claim. It’s not just a ­matter of tolerating Shandong-­style seating arrangements b­ ecause they do not express unjust or vis­i­ble hierarchies. Th ­ ese arrangements are endorsed, and even enjoyed, b­ ecause they express several of the virtues of what we call “just hierarchies,” that is, morally justified rankings of ­people or groups with re­spect to valued social dimensions. Consider the distribution of seats for persons of higher social status in Shandong. The usual “fight” among t­ hose in the know is to refuse a seat with more social prestige. So ­there is a toss and strug­gle, and fi­nally the “loser” of the ­battle ­will give in and reluctantly take the seat that expresses a higher social

4  I n t r o du c t i o n

position. Most often, both sides know the outcome of the strug­ gle—­for example, the dean ­w ill take the seat of the principal host if he or she has the highest rank in the university hierarchy, the guest with the most academic prestige and/or the greatest number of years of physical existence (i.e., the oldest person) ­will take the seat of the guest of honor, and so on—­but it would seem immodest to immediately claim one’s “rightful” position (the foreigners who are ignorant of such rituals often take their assigned seats without putting up a strug­gle, but they are forgiven for their moral transgressions b­ ecause they are not expected to know Shandong-­style norms of civility). Put differently, the strug­gles, however hypocritical they may seem to the critical outsider, express Confucian-­style virtues of humility and modesty.4 Not only that, but the occupants of the seats with more social prestige have more responsibility. They must foot the bill: To be more precise, the assistant host must pay for the meal, but the funds come from the university (since the start of the anticorruption campaign in 2012, funds from public institutions do not cover alcohol, and often the principal host must bring the liquor at his or her own expense). The hosts are responsible for treating the guests well, and each host is supposed to take care of a par­tic­u­lar guest corresponding to her or his hierarchical role. The hierarchical seating arrangement ensures not just that the most honored guest is treated well, but that the next three most socially impor­tant guests ­w ill also get some personal care. The hosts serve the o­ thers from communal dishes in the ­middle, starting with the principal host serving the most impor­tant guest on the right and then the second-­most-­ important guest on the left, and then the assistant host d­ oes the same with the third-­most-­important guest on the right followed by the fourth-­most-­important guest on the left (foreigners might start serving themselves first, but again they are forgiven

I n t r o du c t i o n  

5

for their moral transgressions on the grounds that they may not be familiar with Shandong-­style norms of civility). Then the hosts must give repeated toasts to welcome the guests, anywhere from eight times in Qufu (ground zero for Confucian culture) to three toasts in other parts of Shandong province. The assistant host must then deliver some toasts (usually fewer in number than the toasts by the principal host), then the third-­ most-­important host delivers some toasts (fewer in number than the toasts delivered by the assistant host), then the fourth-­ most-­important host delivers some toasts (fewer in number than the toasts delivered by the third-­most-­important host), and so on. Th ­ ese toasts often express warm feelings of greeting and affection for the visitors, but ideally they are also accompanied by literate references to Chinese history and culture, leavened with some humor. When the “official” toasting is over, the occupants of the most prestigious seating positions must go around the ­table and individually toast and greet each visitor. From the perspective of shy or socially reticent ­people, ­these arrangements are more beneficial to the occupants in the less prestigious seats, who can enjoy the proceedings without any responsibilities. In any case, the formalities usually break down ­toward the end of the eve­ning, with semi-­inebriated participants roaming around the t­ able almost at random, ­either joking or exchanging serious information that could not be shared with the ­whole group. Last but not least, the social hierarchies can shift on dif­fer­ent occasions. If the same p­ eople (or a similar group) meet on dif­fer­ent occasions, the roles may shift, with the guests playing the role of hosts, and vice versa, regardless of who has the most social status in society at large. And what counts as social status is not itself rigid: Sometimes it’s age, sometimes it’s government rank, sometimes it’s academic achievement, sometimes it’s perceived level of virtue, and so on.

6  I n t r o du c t i o n

In this sense, the social hierarchies are not fixed and can shift depending on the context. If the visiting professor is the guest, then he or she may occupy the guest of honor seating position, even if he or she does not have the highest social status outside of the university context. So yes, Shandong-­style seating and drinking rituals are hierarchical, but what’s wrong if they provide hospitality for the guests and generate a sense of harmony among participants? Perhaps such hierarchies are morally justified if they shift over time and if t­ hose with more social power end up caring about the needs of ­those with less power and eventually do more to serve their interests? Not to mention that ­these hierarchical rituals are often aesthetically pleasing (the food is usually varied and delicious) and thoroughly enjoyable for the participants . . . As an ideal, we defend hierarchical Shandong-­style seating arrangements, but in practice they often have a downside. Most worrisome, t­ here is often a fine line between semi-­inebriation and total (if not fatal) inebriation.5 Surely it’s no coincidence that Shandong province has the highest per capita consumption of alcohol in mainland China (Shanghai, perhaps the most Westernized and socially egalitarian part of China, is among the lowest).6 But ­there are social mechanisms that have the effect of moderating alcohol consumption: It is the highest form of rudeness to serve oneself alcohol and to drink alone in a group setting (it is similar in Confucian-­influenced ­Korea; in the West, the first toast is often communal, but then p­ eople often serve themselves and drink without toasting ­others). And ­there is usually accommodation for ­those who do not drink alcohol: Their glasses are filled with ­water, which looks like fiery “white” alcohol (白酒), and they can join participants in group toasts without drinking alcohol. That said, we need to recognize that the social pressure to drink alcohol may not always be welcome.

I n t r o du c t i o n  

7

Lower-­ranking ­people often find it hard to refuse the toasts of their superiors, even if they have exceeded their “normal” levels of inebriation. Even higher-­ranking ­people might feel unwelcome pressure to drink, as a way of showing authority over lower-­ranking ­people or to impress guests. And guests themselves often feel unwelcome pressure to drink.7 Chinese w ­ omen who typically drink far less than men may feel somewhat alienated from the drinking rituals.8 And sometimes the p­ eople from China’s more socially egalitarian southern provinces find that Shandong-­style rituals, even when they work well according to modernized social norms, are not as desirable as they might be. It may be true that universities in Shandong have shed the most egregious patriarchal norms, but t­ here is no serious effort to honor ­women or recognize their special contribution to society. In Zhejiang, by contrast, the wives and c­ hildren of invited guests are sometimes asked to sit in the principal guest of honor position: As a child, Pei recalls being honored as the principal guest in banquets with her parents and ­family friends, a practice that would be nearly inconceivable in Shandong. That said, it ­doesn’t follow that Shandong ­people should shed hierarchical seating arrangements and hierarchical drinking and eating rituals. The task is to modernize the hierarchical rituals according to progressive social values while maintaining the advantages that make them so enjoyable, if not morally uplifting, for the participants. The example of Shandong-­style seating arrangements is meant to shed light on our theoretical concerns. Let’s now turn directly to t­ hose concerns. Equality is clearly an impor­tant value—­ recognized and endorsed by social and po­liti­cal progressives in the modern world—­and much has been written on the ideal and practice of equality as well as the need to equalize relations between ethnic groups, genders, and classes. We generally share ­these egalitarian outlooks and concerns. But hierarchy, arguably,

8  I n t r o du c t i o n

is equally impor­tant, and research on hierarchy has lagged ­behind. All complex and large-­scale socie­ties need to be or­ga­ nized along certain hierarchies, but the concept of hierarchy has become almost taboo in po­liti­cally progressive circles. This is a huge ­mistake. It is impor­tant to think about which forms of hierarchy are justified and how they can be made compatible with egalitarian goals. We need to distinguish between just and unjust forms of hierarchy and think of ways to promote the good forms and minimize the influence of bad forms. But what exactly do we mean by “hierarchy” and why does it m ­ atter t­ oday? What do we mean by “bad hierarchies” that worry p­ eople with po­liti­cally progressive sensibilities? Most challenging from a theoretical perspective—­and the main question we try to answer in this book—­is, which forms of hierarchy are morally justified ­today and how can they be promoted in the ­future?

1. What’s Wrong with Hierarchy? In a purely descriptive sense, a hierarchy is a relation that is characterized by (a) difference and (b) ranking according to some attribute. Social hierarchies tend to have a normative dimension: They are social systems in which ­there is “an implicit or explicit rank of individuals or groups with re­spect to a valued social dimension.”9 But we need further normative justification to argue that socie­ties should value ­those dimensions. In En­glish, the word “hierarchy” has come to have pejorative connotations b­ ecause we now think that most traditional ways of ranking p­ eople or groups are not justified from a moral point of view. Biologists tend to speak of hierarchy in the neutral sense, and they study its origin and evolution without passing any moral judgments.10 Hierarchy is a ubiquitous organ­izing princi­ple in biology and a key reason evolution produces complex, evolvable

I n t r o du c t i o n  

9

organisms. Why did hierarchy evolve? At the level of biological neural networks, the key ­factor is the cost of connections: According to an influential study by computer simulation, networks without a connection cost do not evolve to be hierarchical, whereas t­ hose with a connection cost evolve to be hierarchical, and such networks exhibit higher overall per­for­mance and adapt faster to new environments.11 Put simply, with a degree of centralization in connection-­making, complex biological systems need fewer connections and ­things can run more efficiently. A similar mechanism seems to explain the evolution of hierarchy in larger-­scale social organ­izations. As Peter Turchin explains, “The only way that large ­human groups can arrive at a common course of action is by [hierarchically] structuring interpersonal connections. . . . ​Socie­ties that ­were larger and better or­ga­nized outcompeted smaller and more shambolic ones. Hierarchical organ­ization was one of the cultural traits that was heavi­ly favored by the new se­lection regime in the Holocene [which started roughly 12,000 years ago with the end of the ice age]. . . . ​ It’s a pipe dream to imagine that a large-­scale society (e.g., a million or more—­a small nation by ­today’s standards!) can be or­ ga­nized in a nonhierarchical, horizontal way. Hierarchy (in a neutral sense) is the only way to or­ga­nize large-­scale socie­ties.”12 Just as it’s impossible to efficiently connect large numbers of neural networks without hierarchy, so it’s impossible to connect large numbers of p­ eople in an efficient way without a hierarchically structured social organ­ization. In short, efficiency is a clear benefit of hierarchy. The efficiency of hierarchy may help to explain why we like hierarchies at some unconscious level. According to one study, an abstract diagram representing hierarchy was memorized more quickly than a diagram representing equality, and the faster pro­ cessing led the participants to prefer the hierarchy diagram.

10  I n t r o du c t i o n

And participants found it easier to make decisions about a com­ pany that was hierarchical and thus thought the hierarchical organ­ization had more positive qualities.13 What­ever the negative feelings about hierarchy at the conscious level, it seems that the efficiency benefits of hierarchy in our evolutionary history often prompt us to like hierarchy.14 But efficiency per se is not morally justified. It depends on the ends being pursued. The Nazis built superefficient concentration camps, but they ­were put to use for despicable purposes. Or consider the workings of natu­ ral se­lection. To an impor­tant extent, we are what we are ­because of natu­ral se­lection. The mission of natu­ral se­lection is to get genes into the next generation in an efficient way, and we tend to like what’s helpful for this purpose and dislike what’s not. As Robert Wright puts it, “We w ­ ere ‘designed’ by natu­ral se­ lection to do certain ­things that helped our ancestors get their genes into the next generation—­things like eating, having sex, earning the esteem of other ­people, and outdoing rivals.”15 But we can decide that ­doing some of the ­things that made us effective gene propagators in the past are no longer desirable ­today. Evolution may have prompted us to value our own interests above ­those of ­others, but the costs of excessive self-­regard may now outweigh the benefits. For example, natu­ral se­lection designed ­human minds to size ­people up in a way that would lead to interactions that benefited the genes of the h­ umans ­doing the sizing up, not to size p­ eople up accurately. Hence we tend to exaggerate the virtues of our friends and the vices of our enemies. That may be efficient for purposes of reproduction, but it also provides the psychological roots for tribalism and demonization of the “other.” Upon reflection, we can decide that the social and po­liti­cal consequences of tribalism and warfare threaten our species, if not the ­whole world. If we agree that it’s better to let go of t­ hings like lust and conceit and ill-­will that w ­ ere “programmed”

I n t r o du c t i o n  

11

into us to perpetuate our genes in an effective way, then we can promote practices such as meditation that promote compassion for all sentient beings and help to erode the psychological roots of what we now consider to be immoral be­hav­ior.16 It may well turn out that what’s efficient from the point of view of natu­ral se­lection is morally wrong, and we can and should strive to challenge much of what seems “natu­ral.” In the same vein, t­ here are good reasons to challenge many of the social hierarchies that seem natu­ral to us. Th ­ ese hierarchies may have arisen for reasons of efficiency, but we need not endorse them from a moral point of view. This is not pure theory: Upon reflection, it seems obvious that many of the hierarchies from the past are morally problematic ­today. As historian Yuval Noah Harari puts it, “complex ­human socie­ties seem to require i­ magined hierarchies and unjust discrimination. . . . ​Time and again p­ eople have created order in their socie­ties by classifying the population into i­ magined categories, such as superiors, commoners, and slaves; whites and blacks; patricians and plebians; Brahmins and Shudras; or rich and poor. Th ­ ese categories have regulated relations between millions of ­humans by making some ­people legally, po­liti­cally or socially superior to ­others.”17 But we have made moral pro­gress: ­Today, most educated ­people recognize and condemn the seemingly “natu­ral” hierarchies of our past history.18 Most Americans, for example, now endorse statements about equality and reject statements about the value of hierarchy19 and complain that hierarchies are inhumane, immoral, and undemo­cratic.20 Why do we now reject most traditional hierarchies? A key reason, arguably, is our unhappy experience with morally bad hierarchies in the form of racism, sexism, and caste-­like distinctions between ­people. Few if any progressive and educated ­people living in modern socie­ties defend hierarchies among classes of ­humans who are inherently superior or inferior based on noble

12  I n t r o du c t i o n

birth, race, sex, or religion, although such hierarchies ­were commonly endorsed in the past.21 In ancient Rome, the penalty for assault on a slave was half the penalty for assault on a ­free man,22 but ­today slavery is (fortunately) regarded as morally obscene. Ancient Chinese thinkers argued that scholar officials should be exempt from criminal punishment,23 but no con­temporary Confucian seeks to revive such forms of in­equality before the law. At some level, then, we are all egalitarians who endorse the princi­ple of equality of basic moral and ­legal status for citizens. And with the pos­si­ble exception of crazed terrorists, we all endorse the view that h­ uman beings, regardless of background, are equally entitled to what Michael Walzer terms “thin” ­human rights: rights not to be tortured, enslaved, murdered, and subject to systematic racial discrimination.24 But we—­the co-­ authors of this book—do more than endorse equality before the law in criminal cases and basic ­human rights. Our book is informed by what we might call a “progressive conservative” perspective. On the one hand, we are sympathetic to the traditional egalitarian c­ auses of the po­liti­cal left, including an aversion to extremes of wealth distribution, more rights for the productive classes, more support for poor countries that unduly suffer the effects of global warming, equality between men and ­women, as well as equal rights for same-­sex ­couples. In our view, many of the social hierarchies traditionally viewed as natu­ral and just are neither natu­ral nor just, and we can and should challenge t­ hose hierarchies: by revolutionary means, if necessary. On the other hand, we share a conservative attachment to, if not reverence for, tradition, and we recognize that some traditional hierarchies—­ among f­ amily members, citizens, states, h­ umans and animals, and ­humans and machines—­are morally defensible. We do not argue for blindly reaffirming and implementing hierarchies that

I n t r o du c t i o n  

13

may have worked in the past. But suitably reformed—so we ­will argue—­they can be appropriate for the modern world.

2. In Defense of Hierarchy What­ever the drawbacks of traditional forms of hierarchy, the effort to combat all forms of hierarchy is neither pos­si­ble nor desirable. Complex organ­izations and socie­ties need some form of hierarchy and w ­ ill outcompete and outlast t­ hose that seek to abolish all forms of hierarchy. History bears out this prediction: Efforts to consciously build large-­scale organ­izations or socie­ ties without hierarchies have failed miserably. Edmund Burke famously criticized the French revolutionaries for seeking to equalize relations of command and obedience in the military and predicted such efforts would lead to the rise of “some popu­lar general, who understands the art of conciliating the soldiery, and who possesses the true spirit of command, [and who would] draw the eyes of men upon himself [and become] the master of the ­whole republic.”25 In China’s Cultural Revolution, the effort to stamp out social hierarchies similarly led to mass vio­lence and populist tyranny. In con­temporary China, the populist legacies of the Cultural Revolution still poison the po­liti­cal atmosphere, aided by the internet that allows anonymous masses to hound social undesirables into submission. In the United States, the populist backlashes against elites empower strongmen such as Donald Trump with scant regard and re­spect for traditional constraints on po­liti­cal power. So the effort to combat all forms of hierarchy ­will not only fail; it may lead to something even worse from a moral point of view. In short, the choice t­ oday is not between a society with no hierarchies and one with hierarchies, but rather between a

14  I n t r o du c t i o n

society with unjust hierarchies that perpetuate unjust power structures and one with just hierarchies that serve morally desirable purposes. Perhaps the idea of just or morally justified hierarchies seems difficult to digest at the conscious level, especially from a modern perspective. We have suggested that Shandong-­style hierarchical seating arrangements can be morally justified for formal dinner occasions, but other examples readily come to mind. We generally take hierarchies of esteem for granted: Nobody doubts that LeBron James deserves his trophy as the Most Valuable Player in the 2016 NBA playoffs by virtue of his achievements on the basketball court. And what­ ever the disputes about the moral worthiness of par­tic­u­lar Nobel Peace Prize winners, few object to the princi­ple that we can and should reward ­those with ­great moral achievements of some sort. In China, the government honors adults who are filial to their el­derly parents; we can argue about the choices, but it seems hard to object to the princi­ple of honoring ­those who can set a good model for ­others.26 What’s more controversial is the claim that morally justifiable social hierarchies should structure our social lives on an everyday basis, including our relations with loved ones. That’s the claim we’d like to defend in this book. Our target is the view that all social relations should be equal. The flip side of this view is that unequal relations are fundamentally unjust: As Jean-­Jacques Rousseau lamented in his Confessions, “I felt, more than ever, from repeated experiences, that associations on unequal terms are always to the disadvantage of the weaker party.”27 So t­ hose who care about the interests of the weak—­that is, all sensitive, progressive-­minded people—­ should affirm the ideal of equal social relations at all times in all walks of (social) life. In the con­temporary world, this ideal is often expressed in everyday (En­glish language) speech: Think

I n t r o du c t i o n  

15

of nine-­year-­olds who want to be treated as equals.28 More surprising, perhaps, the blanket defense of social equality is increasingly defended by sophisticated po­liti­cal theorists. In the first few de­cades a­ fter the publication of John Rawls’s groundbreaking book A Theory of Justice (1971), Western po­liti­cal theorists ­were mainly concerned about the nature of t­ hings to be distributed equally (is it income, resources, welfare, capabilities, or something ­else?) and debates about the most defensible egalitarian distributive princi­ple (should it be pure equality, the difference princi­ple, sufficiency, or something ­else?).29 More recently, some theorists—­let’s call them “social egalitarians”—­ argued that this focus on distributive princi­ples is too narrow and neglects the broader agendas of ­actual egalitarian po­liti­cal movements. As Elizabeth Anderson put it, “What has happened to the concerns of the po­liti­cally oppressed? What about inequalities of race, gender, class, and caste?”30 Nor did po­liti­cal theorists obsessed with the just distribution of privately appropriated goods, such as income, or privately enjoyed goods, such as welfare, pay attention to the concerns of gay and lesbian ­people who seek the right to get married and the disabled who seek access to reconfigured public spaces and campaign against demeaning ste­reo­types. To remedy the prob­lem, social egalitarians argue that equality should refer first and foremost to an egalitarian ideal of social relations: Vari­ous goods should be distributed in order to secure a society in which p­ eople are related as equals. The focus on social in­equality allows po­liti­cal theorists to critique the unjust social hierarchies that have plagued and continue to plague ­human socie­ties, “including slavery, serfdom, debt peonage, feudalism, monarchy, oligarchy, caste and class in­ equality, racism, patriarchy, colonialism, and stigmatization based on sexuality, disability, and bodily appearance.”31 So far, so good. As po­liti­cal progressives, we welcome this focus on

16  I n t r o du c t i o n

social relations and applaud the critique of the unjust social hierarchies that have oppressed and stigmatized the large majority of ­people in history.32 But it ­doesn’t follow that equal social relations are necessarily just and that hierarchical social relations are necessarily unjust. As Joseph Chan explains, one could argue that traditional hierarchies “are problematic not ­because they undermine equality, but b­ ecause they deprive ­people in the lower ranks of such hierarchies of the opportunities to pursue wellbeing and develop virtue, and they do so on ascriptive grounds that are morally irrelevant and hence unfair. Rejection of ­these hierarchies may not necessarily lead to endorsement of equal social relationships or rejection of other hierarchies. One could imagine hierarchies that are relatively ­free from the ills of t­ hese historical examples and capable of promoting the wellbeing and virtue of the lower ranked.”33 We’d like to add that not all historical hierarchies are necessarily unjust. We should be open to the possibility that some traditional forms of hierarchy w ­ ere morally justified and they can serve as inspiration for thinking about just hierarchy in the modern world. But which hierarchical relations are justified and why? In our view, it depends on the nature of the social relations and the social context. As a method, we are inspired by Michael Walzer’s call for a pluralistic approach to justice.34 ­There is no one princi­ ple of justice appropriate for all times and places. Our main argument is that dif­fer­ent hierarchical princi­ples o­ ught to govern dif­fer­ent kinds of social relations: What justifies hierarchy among intimates is dif­fer­ent from what justifies hierarchy among citizens; what justifies hierarchy among citizens is dif­fer­ent from what justifies hierarchy among countries; what justifies hierarchy among countries is dif­fer­ent from what justifies hierarchies between ­humans and animals; and what justifies hierarchies

I n t r o du c t i o n  

17

between ­humans and animals is dif­fer­ent from what justifies hierarchies between ­humans and (intelligent) machines. The sum total of our argument is that morally justified hierarchies can and should govern dif­fer­ent spheres of our social lives, though ­these hierarchies ­will be very dif­fer­ent from the unjust hierarchies that have governed much of our lives in the past. We support our arguments with a broad range of philosophical arguments and historical examples from dif­fer­ent cultural traditions, as well as with extensive social scientific evidence and anecdotes from our personal experience. But we freely concede that our arguments are ultimately supported by the considered moral and po­liti­cal intuitions of readers sympathetic to our progressive conservative outlook. We have neither the desire nor the ability to persuade terrorists, white supremacists, antifeminists, misanthropes, narrow nationalists, warmongers, China-­bashers, religious fundamentalists, climate-­change deniers, die-­hard conservatives, homophobes, and ­human carnivores with no moral qualms. Nor can we persuade leftists who dogmatically assert the value of equality in all realms of social life. Our hope is that progressive conservative thinkers w ­ ill come to see the merits of just hierarchical relations in dif­fer­ent kinds of social relations, not just ­because they are philosophically defensible, but also ­because they can help us think about solutions to the leading po­liti­cal challenges of our day. We develop our argument in five separate chapters that correspond to five dif­fer­ent forms of social relations and five dif­ fer­ent corresponding princi­ples of hierarchy. ­These five hierarchical social relations are not meant to be exclusive, but they can and should govern much of our social lives. Chapter 1 focuses on relations between intimates that are characterized by emotions of love and care based on prolonged experience with face-­ to-­face interaction. Much po­liti­cal theorizing, both in the West

18  I n t r o du c t i o n

and (less so) in China, idealizes friendship between equals as the most desirable form of social relation. We do not dispute the desirability of friendship between equals, but we argue that an even higher form of social relation would include shifting hierarchies between intimates. Of course, hierarchies should not include vio­ lence, nor should they be fixed for eternity. But shifting hierarchies between lovers and f­ amily members are not just tolerable; they add much to the color and humor of social interaction. Even hierarchical relations between employers and h­ ouse­keepers can be morally justified if they allow for role changes over time, though it might take a generation for such reversals to occur. Con­temporary po­liti­cal theory does not provide the intellectual resources to develop our arguments on morally justifiable hierarchies between intimates, so we seek intellectual inspiration from ancient Chinese, Indian, and Greek thinkers. In chapter 2, we turn to a discussion of just hierarchies between citizens—­mainly strangers to one another—in modern large-­scale po­liti­cal communities. It is a special challenge to justify hierarchies in po­liti­cal systems without voting mechanisms that (equally) empower citizens to change their rulers ­every few years. We argue that hierarchies between rulers and ruled in such communities are justified if the po­liti­cal system selects and promotes public officials with above-­average ability and a willingness to serve the po­liti­cal community over and above their own private and ­family interests. We have the Chinese po­liti­cal context in mind, and we argue that this kind of ideal—­what we call “po­liti­cal meritocracy”—­helped to inspire the imperial po­ liti­cal system in China’s past and Chinese po­liti­cal reformers in the early twentieth c­ entury, and may help to justify the po­liti­cal system in China t­ oday. However, the meritocratic system needs to be accompanied by demo­cratic mechanisms short of competitive elections at the top that allow citizens to show that they

I n t r o du c t i o n  

19

trust their rulers and provide a mea­sure of accountability at dif­ fer­ent levels of government. In the Chinese context, however, ­there is a large gap between the ideal and the real­ity, and we argue that a judicious mixture of Confucian-­style “soft power” combined with demo­cratic openness, Maoist-­style mass line, and Daoist-­style skepticism about the ­whole po­liti­cal system can help to reinvigorate po­liti­cal meritocracy in China. Chapter 3 discusses relations between states. Whereas relations between rulers and citizens in countries should be characterized first and foremost by actions that benefit the citizens, relations between countries need to be mutually beneficial for both countries. Notwithstanding lip ser­vice paid to the ideal of equality between sovereign states in the modern world, we argue that hierarchy between power­ful and weaker states is the norm in international relations. Such hierarchical relations can be justified if they benefit both power­ful and weaker states. We draw on a mixture of philosophy and history to argue that justifiable hierarchical relations can be characterized by ­either weak reciprocity—­with both countries deriving instrumental benefits from hierarchical relations—or strong reciprocity—­w ith decision makers in stronger and weaker states thinking of their relations from the perspective of both states, not just from the perspective of their own state. Strong reciprocity is more difficult to achieve, but it is more stable and long lasting than weak reciprocity. In terms of the ­future, we argue that an ideal of “one world, two hierarchical systems” may be appropriate for ­future forms of global order. ­Here too, modern theorizing is not sufficient, and we draw on the insights of ancient Indian and Chinese thinkers to make our points. In chapter 4, we consider our relations with the animal kingdom. Throughout much of h­ uman history, most cultural and religious traditions—­w ith some notable exceptions, such as

20  I n t r o du c t i o n

Daoism—­have valued ­humans over animals. We argue that it is morally justifiable to posit a moral hierarchy with h­ umans on top, but only if accompanied by the princi­ple that h­ umans should not be cruel to animals. But the princi­ple of “subordination without cruelty” is not sufficient to spell out the kinds of obligations we owe to animals. We have dif­fer­ent kinds of relations with dif­fer­ ent animals, and we owe the strongest obligations of care to animals with human-­like traits and that contribute most to our well-­being. In the case of animals bred for ­human consumption, we argue that such subordination is only justified if the animals are bred in humane conditions that are exceptionally rare in the modern world. We owe least to ugly animals that harm ­humans, but the princi­ple of subordination without cruelty applies even in the case of the nastiest animals. In chapter 5, we turn to perhaps the greatest challenge of our times: the need to maintain dominance over increasingly intelligent machines. We argue that machines can and should serve ­human interests—in that sense, they should be our slaves—­and it is impor­tant to maintain such hierarchical relations of dominance. ­Here Marxism provides intellectual inspiration: The ideal of higher communism, with artificially intelligent machines ­doing socially necessary ­labor and h­ umans freed to realize their creative essences, may be feasible several de­cades from now. But the state cannot and should not “wither away”: A strong state ­will always be necessary to ensure that artificial intelligence does not invert the human-­machine relation with ­humans on top and machines on the bottom. But worrisome science-­fiction scenarios, with machines that seek to make h­ umans into slaves, are challenges for the long term. In the short to medium term, we argue that Confucianism can help us to think of how to meet the challenge of artificial intelligence so that machines continue to serve ­human purposes.

I n t r o du c t i o n  

21

The online appendix to our book is a joint statement—­a kind of manifesto—­signed by dif­fer­ent po­liti­cal thinkers (including Daniel) in defense of the ideal of just hierarchy (https://press​ .princeton​.edu​/titles​/30674​.html). It is the product of a Berggruen Institute workshop on equality and hierarchy at Stanford University and was penned primarily by Julian Baggini. The manifesto helped to inspire this book (the detailed arguments ­were inspired mainly by conversations between Pei and Daniel over the past few years), and it also shows that t­ here is potentially wide support for the ideal of just hierarchy in the modern world among ­people willing to question the received prejudice that ­social hierarchy is always a bad ­thing.

3. From China to the World We expect that our defense of “just hierarchy” w ­ ill resonate with the considered po­liti­cal intuitions of readers who share our progressive conservative perspective, with the implication that traditional hierarchies, properly reformed and updated for modern socie­ties, can serve progressive po­liti­cal goals. But we recognize that the progressive conservative perspective may sound paradoxical to Western readers.35 How can one be committed to both traditional values rooted in the past and to progressive values that point to a dif­fer­ent (and better) way of ­doing ­things in the ­future? The mainstream narrative of modernity in Western socie­ties is that traditional hierarchies expressed and institutionalized unjust values such as racism, sexism, and aristocratic privilege. Modern enlightened thinkers criticized traditional hierarchies and put forward strong arguments in ­favor of social equality and individual freedom that set the moral standard for ­future pro­gress. ­There remains a large gap between the ideal and the real­ity, but hardly anybody openly argues for a return to the

22  I n t r o du c t i o n

bad old days of rule by white men from aristocratic families. The default moral position, in the eyes of most Westerners, is a commitment to social equality and deep skepticism of the value of traditional hierarchies. In China, it’s a dif­fer­ent (hi)story. Early Confucian thinkers criticized rulers on the grounds that they oppressed and impoverished ordinary p­ eople. In this sense they w ­ ere po­liti­cal progressives. But rather than invoking new or future-­oriented values as a moral standard for criticizing present-­day injustices, they invoked standards from a golden age in the past that expressed morally desirable hierarchies in a harmonious society. The self-­ declared First Emperor of China, inspired mainly by Legalist ideas, implemented harsh policies that destroyed aristocratic privilege and built up a complex bureaucracy that expressed a commitment to social mobility based on merit. Subsequent imperial history was largely informed by Confucian commitments to both traditional social hierarchies and proto-­socialist po­liti­ cal ideals such as poverty reduction, equality of opportunity, and infrastructure proj­ects designed to benefit the large majority of ­people. The imperial system broke down in 1911, and Western-­ influenced intellectuals blamed Confucian-­style hierarchies for China’s backwardness.36 The tradition of antitraditionalism culminated in the Cultural Revolution, a disastrous attempt to abolish all forms of hierarchy from social life. ­Today, it is widely recognized by both government officials and leading intellectuals that China’s way forward needs to draw on both conservative and progressive values: The default moral position often ­favors social hierarchy, and the question is how to make t­ hose hierarchies serve socially and po­liti­cally progressive goals. In terms of our book, it means that our arguments in ­favor of morally justified social hierarchy might find a more ready audience in China and

I n t r o du c t i o n  

23

other East Asian socie­ties influenced by Chinese culture such as ­Korea, Vietnam, and Japan.37 ­Needless to say, this somewhat crude sketch contrasting dominant po­liti­cal values of East Asia and the West overlooks impor­ tant countercurrents.38 But the default moral positions for or against social hierarchy continue to have ­great influence t­ oday. In Sweden, c­ hildren often address all adults by their first (given) names,39 the kind of lesson in social equality that would be inconceivable in China, not to mention socie­ties such as Japan and South ­Korea that institutionalize social in­equality by means of practices such as bowing at differential ­angles depending on a person’s age and social status. In China, the supposedly egalitarian ideals of communism became transmuted into hierarchical social forms without much controversy: Even three-­member party cells of the Chinese Communist Party are expected to appoint a leader in the form of a party secretary.40 It would not be a gross simplification to assert that the norm of social equality has become the default moral position in almost all Western socie­t ies, which may not be the case in China and other Confucian-­influenced East Asian socie­ties. That’s not to say Western socie­ties have eliminated the need for hierarchy, but it takes a dif­fer­ent form. In the United States, ­people feel valued by being treated as social equals, but the expression of superior status (and power) takes the form of wealth. It is fine to address Bill Gates by his first name, but it is also fine for the rich to separate themselves from the poor by means of living in gated communities. Libertarian arguments in defense of stark material in­equality may be widely shared in the United States, but such views have almost no resonance in East Asian socie­ties governed by hierarchical rituals that express differences in social status. Perhaps power­ful members of East Asian socie­ties need not rely

24  I n t r o du c t i o n

on material wealth to show their superiority to the same extent.41 It seems that the power­ful members of almost all complex socie­ ties need to express some form of hierarchy, and the choice comes down to Western-­style economic hierarchy with a commitment to social equality versus East Asian–­style social in­ equality with a commitment to economic equality.42 Such cultural differences are expressed in dif­fer­ent languages: Although the most common word for hierarchy (dengji 等级) in Chinese is nearly as pejorative as the word “hierarchy” in En­glish, it is easier to talk about morally justified social hierarchies in Chinese ­because the language has words such as chaxu (差序) that more readily lend themselves to the idea that not all social hierarchies are bad.43 ­These differences are learned and reinforced in dif­fer­ ent childhood educational practices44 and express dif­fer­ent cognitive orientations.45 Perhaps the cultural differences are most evident in the po­liti­cal sphere, and we do not expect that our arguments in f­ avor of po­liti­cal meritocracy (chapter 2) or a China-­led po­liti­cal hierarchy of states in East Asia (chapter 3) or for a strong Communist Party with the power to combat malevolent artificial intelligence (chapter 5) ­will have much persuasive power outside of China. Cultural differences also ­matter when it comes to prioritizing dif­fer­ent princi­ples of hierarchy that inform dif­fer­ent social spheres. Even if we agree that we can usefully posit the existence of dif­fer­ent spheres informed by dif­fer­ent princi­ples of social hierarchy, we cannot assume that all ­these princi­ples can be si­ mul­ta­neously implemented in some sort of harmonious way. That is, the successful implementation of a princi­ple of hierarchy in one social sphere might conflict with, or undermine, the successful implementation of a princi­ple of hierarchy in another sphere. It is entirely pos­si­ble, for example, that a commitment to serving citizens by meritocratically selected rulers in a strong

I n t r o du c t i o n  

25

state (chapter 2) may conflict with the need to promote ties of strong reciprocity with weaker states (see chapter 3) since citizens of the stronger state may not be willing to share benefits with citizens of a weaker state. Even more worrisome, the commitment to ward off the potential challenge of “machine-­ masters” (chapter 5) may undermine the need for more demo­ cratic checks on the power of the state (chapter 2). In this case, which princi­ple should have priority? Chinese thinkers steeped in a tradition of concern for tianxia (“All-­under-­heaven”) may argue that the first princi­ple should have ultimate priority ­because our very existence is at stake. But Americans are far less likely to accept the potential cost of a totalitarian state that leaves hardly any space for personal privacy or intimacy. If the license plate slogan in New Hampshire—­live ­free or die—­expresses a widely held view in that part of the world, then we do need to take seriously the question of how to prioritize the dif­fer­ent hierarchical princi­ples in cases of conflict, with potentially dif­fer­ ent rankings in dif­fer­ent social contexts. In short, we usually have the Chinese po­liti­cal context in mind. Some of our ideas may seem strange, if not morally outrageous, to ­people in socie­ties far removed from the influence of Chinese culture. Our ideas originate from China: We support our arguments mainly (but not exclusively) with references to China’s history and philosophical traditions such as Confucianism, Buddhism, Daoism, and stories from our personal experience living and working in China. And we write for China: We try to provide a coherent and rationally defensible account of the leading social and po­liti­cal ideas of China’s public culture that can be used to critically evaluate the po­liti­cal real­ity in China. We do not mean to imply that our ideas only have validity in China. But what we say in ­favor of hierarchy needs to overcome a higher cultural hurdle in Western socie­ties that strongly ­favor

26  I n t r o du c t i o n

social equality in all spheres of social life. We hope that Western readers with a strong commitment to social equality w ­ ill learn from this book if they seek to better understand China, but we do not expect that Western readers w ­ ill be persuaded by many (or any) of our China-­centered arguments. That said, we do not entirely forsake the aspiration to universality. Default positions in f­ avor of social equality are difficult to change, but they are not fixed for eternity. The field of business studies provides some evidence that biases in ­favor of social equality can be changed if need be: Man­ag­ers from Western socie­ties that value social equality can perform well if they adapt to the preference for social hierarchy in East Asian workplaces.46 Nor is it hard to imagine po­liti­cal scenarios that allow for the implementation of morally justified hierarchies. In demo­cratic countries, citizens are likely to become disillusioned with populist leaders who fail to deliver on extravagant promises (“Mexico ­w ill pay for the wall”), and t­ here ­w ill be po­liti­cal pressure for meritocratic checks on populist excesses. So which parts of our book may seem more plausible, if not po­liti­cally influential, to readers outside of China? Readers who share our progressive conservative perspective—an attachment to tradition and to progressive po­liti­cal ­causes—­may more readily accept our five­fold division between the forms of morally justified hierarchies that inform dif­fer­ent forms of social relations. The idea that dif­fer­ent hierarchical princi­ples should inform dif­fer­ent social spheres—­ what works in the ­family may not work at the level of the state; what works between citizens may not work between states; what works between states may not work between ­human and animals; and what works between ­human and animals may not work between ­humans and machines—­may resonate with the considered intuitions of ­people in all modern complex socie­ties that allow for dif­fer­ent forms of social hierarchy.

I n t r o du c t i o n  

27

One impor­tant caveat: We do not mean to claim that ­there are completely separate princi­ples justifying dif­fer­ent kinds of social relations and that they all fit together in some seamless ­whole. For one t­ hing, ­there may be social relations not discussed in depth in our book, such as the relation between employer and employee, teacher and student, commander and soldier, or religious leader and follower, that are informed by dif­fer­ent princi­ ples of social hierarchy or that overlap with the princi­ples discussed in this book. Social relations are not Platonic-­like social spheres endowed with mystical autonomy: The social real­ity is far more complex in p­ eople’s minds. Even if we agree, for example, with the argument (in chapter 1) that hierarchical relations between intimates are justified if they involve shifting roles, we might also agree that the princi­ple invoked (in chapter 2) to justify hierarchies between rulers and citizens—­those with power must care for t­ hose with less power—­also applies to the relation between parents and ­children. Or ­else we might agree that the princi­ple invoked (in chapter 3) to justify hierarchies between states—­the relations should be mutually beneficial for both the power­ful and the weaker parties—­could also be invoked to justify our relations with pets (see chapter 4). The bound­aries between social spheres and under­lying hierarchical princi­ples, in other words, are fluid. At best, we might be prepared to defend the claim that we identify dif­fer­ent princi­ples that primarily justify five dif­fer­ent kinds of social relations in dif­fer­ent social spheres in modern complex socie­ties, but we do not mean to imply that ­those princi­ples are exclusive or that the bound­ aries between social spheres are air-­tight. One final methodological point. We do not draw exclusively from Chinese history or philosophy to make our arguments. Our approach is closer to what Stephen ­Angle terms “rooted global philosophy: that is, taking one’s own philosophical tradition as

28  I n t r o du c t i o n

a point of departure, but being open to stimulus from other philosophical frameworks as one strives to make pro­gress (as pro­g­ ress is mea­sured from one’s own, current vantage point).”47 So we draw on ancient Greek and Indian philosophy and con­ temporary French and Anglophone philosophy, as well as social science studies and the history of socie­ties outside of East Asia, if they help to strengthen our arguments. As a general rule, the more we draw on international intellectual resources, the more exportable our arguments. What we say about shifting roles that justify “nighttime hierarchies” (chapter 1) or the hierarchical princi­ple of “subordination with care” that justifies our relations with domesticated animals (chapter 4) draw heavi­ly on intellectual resources outside the Chinese context and may have more persuasive power at the global level. To summarize, our arguments are mainly rooted in the Chinese context and ­will have more persuasive power in that context. But the progressive conservative perspective is not absent from modern socie­ties outside of China, and some of our arguments in ­favor of morally justified hierarchies and the bound­aries between them may also persuade readers in ­those socie­ties. At the end of the day, it’s up to the reader to decide which arguments are persuasive and which ones a­ ren’t. Th ­ ere is one universal value that we w ­ holeheartedly endorse: the need to read with a critical eye. We encourage readers to always ask themselves what’s wrong with our arguments and to think how they can be improved (or rejected). Our book is preliminary—to be more positive, it is the first systematic exploration of just hierarchies in modern socie­ties—­and we look forward to critical comments that ­will allow somebody ­else to write a better book on the topic. J

1 Just Hierarchy between Intimates on t h e i m p or ta nc e of s h i f t i ng r ol e s Confucius said: “In ancient times, the enlightened rulers served their ­fathers with filial piety, and therefore, served the heavens with clarity. They served their ­mothers with filial piety, and therefore served the earth with insight. When the young follow their elders, relations between higher and lower ranking ­people are well regulated.” — ­c l a s sic of f i l i a l pi et y, ch. 16

we spend much of our time interacting with our intimates. Lovers, ­family members, and friends give meaning to life, and it’s almost unbearable to imagine a life (only) with strangers. Yet it takes only a moment’s thought to realize that we do not often interact with intimates on a basis of equality. A parent can freely criticize a five-­year-­old child who does the wrong ­thing, and while the parent should not be immune to criticism, nobody thinks the child is an equal when it comes to judgments about what kind of life to lead. In China and other countries that value 29

30  c h a p t e r 1

filial piety, it’s common for adult c­ hildren to defer to the views of an experienced and wise el­derly parent. Lovers have intimate nighttime rituals of power displays that accompany (if not contribute to) the highest forms of passion and compassion, and it’s not the business of the state to intervene in such m ­ atters. In countries such as India and China, long­time ­house­keepers are not social equals but they can be loved and treated almost like ­family members. That’s not to say hierarchies between intimates are always justified. We oppose any relation that involves involuntary physical vio­lence ­toward another person, no ­matter how loved the recipient, and we ­w ill not try to argue for such relations. We do want to argue that nonviolent hierarchies can be justified if they involve shifting roles. What makes the caste system so morally repugnant is that hierarchical social roles are fixed for eternity.1 But ­there’s nothing wrong with shifting hierarchies. More than that, changing hierarchical roles between f­ amily members and lovers add much to the color and humor of social interaction. Just as it’s hard to imagine a life (only) with strangers, so it’s hard to imagine a life with intimates that prescribe equal treatment at all times. Nothing would be more boring! The one exception is the interaction between friends. It’s the one social relation that does assume equal status, except in extreme circumstances that deviate from the default position of social equality. Even conceptions of friendship that allow for differences in personalities prescribe comparable treatment, without any differences of power or ranking. Friends are both moral and social equals. Perhaps that’s why prominent thinkers in both Western and Chinese traditions have valued friendship as the highest form of social relation. But are they correct to do so?

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  31

1. Relations with Friends In the waning years of the Ming dynasty, an influential Confucian scholar and radical social critic, He Xinyin (1517–1579), spent two years trying to avoid arrest by the imperial court. He was helped by his close friends, but the authorities caught up with him, and he died in prison in the fall of 1579. His last wish was to be buried next to his close friend, the scholar Cheng Xu. This “final act” created an uproar at the time b­ ecause it contravened the Confucian norm that the dead should be buried next to their relatives, thus earning He the epithet “crazy Confucian” (狂儒). In his own mind, however, He was not “crazy”; rather, he aimed to provide a morally desirable interpretation of Confucian ethics. He did not reject the five cardinal interpersonal relations of Confucian ethics: between parent and child, t­ here should be affection; between sovereign and minister, righ­teousness; between husband and wife, attention to their separate functions; between old and young, proper order; and between friends, fidelity. But whereas traditional Confucians valued the first four hierarchical relations—­with special value on the parent and child relation as the most “natu­ral” and the starting point for all the ­others2—­He argued that the relation between friends, founded on equality, was the highest form of social relation, which should serve as the standard for structuring other ­human relations. Even seemingly hierarchical relations should be infused with the spirit of egalitarian friendship. For example, the relation between teacher and student should be a mutual learning pro­cess, with both interacting as equal teachers (相师) and friends (相友).3 And what made the sage rulers of the past so successful is that they interacted with ministers, younger siblings, and even their own ­children as though they ­were equal friends and teachers of each other.4 He traces the origin of his theory to Confucius

32  c h a p t e r 1

himself: “Only friendship could assem­ble all the talented u­ nder Heaven. . . . ​That’s why the Confucian orthodoxy could be traced back to Confucius in the Spring and Autumn period. . . . ​The Dao of friendship was revealed to Confucius by Heaven, so that he could reach the highest good.”5 But He was an innovator, not just a transmitter. He was the first Confucian thinker to argue that friendship is the highest form of social relation and that friends should be both teachers and students learning equally from each other. And he was the first to argue that the ideal of equal friendship should serve as the standard to evaluate other social relations.6 That’s why con­temporary scholars praise He as the pioneer of the “Chinese enlightenment.”7 But is the ideal of equality between friends ­really the highest—­most desirable—­ form of social relation? He’s idealization of friendship is more common in the history of Western philosophy. A recent book on the relation friendship between the eighteenth-­century thinkers David Hume and Adam Smith paints a moving depiction of their deep friendship. Both Hume and Smith valued friendship as the highest social good: Hume held that “friendship is the chief joy of ­human life,” and Smith proclaimed that the esteem and affection of one’s friends constitutes “the chief part of human happiness.”8 Moreover, their own friendship constituted the very highest form of friendship, with two friends motivated by virtue and excellence: “a stable, enduring, reciprocal bond that arises not just from serving one another’s interests or from taking plea­sure in one another’s com­pany, but also from shared pursuit of a noble end: in this case, philosophical understanding.”9 In the Western tradition, the archetype of ideal friendship can be traced to Aristotle. In Aristotle’s view, t­ here are numerous kinds of friendship in the real ethical world. He used the word “friendship” as a meta­phor to discuss (1) the relation in a ­family, for example friendship

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  33

between parents and c­ hildren and between husband and wife; (2) unequal relations, such as the friendship between rulers and subjects and between elders and the young; and (3) friendship in a community, for example, friendship among travel companions and friendship among soldiers.10 But the dif­fer­ent kinds of friendship are not equal in moral worth. Some are motivated by utility, some by plea­sure, and the highest and rarest form—­such as the friendship between Hume and Smith (Aristotle himself does not offer any examples)—is motivated by virtue or excellence, with two equals who seek the ethical life in common: “Perfect friendship is the friendship between good p­ eople and similar in virtue.”11 For both Aristotle and He, the highest ­human relation is characterized by two equal friends sincerely committed to the other’s good and who jointly partake of the ethical life.12 The sixteenth-­century philosopher Michel de Montaigne endorsed Aristotle’s ideal of friendship between equals, but he denied that other forms of friendship are pos­si­ble. For Montaigne, two p­ eople who are not equals cannot be friends: Given the big gap between ­father and son, for example, it is more appropriate for the son to re­spect his f­ ather.13 Montaigne also reversed Aristotle’s view that love, as an excess of emotion, has a character similar to perfect friendship. For Aristotle, love is as extreme as “perfect friendship,” and it is equally rare: One cannot have many true friends, just as it is only pos­si­ble to love one person.14 For Montaigne, however, love is inferior to friendship. The intensity of love and friendship is not the same, and the “chaleur douce” of friendship is more long lasting than the strong erotic passion of love: “As soon as love enters the territory of friendship (where w ­ ills work together, that is), it languishes and grows faint. To enjoy it is to lose it: its end is in the body and therefore subject to satiety. Friendship on the contrary is enjoyed in proportion to our desire: since it is a ­matter of the mind, with

34  c h a p t e r 1

our souls being purified by practicing it, it can spring forth, be nourished and grow only when enjoyed.”15 Friends are soul-­ mates, not body-­mates, and body and soul should be kept separate. That’s not to say soul-­mates ­can’t share their souls, just as lovers share their bodies. For Montaigne, the highest form of friendship involves the merging of souls, so that equality becomes sameness. Montaigne was speaking from personal experience with his dear friend Étienne de la Boétie: “I know his soul as well as mine. . . . ​Every­thing actually being in common between [us]—­wills, thoughts, judgments, good, wives, ­children, honor, and life—­and [our] relationship being that of one soul in two bodies, according to Aristotle’s very apt definition, [we] can neither lend nor give anything to each other.”16 From a normative point of view, however, the merging of souls is questionable. The twentieth-­century French phi­los­o­pher Emmanuel Levinas agrees with Montaigne that friendship between equals can merge into sameness: “My friend and I, we define each other by the relation between us. He is a soul mate, another me (alter ego). Oreste and Pylade are a related pair. The pair of friendship is like the marble and the groove where it rests. According to the my­thol­ogy of Aristophanes in Plato’s Symposium, being double [l’être double] would rather be friendship than love. In friendship, ­every dynamic is absent: friends possess each other.”17 True friends speak with one voice, one heart and one mind, and experience the same emotions: “In his position, I feel my friend’s emotions again through my own sentiment, feel happy for his happiness, and mourned for his pain.”18 Like the person who has found the other half in the my­thol­ogy of Aristophanes, two friends do not have separate identities. But for Levinas, sameness is not something to be celebrated. Quite the opposite. When “friends possess each other,” they exercise power (“puissance”) over each other, which can easily lead to vio­lence.

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  35

­ ere Levinas inverts Montaigne. Far from being inferior to the H relation of friendship, the relation of love (what Levinas terms “eros”) is superior precisely b­ ecause it preserves a distance between self and other that can never be conquered. As long as the distance exists, the relation between self and other can never be a relation of power. Lovers do not seek to complete each other, but to deepen their relation.19 So who’s right? We can accept Levinas’s critique of Montaigne’s ideal: Identities should not be merged. Who wants to be controlled by a “friend” who claims to think the same ­things and experience the same emotions as me? But good friends need not—­and should not—­have the same identity (if anything, lovers are more likely to merge their identities in unhealthy ways). They can be equal but dif­fer­ent, each helping the other to improve her-­or himself and to flourish in her/his own distinctive way. We can also accept Levinas’s point that lovers should maintain their separate identities. But h­ ere too, he goes too far. What would it mean to completely do away with power relations? Drawing on a wide range of findings from the animal world and ­human socie­ties, the cultural anthropologist Christopher Boehm argues that h­ umans have an innate tendency to dominate as well as an innate tendency to resent being dominated.20 Hierarchy is a ubiquitous feature of h­ uman relations, and it is unrealistic to wish it away. That’s not to say we c­ an’t have the equality of friendship, but any social relation also needs to make room for hierarchy, and the task is to distinguish between good and bad forms of hierarchy and to promote social relations that have more of the former. In any case, it makes sense to consult one’s own experience, which might ultimately have more persuasive power than what­ ever ­great phi­los­o­phers have said in the past. It seems hard to object to the ideal of friendship between equals. And we can

36  c h a p t e r 1

learn from friends precisely ­because they are dif­fer­ent. But it ­doesn’t take long to come to the conclusion that friendship is not the highest or most ideal social relation in many ­people’s minds: Think of the “goodbye” letter from a boyfriend (or girlfriend) to another with the devastating line, “let’s just be friends.” This kind of letter implies that loving intimacy is even more valued than ideal forms of friendship.21 But what makes the relation between loving intimates so ­great? ­Here’s our take. Lovers are usually more other-­regarding: they typically care and sacrifice for each other more than ordinary (or even extraordinary) friends. A less obvious—­but equally impor­tant—­reason is that lovers can treat each other as equal friends in the day, while incorporating nighttime hierarchical relations that add color and joy to their lives.22 But what kind of nighttime hierarchical relations between intimates are morally justified? Let’s explore this in the next section.

2. Relations with Lovers The relation between lovers is the most intimate form of social relation, b­ ecause it involves not just intellectual, moral, and spiritual love, but also physical love. And the highest form of love between intimates involves mutual surrender, when both “sides” let go of the defenses that normally accompany social interaction. The lovers let their emotions flow freely, caressing and kissing without constraint, and let their selves merge to the point of self-­annihilation. Even the usual rules of self-­love prescribed by evolution do not apply to the highest form of love between intimates: a lover can sincerely wish for a child that looks and thinks just like her intimate (a thought that would never occur to animals). But love between intimates is also extremely dangerous. The lover, who has surrendered his or her self to the other without

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  37

any mode of self-­protection, is particularly vulnerable to abuses of power, not just physical but also emotional and social. And much of the interaction between lovers takes place when nobody is watching, in the privacy of the bedroom. Hence it is doubly impor­tant that hierarchies between lovers are not fixed and allow for changes over time. We worry most about hierarchies between intimates that mirror and reinforce hierarchies in other spheres of social life. Just about ­every society in h­ uman history has been informed by strongly hierarchical social relations that express patriarchal dominance over w ­ omen. Men hold unjust power over ­women in work and politics. Feminists have long argued that patriarchal relations within the home have negative implications for ­women outside the home: ­Women’s opportunities in the “public” sphere ­will be ­limited so long as ­women do a large share of child-­rearing and ­house­work in the “private” sphere (hence the slogan, “the personal is the po­liti­cal”). When it comes to hiring and promotion, most bosses w ­ ill prefer (male) workers who can (supposedly) devote themselves to the job without worrying too much about f­ amily obligations.23 Less discussed (and admittedly, more speculative) is the view that “nighttime hierarchies” between lovers may exacerbate ­these negative effects on ­women. If “private” sexual relations between lovers (­w hether married or not) are characterized by male dominance—­with the male on top and playing the more active role—­it’s hard to believe that the psychological effects of male dominance ­won’t be transferred to other realms of social interaction, including daytime interaction between lovers. But what if the female is the dominant partner in nighttime hierarchies? Surely that can challenge patriarchy in other realms of social life? Not so sure. In Japan and South ­Korea, patriarchal power relations outside the home are far more deeply entrenched than in mainland China (what­ever we think of the Chinese

38  c h a p t e r 1

revolution, one of its effects has been to equalize gender relations compared to other large-­scale Confucian-­influenced socie­ ties in East Asia). Japa­nese and Korean wives typically control the f­ amily finances, but such practices can reinforce patriarchal power relations outside the home. For one ­thing, the expectation that w ­ omen “run the home” means that employers are less likely to consider female job candidates who (employers think) ­won’t be able to devote themselves wholly to work. And the fact that w ­ omen have power at home may also limit their aspirations outside the home. If ­women get some satisfaction from controlling (most of) what happens in the home, they may have less desire to push for equalizing power relations outside the home. We do not know what happens in the bedrooms of Japan and South K ­ orea, but a similar mechanism may be at work if the w ­ oman is the dominant partner in nighttime hierarchies. If ­women can get a high degree of satisfaction from being in control and exercising power during sexual intimate relations, it may also limit their desire to challenge patriarchal relations in the worlds of work and politics. In short, nighttime hierarchies are problematic from a feminist point of view when lovers habituate themselves to unchanging habits of dominance and subordination, even if the ­woman is the dominant partner. The solution may then seem obvious: Intimate lovers should change their hierarchies over time. Nighttime hierarchies need to be unfrozen and roles (ex)changed over time. But what can be done to realize this ideal? We do not think it’s a ­matter for state intervention or ­legal regulation. However much we may object to the social and po­liti­cal effects of nighttime hierarchies, we agree with former Canadian prime minister Pierre Trudeau (current prime minister Justin Trudeau’s ­father) that the state has no business in the bedrooms of the nation. But it’s still something to worry about, and po­liti­cally

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  39

progressive intimate lovers can learn techniques designed to unfreeze nighttime hierarchies. The ancient Chinese classic The Rec­ords of Rites, written between the fourth and second centuries BCE,24 offers an intriguing example of a role reversal. In a ritual that follows the death of a ruler, the ruler’s son becomes the new ruler, who then performs the sacrifices to his deceased ­father, while an impersonator plays the role of the role of the deceased, receiving sacrifices from the living. What’s in­ter­est­ing is that the living ruler’s own son—­who is normally supposed to be deferential and subordinate to his ­father—­would play the part of the impersonator for the ghost of his grand­father (the ruler’s deceased ­father): Now, according to the way of sacrificing, the grand­son acted as the impersonator of the king’s f­ ather. He who was made to act as the impersonator was the son of the one who made the sacrifice. The f­ ather faced north and served him. By means of this, he made clear the way of a son serving his ­father. This is the relation of ­father and son.25 The stated goal of the ritual, Michael Puett explains, “is to inculcate in each performer the proper dispositions that should hold in the relationship between ­father and son.”26 The learning takes place by means of role reversals, with the ruler behaving as a proper son to his own son, and the son as the honored elder receiving sacrifices from his own f­ ather. The assumption is that the participants w ­ ill learn to think and act from the perspectives of ­those who are on the other ends of the hierarchies in everyday life—­the power­ful ­will learn to think and feel from the perspective of the subordinate, and vice versa—­hence softening the anger, jealousy, and resentment that normally accompany fixed and rigid hierarchies. Carnivals in Eu­ro­pean cultures serve a similar function, with role reversals and impersonations that

40  c h a p t e r 1

release participants from the fixed and rigid hierarchies of everyday life. Some corporations t­ oday rely on such role reversals—­ the CEO takes a job at the lower levels for a day or two—in order to break down the deleterious effects of fixed hierarchies and encourage a certain degree of empathy and understanding for ­those on the other ends of hierarchies. Intimate lovers can also engage in such role reversals of nighttime hierarchies, with the active, relatively power­ful lover playing the part of the subordinate, and vice versa.

The Kamasutra But role reversals between intimates cannot be one-­off games. The changes must be frequent to be long-­lasting, and they must be relatively spontaneous, almost unselfconsciously performed during acts of passion and compassion, to truly engage with the emotions of the participants. ­Here the Indian classic the Kamasutra offers particularly valuable insights. The Kamasutra, or the Treatise on Plea­sure, composed in the third ­century CE, is the world’s most influential textbook on erotic love. It is famous ­today largely ­because of its explicit sexual content (often accompanied by pictures added by translators and publishers to boost sales), but “it is a book about the art of living—­about finding a partner, committing adultery, living as or with a courtesan, using drugs—­and also about the positions in sexual intercourse.”27 The book opens with a brief discussion of the three aims of h­ uman life valued in ancient Hindu texts—­religion/moral duty (dharma), power/material success (artha), and plea­sure/love/ desire (kama). When the three aims compete, “each is more impor­tant than the one that follows,”28 but the rest of the book is a discussion of kama (the least impor­tant). The book is a joy to read, not just ­because, as Wendy Doniger explains, it “reveals

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  41

attitudes to ­women’s education and sexual freedom, and non-­ judgmental views of homosexual acts, that are strikingly more liberal than ­those of other texts in ancient India—or, in many cases, con­temporary India,”29 but also ­because of its somewhat lighthearted, almost cynical tone. The text is largely written from a male point of view, to maximize the man’s plea­sure. But parts of the text are explic­itly directed at w ­ omen: “Scholars [pedants] say: ‘Since females cannot grasp texts, it is useless to teach w ­ omen this text.’ Vatsyayana [author of the Kamasutra] says: But w ­ omen understand the practice, and the practice is based on the text. . . . ​ And ­there are also ­women whose understanding has been sharpened by the text: courtesans and the ­daughters of kings and state ministers.”30 The Kamasutra views sex “as a form of quarrelling, ­because the very essence of desire is argument, and its character is competitive.”31 But the power strug­gles are told from both male and female perspectives. It is clearly sympathetic to ­women, particularly when they suffer from inadequate lovers. The amusing discussion of how a courtesan can get rid of an unwanted (male) lover is worth quoting at length: She does for him what he does not want, and she does repeatedly what he has criticized. She curls her lip and stamps on the ground with her foot. She talks about t­ hings he does not know about. She shows no amazement, but only contempt, for the t­ hings he does know about. She punctures his pride. She has affairs with men who are superior to him. She ignores him. She criticizes men who have the same faults. And she stalls when they are alone together. She is upset by the ­things he does for her when they are making love. She does not offer him her mouth. She keeps him away from between her legs. She is disgusted by wounds made by nails or teeth. When he

42  c h a p t e r 1

tries to hug her, she repels him by making a “needle” with her arms. Her limbs remain motionless. She crosses her thighs. She wants only to sleep. When she sees that he is exhausted, she urges him on. She laughs at him when he cannot do it, and she shows no plea­sure when he can. When she notices that he is aroused, even in the daytime, she goes out to be with a crowd. She intentionally distorts the meaning of what he says. She laughs when he has not made a joke, and when he has made a joke, she laughs about something e­ lse. When he is talking, she looks at her entourage with sidelong glances and slaps them. And when she had interrupted his story, she tells other stories. She talks in public about the bad habits and vices that he cannot give up. Through a servant girl, she insults him where he is vulnerable. She does not see him when he comes to her. She asks for ­things that should not be asked for.32 One won­ders if the author speaks from his own unhappy experience, but what’s certain is that w ­ omen (not just courtesans) are provided with an exhaustive list of tips for puncturing the male ego. The Kamasutra’s account of shifting roles in nighttime power games helps us think about morally justifiable hierarchies between intimate lovers.33 The text pays lip ser­v ice to conventional (patriarchal) ideas of gender: “By his physical nature, the man is the active agent and the young ­woman is the passive locus. . . . ​The man is aroused by the thought, ‘I am taking her,’ the young w ­ oman by the thought, ‘I am being taken by him.’ . . . ​ A man’s natu­ral talent is his roughness and ferocity; a w ­ oman’s is her lack of power and her suffering, self-­denial and weakness.”34 Intimate lovers, however, can (and should) deviate from ­these norms: “Their passion and a par­tic­u­lar technique may sometimes

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  43

lead them to exchange roles, but not for very long.”35 ­Women need to take the lead in initiating such role reversals: When she sees that the man has become exhausted by continuous repetition, but that his passion is still not quenched, she may, with his permission, roll him ­under her and give him some help by playing the man’s part herself. Or she can do it out of her own desire to do something she has only ­imagined ­doing, or to satisfy the man’s erotic curiosity. To play the man’s part, when he is inside her, she gets on top and puts him under­ neath her. . . . ​She says, “You threw me down, and now I am throwing you down in return,” laughing and threatening him and hitting him. And at the same time, she indicates that she is embarrassed and exhausted and wishes to stop.36 The power games and role reversals may get out of hand, especially for the “power­ful” man: “What­ever wound a man inflicts on a ­woman, even when she tries to restrain him and cannot bear it, she should do that very t­ hing to him twice as hard. . . . ​She grabs him by the hair and bends down his face and drinks from his mouth; she pounces on him and bites him ­here and ­there, crazed with passion.”37 But role reversals can help the man learn about his lover’s needs: “Even when a sensual ­woman covers up her own feelings and hides her signals, she unveils her own feelings completely when her passion drives her to get on top. A man can learn every­thing—­a ­woman’s personality, what sort of sex excites her—­from the ways she moves on top.”38 Role reversals contribute to the diversity that sustains passion: “For even passion demands variety. And it is through variety that partners inspire passion in one another.”39 Hence, role reversals can help to stabilize long-­term relationships between intimate lovers: “When two ­people behave in this way with modesty and concern for one another’s feelings, their love ­w ill never wane, not

44  c h a p t e r 1

even in one hundred years.”40 Conversely (by implication), fixed hierarchies between intimates undermine love and mutual concern, and kill off passion. It might seem strange, perhaps even counterproductive, to advocate detailed techniques for role reversals with the aim of deepening passion between intimates. The more one thinks about such techniques, the less unbridled the passion. But h­ umans are not animals when it comes to sex: “Scholars [pedants] say . . . ​Since even animals manage to have sex by themselves, and since it goes on all the time, it should not be handled with the help of a text.” Vatsyayana says: “­Because a man and a ­woman depend on one another in sex, it requires a method, and this method is learned from the Kamasutra. The mating of animals, by contrast, is not based on any method: ­because they are fenced in, they mate only when the females are in their fertile season and ­until they achieve their goal, and they act without thinking about it first.”41 The method involves role reversals so that both lovers learn more about each other: “The best alliance plays the game so that both sides taste one another’s happiness and treat one another as unique individuals.”42 Once the techniques are learned, they can be unselfconsciously performed, with frequent changes of roles motivated by passion rather than the desire to perfect a technique: “The territory of the texts extends only in so far as men have dull appetites; but when the wheel of sexual ecstasy is in full motion, ­there is no textbook at all, and no order.”43 The Kamasutra aims to make itself obsolete in the long term. Intimate lovers w ­ ill have learned the techniques that contribute to lasting passion, including frequent role reversals in nighttime hierarchies, and then unselfconsciously apply ­those techniques without need of any text. In short, what’s distinctive about morally justifiable hierarchies between intimate lovers is not just that they involve role

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  45

reversals, but that, ideally, ­those role reversals are frequently and unselfconsciously applied in the midst of passionate embraces. Not only ­will t­ hese role reversals contribute to the diversity and empathy that help sustain passion over the long term, but they can also help to challenge the patriarchal relations that typically characterize other spheres of social and po­liti­cal life (both in ancient India and in the con­temporary world). Let us now turn to other forms of morally justifiable hierarchies between intimates. ­These also involve role reversals, but with longer time spans in family-­based hierarchies, and even longer time spans in hierarchical relations with ­house­keepers.

3. Relations with ­Family Members The idea that t­ here is a hierarchy between elders and younger ­people is central to Chinese culture. Mencius (known as Mengzi in Chinese) and Xunzi, for example, notwithstanding totally dif­ fer­ent starting points about ­human nature, both agree that ­there should be age-­based hierarchical rankings between ­people.44 Throughout Chinese history, this idea was institutionalized by means of laws and informal norms that empowered the el­derly in a variety of economic, social, and po­liti­cal ways.45 Still t­ oday, the Chinese po­liti­cal system is structured so as to ensure that top decision makers have de­cades of experience: since 1989, nobody ­under fifty has assumed positions in the Standing Committee of the Politburo. Age-­based hierarchies are rooted in the idea of filial piety (孝). We ­ought to revere el­derly members of the ­family, and then extend that reverence to el­derly ­people as a ­whole.46 Note that filial piety does not simply refer to the idea that c­ hildren ­ought to (generally speaking) obey their parents. E ­ very known society accepts a hierarchy between parents and c­ hildren. Parents,

46  c h a p t e r 1

due to superior knowledge and morality, have power over ­children, but parents are supposed to use that power to help children develop into flourishing adults. What happens when ­children reach adulthood, say, around eigh­teen years of age? In the West, the assumption is that adult c­ hildren and their parents are then equals, and hierarchical relationships are no longer justified. In China, the assumption is that parents continue to have some form of unequal authority over adult c­ hildren and adult ­children are supposed to serve their el­derly parents. They are not equals, even when both are adults. What justifies age-­based hierarchies between adult ­children and el­derly parents? Confucian thinkers often invoke the argument that filial piety is distinctive to h­ uman beings.47 Animals, like ­humans, often care for their ­children, but they typically do not care for their el­derly. Even if the claim that filial piety is distinctively ­human is correct, we still need to justify a move from descriptive fact to normative injunction (the ability to develop intricate means of inflicting pain may also be distinctive to ­humans, but it’s not a good ­thing). It’s also worth asking what, if anything, hinges on the descriptive fact: If filial piety is a good ­thing, and if it turns out that some animals do care for their el­ derly, we should praise ­those animals rather than give up our commitment to filial piety. 48 So Mencius’s argument c­ an’t be the ­whole, or even the main, story. Another argument is more recent: Filial piety is central to Chinese culture and if we care about maintaining Chinese culture, we should care about maintaining filial piety.49 But clearly filial piety has changed over the years. The days when el­derly parents controlled the ­family income and property are long gone; now, adult c­ hildren typically support their el­derly parents. And fewer and fewer ­people live with their el­derly parents. El­derly parents have far less say in ­family decisions compared to families in imperial China. Clearly el­derly

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  47

parents are becoming more and more disempowered. W ­ hether this trend is good or bad is a separate question, and we need more arguments to defend the claim that (what’s left of) age-­based hierarchies should be maintained, if not reinforced in the ­future. What, then, are the arguments in ­favor of age-­based hierarchies between adult ­children and their parents? ­There are six arguments that, together, constitute a strong case for age-­based hierarchies. The first argument, more closely tied to the ­family, invokes the value of reciprocity. Our parents cared for us when we ­were c­ hildren, and adult ­children have an obligation to care for el­derly parents when they are older and frail. That seems fair. In the West, adult c­ hildren often serve their parents, but it’s viewed as a m ­ atter of choice. Not in China: Adult c­ hildren must serve their adult parents, a norm that is often reinforced by ­legal means.50 Yes, the state ­will need to play a greater role in providing el­derly care in the f­ uture (given that many single c­ hildren need to support two adult parents), but the ideal of providing care for el­derly parents is not likely to lose its normative force in the foreseeable ­future. And some social practices that express reverence for the el­derly are still widely followed in China: For example, adult parents typically get first dibs at communal dishes at the f­ amily ­table. Still, the idea that we have an obligation to serve our parents does not necessarily translate into an argument that el­derly parents should assume leadership roles in hierarchical relations with their adult ­children: We also have an obligation to care for our ­house pets, but the pets are not our masters. So we need more arguments to justify age-­based hierarchies within the f­ amily. The second argument invokes the value of learning from experience. This argument is not distinctive to ­humans. Female red deer, for example, do not fight for dominance of the herd. The leaders are usually older, and they lead ­because the followers

48  c h a p t e r 1

“recognize the experience that comes with age.”51 But ­humans can do even better: They can consciously commit themselves to learning in a way that broadens their intellectual horizons. Learning is a never-­ending pro­cess of accumulating knowledge: As Confucius put it, “A person who is constantly aware of what has yet to be learned and who, from month to month, does not forget what has been learned, can be said to truly love learning” (19.5).52 Since reading and studying are time-­consuming pro­ cesses, the el­derly are more likely to have had the time to read and study with a view to improving their lives. Hence, adult c­ hildren need to defer to the intellectual judgments of el­derly parents, other ­things being equal. But other ­things are rarely equal. In disciplines such as mathe­matics or physics, the best intellectual work is often done by thinkers in their twenties or thirties. And ­today, youngsters often need to teach el­derly parents (and grandparents) about uses of modern gadgets essential for navigating the social and virtual worlds. What­ever intellectual authority the el­derly may have had in the past has been eroded by the youthful virtuosos of Silicon Valley and Shenzhen. So we need still more arguments for age-­based hierarchies within the ­family. The third argument invokes the value of emotional intelligence.53 From a Confucian standpoint, it seems obvious that emotional intelligence—­meaning social skills such as self-­ awareness, self-­regulation, and the ability to understand others—­ normally increases over time. As we age, we experience dif­fer­ent roles (such as dealing with bosses, colleagues, and subordinates in the workplace) and deepen our experience in par­tic­u­lar roles (a community or­ga­nizer with ten years’ experience should be more effective than a brand-­new or­ga­nizer), and thus we increase our ability to understand and cooperate with dif­fer­ent kinds of ­people for the purpose of achieving desired ends, so long as we maintain the quest for self-­improvement and our

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  49

desire for social interaction. As it turns out, scientific research bears out this Confucian insight: “One t­ hing is certain: Emotional intelligence increases with age.”54 Fredda Blanchard-­ Field’s research compares the way young adults and older adults respond to situations of stress and “her results show that older adults are more socially astute than younger p­ eople when it comes to sizing up an emotionally conflicting situation. They are better able to make decisions that preserve an interpersonal relationship. . . . ​And she has found that as we grow older, we grow more emotionally supple—we are able to adjust to changing situations on the basis of our emotional intelligence and prior experience, and therefore make better decisions (on average) than do young ­people.”55 Other research shows that older adults seem particularly good at quickly letting go of negative emotions ­because they value social relationships more than the ego satisfaction that comes from rupturing them.56 In short, we have good reason to empower el­derly parents in the ­family context— to give them more voice, and let them decide in moments of emotional conflict—­because they are more likely to have superior social skills. On the other hand, a sociopath with superior social skills can “read” ­people and manipulate them for immoral purposes more effectively than an incompetent and insensitive person. So we need still more arguments to justify age-­based hierarchies within the ­family. The fourth argument invokes the value of moral growth.57 One of the most widely quoted sayings from The Analects of Confucius is the brief account Confucius gives of his own life: “At fifteen, I set my mind upon learning; at thirty, I took my stance; at forty, I was no longer perplexed; at fifty, I realized the ‘ways of the universe’; at sixty, my ear was attuned; at seventy, I followed my heart’s desire without overstepping the bound­aries” (2.4). In con­temporary China, the saying has been somewhat

50  c h a p t e r 1

distorted: for example, thirty-­year-­olds take Confucius to be saying that they should be established in their ­careers. But Confucius himself is tracing his own pro­gress of moral growth: His capacity for moral judgment improves, and he can act better, morally speaking. Why does he think morality improves with age? The text is not so clear, but one reason for believing that the el­derly have greater capacity for moral judgment is that they are less likely to be enslaved by sexual desire. Confucius notes that he can give ­free rein to his heart’s desires at the age of seventy, meaning that ­there is less of a conflict between what he wants to do and what he should do. Why would Confucius say that? Elsewhere in the Analects, Confucius notes despairingly that he “has yet to meet anybody who is fonder of virtue than of sex” (15.13). But Confucius is addressing his own students, and he may not say the same t­ hing to an older crowd. That is, as sexual desire diminishes with age, ­there may be less conflict between the desire for sex and the desire to do good. This is not to imply that the desire for sex is entirely extinguished for el­derly ­people, but it is easier to control and subordinate to moral princi­ples (compared to male adolescents!). That said, the phenomena of sexual abuse and harassment do not, sadly, have an age limit, so we need still more arguments to defend age-­based hierarchies among adult ­family members. The fifth argument invokes the value of economic equality. Zhang Taisu argues that seniority-­based social hierarchies in early modern China served to safeguard remarkably per­sis­tent socioeconomic equality. Qing and Republican property institutions often gave greater economic protection to the poorer segments of society than comparable institutions in early modern ­England: “the comparatively ‘egalitarian’ tendencies of Qing and Republican property institutions stemmed from the dif­fer­ent ways Chinese and En­glish rural communities allocated social

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  51

status and rank. Hierarchical ‘Confucian’ kinship networks dominated social and economic life in most Chinese villages. Within t­ hese networks, an individual’s status and rank depended, in large part, on his age and generational se­niority, rather than personal wealth. This allowed many low-­income ­house­holds to enjoy status and rank highly disproportionate to their wealth.”58 In other words, age-­based hierarchies at the ­family and village levels actually promote economic hierarchy at the societal level: If ­people are given power ­because of their age, not their wealth, the state is less likely to enact policies that benefit the rich, and the overall effect w ­ ill be to equalize the distribution of wealth (relative to socie­ties that do not empower the el­derly). Of course, such policies ­were based on patriarchal assumptions that we reject t­ oday: Only the el­derly men had substantial power in traditional China. But a gerontocracy without gender bias would still equalize wealth, so in princi­ple it’s a good idea to strive to empower el­derly men and ­women in the ­family and the local community (assuming that we worry about radical inequalities of wealth distribution). The bigger prob­lem is practical. Zhang objects to the “socialist” road to economic equality on the grounds that it involves too much coercion in the form of strong state regulation and aggressive wealth distribution from the top. Empowering the el­ derly in families and local communities could achieve the same end without nearly as much coercion. But it’s hard to imagine the Chinese Communist Party reviving such “feudal” practices from the past. And it’s even harder to imagine such pro-­gerontocracy reforms in demo­cratic Western socie­ties without a history of empowering the el­derly. It is abstractly conceivable that, say, extra votes for the el­derly could mitigate the power of wealth in the United States,59 but the rest of the society is almost certain to object strongly to such mea­sures.

52  c h a p t e r 1

The sixth argument invokes the value of harmony. Harmony in the Confucian sense of peaceful order and re­spect for diversity depends on the idea that ­there is a “decider” who has the authority to make the final call in cases of conflict.60 Without a “decider,” conflicts that cannot be solved by love and peace are likely to spin out of control, as factions fight for victory, thus undermining social harmony. In politics, the “decider” would be, ideally, a public official with superior ability and virtue and a proven track rec­ord of good per­for­mance. Families are not (and should not be) meritocracies to the same extent: They are supposed to be run on love and informal norms, and it would be absurd to subject el­derly parents to a battery of tests before they can assume positions of power in the f­ amily.61 But it does make sense to empower the ­family member who is (1) owed thanks due to previous love-­infused work on behalf of the f­ amily, (2) more likely to have the most knowledge, (3) more likely to have superior emotional intelligence, and (4) more likely to be in control of his or her sexual urges. Elder members of the ­family are more likely to instantiate t­ hese desiderata (counter-­examples may readily come to mind, but it’s a question of tendencies). Empowering ­family elders is also likely to promote more economic equality in the po­liti­cal community. Hence, on balance, t­ here is a good case for empowering the el­derly members of the ­family, with the consequence that adult ­children typically need to defer to their el­derly parents in cases of conflict. Still, we need to reiterate our opposition to hierarchies that become frozen and impervious to change. So do we need to worry about fossilized age-­based hierarchies in the f­ amily? Not ­really. For one ­thing, younger f­ amily members on the bottom of the f­ amily hierarchy can and should criticize power­ful elders who commit moral wrongs. In the Confucian tradition, even young ­children have an obligation to criticize parents who commit

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  53

moral wrongs, though it must be done in an appropriate context. The Qing dynasty classic “Rules for Students” (弟子规)—­still widely taught in China’s primary schools t­ oday—­suggests three steps to the young child: first, use persuasion; if that ­doesn’t work, then wait ­until the parent is happy, and try again; if that ­doesn’t work, then try weeping and wailing, working on the parent’s emotions; and if that d­ oesn’t work, then the child needs to accept the parent’s ­will (including the use of physical punishment, which we do not endorse).62 ­There is still in­ equality between parent and child—­these conditions of critique do not apply in the case of the parent criticizing the child who commits a wrong—­but clearly ­there is no injunction of blind obedience on the part of the child. Plus, it makes sense to say that the child needs to obey the parent at the end of the day, if none of the tactics are successful (other than extreme situations, the final option should not be to run away or change families). But does it make sense to say that adult ­children have to defer to el­derly parents in cases of conflict? In the past, el­derly Chinese parents arranged the marriages of their adult ­children, and few young ­people want to bring back ­those days. But it’s still common, and widely ­accepted, for el­derly parents in China to set up meetings with potential marriage mates for their adult ­children, and ­there is also an expectation that parents be consulted before the marriage deal is closed. Being the “decider” does not necessarily mean exercising veto power in all aspects of ­family life. It might mean that adult c­ hildren have an obligation to consult el­derly parents when they make decisions, whereas el­derly parents do not have the same obligation to consult with adult c­ hildren (e.g., if el­derly parents decide to get remarried). Such unequal social practices may seem dubious in the West, but they are taken for granted in Confucian-­ influenced China.

54  c h a p t e r 1

Another reason not to worry is that roles in age-­based hierarchies, by definition, change over time. The child w ­ ill become an adult, and then an elder, who ­will eventually have the same authority over adult ­children that her or his own parents had.63 In that sense, age-­based hierarchies are fundamentally dif­fer­ent, and more legitimate, than race-­or gender-­based hierarchies that are fixed for eternity. Moreover, the hierarchy between adult ­children and el­derly parents often ends up with a complete role reversal. Beyond a certain age, the el­derly parent often loses the capacity to make decisions due to physical and m ­ ental deterioration. In the case of Alzheimer’s patients, el­derly parents literally regress over time, to the point that they become like helpless babies. At that point, ­there is a complete role reversal, with adult ­children taking charge of the decision making.64 In traditional China, most ­people did not live long enough for Alzheimer’s to manifest itself, which may help to explain why role reversal in the f­ amily was rarely discussed in ancient texts. And in the ­f uture, we ­w ill hopefully discover ways of slowing down, if not eliminating, cognitive and physical decline. But for the moment, we can expect that the hierarchical relation between adult ­children and their el­derly parents w ­ ill often be characterized by complete role reversals over the long term. We do not mean to imply that the traditional Confucian justifications for hierarchical relations between f­ amily members should serve as the standard for assessing hierarchical relations in the ­family t­ oday. Traditional Confucians also tried to justify a hierarchical relation between wife and husband that is morally indefensible from a progressive perspective.65 We do think that Confucian-­inspired justifications for age-­based f­ amily hierarchies, properly adjusted to con­temporary norms of gender equality, are still relevant t­ oday. But t­ here may be an additional worry. If age-­based hierarchies are justified, does it follow that

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  55

we should revive the Confucian-­style hierarchical bond between elder and younger b­ rother, modernized so that it refers to the hierarchical bond between elder and younger sibling, regardless of gender? Our response is that it depends on the age gap. Daniel is eleven months older than his younger ­sister, and while he occasionally invokes the Confucian need for deference to elder siblings, he has not been effective in asserting his authority over his ­sister. Perhaps the age gap with his s­ ister is too small for the advantages of age difference to kick in. But the elder sibling may have a legitimate claim to unequal power in the f­ amily if t­ here is, say, a twenty-­year gap between the two siblings. An elder sibling with de­cades of experience is more likely to have superior knowledge, social skills, and virtue, and is owed thanks if he or she cared for the younger one. So age-­based hierarchies in the ­family may be extended beyond the child-­parent relation, but only if the age gap approximates the same age gap as that between child and parent. In short, age-­based hierarchies between f­ amily members, like hierarchical relations between intimates, may involve role changes if not complete reversals: That’s key to their moral legitimacy. The age-­based role changes do occur, though at a much slower pace than role changes between intimate lovers. As we w ­ ill see, however, one role change between hierarchical intimates takes even longer: the change in the relation between employer and ­house­keeper.

4. Relations with House­keepers Perhaps the most infamous argument in the history of (Western) po­liti­cal theory is Aristotle’s defense of slavery: Some p­ eople are born to be slaves, therefore slavery is justified. How can someone so smart—­the genius among geniuses—­get it so wrong?

56  c h a p t e r 1

Well, for one t­ hing, he recognizes that slavery would not be necessary in a world of advanced technology: “if ­every instrument could accomplish its own work, obeying or anticipating the ­will of ­others, like the statue of Daedalus, or the tripods of Hephaestus, which, says the poet, ‘of their own accord entered the assembly of the Gods,’ if in like manner, the shut­tle would weave and the plectrum the lyres, chief workmen would not want servants, nor masters slaves.”66 Clearly slavery is not something to celebrate. Aristotle ­couldn’t foresee the development of modern machines and artificial intelligence, but he’d prob­ably be overjoyed by the prospects that all h­ umans, including slaves, could potentially be freed from the need to engage in menial ­labor (see chapter 5). Karl Marx did foresee a world where machines do all the drudge ­labor—he called it higher communism—­but he also recognized that exploitation of workers is a necessary evil ­until we get to that stage. Aristotle also argues that the use of raw power cannot justify slavery: “­There is slavery by convention as well as by nature. The convention is a sort of agreement—­the convention by which what­ever is taken in war is supposed to belong to the victor. But this right many jurists impeach, as they would an orator who brought forward an unconstitutional mea­sure: they detest the notion that, b­ ecause one man has the power of ­doing vio­lence and is superior in brute strength, another ­shall be his slave and subject.”67 Enslaving prisoners of war is not a justification for slavery, particularly if the war is unjust: “­Others, clinging as they think simply to a princi­ple of justice (for convention is a sort of justice), assume that slavery in accordance with the custom of war is just, but at the same time deny this. For what if the cause of war be unjust?”68 Nor can success in war justify enslaving the ­children of prisoners of war: “And again, no one would ever say that he is a slave who is unworthy to be a slave. W ­ ere this the case,

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  57

men of the highest rank would be slaves and the ­children of slaves if they or their parents chanced to be taken captive and sold.”69 In short, Aristotle’s defense of slavery is somewhat half-­hearted. It only applies in a world where h­ uman intelligence is necessary to provide the necessities of life, and it does not apply to “worthy” ­people who happen to have been enslaved in war, nor does it extend to the next generation. So what exactly is Aristotle’s argument for “natu­ral” slavery? He says: “For that some should rule and ­others be ruled is a ­thing not only necessary, but expedient, from the hour of their birth, some are marked out for subjection, ­others for rule.”70 Even if true, however, that is not a sufficient argument for slavery. Mencius, writing around the same time, made a similar argument: “Some ­labor with their minds, and some ­labor with their strength. ­Those who ­labor with their minds rule ­others, and ­those who ­labor with their strength are ruled by o­ thers. Th ­ ose who are ruled by ­others feed p­ eople, and t­ hose who rule p­ eople are fed by ­others.”71 This passage was frequently cited in defense of social hierarchies in traditional China, but never as a justification for slavery. It referred to the idea that t­ hose with superior intelligence (and virtue) should be put in positions of power, but their obligation is to serve the o­ thers with compassion (a notion that is largely absent from ancient Greek thought), not to exploit them.72 ­Those who are ruled have an obligation to work the fields to feed the rest of society, but certainly not to serve rulers qua slaves.73 So Aristotle needs another argument to defend “natu­ral” slavery. His argument depends on a strong distinction between the soul and the body and their respective roles: A living creature consists in the first place of soul and body, and of the two, the one is by nature the ruler and the other the subject . . . ​the soul rules the body with a despotic rule,

58  c h a p t e r 1

whereas the intellect rules the appetites with a constitutional and royal rule. And it is clear that the rule of the soul over the body, and of the mind and the rational ele­ment over the passionate, is natu­ral and expedient; whereas the equality of the two or the rule of the inferior is always hurtful. . . . ​W hen then ­there is such a difference as that between soul and body, or between men and animals (as in the case of ­those whose business is to use their body, and who can do nothing better), the lower sort are by nature slaves, and it is better for them as for all inferiors that that should be ­under the rule of a master. For he who can be, and therefore is, another’s, and he who participates in reason enough to apprehend, but not to have, is a slave by nature. Whereas the lower animals cannot even apprehend reason, they obey their passions.74 The first ­thing to note is that Aristotle’s argument has no purchase in a philosophical culture that rejects, or ­doesn’t work with, the soul-­body distinction. In ancient China, it was a nonissue: the “soul” ­hadn’t been in­ven­ted as a concept and, more generally, ­there was no mind-­body dichotomy (the word 心 xin in classical Chinese is usually translated as “heart-­mind”). Even assuming the distinction between soul and body, however, the bound­aries are unclear. Aristotle argues that rulers are ­those whose (rational) souls rule over their (passionate) bodies. Slaves are the opposite, but they have the ability to apprehend the dictates of reason (unlike animals). But how can we confidently identify ­those with worthy souls, given that, as Aristotle admits, “the beauty of the body is seen, whereas the beauty of the soul is not seen”?75 And how can we be sure that ­those with not-­so-­ beautiful souls cannot be improved with education? Aristotle seemed to have recognized the limits of his own argument at the end of his life: “in Aristotle’s w ­ ill, the only genuinely personal

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  59

document from his pen that we possess . . . ​we find Aristotle freeing his slaves, an unnecessary and generous gesture for a man of his time.”76 If Aristotle had persuaded himself that some ­people ­were born to be slaves, why did he go out of his way to grant freedom to his own slaves? What’s in­ter­est­ing for our purposes is that Aristotle’s argument for slaves is made in the context of a discussion about ­house­hold management and the work of slaves includes “cookery and similar menial arts.”77 And h­ ere ­there may be more parallels with other cultures. In ancient China, for example, ­there was not a formal system of slavery, but h­ ouse­keepers ­were often bound to h­ ouse­holds for life, and had to serve their masters in ways that may not have been so dif­fer­ent in substance than Aristotle’s ­house­hold slaves. However, ­house­keepers ­were also bound within a Confucian ethical system that often mitigated the worst abuses of domestic servitude.78 A basic assumption of Confucian ethics is that the moral life is pos­si­ble only in the context of particularistic personal ties. For the general population, the most impor­tant relationship by far is the ­family. It is by fulfilling our responsibilities to ­family members that we learn about and practice morality (for Aristotle, in contrast, the good life lies outside the home). The value of caring for ­children is widely shared in other cultures, but Confucianism places special emphasis on reverence for el­derly parents (aka filial piety). Moreover, reverence for elders is not simply a m ­ atter of providing material comfort. As Confucius put it, “It is the attitude that ­matters. If young ­people merely offer their ser­vices when t­ here is work to do, or let their elders drink and eat when ­there is wine and food, how could this be [sufficient for] filial piety?”79 We need to serve our parents, and other ­family members, with love.80 Morality within the immediate, blood-­related ­family, however, is not sufficient. In Confucianism, t­ here is a firm distinction between

60  c h a p t e r 1

f­ amily insiders and nonfamily outsiders, but the concept of ­family is relatively flexible, and family-­like concern and care are supposed to be extended to ­others. Mencius explic­itly asks us to “Treat the aged of our own f­ amily in a manner befitting their venerable age and extend this treatment to aged of other families; treat our own young in a manner befitting tender age and extend this treatment to the young of other families.”81 One mechanism for extending such care is to treat ­house­keepers, or longtime domestic servants, almost like ­family members. In China’s past, family-­like labels and norms ­were often extended to ­house­keepers and longtime h­ ouse­keepers w ­ ere sometimes treated with care and re­spect that went beyond treatment accorded to ­family members.82 Of course, ­house­keepers in China’s past ­were not ­really ­family members. Even in the best cases when they w ­ ere treated with family-­like love and care and their full personalities could develop and shine, ­house­keepers ­were still bound to the home as domestic servants and ­were not included in the ­family line. ­There was always a sharp differentiation of roles between ­family members and h­ ouse­keepers. ­Today, fortunately, the practice of bound h­ ouse­keepers has been relegated to China’s “feudal” past. But the system of extending family-­like love and care still shapes interactions between employers and ­house­keepers in con­ temporary China. Day care and nursing-­home systems are relatively undeveloped, even in wealthy Chinese cities. ­People worry that strangers entrusted with caring duties w ­ on’t show the right “attitude,” hence the reluctance to commit one’s ­children and el­derly parents to state (or private) institutions. It’s much better for ­family members to provide care for other needy ­family members, and if that’s not pos­si­ble, to hire somebody to provide personal care in the home. So middle-­class and upper-­class families often hire ­house­keepers to help with caring duties. In

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  61

mainland Chinese cities, they typically hire mi­grant workers from the impoverished countryside, and in Hong Kong, they hire mi­grant workers from the Philippines, Indonesia, and other relatively poor Southeast Asian countries. In the best cases, the ­house­keepers are treated like ­family members: It is not uncommon for c­ hildren of employers in modern China to refer to ­house­keepers as “auntie” (阿姨). One recent study provides a good example of family-­like treatment by a Chinese employer in Hong Kong. A Filipina ­house­keeper (aka domestic helper or domestic worker) valued her employer’s parents b­ ecause she was treated as the d­ aughter they never had. The ties between the employee and the employer’s f­ amily ­were based on mutual concern and caring, not simply fairness and re­spect (which was more characteristic of the “good” Western employers): They watched TV together, engaged in mutual teasing, and the employer showed sincere concern for the ­house­keeper’s ­family in the Philippines. Interviews with other h­ ouse­keepers in Hong Kong revealed similar reactions. One ­house­keeper praised her former boss in Singapore for her use of affectionate family-­like appellations and for including her in weekend ­family outings. Another ­house­keeper was made the godmother of her employer’s child, and they would go to church together. Such cases of family-­like treatment are rare in Hong Kong, but they may be more common in mainland China where t­ here are fewer linguistic and cultural barriers between employer and ­house­keeper. ­Here too, however, it’s worth keeping in mind that ­house­keepers are not ­really (or wholly) treated like ­family members. The hope is that treating h­ ouse­keepers like f­ amily members increases the likelihood that they ­will care for the employer’s ­family members with love and care, but even in the best of cases the roles of h­ ouse­keeper and ­family member are kept distinct.83

62  c h a p t e r 1

We w ­ ill simply assume that blatantly unfair treatment of housekeepers—­paying unreasonably low salaries, forcing them to work unreasonably long hours, verbal abuse, not to mention physical vio­lence and sexual exploitation—is bad and should be curbed to the extent pos­si­ble.84 It is difficult for the state to monitor such abuses ­because interaction with ­house­keepers takes place in the “privacy” of the home, but it can enforce strict punishments for publicly exposed abuses. Other forces for decency in society, including ­family members of immoral employers, can and should help to rein in such abuses. But what if the ­house­keepers are paid well above the market rates, given substantial time o­ ff, and treated with both re­spect and family-­like care, and both employers and h­ ouse­keepers are satisfied with such arrangements? ­There is still a clear hierarchy, but perhaps it’s justified. Before we go that route, we need to recognize that the life of a ­house­keeper should never be viewed as more than a necessary evil, similar to Aristotle’s slaves and Marx’s proletarians in technologically undeveloped society. House­keepers do it for the money, and overseas domestic workers and mi­grant workers from the countryside are deprived of interaction with their own ­family members while they live and work for their employers, so the system is a double evil. Ideally, no one would be forced to travel long distances and be deprived of key ­family relations simply to make a decent living. In the long term, assuming an optimistic scenario, economic necessity ­will no longer influence what ­people do. We ­will have overcome the prob­ lem of global poverty, robots w ­ ill help to care for needy ­family members, and nobody ­will need to take jobs as ­house­keepers. But in the meantime, we need to consider under what conditions the intimate hierarchy between employer and housekeeper—­ the least intimate of intimate hierarchies, but intimate nonetheless ­because interactions take place inside the home and (in the

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  63

best cases) can take the form of family-­like relations—­can be justified from a moral point of view in a nonideal world. At minimum, the employer needs to refrain from all forms of mistreatment and humiliation. But beyond that, the employer has a positive obligation to treat the ­house­keeper with re­spect and, especially in cultures that value extending family-­like treatment, like a ­family member.85 And beyond that, it depends on the nature of the ­house­keeper. H ­ ere we can (re)turn to Aristotle’s (other­wise horrible) argument about “natu­ral” slaves for inspiration. Consider two kinds of ­house­keepers: House­keeper 1 is a Beijing-­born resident. She is not poor but works as a ­house­keeper to earn some extra funds for her ­family. She is neither intelligent nor ambitious, lacks any sense of humor, does her job in a rigid but effective way, and her interactions with her employer and f­ amily members are kept to a bare minimum. House­keeper 2 is from the Chinese countryside. She is from a poor ­family and could not even finish primary school due to economic constraints. But she goes to Beijing of her own accord with an impressive desire to improve herself. She learns how to drive and works as a “black car” (illegal taxi) driver ­because only ­those with a Beijing ­house­hold registration (hukou) can work as taxi d­ rivers. She is hired by a Beijing-­ based employer without any formal contract. The employer hires her as a driver and a cook (she is an excellent and highly imaginative cook) and all-­around helper. The employer’s son is very fond of the h­ ouse­keeper, and vice versa. They eat meals as a ­family, discuss intimate ­family affairs, and the ­house­keeper’s moral insights help to smooth out f­ amily conflicts. The employer is impressed by the intelligence and drive of her h­ ouse­keeper and gives her books on the Chinese

64  c h a p t e r 1

classics, which the h­ ouse­keeper devours in her spare time. Meanwhile, the ­house­keeper saves enough to buy two apartments in the Beijing suburbs, which she rents out to other mi­grant workers for a healthy profit. ­After eleven years, the ­house­keeper leaves the employer’s f­ amily, taking up a new job as a teacher in an informal school that teaches the classics to young ­children. In the case of House­keeper 1, the employer has no obligation other than to refrain from committing moral harms, treating the ­house­keeper with re­spect, and making some effort, however minimal, at family-­like treatment. But House­keeper 2 is dif­fer­ ent. Clearly she would have done well had she been given decent opportunities to flourish, and she ended up as a ­house­keeper only ­because she had the bad luck of being born in a very poor ­family. So the employer, even though it’s against her long-­term interests (­because it’s difficult to find an honest, loving, and intelligent ­house­keeper), has an obligation, if she has any sense of decency, to plan for a role change: to provide the resources and opportunities for House­keeper 2 to assume another role that provides more re­spect and income and allows for full flourishing of her natu­ral talent. The employer, to go back to Aristotle’s example, must plan for the freedom of the ­house­keeper. But what about a role exchange between employer and h­ ouse­keeper, with each taking the role of the other? That may be asking too much of the employer b­ ecause it’s hard to imagine any employer of h­ ouse­keepers being willingly downgraded to the role of ­house­keeper, except in extraordinary times such as revolution or economic collapse.86 But the employer needs to be prepared for the possibility of complete role reversals in the next generation. If the employer gives birth to a dull-­w itted child or a child with talents not valued by the marketplace, and

J u s t H i e r a r c h y b e t w e e n I n t i m a t e s  65

House­keeper 2 gives birth to an intelligent and energetic child with eco­nom­ically valuable skills, then it is pos­si­ble, and fair, that House­keeper 2’s child could end up as the employer of her employer’s child.87 Role reversals between employers and ­house­keepers take longer than reversals in other intimate hierarchies, but if they do occur we can conclude that the intimate hierarchical relation between employer and ­house­keeper is morally justified. A similar dynamic of shifting roles helped to justify direct democracy in ancient Athens, with rulers and ruled taking turns. ­Today, most citizens in liberal democracies do not take an active part in ruling, but the voting system provides a form of equality, giving citizens the right to change their rulers ­every few years in ­free and fair elections. How might it be pos­si­ble to justify hierarchical relations in large modern socie­ties without a voting mechanism? ­Here we need to consider a dif­fer­ent dynamic.

2 Just Hierarchy between Citizens on t h e i m p or ta nc e of s e r­v ic e If we gather as a group for a meal, then it is the concern of every­one, and we naturally call for every­one’s idea and decide by majority vote. . . . ​If we regard the aim of life as satisfying the desires of h­ uman beings, and regard the aim of politics as fulfilling every­one’s desires, we have to decide by majority vote. . . . ​However, if we regard life as a pro­cess of moral improvement, rather than the satisfaction of ordinary desires, it would be dif­fer­ent. . . . ​It means we all need to pay re­spect to our teachers and to continuously ask for advice from ­people, and we would naturally choose the leadership of a minority [with above average ability and virtue], and not decide by majority vote. —­l i a ng shu m i ng, ru r a l r econst ruct ion t h eory, 1937

r elations between intimates are characterized by strong emotions of love and care based on prolonged experience with face-­to-­face interaction. In the previous chapter, we argued that hierarchies between intimates can be justified if they are 66

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  67

characterized by changing roles that break down ossified power relations. In small po­liti­cal communities composed of citizens who trust and know one another, the justification for hierarchies may be similar. In ancient Athens, for example, a few thousand (male) citizens took turns as ruled and rulers via such mechanisms as regular rotation and sortition meant to equalize chances of holding office in the executive branch that set the agenda for the assembly, deliberation and voting on policies in frequent meetings in the assembly, popu­lar law courts with juries selected by lot, as well as local governments that modeled their organ­ ization and deliberations upon that of the city and created strong bonds of solidarity among citizens.1 While the ­whole system was founded on morally unjustified hierarchies (slaves and ­women did most of the socially necessary ­labor), the hierarchical relation between citizens was morally justified b­ ecause e­ very citizen had an opportunity to rule at some point in his life and no one could dominate ­others for extended periods of time. In large-­scale po­liti­cal communities composed mainly of strangers, however, most ­people ­will not have the opportunity to rule over ­others. In China, for example, it is simply not feasible for 1.4 billion citizens to take turns ruling and being ruled. Any attempt to implement an Athenian-­type democracy would be a r­ ecipe for chaos in huge po­liti­cal communities of this sort. So po­liti­cal hierarchies composed of rulers who exercise power over large groups of strangers for extended periods need a dif­ fer­ent justification. The most influential alternative for large-­scale po­liti­cal communities has been indirect representative democracy in the form of voting in f­ ree and fair periodic elections. Although most citizens w ­ on’t have a chance to rule in large po­liti­ cal communities, at least they have the equal opportunity to select, and deselect, their rulers ­every few years. Social science studies that draw pessimistic conclusions about the rationality

68  c h a p t e r 2

of voters,2 as well as the election of demagogic leaders with authoritarian inclinations in the mold of United States president Donald Trump, Turkish president Recep Tayyip Erdogan, and Hungarian prime minister Victor Orban, have cast doubt on the moral legitimacy and long-­term viability of “actually ­existing” electoral democracy. Po­liti­cal theorists such as Yves Sintomer and Wang Shaoguang have written persuasive works arguing that modernized interpretations of sortition are more likely (compared to elections) to realize demo­cratic values in both small-­ and large-­scale po­liti­cal communities.3 But let us leave aside ­these debates. We would like to focus the discussion on the possibility of justifying hierarchies between rulers and citizens in large-­scale po­liti­cal communities (such as China) that do not use periodic elections to select rulers.

1. Justifying Hierarchical Po­liti­cal Rule in the Chinese Context The most obvious justification for hierarchy in large communities is efficiency. According to historian Niall Ferguson, “The crucial incentive that favored hierarchical order was that it made the exercise of power more efficient: centralized control in the hands of the ‘big man’ eliminated or at least reduced time-­ consuming arguments about what to do, which might at any time escalate into internecine conflict.”4 Moreover, the larger the po­liti­cal community, the more efficiency considerations f­ avor centralized, hierarchical rule in the hands of the few: “The core lesson of classical po­liti­cal theory was that power should be hierarchically structured and that power naturally became more concentrated in fewer hands the larger a po­liti­cal unit became.”5 Chinese po­liti­cal history bears out this insight. Pre-­imperial China was characterized by fierce military competition between

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  69

small warring states. A ­ fter the late fifth c­ entury BCE, however, “a synergism of the necessities of war, the power of the state, and Legalist ideology [a philosophy that justifies heavy-­handed state power and harsh punishments to secure social order] held sway: increasingly the power of ferocious warfare favored ­those states that ­were more instrumental in organ­ization and action; the warfare of ordinances imposed by the Legalists enhanced state capacity to harness aristocratic power and exact resources from the population; and the states that w ­ ere more able to act instrumentally by more thoroughly implementing Legalist regulations ­were likely to triumph in the fierce military competition.”6 The Qin state proved to be most efficient at centralizing power and promoting a ruthlessly efficient military meritocracy (soldiers ­were promoted based on the number of decapitated heads of ­enemy soldiers),7 and the Qin successfully unified China ­under the one-­man rule of Qin Shi Huang, the self-­proclaimed First Emperor of Qin. But we have to ask: What’s the point of pursing efficiency in politics? Private companies prioritize efficiency ­because it’s necessary to maintain an edge vis-­à-­vis other profit-­seeking competitors, but states need to have a moral mission and efficiency can be justified only if it helps with that mission. The core justification for the state—­whether the state is governed by one ruler, a few rulers, or many rulers, as Aristotle famously argued in Politics III.7—is that policies must benefit the ruled (rather than the rulers). Hence, the pursuit of efficiency cannot be morally justified if it conflicts with that aim. The Qin empire lasted for only fifteen years—­the shortest-­lived major dynasty in Chinese history—­precisely b­ ecause it lost sight of the state’s moral mission. In line with Legalist thinking, Emperor Qin Shi Huang aimed to increase state power and employed ruthlessly efficient means for that end. He developed the world’s first sophisticated

70  c h a p t e r 2

bureaucracy, unified the Chinese script, and built an advanced transportation and communication system, but still went down in history as cruel dictator. The next major dynasty—­the Han—­found the normative solution that lasted for nearly two thousand years. The Han dynasty was still willing to use ruthless officials: The Book of Han (汉书) even had a special chapter titled “Biography of Cruel Officials.”8 But the Han ­adopted the po­liti­cal thought of Confucianism—­a philosophy that justifies using compassionate politics to benefit the ­people—as the governing ideology. Qin Shi Huang was famous for burning books and burying Confucian scholars. By contrast, Emperor Wu Di ­adopted Dong Zhongshu’s (179–104 BCE) interpretation of Confucian thought to educate the ­people and train officials with a unified Confucian ideology. Emperor Wu Di did not forsake the use of Legalist-­style severe laws and punishments—­five out of fourteen ministers during his fifty-­year reign ­were executed—­but he used Confucian thought to provide legitimacy for his rule, setting the dynamic for subsequent imperial po­liti­cal history. As Zhao Dingxin explains, In the Confucian-­Legalist state, the emperors accepted Confucianism as the ruling ideology and subjected themselves to the control of a Confucian bureaucracy, while Confucian scholars both in and out of the bureaucracy supported the regime and supplied meritocratically selected officials who administered the country using an amalgam of Confucian ethics and Legalist regulations and techniques. This symbiotic relationship between the ruling ­house and Confucian scholars gave birth to what is by premodern standards a power­ful po­liti­cal system—­a system so resilient and adaptive that it survived numerous challenges and persisted up ­until the Republican Revolution in 1911.9

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  71

The Legalist legacy is less evident ­because Legalism largely dis­appeared from official discourse for nearly two thousand years—­there w ­ ere no card-­carrying Legalists from the Han dynasty u­ ntil Mao’s invocation of Legalism in the Cultural Revolution. But Legalist ideas w ­ ere employed to improve the state’s capacity and ensure administrative efficiency. What­ever the official rhe­toric, the po­liti­cal system often relied on a Legalist standard for the se­lection of competent public officials, namely, the se­lection of officials with the ability to carry out strong and effective execution and the willingness to use brute power to solve prob­lems for the emperor. But the Legalists w ­ ere not overly concerned with the question of ­whether the aim itself was just or moral. So Confucianism set the aim of politics—to persuade the emperor to “Rule for All” (天下为公). Confucians favored the se­lection and promotion of public officials who could grasp the moral Way (道), implement benevolent policies that benefit the ­people, and protect civilians from cruel policies. The Chinese term for po­liti­cal meritocracy—­the se­lection and promotion of public officials with superior (Confucian-­style) virtue and (Legalist-­style) ability (贤能政治)—­well captures the ideal of the public official with an ability to grasp practical issues with the aim of efficiently implementing the princi­ple of “Rule for All.” In real­ity, however, Legalism and Confucianism often pulled in dif­fer­ent directions. From a Legalist perspective, Confucians often selected exemplary men who lacked the ability to deal with practical politics and efficient administration. From a Confucian perspective, Legalists often selected capable villains with no desire for justice or morality. Legalists deferred to the emperor’s wishes as the final court of appeal, whereas Confucians relied on the moral Way to evaluate the status quo, with a responsibility to admonish the emperor who implemented immoral policies. Legalists cynically dismissed the possibility of morality and

72  c h a p t e r 2

criticized Confucians as hypocrites who sowed po­liti­cal chaos, whereas Confucians doubted that a po­liti­cal system could survive for long without a moral foundation. This kind of dynamic between Confucianism and Legalism, as we w ­ ill see, continues to influence Chinese politics ­today. What­ever its internal tensions, what we can term the “Legalist Confucian” ideal of po­liti­cal meritocracy not only informed Chinese politics for over two thousand years,10 more surprisingly, it has also inspired po­liti­cal reform in China over the past four de­cades or so. A typical trope in the Western media is that ­there has been substantial economic reform in China, but no po­liti­cal reform. But that’s b­ ecause electoral democracy at the top is viewed as the only standard for what counts as po­liti­cal reform. If we set aside this dogma, it’s obvious that the Chinese po­liti­cal system has under­gone substantial po­liti­cal reform over the past few de­cades, and the main difference is that t­ here has been a serious effort to (re)establish po­liti­cal meritocracy.11 The country was primed for rule at the top by meritocratically selected officials following a disastrous experience with radical pop­u­lism and arbitrary dictatorship in the Cultural Revolution, and China’s leaders could reestablish ele­ments of its meritocratic tradition, such as the se­lection of leaders based on examination and promotion based on per­for­mance evaluations at lower levels of government—­almost the same system, in form (but not content) that ­shaped the po­liti­cal system in much of Chinese imperial history—­w ithout much controversy. And since then, po­liti­cal meritocracy has inspired po­liti­cal reform at higher levels of government, with more emphasis on education, examinations, and po­liti­cal experience at lower levels of government. T ­ here remains a large gap between the ideal and the practice, but the under­lying motivation for po­liti­cal reform is still the ideal of po­liti­cal meritocracy.

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  73

From a normative perspective, the ideal of po­liti­cal meritocracy is most compelling at higher levels of government in large-­ scale po­liti­cal communities. The reason is that it is much more difficult to rule and manage huge and incredibly diverse countries such as China, and it is not helpful to compare China with small, relatively homogenous countries endowed with plentiful natu­ral resources.12 Moreover, at higher levels of government of large countries, prob­lems are complex and often impact many sectors of society, the rest of the world, and ­future generations. In large countries, po­liti­cal success is more likely with leaders who have po­liti­cal experience at lower levels of government and a good rec­ord of per­for­mance. Electoral democracy may be appropriate for small countries or at lower levels of government of large countries; even if t­ hings go wrong—­say, too much pop­ u­lism and small-­minded navel-­gazing at the cost of neglecting long-­term planning and concern for f­ uture generations and the rest of the world—­it’s not the end of the world. But it may well be the end of the world if ­things go drastically wrong at the top of big and power­ful countries. The policies of leaders at the top of huge po­liti­cal communities shape the lives of hundreds of millions ­people, including ­future generations and the rest of the world. Hence, the ideal of po­liti­cal meritocracy is more appropriate to assess the higher levels of po­liti­cal systems of large countries like China. Of course, the ideal of po­liti­cal meritocracy would be a nonstarter in po­liti­cal communities where ­there is widespread aversion to the idea that is impor­tant to select and promote public officials with superior ability and virtue. In such communities, the best (or least bad) option might be to seek to improve the quality of decision making of elected politicians, what­ever the theoretical and practical challenges. In China, however, po­liti­ cal meritocracy has been consistently supported by po­liti­cal

74  c h a p t e r 2

reformers since the turn of the twentieth c­ entury. As Liang Shuming put it, “if we regard life as a pro­cess of moral improvement, rather than the satisfaction of ordinary desires . . . ​we would naturally choose the leadership of a minority [with above-­average ability and virtue] and not decide by majority vote.”13 More recently, survey results consistently show widespread support for the ideal of po­liti­c al meritocracy (aka “guardianship discourse”).14 The ideal is widely shared, much more so than the ideal of selecting leaders by means of elections. And the ideal of po­liti­cal meritocracy is also widely used to evaluate the po­liti­cal system. Corruption became such a big issue in the popu­lar mind at least partly ­because of the expectation that meritocratically selected leaders are supposed to have superior virtue. But the ideal of po­liti­cal meritocracy may not be an appropriate standard for evaluating po­liti­cal pro­gress (and regress) in socie­ties where the ideal is not typically used by the ­people to evaluate their po­liti­cal leaders.15 In short, the ideal of po­liti­cal meritocracy is an appropriate standard for assessing po­liti­cal pro­gress and regress at higher levels of government in China ­because the ideal has been central to Chinese po­liti­cal culture, it has inspired po­liti­cal reform over the past few de­cades, it is appropriate for large-­scale po­liti­cal communities, and it is endorsed by the vast majority of the ­people. ­These reasons may be par­tic­u­lar to the Chinese context, but ­there are also more general reasons to support the ideal of po­liti­cal meritocracy in the modern world. For one ­thing, po­ liti­cal meritocracy, with its emphasis on high-­quality leaders with wide and diverse po­liti­cal experience and a good track rec­ ord of responding and adapting to changing circumstances, may be particularly appropriate in a time of fast technological change and unpredictable global shocks. It may have made sense for the eigh­teenth-century found­ers of the United States

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  75

to enshrine a rigid constitutional system b­ ecause they could be quite sure that society would not change much over the course of the next few de­cades. It was more impor­tant to fix a good po­ liti­cal system than to allow for an ever-­evolving po­liti­cal system that aims to select and promote dif­fer­ent kinds of high-­quality leaders appropriate for dif­fer­ent times. But ­today, the only t­ hing we can predict about the next few de­cades is that ­there w ­ ill be radical changes to our current way of life,16 and the quality of leaders ­will ­matter even more than the quality of our po­liti­cal institutions (more precisely, our po­liti­cal institutions should be designed with the aim of selecting and promoting leaders with wide and diverse po­liti­cal experience and a good track rec­ ord of responding and adapting to changing circumstances). That said, the ideal of po­liti­cal meritocracy needs to be complemented by demo­cratic values and practices. In China, the ideal of democracy is widely deployed in both official and unofficial po­liti­cal discourse.17 So perhaps the most compelling argument for po­liti­cal meritocracy is that it is compatible with most demo­cratic values and practices, unlike, say, fascism or communist totalitarianism. Po­liti­cal meritocracy can and should be complemented by such demo­cratic practices as sortition, referenda and elections, consultation and deliberation, as well as the freedom of speech. That’s not to say all po­liti­cal goods go together. Po­liti­cal meritocracy is not compatible with competitive elections at the top b­ ecause electoral democracy for top leaders would undermine the advantages of a system that aims to select and promote leaders with experience, ability, and virtue: An elected leader without any po­liti­cal experience (such as Donald Trump) could rise to the top and make many beginner’s ­mistakes, an elected leader would have to spend valuable time raising funds and giving the same speech over and over again instead of thinking about policy, and an elected leader would be

76  c h a p t e r 2

more constrained by short-­term electoral considerations at the cost of long-­term planning for the good of the po­liti­cal community and the rest of the world.

Is Transparency Desirable? ­ ere may also be limits to transparency in a po­liti­cal system that Th values po­liti­cal meritocracy. In China, the power­ful organ­ization department—like the ­human resources department of the world’s largest communist organ­ization—­has the task of selecting and promoting public officials with above-­average ability and virtue. The organ­ization department has become somewhat more transparent of late: Its criteria for se­lection and promotion (and demotion) are more transparent, and it has put on mock interviews for visiting dignitaries from abroad, showing how candidates are selected in the interview pro­cess, though without naming real ­people. But we still do not have any clear idea of why some candidates get promoted over o­ thers who appear to be equally well qualified. Daniel put this question to leaders of the organ­ization department in Shanxi in June 2017. Shanxi was perhaps China’s most corrupt province, and Daniel was invited for a government-­led tour. The point of the tour was prob­ably to show that the organ­ization department had successfully replaced corrupt cadres with a new group of clean and hard-­working leaders. Daniel took this opportunity to ask a leader of the province’s organ­ization department why the se­ lection pro­cess ­can’t be more transparent. If their leaders are so ­great, surely it would help them make the case, both to fellow Chinese and to the outside world, to show that the leadership se­lection pro­cess is, in fact, rigorous and meritocratic. The organ­ization department leader asked how professors select candidates in academia. Daniel replied that the relevant

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  77

department establishes a committee that aims to select the best candidates, and committee members deliberate among themselves. The leader asked if the deliberations are open. Daniel replied of course not: Open deliberations would set constraints on what’s said, nor would it be fair to the candidates who are not selected. The leader smiled and said that “the same goes for us.” And he explained that the organ­ization department—­one of the most selective and prestigious departments in the Chinese po­liti­cal system—­ selects candidates partly according to their ability to keep secrets. So while po­liti­cal meritocracy may be compatible with most demo­cratic values and practices, we should just accept that lack of transparency is an inevitable cost of any organ­ization that aims to select the best candidates. It’s true not just of the Chinese Communist Party (CCP) and academia, but also of Goldman Sachs and the Catholic Church. That’s not to say we c­ an’t hope for more transparency in the Chinese po­liti­cal system—­the words and actions of emperors ­were tracked by official court historians for posterity in imperial China, and ­today we can imagine, say, video recordings of the deliberations of CCP leaders to be released fifty years from now. But full transparency is both unrealistic and unfair to the “losers” in the Chinese po­liti­cal system.18 To sum up. The hierarchical ideal of po­liti­cal meritocracy is an appropriate standard by which to assess the po­liti­cal real­ity at higher levels of government in China for a mixture of contextual and general reasons. The practical question is how to close the gap between the ideal and the real­ity. On the one hand, it means thinking about how best to maximize the advantages of the system. But any defense of po­liti­cal meritocracy also needs to address the question of how to minimize the disadvantages of the system. If the disadvantages are so g­ reat as to outweigh the advantages, then we can no longer be confident that po­liti­cal meritocracy should be used to assess the po­liti­cal real­ity in China

78  c h a p t e r 2

(or elsewhere). The next three sections ­will discuss prob­lems associated with any attempt to implement po­liti­cal meritocracy: (1) Rulers chosen on the basis of superior ability are likely to abuse their power; (2) po­liti­cal hierarchies may become frozen and undermine the ideal of equal opportunity; and (3) it is difficult to legitimize the system to ­those outside the power structure.19 Given that electoral democracy at the top is neither desirable nor feasible in China, we ­will ask if it is pos­si­ble to address ­these prob­lems without demo­cratic elections.

2. Is It Pos­si­ble to Limit Po­liti­cal Power without Competitive Elections? At the end of the day, what justifies po­liti­cal meritocracy (or any other po­liti­cal system) is that its leaders serve the p­ eople. This means, at minimum, that the leaders should re­spect the basic ­human rights of ­people: prohibitions against slavery, genocide, murder, torture, prolonged arbitrary detention, and systematic racial discrimination, as well as the idea that all citizens should be equal before the law in criminal cases. Leaders should also strive to secure the basic material needs of the ­people, such the right to food and decent health care. Such values are shared at the level of princi­ple in most modern socie­ties, including con­ temporary China. Of course, po­liti­cal leaders should do much more than re­spect basic h­ uman rights, but po­liti­cal priorities w ­ ill vary with the circumstances (e.g., t­ here is stronger need to place emphasis on poverty reduction at very low levels of GDP per capita and more need to combat environmental degradation once the country becomes wealthier). What ­won’t vary, however, is the need for leaders who put the interests of the po­liti­cal community above their own or f­ amily interests. Such requirements are most obvious in the case of the military: Soldiers are expected

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  79

to be prepared to die if necessary to protect the po­liti­cal community. But po­liti­cal leaders who systematically misuse power for their own or ­family benefit ­will also undermine the well-­being of the community. Leaders need not be totally self-­sacrificing or virtuous, but a degree of virtue is indispensable: Without any desire to serve the public, a po­liti­cal leader can put his or her ability to disastrous uses. So a po­liti­cal meritocracy must aim to select and promote leaders who are likely to use their power to serve the public. The prob­lem, of course, is that meritocratically selected leaders have few constraints on their power. If rulers are not chosen by the ­people, and if the ­people cannot change their rulers (other than by extreme means such as violent rebellion), what prevents the rulers from serving their own interests instead of the interests of the community? We ­don’t have to be full-­blown Legalists to worry about bad apples in politics. And what­ever we think of the arguments against electoral democracy, it is a good means of checking the power of rulers ­because they can be changed at election time. In imperial China, informal constitutional norms such as censors, court historians, and Confucian educators served to limit the power of the emperor. But emperors could, and sometimes did, ignore informal constitutional norms if it was in their interests to do so.20 Since the period of reform in the late 1970s, the princi­ple of collective leadership, terms limits, and the introduction of mandatory retirement age has served to limit abuses of power. The constitutional amendment abolishing term limits for the presidency, however, has revived fears of the “bad emperor” prob­lem: how to get rid of a ruler who abuses power without constraints and/or remains in office a­ fter becoming physically and mentally disabled. To be fair, collective leadership can still help to restrain, or even overthrow, leaders who become physically or mentally disabled. As a prominent Chinese

80  c h a p t e r 2

po­liti­cal analyst put it, the Chinese president may be “first among equals,” but he is not “first above equals.”21 Even if collective leadership still serves to constrain the top leader, however, t­ here are serious risks of abuse of power. At this stage, t­ here are dif­fer­ent possibilities. The most pessimistic scenario is that the next few years ­will be characterized first and foremost by increased po­liti­ cal repression at home and morally unjustified military adventures abroad. More optimistic scenarios are based on the assumption that somewhat longer terms of rule may be necessary to take on the vested interests that block economic, po­liti­cal, and educational reforms (Franklin Delano Roosevelt, perhaps the most successful U.S. president of the twentieth c­ entury, also broke the pre­ce­ dent of two-­term limits for the presidency). It would be foolish to make predictions, but one reason for optimism is that the anticorruption drive has succeeded beyond initial expectations, and we can expect more such difficult reforms in the f­ uture. The prob­ lem, however, is that the Legalist means employed to clamp down on corruption—­instilling fear of harsh punishment—­has led to negative consequences that may be difficult to remedy. The main advantage of Legalism is that instilling fear of harsh punishment is an efficient way to swiftly carry out necessary social change. In times of war, few generals would get very far without, say, harsh punishments for deserters. In times of peace, Legalist means may also be necessary to carry out rapid social change. One clear example is the campaign against drunk driving in China. Ten years ago, it would have been almost rude not to serve fiery white liquor to guests in Chinese restaurants.22 Drunk ­drivers would head back home, with predictably disastrous consequences. Alarmed by data that showed at least 20 ­percent of serious road crashes ­were alcohol related, the Chinese government de­cided to crack down on drunk driving. Almost overnight, the authorities set up frequent random roadside sobriety checks

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  81

and rigorously enforced harsh penalties, including compulsory jail time for first offenders and an automatic six-­month driving ban. It worked: Attitudes have changed, and drinking and driving is almost universally frowned upon. Death rates caused by drunk ­drivers have plunged nationwide,23 and random checks, now few and far between, have become almost superfluous.24

The Prob­lem of Corruption  Similar Legalist means w ­ ere employed to deal with corruption.25 Before President Xi Jinping assumed power in 2012, paying bribes and showering cadres with lavish meals and gifts w ­ ere seen as part of ordinary public life. It was difficult to get into good schools and hospitals without greasing the palm of authorities who ­were supposedly t­ here to safeguard the public good. Promotion to higher-­level posts often involved paying bribes to superiors. But corruption reached a tipping point, inflaming public attitudes to the extent of endangering the legitimacy of the po­liti­cal system.26 In response, the government launched what has turned out to be the longest and most systematic anticorruption campaign in Communist Party history. The vari­ous and largely in­effec­tive anticorruption agencies ­were centralized into one agency: the Central Commission for Discipline Inspection, led by Wang Qishan, and since 2018 the National Supervision Commission, led by Yang Xiaodu. As of 2018, more than one million officials have been netted for corruption, including a dozen high-­ranking military officers, several se­nior executives of state-­owned companies, and five national leaders.27 Cynical observers claim that the w ­ hole ­thing is a means of ­going ­after po­liti­cal enemies, but what distinguishes this anticorruption drive from previous ones is that it also creates many po­liti­cal enemies, which seems irrational from the point of view

82  c h a p t e r 2

of po­liti­cal self-­preservation.28 What­ever the motivation, the effect is clear: The anticorruption drive has worked. Anybody who has dealt with public officials has noticed the changes: Corrupt practices are now almost universally frowned upon, similar to drunk driving. The advantages are obvious: The profits of companies are up b­ ecause t­ here’s no longer any need to pay extras to corrupt public officials.29 Ordinary citizens perceive the system as less unfair ­because it’s now pos­si­ble to access public ser­v ices without paying bribes and gifts to bureaucrats. The prob­lem is that the anticorruption drive has worked almost too well. It’s not just that government officials think twice before engaging in corrupt practices. They think almost all the time about what can go wrong, to the point that decision making has become almost completely para­lyzed. Few officials are willing to take risks or consider innovative solutions necessary to solve new and unexpected prob­lems. The procedures for using public funds have become bafflingly complex, and it’s safer not to spend any money. It’s as though p­ eople stop driving cars in response to random checks against drunk ­drivers on ­every city block. It’s good that bad officials find it harder to misuse public funds, but it’s bad that good officials ­can’t use public funds necessary to serve the community. Th ­ ese costs are huge, and growing. China’s success over the past three de­cades is partly explained by the fact that government officials w ­ ere encouraged to experiment and innovate, thus helping to propel China’s reform.30 But ­today’s conservatism means innovative public officials ­won’t get promoted, and prob­lems ­won’t get fixed. In the long term, paralyzing talented and public-­spirited public officials can be deadly. The second prob­lem related to the anticorruption drive also stems from the fact it has been almost too effective.31 For each high-­level public official who has been brought down by the anticorruption drive, t­ here may be dozens of allies and

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  83

subordinates who lose prospects of mobility in an ultracompetitive, decades-­long race to the apex of po­liti­cal power. ­These real enemies make the leaders even more paranoid than usual and lead the government to ramp up censorship and curb civil and po­liti­cal rights even more aggressively. So it’s not just the po­liti­cal outcasts who feel estranged from the system, but also intellectuals and artists who object to curbs on what they do, as well as business­people who worry about po­liti­cal stability and flee abroad with their assets. With yet more social dissatisfaction among elites, leaders further clamp down on real and potential dissent. Knowing their enemies are waiting for the opportunity to pounce,32 the current leaders are even less likely to give up power (el­derly leaders may not worry so much about their own fate b­ ecause they w ­ ill soon “visit Karl Marx,” but they worry about c­ hildren and f­ amily members). So it’s a vicious circle of Legalist means and po­liti­cal repression. Ironically, the most efficient and effective drive to limit abuses of power in recent Chinese history (in the form of the anticorruption campaign) may also have led the leaders of the campaign to remove the most impor­tant constraint on their own power (in the form of term and age limits). What can be done? In retrospect, it may have been a m ­ istake to rely almost exclusively on Legalist means to combat corruption. Legalism can bring short-­term po­liti­cal success, but it can also lead to long-­term doom, similar to the fate of the Qin dynasty. Chinese history does point to other possibilities, including amnesty for corrupt officials. As the current anticorruption drive was getting u­ nder way, reformers argued that general amnesty should be granted to all corrupt officials, with serious policing of the bound­aries between private and public, and resources provided to allow them to start afresh.33 To deal with the prob­lem of mai guan (买官) (buying of government posts),

84  c h a p t e r 2

public posts could be distributed by lot once officials pass a certain level of qualification, as was done u­ nder Emperor Zhu Yijun.34 But it’s too late to start over. What can be done is to wind down the anticorruption drive.35 Wang Qishan himself said that the anticorruption drive ­will need to move from an initial deterrent stage to a point where the idea of acting corruptly would not even occur to officials as they went about their business. The next stage ­can’t rely first and foremost on fear of punishment. It must rely on mea­sures that reduce the incentive for corruption, including higher salaries for public officials and more clear separation of economic and po­liti­cal power. It also ­matters what officials do when nobody is looking: Moral education in the Confucian classics can help to change mindsets in the long term.36 It’s now time to put more trust in public officials and to empower them to do their work on behalf of the public. Any po­liti­cal system must balance the need to constrain government officials from ­doing bad and empowering them to do good, and the balance in China needs to swing back to the latter. ­There may be too few constraints on the power of top leaders, but ­there are too many constraints on the ­others.37

3. Po­liti­cal Meritocracy as the Prob­lem, Po­liti­cal Meritocracy as the Solution What­ever the constraints on po­liti­cal leaders, however, the meritocratic system w ­ on’t work well if t­ hose leaders tend to come from the same social background. According to the meritocratic ideal, a po­liti­cal system should aim to select and promote leaders with superior ability and virtue. In the Warring States period (475–221 BCE), the market for po­liti­cal talent was basically international. Scholars roamed from state to state hoping to persuade rulers of their superior qualities and to be chosen to serve

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  85

as ministers and advisors. In princi­ple, the po­liti­cal system should aim to choose the most worthy (and demote the unworthy), regardless of class background. Xunzi expressed this idea: “Promote the worthy and capable without regard to se­niority; dismiss the unfit and incapable without hesitation. . . . ​Although they may be descendants of kings and dukes or scholar-­officials and counselors, if they are incapable of devotedly observing the requirements of ritual princi­ples and justice, they should be relegated to the position of commoners. Although they may be descendants of commoners, if they have acquired learning, are upright in conduct, and can adhere to ritual princi­ples, they should be promoted to the post of prime minister, scholar-­ official, or counselor.”38 Around the same period in a dif­fer­ent part of the world, Plato (in The Republic) extended meritocratic princi­ples to two other groups. First, the very top ruler(s) should also be chosen according to merit (Xunzi did not extend meritocracy to the position of king). Second, w ­ omen should also have the opportunity to be rulers. In short, the meritocratic ideal is that every­one should have equal opportunity to serve as a po­ liti­cal official regardless of social background, and the po­liti­cal system should aim to choose the most able and virtuous among the contenders. In real­ity, however, po­liti­cal hierarchies tend to ossify, and the po­liti­cal se­lection pro­cess often misses out on talent from broad sectors of the population. Eventually the gap between the meritocratic ideal and the frozen real­ity becomes so large that it threatens the legitimacy of the w ­ hole system. In Chinese history, the debates about po­liti­cal meritocracy tend to reappear, with new iterations and interpretations, precisely when the old ­po­liti­cal hierarchies become ossified, and the governing body could not function well. In the minds of the critics, it is not meritocracy itself that leads to the freeze of po­liti­cal hierarchies.

86  c h a p t e r 2

Quite the reverse, po­liti­cal meritocracy, with new interpretations and practical innovations, is the only solution for improving the po­liti­cal impasse. Historically, the debates about the meritocratic ideal centered around two methods of recommendation and examination, with dif­fer­ent emphasis in dif­fer­ent times. The most impor­tant ways of selecting officials in Chinese history ­were the recommendation system, the nine-­rank system, and the imperial examination system. We w ­ ill discuss t­ hese vari­ous methods, in (roughly) chronological order, showing how each innovation occurred in response to the perceived ossification of po­liti­cal hierarchies.39

The Recommendation System The recommendation system made its appearance and was perfected in the Han dynasty (206 BCE–220 CE). At the beginning of the Han, many officials ­were appointed to participate in po­liti­ cal governance ­because of their meritorious military ser­vices, similar to the Legalist-­inspired Qin dynasty (221–207 BCE). As the po­liti­cal environment shifted from a turbulent state of war to a period of relative peace, however, the division of work between ministers and generals became clearer and ­there was a need for more ministers. Therefore, Emperor Han Wu Di (157–87 BCE) further developed the recommendation system, which originated in the period of his f­ ather, Han Wendi. Emperor Wu Di named it “以儒取士” (selecting scholars according to the criteria of Confucianism).40 Officials w ­ ere selected according to four criteria or “specialties” (四科): virtue and conduct, the study of Confucian classics, a clear understanding of laws and regulations, and a problem-­solving mindset.41 The rulers of the Han dynasty declared themselves dif­fer­ent from the rulers of the Qin dynasty ­because they aimed to

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  87

govern the country according to 孝 (filial piety, or reverence for el­derly ­family members). The most impor­tant of the four specialties was the first, virtue and conduct, and the most impor­tant part of moral conduct was filial piety.42 As the History of the L ­ ater Han Dynasty (后汉书) puts it, “Han is a virtue of fire, fire is born from wood, wood promotes fire, so the virtue of fire is filial piety, the image of which in The Book of Changes (易经) is Li (礼). . . . ​ Thus the system of Han Dynasty ordered the w ­ hole country to learn The Book of Filial Piety (孝经), and selected officials by recommendation according to filial piety and a clean rec­ord.”43 Such ideas, inspired by The Book of Changes, can be seen as metaphysical justifications for the po­liti­cal importance of filial piety in the Han dynasty. Filial piety gradually became the most impor­tant virtue at that time, in both a po­liti­cal and a metaphysical (or religious) sense. As The Book of Filial Piety put it: “孝，始于事亲，中于事君, 终于立身” [Filial piety, starts with serving parents, u ­ nfolds by serving the emperor, and completes itself by establishing oneself in society]. . . . ​“故当不义，则争之” [When the ones you serve are unjust, argue with them]. . . . ​“孝悌之至，通于神明, 光于四海，无所不通” [When filial piety grows utmost, it could access the spirits and illuminate the four seas—it reaches everywhere].44 Filial piety was deployed as a critical standard grounded in the moral order of the universe to evaluate the po­ liti­cal real­ity. It demanded serving parents and the emperor by means of upright conduct, in contrast to Legalist-­style blind obedience to the status quo. It was also filial to correct the faults of parents and the emperor. That said, the po­liti­cal importance of filial piety can be traced to the social real­ity at the time. In the Han dynasty, many emperors took over the Court at a very young age, and they needed their m ­ other to help them govern the country. As a consequence, filial piety became an extremely

88  c h a p t e r 2

impor­tant ele­ment to justify obedience to the m ­ others who acted as de facto rulers.45 Although filial piety was most impor­tant in the Han dynasty, the other three specialties also influenced the se­lection of scholar-­ officials. Whereas filial piety relied mainly on recommendations, the other three specialties relied more on examinations. The words 察举 refer to recommendation by examination. Local officials recommended a certain number of ­people to the central government ­every year. The Court gave recommended ­people suitable official positions, and students at all levels of school underwent some sort of assessment or interview to get po­liti­cal positions. The original motivation of the recommendation system was to select and promote virtuous and capable candidates for public ser­v ice, emphasizing the examination of their virtue and capacity rather than 门第(­family status). A ­ fter the m ­ iddle of the Eastern Han dynasty (25–220 CE), however, the more power­ful families controlled the main power of recommendation, and most of the recommended ­people came from noble families. The recommendation system lost its initial function of selecting and promoting competent and virtuous officials and degenerated into a way to protect the interests of certain families. Po­liti­cal hierarchies ossified, and rulers ruled for the interests of established elites.

The Nine-­Rank System To solve the prob­lem of ossification, the nine-­rank system appeared in the Wei, Jin, Southern, and Northern dynasties (220– 589 CE).46 The court appointed several local recruiters (中正) to recommend talented candidates, and the competent men ­were

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  89

ranked by three criteria: ­family status, moral conduct, and capacity. A ­ fter ranking they would be appointed to suitable positions. The original intention of setting up the nine-­rank system was to correct a series of prob­lems raised by the recommendation system. As The History of Liu Song Dynasty (宋书) put it, “The nine-­rank system aims to classify the competent men by merit, rather than f­ amily status.”47 At the time of implementation, several noble families dominated the recommendation system, and ­those from ­humble families hardly had a chance to be recommended. The most impor­ tant reform of the nine-­rank system was to empower local recruiters (中正), rather than power­ful families, with the “right” to recommend public officials. To a certain extent, men from ­humble families had more opportunities in the new system, compared to the recommendation system. Unfortunately, the new se­lection system gradually became frozen. In the intellectual history of ancient China, the nine-­rank system is often criticized as backward. The most famous criticism was put forward in The History of the Jin Dynasty (晋书): “In upper ranks, no one came from the h­ umble families; in lower ones, no one came from the power­ful families.”48 Power­ful families gradually took charge of the positions of local recruiters, thus entirely controlling the ranking system. Still, it’s worth noting the po­liti­cal context of education in the Wei and Jin dynasties. Due to incessant warfare, ­people ­were constantly on the run. Official and private schools decayed, and aristocratic families preserved their par­tic­u­lar ­family education by homeschooling or paternal teaching and influence. Hence, the educational situation of aristocratic families often fit their good reputations. But when the official and private schools ­were revived, the nine-­rank system declined and was replaced by the imperial examination system.

90  c h a p t e r 2

The Imperial Examination System The imperial examination system was founded in the Sui dynasty (581–618) and perfected in the Tang (618–907) and Song dynasties (960–1276). The court selected competent men by imperial examination regardless of f­ amily background and without any need for references or recommendations. As a consequence, the mono­poly of aristocratic families on the se­lection of the competent men was almost completely broken, reflecting the spirit of fairness. In the Song dynasty, the imperial examination system was substantially improved. But some Confucian thinkers criticized the scholars’ motivation for taking the examinations. If scholars lost their original commitment to rightness and truth, and instead took the exam as a profitable way to gain fame and material interests, t­ hose selected as officials would have no virtue at all. Thus they argued for reforming or even abolishing the imperial examination system. For example, Zhu Xi (1130–1200) wrote to the emperor arguing for establishing a separate discipline, called 德行 (virtue and moral conduct), and for abolishing the discipline of 词赋 (composition and rhe­ toric).49 The discipline of “virtue and moral conduct” would be dif­fer­ent from the other examined disciplines of 经, 子, 史, 时务 (Confucian classics, philosophical writings, history, and current affairs). This new discipline of “virtue and moral conduct” would use recommendations to select virtuous officials and was designed to complement the examined disciplines. Zhu Xi regarded recommendation as a necessary supplement of the imperial examination system and for reminding scholars of the significance of virtue and moral conduct. Zhu Xi’s teachers w ­ ere even more radical. Cheng Hao (1032–1085) and Cheng Yi (1033–1107), known as “The Two Chengs,” ­were completely against the system of the imperial

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  91

examinations.50 They argued that examinations could not test for virtue and ­were not po­liti­cally practical.51 On the one hand, ­there ­were too few officials selected by examination, leading to insufficient numbers for the governing body; on the other hand, ­people selected through examinations would only “博闻强记” (“have encyclopedic knowledge”) and would lack the practical ability to deal with politics. Hence they suggested replacing the imperial examination system entirely by recommendation. First, the elders of the counties and the students of the Imperial College would recommend some candidates. Then, the Court would inspect them, assess their abilities as erudite scholars, and appoint them as provisional officials in order to review their po­liti­cal capacities. In the end, the Court would rank scholars through a debate. A ­ fter all ­these procedures, the Court would officially appoint ­every selected scholar to a proper position. Their most innovative suggestion was that the Court could select scholars who prove their po­liti­cal competence through ­actual po­liti­cal ­trials. The suggestions of the Two Chengs ­were not ­adopted at the time, but they ­were implemented to a certain extent in ­later dynasties, and the princi­ple of per­for­mance evaluations at lower levels of government still informs po­liti­cal meritocracy in con­temporary China. In short, social critics argued against the recommendation system ­because it could not give equal opportunities to candidates without power­ful ­family backgrounds and against the examination system ­because it could not test for virtue or practical po­liti­ cal skills. Th ­ ese debates about the prob­lems of actually existing meritocracy took place precisely when po­liti­cal hierarchies became ossified, and ­there was a need for both new thinking about po­liti­cal meritocracy and new innovations designed to close the gap between the meritocratic ideal and real­ity. The prob­lem was po­liti­cal meritocracy, and so was the cure.52

92  c h a p t e r 2

In fact, con­temporary debates about po­liti­cal meritocracy appear against a similar background. For one t­ hing, the gap between rich and poor has exploded during the four de­cades of economic reform, with the consequence that candidates from wealthy and po­liti­cally power­ful families have unfair advantages in the meritocratic competition for po­liti­cal power. The fact that almost all top leaders are men further exposes the gap between the ideal and the real­ity of po­liti­cal meritocracy. And the corruption prob­lem threatens to undermine the moral legitimacy of the public officials who do make it through the system. Similar to imperial times, social critics and po­liti­cal reformers in China argue that the solution is not to abandon the ideal of po­liti­cal meritocracy, but rather to propose innovations designed to reduce the gap between the ideal of meritocracy and the real­ity.53 What’s dif­fer­ent this time, however, is that po­liti­cal meritocracy, even in ideal form, is not sufficient to legitimatize the ­whole po­liti­cal system. The value of democracy is widely advocated by both the Chinese government and its critics, and nobody argues for establishing a purely meritocratic po­liti­cal community ­today. It would be hard to persuade ­people that they should be totally excluded from po­liti­cal power. Plato himself recognized (in The Republic) the need to propagate a “Noble Lie” that the guardians deserve absolute power ­because they have gold in their souls, unlike every­body ­else. Regimes like North ­Korea can propagate such myths about the quasi-­divine status of their rulers ­because they are closed to the rest of the world, but no modern open society can get away with it. In short, it’s hard to imagine a modern government ­today that can be seen as legitimate in the eyes of the p­ eople without any form of democracy. So the question is how to inject an ele­ment of democracy into the po­liti­cal system. In a Chinese context, the special challenge is how to legitimize a hierarchical po­liti­cal

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  93

system informed by the princi­ple of po­liti­cal meritocracy to the majority of ­people who are formally excluded from po­liti­cal power. With 90 million members, the Chinese Communist Party is the world’s second-largest political party, but it’s still composed of a small fraction of the 1.4 billion ­people.

4. Justifying a Hierarchical Po­liti­cal System to Th ­ ose outside the System The key justification for po­liti­cal rule is that rulers serve the ­people. But who decides if the p­ eople are being served? In the Confucian tradition, contrary to popu­lar perception, the ­people have an impor­tant role. As Joseph Chan explains, “Confucians are not interested in authority merely as an institution justified externally by certain objective reasons such as the promotion of ­people’s well-­being. Authority is also a kind of relationship or bond between the ruler and the ruled (or in con­temporary terms, between ­those who govern and the governed). What makes the relationship truly authoritative is not just the ruler’s ability to protect and promote the ­people’s well-­being, but the willing ac­ cep­tance of his rule by the ­people.”54 Confucius himself asserted the importance of winning the hearts of the p­ eople (20.1), and gaining the willing compliance of the ­people is a recurring theme in the Confucian tradition. The Song dynasty thinker and politician Su Dongpo (1037–1101) spells out the flip side in his counsel to Emperor Shenzong: “He who is able to command the support of the millions becomes a king, while he who alienates their support becomes a solitary private individual (独夫). The basis of the ruler’s authority (人主) lies, therefore, entirely in the support of the p­ eople in their hearts. . . . ​And when an emperor loses the support of the p­ eople, it spells his ruin.”55 In short, the legitimacy of po­liti­cal rule is based both on the (objective)

94  c h a p t e r 2

idea that the ruler promotes the well-­being of the p­ eople and the (subjective) idea that the ­people willingly accept the ruler’s authority. Without the compliance of the p­ eople, the ruler’s rule, and the ­whole po­liti­cal system, is ­under threat of collapse. But how can we know that the ­people accept the ruler’s authority? The answer in modern-­day electoral democracies, of course, is that winning the ­people’s support by means of ­free and fair competitive elections is the way to mea­sure willing ac­cep­tance by the p­ eople. As argued above (section 1), however, competitive elections at the top would wreck the advantages of the meritocratic system. And what­ever we think of the normative arguments for and against elections, we can safely assume that the Chinese po­liti­cal system is not about the selection of its top rulers by means of ­free and fair competitive elections in the foreseeable ­future. So we are back to the question of how to secure the ­people’s endorsement of a hierarchical po­liti­cal system without demo­ cratic elections. In the case of China, the Chinese Communist Party has drawn on three sources of legitimacy: nationalism, per­ for­mance legitimacy, and po­liti­cal meritocracy.56 Although all three sources of legitimacy have been impor­tant at dif­fer­ent times to a certain extent, perhaps the strongest source of legitimacy in the reform era has been per­for­mance legitimacy in the form of economic growth. Over the past four de­cades, ­there has been a widespread consensus that the government should strive for high growth rates b­ ecause growth was seen as key to poverty reduction. Hence, government officials could be promoted based on economic per­for­m ance above all e­ lse without much controversy. ­Today, however, the prob­lems are much more diverse, some directly attributable to the lopsided emphasis on economic growth: rampant pollution, a huge gap between rich and poor,

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  95

precarious social welfare, and an explosion of government debt, not to mention massive corruption. In the f­ uture, the government w ­ ill lose support if it d­ oesn’t deal with ­those prob­lems, what­ever the rate of economic growth. ­Here ­things become more complicated for a po­liti­cal system that prides itself on meritocratic mechanisms for the se­lection and promotion of leaders. Should government officials be assessed according to their ability to deliver economic growth, to improve social welfare, to reduce corruption, to protect the environment, to reduce the gap between rich and poor, to reduce government debt, or to achieve some combination of t­ hese goals? It is impossible to resolve ­these issues in a noncontroversial way, and ­there are bound to be many winners and losers no ­matter the decision. Hence, the government needs more input from the ­people, not just to help decide on priorities, but also to take the heat off when large constituencies are unhappy with some policies. More generally, as China modernizes, ­there ­will be more demands for po­liti­cal participation by the ­people.57 In short, democracy is necessary to save po­liti­cal meritocracy in China. Legalist-­style repression can work in the short term, but t­ here must be more deliberation and participation in the long term. This is not to deny that the Chinese po­liti­cal system already incorporates strong ele­ments of democracy. Over nine hundred million farmers have participated in village-­level elections since 1988.58 Cadres are selected through a three-ticket system that includes demo­cratic evaluation and the election of the party’s Standing Committee in addition to meritocratic methods such as examinations. He Baogang argues that public deliberation is increasingly becoming an indispensable ele­ment in decision-­ making pro­cesses.59 ­There is already a degree of consultation in the po­liti­cal system—­for example, it took five years for the Property Law to pass the National P ­ eople’s Congress following

96  c h a p t e r 2

almost endless rounds of expert advice and public debate.60 But all ­these demo­cratic practices are not sufficient: More citizens ­will need to argue about what works in a wide range of domains and to have a greater say. Such openness is necessary not just to improve decision making but also to diffuse the sense of responsibility for ­those decisions. This ­w ill entail more freedom of speech and association and more mechanisms for consultation and deliberation within and outside the party, as well as transparent mechanisms to remove public officials who perform badly. All the innovations of modern demo­cratic socie­ties, such as sortition, open public hearings, deliberative polling, and referenda on key issues, could help to stabilize the po­liti­cal system. And more firmly establishing the rule of law is necessary to protect basic ­human rights.61 Even with more participation and deliberation at lower levels of government, however, it ­will be a challenge to legitimize the po­liti­cal system to t­ hose outside the power structure, especially to citizens who seek to make a positive difference in Chinese society without g­ oing through the official se­lection and promotion pro­cess for po­liti­cal leaders. Competitive elections at the top can give all citizens the hope (or illusion) that they can participate in po­liti­cal power, but this option is not open to a po­liti­cal system informed by the ideal of po­liti­cal meritocracy.62 So what can be done to expand po­liti­cal opportunities to the vast majority of citizens? The Confucian tradition, it should be recognized, may not have much to offer in this re­spect. As Zhang Yongle argues, Confucian education emphasized virtue “in order to maintain a sense among the p­ eople that the c­ areer of a politician requires special talents and training, and perhaps only suits a minority of p­ eople.”63 However, t­ here is also a need to affirm the idea that “the average person can also participate in public affairs at the grassroots level, and even realize outstanding

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  97

achievements and achieve recognition by the state.” Zhang argues that the Maoist tradition offers po­liti­cal insights. China’s revolution, inspired by Mao’s thought, allowed for the possibility that average p­ eople could achieve recognition by the state. For one t­ hing, ­there was a less intellectual view of what constitutes po­liti­cal merit: “­those who ­were selected as model workers often ­were able to use the opportunity to enter the po­liti­cal stage, which helped to forge the common belief that ordinary jobs can offer valuable contributions to society and even the possibility of being rewarded with a leadership position.”64 In Mao’s time, however, valuing workers was accompanied by a radical form of anti-­intellectualism. ­Today, the challenge is to value dif­fer­ent forms of po­liti­cal merit without radical critiques of forms that fall outside t­ hose valued by the state (and without vio­lence directed at ­people from “bad” class backgrounds). The greatest resource for maintaining legitimacy, Zhang argues, is the “mass line,” which stems from the revolutionary era: The mass line, the Party term for a policy aimed at cultivating contacts with the common ­people, emphasized the idea of coming from the masses and ­going among the masses. It represents opposition to the idea that a minority or elites should be able to pursue top-­down policies. Instead it argues that the understanding of truth is a pro­cess that is constantly being revised by collective practice and that close contact with the masses is necessary to reach a more realistic understanding of the country’s situation, which in turn is critical for formulating the correct party line and policies. . . . ​To put the concept of mass line into practice, it’s necessary to “find the masses.” This not only requires cadres to go out into the masses, but also necessitates a certain level of organ­ization on the part of grassroots society, in order to create connections

98  c h a p t e r 2

between the grassroots, policymakers, and the executive branch of government. With ­these connections, policymakers can hear the voice of ordinary ­people and improve the responsiveness of their policies, which helps policymakers.65 Zhang’s argument is thought provoking. The mass line put into practice is a way of securing widespread legitimacy for the po­ liti­cal system without a system of competitive elections. ­Today, it is encouraging that up-­and-­coming public officials in China typically need to spend extended periods in poor rural regions to help to sensitize cadres to the needs of the worst-­off members of the community.66 That said, ­there may be a need to further de­moc­ra­tize the mass line. Similar to the po­liti­cal practice in the Ming and Qing dynasties, sortition can be used as a method to distribute cadres to local communities, which can also minimize corruption and favoritism. ­There is also a need to increase opportunities for self-­organization at the grassroots level. Perhaps the deepest prob­lem with the Confucian tradition, however, is the assumption that the best form of life involves serving the po­liti­cal community. In socie­ties with a Confucian heritage, it seems likely that po­liti­cal leaders w ­ ill continue to have the highest social status and t­ hose without po­liti­cal power may not feel a sense of (equal) social worth. So ­there is a need to affirm the social value of “nonpo­liti­cal” ways of life.67 ­Here we can turn to the Daoist critique of po­liti­cal meritocracy in the pre-­Qin period for inspiration.68 Laozi, the originator of Daoist thought, bluntly put forward the idea of not valuing or employing the virtuous: “Not to value and employ men of superior ability is the way to keep the p­ eople from rivalry among themselves; not to prize articles which are difficult to procure is the way to keep them from becoming thieves; not to show them what is likely to excite their desires is the way to keep them minds from

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  99

disorder.”69 The basic idea is that any sort of competitive society—­including a society that encourages competition according to a conception of po­liti­cal merit—­will make p­ eople, especially the “losers,” envious and miserable, so it’s best to discourage any form of competition and desire for a better life. Hence, “the sage, in the exercise of government . . . ​constantly tries to keep p­ eople without knowledge and without desire, and where ­there are ­those who have knowledge, to keep them from presuming to act on it.”70 The ruler should limit politics driven by competitive feelings and ambition, which means not employing the wise and the virtuous. In the same vein, Zhuangzi discouraged use of the wise and virtuous. He shares Laozi’s view that “elevating the worthy” w ­ ill lead to a competitive and chaotic society: “if you raise the men of talent to office, you ­will create disorder; making the p­ eople strive with one another for promotion; if you employ men for their wisdom, the ­people ­will rob each other.” Zhuangzi goes further by casting doubt on the w ­ hole idea of distinguishing between t­ hose with more worth and less. Every­body has l­ imited talent and biased perspectives: “no one has covered or extended the ­whole range of truth . . . ​­there is a limit to our life, but to knowledge ­there is no limit. With what is ­limited to pursue ­after what is unlimited is a perilous t­ hing; and when, knowing this, we still seek the increase of our knowledge, the peril cannot be averted.” H ­ umans can only dwell in specific places, occupy a ­specific situation, obtain ­limited knowledge, yet they often take their own view as the w ­ hole truth and argue endlessly from and for their l­imited perspective: “So it is that we have the contentions between the Confucians and the Mohists, the one side affirming what the other denies and vice versa.” And however brilliant a sage may be, he cannot avoid becoming entangled in social connections and po­liti­cal plots that lead to disaster: “Long Feng

100  c h a p t e r 2

was beheaded; Bi Gan had his heart torn out; Chang Hong was ripped open; and Zi Xu was reduced to pulp. Worthy as ­these four men ­were, they did not escape dreadful deaths.” So the solution is to abandon the ­whole idea of pursing wisdom: “in the age of perfect virtue, they attached no value to wisdom, nor employed men of ability.”71 The Lao-­Zhuang tradition may seem extreme in its anti-­ intellectualism. But it does remind us of our necessarily l­ imited perspectives and of the need to distrust ­those who arrogantly claim access to the ­whole truth and confidently assert their po­ liti­cal effectiveness. The solution is not to abandon the idea that some perspectives are better than o­ thers—­even Zhuangzi would agree that t­ hose who are aware of their limitations are better than ­those who ­aren’t. Nor is the solution to abandon the po­liti­ cal aim of selecting and promoting ­those with above average talent and virtue. What must be done is to employ officials with diverse talents and dif­fer­ent perspectives to help correct for the necessary limitations of any one person’s perspective. Cao Feng shows how the Huang-­Lao tradition drew on Daoist insights for po­liti­cal purposes: “Since it was a po­liti­cal ideology, Huang-­ Lao thought had to use p­ eople of talent and virtue to be carried out, and therefore could not reject the talented and virtuous as the Lao-­Zhuang tradition did, let alone consider them to be initiators of turmoil. On the contrary, why sages w ­ ere needed, what kind of sages w ­ ere needed, and how to make use of them ­were impor­tant ele­ments of Huang-­Lao po­liti­cal thought.”72 The Huang-Lao tradition emphasized that the monarch needs to recognize he cannot do every­thing on his own and hence needs to employ public officials with superior talents. Even the wisest of the sages has ­limited knowledge and perspectives and needs supplementary assistance (and criticism): “since the ruler’s wisdom and talents are not sufficient to spread his

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  101

splendor across lands and seas, he is surrounded by high ministers who assist him.” To allow for public officials to flourish and make contributions, the monarch must practice inaction: “if the ruler is able to renounce his own wisdom, talent, and accomplishments, he w ­ ill be able to bring fully into play his ­people’s wisdom, talent, and accomplishments.” Given necessarily ­limited knowledge and perspectives, the monarch should strive to employ dif­fer­ent kinds of public officials with dif­fer­ent backgrounds and dif­fer­ent skills: diverse “types of men have opposite characters, however sages can still use them in a tolerant way. . . . ​If guarding just one corner means leaving out the rest of the world, and selecting one species means giving up all other beings, one is sure to achieve very ­little: the reach of one’s administration w ­ ill certainly be very narrow.”73 In short, the monarch should be aware of his limitations and make comprehensive use of public officials with diverse backgrounds and talents. In a po­liti­cal system led by a monarch, the Huang-­Lao school of thought might counsel against a cult of personality that portrays him as all-­wise and benevolent. Other ­things being equal, a system of collective leadership is best to ensure that diverse perspectives can inform the policy-­making pro­cess at the very top. In a large country such as China, collective leadership at the top would also need to be supported by an extensive bureaucracy at dif­fer­ent levels of government staffed with a wide range of public officials from diverse backgrounds with diverse talents. But even this kind of system would not fully alleviate Daoist worries about the downside of po­liti­cal meritocracy: In the modern world, even a well-­functioning po­liti­cal meritocracy that selects and promotes public officials with diverse talents and backgrounds would need to be supported by an ultracompetitive educational system that aims to identify and educate ­those with above-­average ability and talent, and the dominant competitive

102  c h a p t e r 2

ethos of that society ­will lead to endless striving for success that ­causes misery for the “losers” and hence sows the seeds of social disorder. ­These Daoist worries would be further exacerbated in a cap­i­tal­ist economic system that rewards entrepreneurs and companies who successfully invent new needs and desires for consumers who are never supposed to be satisfied with the status quo. So what can be done to soften the deleterious societal effects of po­liti­cal meritocracy in the modern age?74 Perhaps the best way is to emphasize that the role of the professional public official is not the only way to lead a meaningful life. This means attributing more social (and material) value for “non-­political” ways of life that contribute to the social good, such as the work of farmers, f­ amily caretakers, and manual workers. It also means allowing for mechanisms that cast doubt on the ­whole meritocratic system, but without ­really threatening the w ­ hole system. Perhaps the most fascinating social development in con­ temporary China has been the rapid spread of what we can term a “culture of cuteness”: a public affirmation of cute animals, robots, and emojis that inform everyday social interaction. The trend started in Japan in the 1970s75—­when Japan was largely ruled by a meritocratically selected bureaucracy selected from an ultracompetitive educational system. It was led by teenage girls and eventually spread to other sectors of society. Over the past de­cade or so, the culture of cuteness has spread to China almost like wildfire. The streets of Chinese cities are populated with ridiculously cute dogs and cats,76 and the use of cute emojis is the norm for communication on social media, even in official settings such as exchanges between university administrators.77 It’s worth asking why the culture of cuteness has planted social roots so quickly and so deeply in China. One explanation is

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  103

figure 2. Cute dog in Shanghai with counter-­revolutionary clothing. Photo­graph by Wang Pei.

that it’s helpful for meritocratic competition: Viewing cute images promotes careful be­hav­ior and narrows attentional focus, with potential benefits for learning and office work.78 But the deeper reason may be both disturbing and encouraging for defenders of po­liti­cal meritocracy. On the one hand, the culture of cuteness represents a kind of rebellion against the w ­ hole system: Instead of affirming the value of boring and hard-­working (largely male) bureaucrats who serve the public good, it affirms the value of playful and somewhat self-­indulgent ways of life. As Simon May puts it, the culture of cuteness articulates “a nascent w ­ ill to repudiate the ordering of ­human relations by power, or at least to question our assumptions about who has power and to what end. This is a ­will that Cute can vividly convey precisely ­because it usually involves a relationship to a vulnerable object or to an

104  c h a p t e r 2

object that flaunts, or flirts with, vulnerability. It is a ­will to liberation from the power paradigm that many, especially in the West and Japan, but perhaps ordinary Chinese ­people too, might be expected to affirm as an antidote to a ­century and more of unparalleled brutality.”79 On the other hand, the culture of cuteness reduces the desire to join the race to the top, which helps to placate the “losers” in the po­liti­cal meritocracy and hence stabilizes the meritocratic system.80 To summarize, if the task is to legitimize hierarchical po­liti­cal meritocracy in a modern society that values equal social worth and participation in politics by ordinary ­people, we can learn much not just from Confucians and liberal demo­crats, but also from Maoists and Daoists. More specifically, both Maoist and Daoist ideas can help to legitimize the system among t­ hose left out from the official power hierarchies in po­liti­cal meritocracies without the safety valve of electoral competition for higher-­level po­liti­cal posts. The Maoist mass line can help to provide ave­nues for grassroots participation in politics and make elites more responsive to the needs of the masses. And Daoist-­style skepticism about the desirability of the ­whole meritocratic system can help to legitimatize alternative ave­nues for socially valued ways of life such as the culture of cuteness that gives meaning to the lives of t­ hose left out of the po­liti­cal hierarchies. At the end of the day, however, we are left with several ironies. Confucians argue that the best life involves ser­vice to the po­liti­ cal community, but the only way to justify Confucian-­inspired po­liti­cal hierarchy ­today is to or­ga­nize a po­liti­cal system that ­doesn’t value serving the public as the highest form of life. Th ­ ere may also be trade-­offs between the dif­fer­ent ways of addressing the main challenges of po­liti­cal meritocracy. Strong, Legalist-­ style mea­sures to deal with corruption may be most effective in the short term, but they can further ossify po­liti­cal hierarchies

J u s t H i e r a r c h y b e t w e e n C i t i z e n s  105

(­because innovation is discouraged and officials become cautious and conservative) and lead to an even more closed society (­because leaders worry about backlash from their growing list of enemies). In the long term, a judicious mixture of Confucian-­ style soft power combined with demo­cratic openness, Maoist-­ style mass line, and Daoist-­style skepticism about the ­whole po­ liti­cal system w ­ ill help to reinvigorate po­liti­cal meritocracy in China. What w ­ on’t change, however, is the need to identify and select public officials with above-­average ability and a willingness to serve the po­liti­cal community over and above their own private and ­family interests. That ideal goes back over two millennia in China, and it w ­ ill continue to inform the po­liti­cal system in the foreseeable ­future. Even if China succeeds in closing the gap between the ideal and practice of po­liti­cal meritocracy, however, it ­doesn’t follow that Chinese rulers should try to export this po­liti­cal ideal abroad. For one ­thing, the ideal might be difficult or impossible to implement in a po­liti­cal culture that lacks a history of complex bureaucratic rule with argumentation about the ideal of po­liti­cal meritocracy. Nor do we mean to imply that po­liti­cal meritocracy is the only morally justified way of securing the hierarchical relation between rulers and the ruled. And when it comes to international relations, dif­fer­ent kinds of princi­ples should inform relations between states, as we ­will see in the next chapter.

3 Just Hierarchy between States on t h e n e e d for r e c i pr o c i t y A humane authority would keep in good order the obligations between small and large countries, between the strong and the weak, and would rigorously maintain them. The impor­tant points of ritual would be observed with the extreme of good form. —­x u n zi, 10.2 0

a mor ally justified hierarchy within a state, as we saw, involves a conception of ser­vice: The rulers are supposed to serve the ­people. The rulers n­ eedn’t be pure altruists, but the state’s policies should aim mainly to benefit the p­ eople rather than the rulers, and such policies are more likely if the rulers are at least partly motivated by the desire to serve the p­ eople. Morally justified hierarchies between states are dif­fer­ent. The rulers of states owe their first obligations to their own p­ eople, and they cannot be expected to systematically sacrifice the interests of their own ­people for the interests of p­ eople in other states. Hierarchical relations between states must be reciprocal: They must benefit 106

J u s t H i e r a r c h y b e t w e e n S t a t e s  107

p­ eople in both power­ful and weaker states. In other words, they must be “win-­win.” But ­there are two kinds of reciprocity. One kind—­let’s call it “weak reciprocity”—is the ideal that hierarchical relations between states should be mutually advantageous. Each state thinks from the perspective of its own state (more precisely, the rulers think of the interests of their own p­ eople), and they strike deals or make alliances if they are beneficial to (the p­ eople of) both states. But weak reciprocity is fragile. Once the situation changes and the deal is no longer advantageous to one of the states, that state can simply opt out of the deal, just as the Trump administration seems to have de­cided to renegotiate or scrap ­free trade accords (and even security alliances) on the grounds that ­those deals no longer benefit the United States (if they ever did). Weaker states are particularly vulnerable ­under the terms of weak reciprocity ­because they are subject to the whims of the stronger states that can decide to change the terms of the deal. Another kind of reciprocity—­let’s call it “strong reciprocity”—is the ideal that both states come to think of their alliances from the perspective of both states, no longer simply from the perspective of their own state. The rulers no longer think simply in terms of benefiting their own p­ eople, and they are willing to stick with deals or alliances even if (temporarily?) the deals may be more beneficial to the ­people of other states. Moreover, what counts as the interest of each state itself comes to be influenced, at least partly, by the interests (and culture and history) of the other state: Th ­ ere is mutual learning that affects how p­ eople think of their own interests and conceptions of the good life. A former e­ nemy state can come to be seen as a friendly state with shared interests and values. One example might be the relations between the United Kingdom and the United States. Strong reciprocity is more demanding (and

108  c h a p t e r 3

perhaps more rare) than weak reciprocity, but it is more stable and beneficial for the weaker states. Does the ideal of reciprocity, w ­ hether strong or weak, between hierarchical states still m ­ atter in the modern world? Not on the (juridical) face of it. We are supposed to live in an age of equal sovereign states. The Peace of Westphalia treaty in 1648 set in stone the ideal of equality between sovereign states who are supposed to re­spect each other’s sovereignty and refrain from interfering in each other’s domestic affairs. This ideal originated in Eu­rope and slowly spread to the rest of the world. In 1945, the United Nations generalized the one person one vote princi­ple to the level of states, with each state given equal repre­sen­ta­tion regardless of size or wealth. Much theorizing in (Western) international relations is based on this ideal of formal and juridical equality between sovereign states. In real­ity, however, states are neither equal nor sovereign. As David A. Lake puts it, “sovereignty is a bundle of rights or authorities that can be divided among dif­fer­ent levels of governance and dif­fer­ent rulers. . . . ​Treating sovereignty as divisible allows authority between states to vary along continua of lesser or greater hierarchy.”1 It takes only a moment’s reflection to realize that the global order consists of a hierarchy between dif­fer­ent states, with some states having more de facto power than o­ thers. Nobody ­really cares about the fact that Nicaragua ­didn’t sign on to the Paris Climate Accord, but President Trump’s decision to withdraw from this accord may be a global disaster ­because of the United States’ disproportionate power to set the global agenda. Even the United Nations expresses the fact of global hierarchy: The most impor­tant decisions are often made at the level of the Security Council, which distinguishes between permanent members, nonpermanent members of the Security Council, and ordinary member states. That’s why rising

J u s t H i e r a r c h y b e t w e e n S t a t e s  109

powers such as India and Brazil fight hard (thus far unsuccessfully) for recognition as permanent members on the Security Council. If theorists of international relations aim to develop theories that explain the be­hav­ior of states and (more ambitiously) predict outcomes in the international system, then theorizing should be more attentive to the real­ity of hierarchy between states. ­There may also be good normative reasons to justify hierarchies between states. If it’s just a ­matter of strong states bullying weaker ones to get what they want, normative theorists can just step aside. But strong states do good ­things for the global order as a ­whole. However much we worry about “rogue” leaders in strong states that sabotage global agreements, it would be much harder to forge agreements for dealing with global challenges such as climate change in an international system characterized by states with equal power to shape and withdraw from global accords.2 Hierarchical systems can also contribute to international peace: As Yan Xuetong puts it, “if we examine recent international history, we can see that in t­ hose areas that implemented hierarchical norms, international peace was better maintained than it was in areas that had norms for equality. During the Cold War, the equal status of the United States and the Soviet Union was such that they undertook many proxy wars in order to compete for hegemony, while their special status in NATO and the Warsaw Pact, respectively, enabled them to prevent the members of ­those alliances from engaging in military conflict with one another.”3 Moreover, hierarchical arrangements can actually benefit weaker states ­because this sense of dominance means that states have extra responsibilities. Security hierarchies, for example, reduce the level of defense expenditure in subordinate states.4 Unequal economic power can also benefit weaker states. Rather than insisting on equal reciprocity with weaker states, strong states can

110  c h a p t e r 3

gain their support by allowing differential international norms that work in their f­ avor: For example “in the cooperation of the 10+1—­the Association of Southeast Asian Nations (ASEAN) and China—­China is required to implement the norm of zero tariffs in agricultural trade before the ASEAN states do. This unequal norm enabled the economic cooperation of the 10+1 to develop more rapidly than that between Japan and ASEAN. Japan’s demand for equal tariffs with ASEAN slowed the pro­gress of economic cooperation with the ASEAN states, which lags far ­behind that of China and ASEAN.”5 With extra powers come extra responsibilities, and it’s not completely utopian to suggest that strong states do occasionally act on ­those responsibilities and should be held accountable if they fail to do so. At the very least, we need theories that can help us distinguish between good and bad forms of international hierarchies and help us think of how to promote the good forms and avoid the bad ones. Hence, as Lane puts it, “like a Gestalt shift picture . . . ​refocusing on hierarchy reveals an alternative real­ity that has always been with us if we would but choose to see it.”6 But we ­don’t need a Gestalt shift as much as a return to ancient ways of thinking. In both classical India and classical China, po­liti­cal thinkers developed rich and diverse theories of international politics that took hierarchy between states for granted. We can mine t­ hese ancient theories for con­temporary insights. Some po­liti­cal thinkers in ancient India and China defended the ideal of weak reciprocity between hierarchical states, and ­others argued for strong reciprocity. This chapter leads off with a discussion of (some) ancient Indian views of hierarchical global order, followed by a discussion of (some) ancient Chinese views of hierarchical global order. The final section ­w ill argue for an ideal of “one world, two hierarchical systems” that may be appropriate for ­future forms of global order.

J u s t H i e r a r c h y b e t w e e n S t a t e s  111

1. Hierarchical Ideals of Global Order in Ancient India In ancient India, the most systematic work in interstate relations is Kautilya’s Arthasastra: The En­glish translation runs over 800 pages, more than half of which are devoted to foreign policy and war.7 Kautilya prob­ably flourished in the first ­century CE, and tradition identified him as the shrewd minister who brought the king to power and established the Maurya dynasty. He makes Machiavelli look like a sentimental idealist; had his work been more influential in Eu­rope, we’d be using the term “Kautilyan” rather than “Machiavellian” to describe amoral realism in international politics. Writing in a time of small kingdoms ruled by monarchs, he assumed a state of warfare as the norm. The ruler should do his best to expand his territory, without moral or religious constraints. More than that, he should go out of his way to prey on ­people’s superstitious beliefs to further his own ends. Consider the following list of tactics for assassinating the ­enemy: During a pilgrimage for worshipping a divinity, t­ here are numerous places that (the ­enemy) ­will visit to pay homage according to his devotion. At ­those places, he should employ trickery on him. Upon him, as he enters a ­temple, he should make a false wall or a stone fall by releasing a mechanical device; or set off a shower of stones or weapons from an upper chamber; or let a door panel plunge; or release a door bar attached to a wall and secured at one end. Or, he should make the statue, banner, or weapons of the god fall upon him. Or, in places where he stands, sits, or walks, he should arrange for poison to be used against him by means of the cow dung that is smeared, the scented ­water that is sprinkled, or the flowers and powders that are offered. Or, he should waft over to him

112  c h a p t e r 3

lethal smoke concealed by perfume. Or, by releasing a pin, he should make him plunge into a well with spikes or a pitfall that is located beneath his bed or seat and whose top surface is held together by a mechanical device.8 Kautilya’s most impor­tant contribution to interstate po­liti­cal thinking is the theory of mandala, the circle of kingdoms. As Patrick Olivelle explains, “A king is surrounded in a circle by other states, and ­because they have common bound­aries with him, they are his natu­ral enemies. Around ­these ­enemy kingdoms is a second circle of kingdoms. ­Because they abut the territories of ­enemy kings of the first circle, they become his natu­ral allies: my ­enemy’s e­ nemy is my friend. ­Those forming the third outer circle would, by the same logic, be the enemies of his allies, and thus his own enemies—­and so on.”9 The theory of mandala assumes rough parity between states in the sense that all states can wage wars against one another, but, to repeat, ­there is no “modern” proviso about the need to re­spect the territorial integrity of states, hence no theorizing that assumes equality of states. Quite the opposite, the constant quest for expansion of territory means that the size, wealth, and power of states shift in accordance with the gains and losses of territory that result from a near-­constant state of warfare. But the princi­ple that the ­enemy of my ­enemy is my friend can also lead to mutually beneficial outcomes: To help justify the CCP’s alliance with the Kuomintang of China (KMT) in the strug­gle against Japa­nese imperialism, Mao Zedong famously said, “[w]e should support what­ever our enemies oppose and oppose what­ever our enemies support.”10 This princi­ple also helped to justify rapprochement with the United States when both countries had the Soviet Union as a common ­enemy. Kautilya himself affirms that kings should strive for mutually

J u s t H i e r a r c h y b e t w e e n S t a t e s  113

beneficial peace pacts: “when the gain is equal, one should conclude a peace pact.”11 Even weaker kings can initiate peace pacts with stronger powers: “When a weaker king is overwhelmed by a stronger king with a superb army, he should quickly submit with a peace pact by offering his trea­sury, his army, himself, or his land.”12 But “weak reciprocity” in the form of a mutually beneficial peace pact is temporary at best. For one t­ hing, a peace pact cannot fundamentally challenge the ally/enemy configuration specified by the theory of the mandala. A peace pact formed by two natu­ral enemies with contiguous bound­aries is pos­si­ble but deeply unstable. And bound­aries can change, so that one’s natu­ral ally can become a natu­ral ­enemy if conquests result in two formerly friendly states with contiguous bound­ aries. More fundamentally, a ruler can—­a nd should—­ disregard the peace pact when it’s no longer in his interest to maintain it. Kautilya takes this point to its profoundly cynical extreme: “When he wishes to outwit an ­enemy who is corrupt, hasty, disrespectful, and lazy or who is ignorant, he should tell him, ‘We have entered into a peace pact’ without fixing the region, time, or task. Through the confidence generated by the peace pact, he should find his vulnerable points and attack him.”13 Hence, a Kautilyan-­style “peace pact” should be viewed as nothing more than a strategy designed “to outsmart, outmaneuver, and fi­nally overpower the king with whom he has concluded the pact.”14 Rulers should never lose sight that the ultimate aim is territorial conquest: Bigger is better, and too bad for the smaller states that end up on the losing side. Some states become so large that they are outside the mandala theory of ally and ­enemy: Large states led by power­ful kings can be neutral. And what happens when a power­f ul “neutral” king conquers much of the (known) world? At that point, is it pos­si­ble to move from “weak reciprocity” to a more stable “strong reciprocity” between

114  c h a p t e r 3

hierarchical states? Kaultilya does speak of the “righ­teous king” as a protector of social harmony,15 but it’s Ashoka who shows the way. Ashoka Maurya, commonly known as Ashoka and also as Ashoka the G ­ reat, was a successful conqueror who ended up ruling almost all of the Indian subcontinent from circa 268 BCE to 232 BCE. He relied on Kautilyan-­style methods to conquer territories, including the brutal war against the Kalingas (­today’s Orissa) with approximately 100,000 killed and 150,000 taken away as captives. At the height of his power, however, Ashoka had a conversion to Buddhism that radically changed his outlook from warmongering to peace-­loving (his experience is perhaps the most striking counterexample to the dictum that power corrupts). He expressed profound regret for the Kalinga war and propounded a commitment to dharma, which can be roughly translated as the moral way, in Rock Edicts throughout his empire. But this commitment to spreading morality was not restricted to his own empire: “In the imperial territories among the Greeks and Kambojas, Nabhakas and Nabhapanktis, Bhojas and Pitinikas, Andras and Parindas, everywhere ­people follow the Beloved of the Gods’ instruction in Dhamma [aka Dharma]. Even where the envoys of the Beloved of the Gods have not gone, ­people hear of his conduct according to Dhamma, his precepts and his instructions in Dhamma, and they follow Dhamma and w ­ ill continue to follow it. What is obtained by this is victory everywhere, and everywhere victory is pleasant. This plea­sure has been obtained through victory by Dhamma.”16 This vision seemed to express an ideal rather than a real­ity, but Ashoka sent his “envoys of the Beloved of the Gods” to faraway lands to spread the moral way. What is the content of the moral way? At minimum, it means a commitment to peace and nonviolence. The commitment to life, Buddhist-­style, extends to all forms of life, not just ­human

J u s t H i e r a r c h y b e t w e e n S t a t e s  115

beings: “I have enforced the law against killing certain animals and many o­ thers, but the greatest pro­gress of righ­teousness among men comes from the exhortation in ­favor of non-­injury to life and abstention from killing living beings.”17 It includes the provision of medical knowledge to foreign countries, prompting Patrick Olivelle to comment that “the intention of Ashoka in sending ­these missions is very clear: it was a missionary effort to spread his dharma philosophy, to get rulers of t­ hese countries to adopt Ashoka’s moral philosophy in their internal administration and external affairs. . . . ​This is very similar to the way Christian missionaries acted in countries they ­were attempting to evangelize.”18 But Ashoka’s moral way refers to the idea of building a common morality that draws on dif­fer­ent moralities while respecting difference. In that sense, his “envoys of the Beloved of the Gods” w ­ ere not like Christian missionaries who tried to spread what they considered to be the truth and (implicitly or explic­itly) downgraded other moral systems. Consider what Ashoka said about intercommunal relations. Ashoka’s aim was not just peaceful co-­existence among deeply divided communities: he also aimed for mutual learning which requires restrained and respectful speech on the part of the “Beloved of the gods”: ­ ere should not be honor of one’s own sect and condemnaTh tion of o­ thers’ sect without any common ground. Such slighting should be for specified grounds only. On the other hand, the sects of ­others should be honored for this ground or that. Thus ­doing, one helps his own sect to grow and benefits the sects of ­others, too. ­Doing other­wise, one hurts his own sect and injures the sects of ­others. For whosoever honors his own sect and condemns the sects of others wholly from devotion to his own sect, i.e., the thought “How I may

116  c h a p t e r 3

glorify my own sect” and acting thus injures more gravely his own sect on the contrary. Hence concord alone is commendable, in this sense all should listen and be willing to listen to the doctrines professed by o­ thers, this is, in fact, the desire of His Sacred Majesty.19 If envoys refrain from excessive self-­glorification and immoderate criticism of the other sects, they can maintain the peace and avoid humiliating other sects. But they must also strive to transform their own views: As Rajeev Bhargava explains, “Ashoka says that ­those seeking improvement in their ethical views should not only communicate with ­others with dif­fer­ent perspectives in order to learn from them but even follow their precepts, ‘obey’ them. Thinking as if you ­were in someone ­else’s shoes may not on occasions be sufficient; you have to act with their shoes on. This practical ethical engagement brings an experiential dimension that could be ethically transformative.”20 Clearly the aim is close to what we termed “strong reciprocity”: Both sides re­spect each other’s differences while attempting to learn from each other and forge a common morality that draws on the morality of both sides. ­There may be unequal power relations and hierarchies between states—­Ashoka sends envoys to less power­ful states, not the other way around—­but we are a long way from Christian missionaries who aim to spread the Gospel to morally backward natives. In ­today’s world, what most grates intellectuals in post-­colonial countries are moralizing sermons by modern-­day envoys—­hectoring politicians, crusading journalists, nongovernmental organization (NGO) activists, culturally insensitive tourists, not to mention card-­carrying religious missionaries—­from Western countries with a dark track rec­ord of racism, colonialism, and imperialism. Surely relations between unequal powers could be improved if representatives from

J u s t H i e r a r c h y b e t w e e n S t a t e s  117

g­ reat powers exercised Ashokan-­style restrained and respectful speech in dealing with weaker countries. This is not to deny the possibility of genuine (economic or security) conflicts of interest between states, but such conflicts would be easier to solve if modern-­day states adhered to Ashoka’s guidelines. Ashoka may be dead, but let his ideals live on! It could be argued that Ashoka put forward ­these edicts on respectful and restrained speech primarily as guidelines for domestic policy, that is, for the sake of peaceful and harmonious communal relations in his own empire.21 But surely Ashoka believed in the universality of his Buddhist-­inspired edicts, and they also provide useful guidelines for foreign policy. As it turns out, his ideals resonate with ideals of strong reciprocity between countries put forward by po­liti­cal thinkers in ancient China as well. Let us now turn to proposals for morally justified hierarchies that w ­ ere developed by Xunzi, perhaps the greatest theorist of international relations in ancient China. Xunzi, as we ­will see, proposed a mechanism—­ritual—­that can help to underpin strong reciprocity between hierarchical powers.

2. Hierarchical Ideals of Global Order in Ancient China As in ancient India, ancient Chinese thinkers took for granted the idea of hierarchy in social life. Xunzi (ca. 310–219 BCE) most explic­itly extolled the virtues of hierarchy. He is widely regarded as one of the three founding f­ athers of Confucianism (along with Confucius and Mencius). He has been tainted ­because of his supposed influence on the Legalists—­the Kautilyans of ancient China22—­but his ideas had g­ reat influence on the ­actual politics of East Asian socie­ties. His writings are clear and systematic, and he deliberately avoids utopian assumptions about h­ uman

118  c h a p t e r 3

nature and society. In fact, he begins with the assumption that “­human nature tends ­toward badness” (23.1). If ­people follow their bodily natures and indulge their natu­ral inclinations, aggressiveness and exploitation are sure to develop, resulting in cruel tyranny and poverty (19.1). Fortunately, that’s not the end of the story. ­Human beings can be “made good by conscious exertion” (23.1). They can learn to contain their natu­ral desires and enjoy the benefits of peaceful and cooperative social existence.

Xunzi on Hierarchical Rituals23 The key to transformation is ritual (23.3).24 By learning and participating in rituals, p­ eople can learn to contain their desires, ­there w ­ ill be a fit between ­people’s ­actual desires and the good available in society, and social peace and material well-­being ­will result (19.1). Rituals provide bonds not based solely on kinship that allow ­people to partake of the benefits of social existence. But what exactly is ritual? Xunzi’s account of ritual contains features that are familiar to con­temporary accounts of ritual: It is a social practice (as opposed to be­hav­ior involving only one person), it is grounded in tradition (as opposed to newly in­ven­ted social practices), it is noncoercive (in contrast to ­legal punishments), and the details can be changed according to the social context. But Xunzi’s account of rituals is driven by normative considerations, and he highlights two considerations that may be less familiar to readers ­today. In En­glish, the term “ritual” tends to connote paying behavioral lip ser­vice to social norms. The word “ritual” is often preceded by “empty,” meaning that it’s devoid of true emotions. But that’s not ritual in Xunzi’s sense: Ritual must involve emotion and be­hav­ior. As Xunzi puts it, “Rituals reach their highest perfection when both emotion and form are fully

J u s t H i e r a r c h y b e t w e e n S t a t e s  119

realized” (19.7). The main point of ritual is to civilize our animal natures, and if ­people are just ­going through the outward routines without any emotion, they are not likely to transform their natures. The ritual needs to involve, or trigger, an emotional response, so that it w ­ ill have an effect on the participants during the ritual and beyond the ritual itself. Hence, rituals often need to be accompanied by ­music that helps to trigger t­ hose emotional responses (Xunzi devoted a ­whole chapter on the moral and po­ liti­cal effects of m ­ usic). Still t­ oday, in Chinese the word “ritual” is often followed by the word for “­music” (礼乐), as though the two ideas are almost inseparable. Second, and equally impor­tant, Xunzi’s account of ritual involves social hierarchies: Rituals specify dif­fer­ent treatment for dif­fer­ent kinds of p­ eople, depending on rank (as opposed to practices that are meant to treat every­one equally). As Xunzi puts it, “The exemplary person has been civilized by ­these t­ hings, and he w ­ ill also be fond of ritual distinctions. What is mean by ‘distinctions’? I say that ­these refer to gradations of rank according to nobility or baseness, differences between the treatment of old and young, and modes of identification to match ­these with poverty and wealth and relative [social] importance” (19.3). Rituals involve p­ eople with dif­fer­ent power in common social practices that treat ­people differently. But why does Xunzi affirm that rituals must be hierarchical? At one level, he recognizes the social fact of hierarchy and that hierarchical rituals can help to secure social peace: By allotting dif­fer­ent responsibilities, privileges, and goods to dif­fer­ent individuals, rituals help to prevent conflict over ­these ­things among ­people with dif­fer­ent social status.25 But it’s not just a m ­ atter of pacifying the potential malcontents and justifying a system that gives more goods to t­ hose with more power. Quite the opposite: Hierarchical rituals are essential for generating a sense of community and the emotional disposition

120  c h a p t e r 3

for the power­ful to care for the interests of ­those at the bottom of hierarchies. Like other Confucians, Xunzi intended to persuade po­liti­cal rulers to adopt his ideas b­ ecause such rulers had the most power to transform society in the desired way. In an ideal society, the wise and humane ruler would implement such rituals, and the ­whole society would be harmonious and prosperous. But what about nonideal society? Xunzi is famously sensitive to context, and he advocated dif­fer­ent prescriptions for dif­fer­ ent contexts. So the question is how to persuade the rulers who have yet to be morally transformed? For such purposes, Xunzi had to appeal to their self-­interest. The prob­lem, however, is that the power­ful have the most to benefit from “uncivilized” society, where the strong can rely on brute force to exploit the weak. Th ­ ose with power need to be persuaded that they benefit from a social system that seems to place constraints on their desires. Hence, much of Xunzi’s discussion of ritual is designed to persuade po­liti­cal rulers that it is in their interest to promote rituals in society. Ritual, he says, is the root of the strength of the state (15.8), and the right sort of ­music can strengthen its military forces (20.5). One would expect rulers to be receptive to this sort of advice. But rituals do not only benefit rulers. Both Marxists and liberal demo­crats have denounced hierarchical rituals ­because they seem designed to benefit the ruling classes of feudal socie­ties and thus are inappropriate for modern times. But this is a misreading of Xunzi’s intentions. For Xunzi, hierarchical rituals also have the effect of benefiting the weak and the poor, t­ hose who would fare worse in a “state of nature”: “Without rituals, desires are unlimited, leading to contention, leading to disorder, and leading to poverty” (19.1). Of course, the tyrant himself w ­ on’t be the worst hit by a system where he can exercise power without

J u s t H i e r a r c h y b e t w e e n S t a t e s  121

constraints. It is the weak and vulnerable who are worst hit by disorder and poverty: In a situation without ritual civility, Xunzi says, “the strong would harm the weak as well as rob them” (23.9). Putting ritual in practice means “being kind to the h­ umble” (27.17). But why does Xunzi emphasize rituals involving ­people with dif­fer­ent power? Hierarchical rituals seem most attractive if they are contrasted with practices that exclude p­ eople of dif­fer­ent status: The rich and power­ful do their own ­thing, as do the poor and the weak (consider the ste­reo­typical account of the Indian caste system, or the dif­fer­ent rituals of the rich and the poor in highly stratified socie­ties such as the United States). The choice, typically, is not between hierarchical and egalitarian rituals, but between rituals that involve the power­ful and the vulnerable and two dif­fer­ent sets of rituals for ­those with power and t­ hose without. Xunzi argues for the former. The village wine ceremony, for example, is praised ­because the (less power­ful) young and (more power­ful) old take a drink from the same wine cup, and “in this way we know it is pos­si­ble for ju­nior and se­nior to drink together without anyone being left out” (20.12). Rituals such as common birth, marriage, and burial practices also have the effect of including the poor and marginalized as part of the society’s culture and common understandings. As Patricia Buckley Ebrey puts it, “Confucian texts and the rituals based on them did not simply convey social distinctions. At another level they overcome them by fostering commonalities in the ways ­people performed rituals. . . . ​[In early modern Eu­rope, by contrast], over time class differences in the per­for­mance of ­family rituals seem to have narrowed rather than widened.”26 In hierarchical rituals, the power­ful are made to think of the powerless as part of the group, and they are more likely to do t­ hings for them (or at least, to refrain from the worst parts of rapacious be­hav­ior).

122  c h a p t e r 3

It is no coincidence that Xunzi devotes a g­ reat deal of attention to the proper treatment of the dead, notwithstanding his aversion to religious thinking and super­natural explanations for changes in the world of the living. The dead, for obvious reasons, are the least capable of protecting their interests. They are the worst off of the worst off. Hence, ­those with power—­the living—­ need to be trained by means of certain rituals to treat them with re­spect. Xunzi carefully specifies the need to adorn the corpse ­because “if the corpse is not adorned, it becomes hideous, and if it is hideous, no grief ­will be felt” (19.12). He also specifies that the corpse must be gradually moved farther away each time it is adorned b­ ecause “if it is kept close at hand, one begins to scorn it; when having it close at hand makes it the object of scorn, one begins to weary of it; when one wearies of it, one forgets one’s duty to it; and if one forgets one’s duties, then one no longer shows proper re­spect” (19.12). The ritual should be gradually phased out so that it allows for a smooth transition to everyday life as well as an extension of the cultivated emotions of proper re­spect and mindfulness of duty to the needy in the world of the living: “With each move he takes it further away, whereby he ensures continued re­spect. With the passage of time he resumes the ordinary course of life, whereby he cares for the needs of the living” (19.12). The real moral value of hierarchical rituals, for Xunzi, is that they generate a sense of community among p­ eople with dif­fer­ent power and status, and benefit both the power­ful and the weak. Put differently, they can help to generate a sense of strong reciprocity among members of a hierarchical relationship, with both the power­ful and the weak coming to think of their fate as a common one.27 The bonds that hold them together are stronger than the fluctuating interests that underpin “weak reciprocity.” Xunzi did not only have “domestic policy” in mind. Hierarchical rituals can work their magic not just between p­ eople in one country, but also between ­people in dif­fer­ent countries.28

J u s t H i e r a r c h y b e t w e e n S t a t e s  123

Xunzi is particularly critical of economic diplomacy between states on the grounds that it can, at most, generate a weak sense of reciprocity that breaks down once the states’ interests are no longer aligned: If you serve them with wealth and trea­sure, then wealth and trea­sure ­will run out and your relations with them ­will still not be normalized. If agreements are sealed and alliances confirmed by oath, then though the agreements be fixed yet they ­will not last a day. If you cut off borderland to bribe them, then ­after it is cut off they w ­ ill be avaricious for yet more. The more you pander to them, the more they w ­ ill advance on you u­ ntil you have used up your resources and the state has given over and then t­ here is nothing left.29 If a rich country aims to gain friends just by throwing money at them, t­ hose friends ­will be fickle indeed. That said, Xunzi does not deny that “weak reciprocity” grounded in mutually beneficial self-­interest between hierarchical powers can be relatively stable and long lasting. In an anarchic world of self-­interested states, what Xunzi calls the “hegemonic state” (霸) can attain interstate leadership by being strategically reliable: Although virtue may not be up to the mark, nor ­were norms fully realized, yet when the princi­ple of all u­ nder heaven is somewhat gathered together, punishments and rewards are already trusted by all ­under heaven, all below the ministers know what they can expect. Once administrative commands are made plain, even if one sees one’s chances for gain defeated, yet t­ here is no cheating the ­people; contracts are already sealed, even if one sees one’s chance for gain defeated, yet ­there is no cheating one’s partners. If it is so, then the troops ­w ill be strong and the town ­w ill be firm and ­enemy

124  c h a p t e r 3

states ­w ill ­tremble in fear. Once it is clear the state stands united, your allies ­will trust you. . . . ​This is to attain hegemony by establishing strategic reliability.30 But strategic reliability must also have a basis in hard power for the hegemon to gain the trust of its allies. A very poor or weak country cannot be trusted to keep its promises. So with a combination of wealth, military might, and strategic reliability, a self-­interested but honest hegemon can establish mutually beneficial interest-­based relations with weaker states. If China’s “­belt and road initiative” provides material benefits both to China and to weaker countries in Central Asia, and China sticks to its contracts even in eco­nom­ically difficult times and shows that it’s a trustworthy partner, the initiative can be successful in the short to medium term. Let’s call this relation “weak reciprocity plus”—­grounded in nothing more than the self-­interest of states, but more stable than Kautilyan-­style peace pacts or naked economic diplomacy. The most stable (and desirable) kind of international leadership, however, is what Xunzi calls “humane authority” (王), meaning a ruler who wins the hearts of the p­ eople at home and abroad. At home, the proper use of rituals, combined with effective policies that secure peace and prosperity, is key to leadership success: “one who cultivates ritual becomes a humane authority; one who effectively exercises government becomes strong.”31 Setting a good model at home is necessary but not sufficient. The humane authority can gain the hearts of t­ hose abroad by institutionalizing interstate rituals: If you want to deal with the norms between small and large, strong and weak states to uphold them prudently, then rituals and customs must be especially diplomatic, the jade disks

J u s t H i e r a r c h y b e t w e e n S t a t e s  125

should be especially bright, and the diplomatic gifts particularly rich, the spokespersons should be gentlemen who write elegantly and speak wisely. If they keep the p­ eople’s interests at heart, who ­will be angry with them? If they are so, then the furious ­will not attack. One who seeks his reputation is not so. One who seeks profit is not so. One who acts out of anger is not so. The state ­will be at peace, as if built on a rock and it ­will last long like the stars.32 Moreover, the content of the rituals depends on the hierarchy of states: “The norms of humane authority are to observe the circumstances so as to produce the tools to work thereon, to weigh the distance and determine the tribute due. How could it then be equal!”33 The Western Zhou dynasty—­regarded by Confucian thinkers as the ideal humane authority—­set the model for hierarchical rituals with surrounding states with its system of Five Ser­vices: Therefore, the vari­ous Chinese states had the same ser­vice and the same customs, whereas the states of the Man, Yi, Di, and Rong had the same ser­vice but dif­fer­ent regulations. Within the pale was the domain ser­vice and outside the pale the feudal ser­vice. The feudal areas up to the border area ­were the tributary ser­vice; the Man and the Yi w ­ ere in the formal ser­ vice; the Rong and the Di ­were in the wasteland ser­vice. The domain ser­vice sacrificed to the king’s f­ ather, the feudal ser­ vice sacrificed to the king’s grandparents, the tributary ser­vice sacrificed to the king’s ancestors, the formal ser­vice presented tribute, and the wasteland ser­vice honored the king’s accession. The sacrifices to the ­father ­were carried out daily, to the grand­father monthly, to the ancestors by season. The tribute was offered once a year. This is what is called observing the

126  c h a p t e r 3

circumstances so as to produce the tools to work thereon, weighing the distance, and determining the tribute due. This is the system of humane authority.34 The general princi­ple, as Yan Xuetong explains, is that “the norm [ritual] of providing offerings at dif­fer­ent frequencies was made according to geographic distance from the throne.”35 At one level, the princi­ple of greater reciprocity among closer po­liti­cal communities was an accommodation to the practical real­ity of the difficulties of traveling long distances at the time. But geography also ­matters b­ ecause territorial proximity generates more security threats. A large power­ful country can afford to be “neutral” (to borrow in Kautilya’s terminology): It can provide security guarantees to surrounding states and peaceful conditions beneficial to all sides in the hierarchical relationship, hence meeting the requirements of weak reciprocity grounded in mutual self-­ interest. But territorial proximity between a strong state and its weaker neighbors also allows for more frequent rituals and interactions between ­people, hence providing the basis for long-­ lasting harmonious relations that help neighboring states weather the changing conceptions of self-­interest. We are closer to strong reciprocity grounded in common values and mutual learning. Such speculation is not mere theory: The ideal of reciprocity between hierarchical po­liti­cal communities informed the tributary system in imperial China, with the M ­ iddle Kingdom at the center and “peripheral” states on the outside. In this system, the tributary ruler or his representative had to go to China to pay homage in ritual acknowledgment of his vassal status. In return, China guaranteed security and provided economic benefits.36 In Ming China, the surrounding po­liti­cal communities w ­ ere divided into five zones corresponding to the Five Ser­vices system of Western Zhou, and the frequency of ritual interaction

J u s t H i e r a r c h y b e t w e e n S t a t e s  127

(roughly) correlated with the degree of closeness to the center (capital) of China, which was also meant to map the cultural achievement. What’s in­ter­est­ing for our purposes is that the system allowed for both weak and strong reciprocity. The security guarantees to the surrounding states allowed for peaceful relations that benefited both China and the vassal states. Students of Korean and Viet­nam­ese history ­will know that ­there w ­ ere repeated incursions/invasions from China, but the big picture was relatively peaceful (again, in comparison to similar periods in Eu­ro­pean history): According to David Kang, ­there was only one war between K ­ orea, Japan, and China in five centuries of the tributary system in the Ming and Qing dynasties.37 And what’s even more in­ter­est­ing is that borders w ­ ere respected even without the notion of re­spect for the sovereignty of equal states: The borders between K ­ orea, Japan, Vietnam, and China ­were relatively fixed and did not significantly change during t­ hose five centuries. The comparison with Eu­ro­pean imperialism is even more striking in terms of the dynamic of economic relations. Whereas Eu­ro­pean imperialism was motivated partly, if not mainly, by the quest for profit, the tribute-­trade system was a net loss for China and generally benefited the tributary.38 The imbalance between tribute received and gifts bestowed helped maintain the hierarchical East Asian po­liti­cal order centered on China ­because it made Chinese vassals understandably ­eager to have their inferior status recognized, thus entitling them to receive tribute.39 Salvatore Babones comments that “the emperor could even punish vassals by refusing to receive tribute from them—­a ‘punishment’ that makes sense only in terms of the disproportionate benefits accruing to the tribute-­giver.”40 Clearly t­ hese hierarchical relations satisfy the conditions for weak reciprocity, since they ­were mutually beneficial, and in some ways even more beneficial to the weaker surrounding states.

128  c h a p t e r 3

More controversially, the tributary system also allowed for a certain degree of strong reciprocity between hierarchically ordered states. China used moral power to spread Confucian norms, while allowing traditional ways of life and practices to flourish.41 ­Korea, Vietnam, and (to a lesser extent) Japan willingly accepted Chinese ideas and institutions (such as the examination system) and sought to model themselves on China.42 This is not to deny that instrumental considerations motivated most of the interaction between states in the tributary system, but Zhang Feng’s empirical analy­sis found that the early Ming’s foreign relations with ­Korea, Japan, and Mongolia ­were motivated by expressive considerations in accordance with Confucian propriety—­a form of what we term “strong reciprocity”—­ about one-­fifth of the time.43 Of course, even weak reciprocity was frequently ­violated in practice. In a study on the Ming dynasty’s g­ rand strategy against the Mongols, Alasdair Ian Johnston is struck by “the prevalence of assumptions and decision axioms that in fact placed a high degree of value on the use of pure vio­lence to resolve security conflicts.”44 ­Others argue that the tributary system itself is largely an invention of Western sinologists and cannot usefully explain China’s interaction with its neighbors over long periods of time.45 In historical practice, Chinese imperial courts did not usually use the idea of tributary relations to interfere in the internal affairs of neighboring states, and the states on China’s periphery often had complete in­de­pen­dence to do as they wished: The tributary system, according to Zhuang Guotu, was an “unreality.”46 Even assuming a large gap between the ideal and the real­ity of the tributary system in imperial China, however, it ­doesn’t follow that the ideal is not worth defending ­today. On the face of it, the tributary system seems like a good r­ ecipe for hierarchical relations between a strong power and weaker surrounding states.

J u s t H i e r a r c h y b e t w e e n S t a t e s  129

The central power offers material benefits and security guarantees to weaker surrounding states, and the weaker states pay symbolic tribute to the leadership of the central power, with frequency of ritualistic interaction depending on geo­graph­i­cal distance from the central power. Such an arrangement can be mutually beneficial, and rituals can help generate a sense of community between the strong and the weak states: what we have termed “strong reciprocity.” So should China try to re-­establish the tributary system with surrounding countries t­ oday? Yan Xuetong answers firmly in the negative: “any effort to restore the tribute system ­will weaken China’s capability for international po­liti­cal mobilization.”47 But why not try?

3. One World, Two Hierarchical Systems? What­ever its advantages in the past, the tributary system is problematic for the modern world, even as an ideal. The most obvious reason is that the tributary system, which symbolically enshrines the secondary status and moral inferiority of the vassal states, is incompatible with the idea of the equality of sovereign states. In real­ity, as mentioned, states are neither sovereign nor equal, but ­there may be a case for paying lip ser­vice to the ideal of equal sovereignty even knowing it’s far removed from the real­ity (and knowing it cannot become anywhere close to the real­ity in the foreseeable ­future). The argument for hy­ poc­risy has a long history in po­liti­cal theory.48 For example, Plato (in The Republic) famously defended the idea of a “noble lie” to persuade ­those at the bottom of the po­liti­cal hierarchy to endorse an ideal republic run by phi­los­op­ her kings and queens. The religious skeptic David Hume mounted a vigorous case in defense of an established church on the grounds that it is essential for social order.49 Xunzi most likely did not believe that

130  c h a p t e r 3

the h­ uman per­for­mance of rituals could have the power to affect super­natural beings such as ghosts or spirits, but he still defends religious rituals ­because of their positive psychological and social effects.50 ­Today, Straussian po­liti­cal theorists knowingly propagate what they consider to be falsehoods such as the idea of natu­ral rights on the grounds that they are necessary to pacify the poorly educated masses who ­can’t deal with disturbing philosophical truths that cast doubt on the ultimate value of their ordinary way of life. Arguably, a similar case can be made for paying lip ser­vice to the ideal of the equality of sovereign states. Notwithstanding a history of informal bullying by power­ful countries, it has served to constrain ­legal takeover of territory in the post–World War II era. Over the past few decades, China has become distinctly obsessed with sovereignty in the form of noninterference in the internal affairs of countries precisely ­because it seeks to avoid a repeat of seeing its territory carved up by foreign powers. That said, t­ here are limits to the idea of paying lip ser­vice to sovereignty. Most obviously, rulers lose the moral right to govern if they engage in massive abuses of basic ­human rights of their own ­people. ­Earlier Confucian thinkers such as Mencius defended the view that what we’d call t­ oday “humanitarian intervention” can be justified if the aim is to liberate ­people who are being oppressed by tyrants,51 and the Chinese government has recently signed up to the international accord that enshrines the “responsibility to protect” p­ eople from genocide and systematic violations of basic ­human rights.52 Secondly, the ideal of equality of sovereign states should not be used by power­ful countries as an excuse to shirk their extra share of responsibility for dealing with global challenges. If we agree that justice requires po­liti­cal leaders to take into account the interests of all ­those affected by their policies, then po­liti­cal leaders in large

J u s t H i e r a r c h y b e t w e e n S t a t e s  131

power­ful countries have a responsibility to consider how their policies affect not just the current generation of p­ eople in the home country, but also f­ uture generations, p­ eople in other countries, and the natu­ral world. If large countries launch major wars or make “­mistakes” on such issues as climate change and artificial intelligence, it can literally be the end of the world. As one author recently put it, China “shakes the world;”53 in contrast, nobody would write a book titled Canada Shakes the World. So it would be frankly immoral if leaders of large countries proclaim that they look out only for the interests of their own ­people; even President Trump claims that he defends the princi­ple of “American first” rather than “Amer­i­ca alone.”54 In short, it’s fine to pay (hypocritical) lip ser­v ice to the ideal of sovereign equal states, but large states should not use that as an excuse to shirk what o­ ught to be an extra share of global responsibilities. ­There’s another fatal flaw with the proposal to re-­establish the tributary system in the modern world: ­Today, power­ful countries are not necessarily the most civilized (or advanced), from a moral point of view. The tributary system was founded on the assumption that China is the center of culture and morality, and that China can and should spread its superior civilization to the rest of the world. The closer the country (or “zone”) to Beijing (the capital in the Ming and Qing dynasties), the more civilized the territory, and conversely, the farther away from Beijing, the more wild the barbarians. Nobody seriously holds this view ­today. That’s not to deny the value of proximity to power­ful countries. Kautilya’s worry that territorial contiguity can generate conflicts still holds true ­today, so major powers need to establish mutually beneficial peace pacts with neighboring countries: For example, strong countries can provide nuclear guarantees to neighboring states on the condition that ­those states do not

132  c h a p t e r 3

manufacture their own nuclear weapons. Proximity also allows for more frequent interaction—­ritual and other­wise—­hence providing the conditions for a stronger form of reciprocity grounded in common values, similar to Ashoka’s effort to promote Buddhist-­inspired values to neighboring states and the spread of Confucianism from China to K ­ orea and Vietnam. In short, the challenge is to reconstitute a de facto form of hierarchy between strong states and neighboring (weaker) states that provides the conditions for weak and (ideally) strong reciprocity while still paying lip ser­vice to the ideal of equal sovereignty of states. A modernized version of the traditional Chinese ideal of tianxia, conventionally translated as “all-­under-­heaven,”55 can inspire thinking about a hierarchical system of states that is both realistic and desirable. The term tianxia is a vague concept that has been used differently in dif­fer­ent times (and differently in the same times).56 In the Tang dynasty, for example, tianxia referred ­either to the area actually governed by Tang dynasty rulers or the ­whole world with Zhongguo (China) at its core surrounded by other countries.57 To further complicate ­matters, the term has sometimes been used in a descriptive sense meaning territory and other times in a normative sense of an ideal that contrasts with the real­ity. In the Mencius, for example, the term is used eighty-­six times58 and often refers to an ideal of a unified world without any territorial bound­aries governed by one benevolent ruler, an ideal that is meant to contrast with the ugly real­ity of small states competing ruthlessly for territorial advantage in the Warring States period. In con­temporary times, tianxia was famously revived by the phi­los­o­pher Zhao Tingyang who gave it a normative definition. According to Zhao’s formulation, tianxia has three meanings: (1) a geo­graph­i­cal meaning referring to the w ­ hole world; (2) a

J u s t H i e r a r c h y b e t w e e n S t a t e s  133

psychological meaning in the sense that the hearts of all the world’s ­peoples are unified, like a big ­family; and (3) an institutional meaning in the sense of a world government with the power to ensure universal order.59 Critics in the West have raised doubts about this proj­ect. William Callahan, for example, has charged that Zhao’s ideal of tianxia masks an effort to replace Western hegemony with Chinese hegemony.60 But Zhao is explic­itly committed to a hierarchically or­ga­nized cosmopolitan ideal. Institutionally, he argues for a world organ­ization that would have more territory and resources than any one state (including the Chinese state): “The world government directly rules a land called King-­land, about twice the size of a large sub-­state, and four times that of a medium sized sub-­state and so on. The military force controlled by the world government is greater than that of large, medium and small sub-­states with a ration of 6 to 3, 6 to 2, and 6 to 1 divisions. This proportional design limits the King-­land of the world government in its advantages over the sub-­states e­ ither in resources or military power.”61 The prob­lem, however, is that Zhao’s interpretation of tianxia is neither desirable nor realistic. Zhao claims that his ideal derives inspiration from the values and practices put forward by the founding f­ athers of the Zhou dynasty approximately three thousand years ago—­the same sage kings who inspired Confucius—­but his ideal is radically inconsistent with the key Confucian value of graded love. Zhao’s global government is supposed to be supported by the world’s ­people psychologically bound like an intimate ­family,62 but this ideal owes more to Mohism and imported traditions like Buddhism, Chris­tian­ity, Marxism, and liberal cosmopolitanism that aim to break down particularistic attachments. E ­ very Chinese intellectual knows the famous passage from the ­Great Learning—­a Han dynasty work subsequently canonized by the

134  c h a p t e r 3

Song dynasty scholar Zhu Xi (1130–1200) as one of the four Confucian ­classics—that lays out the road to tianxia: “when the personal life is cultivated, the f­ amily w ­ ill be regulated; when the ­family is regulated, the state w ­ ill be in order; and when the state is in order, t­ here is peace throughout the world (tianxia).” Starting from the moral ordering of the individual person and the ­family, an impor­tant goal of Confucianism is to bring order to the state and thereby spread peace throughout the world. The ideal is a harmonious po­liti­cal order of global peace. But nowhere does the ­Great Learning state that ties to the ­people outside the state should be as strong as, or stronger than, ties to ­people within the state. The reason is ­simple: Ties should be extended from intimates to o­ thers, but with diminishing intensity as we move beyond the circle of intimates. We owe more to intimates (starting with the ­family) than to strangers, both ­because they are the main sources of happiness and ­because we need to reciprocate for what they have done to us. In other words, the key social relations do not ­matter equally: Our ethical obligations are strongest to t­ hose with whom we have personal relationships, and they diminish in intensity the farther we go from ­those relationships. We do have an obligation to extend love beyond intimates, but it is not expected that the same degree of emotions and responsibilities w ­ ill extend to strangers. The web of caring obligations that binds f­ amily members is more demanding than that binding citizens, the web of obligations that bind citizens is more demanding than that binding foreigners, and the web binding h­ umans is more demanding than that binding us to nonhuman forms of life. The ideal of graded love is not meant to deny that we have obligations to the wider world. Just as we should extend ties beyond the f­ amily, so too should we extend ties beyond the nation. Extending this concern to outsiders, although with less concern as they extend further

J u s t H i e r a r c h y b e t w e e n S t a t e s  135

and further from the po­liti­cal community, is also natu­ral and right. Hence, Confucians should not just view special concern for fellow citizens as a po­liti­cally necessary compromise or a second-­ best deviation from an ideal world. Nor is it just a necessary step on the way to the politics of global love and government. At least some sort of special commitment to the po­liti­cal community is required by the logic of graded love, a commitment that should be extended (in diminishing degrees) to outsiders. Zhao’s interpretation of tianxia, in short, has it backward: Attachments to par­ tic­u­lar po­liti­cal communities should have ethical and po­liti­cal priority over attachments to the world. We do need to think about obligations to the world, but not if it entails systematically overriding obligations to particularistic attachments. The second prob­lem with Zhao’s interpretation of tianxia is that he ­doesn’t provide any plausible mechanism for realizing his ideal. As Zhang Feng puts it, “the critical flaw of Zhao’s thesis is his failure to outline any clear pathway that might lead to the creation of the world institutions of the tianxia system. . . . ​He insists on the priority of the world institution, yet surprisingly fails to provide any description of how it might come about . . . ​and [be] maintained.”63 Zhang wrote ­these words in 2010, and with the rise of nationalist pop­u­lism the gap between Zhao’s ideal and the real­ity has further widened, almost to the point of no return. But we can reformulate Zhao’s ideal so that it is both realistic and desirable: All we have to do is change “the world” to “East Asia” and defend the ideal of attachment to the East Asian region without the implication that this attachment needs to override ethical and po­liti­cal attachments to the state (or other “lower” forms of communal attachments, such as attachment to the ­family). So ­here’s the reformulation of the tianxia ideal: (1) a geo­graph­i­cal meaning referring to East Asia; (2) a psychological

136  c h a p t e r 3

meaning in the sense that the hearts of the East Asian p­ eoples are unified, minimally in the sense of weak reciprocity, with an aspiration for strong reciprocity; and (3) an institutional meaning in the sense of a hierarchical East Asian po­liti­cal order led by China that pays lip ser­vice to the ideal of the equal sovereignty of states. On this modernized account of tianxia, China is the center of East Asia by virtue of its dominant economic status and increasing ability to proj­ect military power, and it has both extra powers and extra responsibilities in the East Asian region. In practice, it might mean setting up East Asian regional institutions with China as the major power, similar, perhaps, to Germany’s role in the European Union. As a nuclear state, China can give security guarantees and economic benefits to weaker states such as North ­Korea in exchange for nuclear disarmament. It might mean common East Asian financial institutions (or even a common currency), with China as the major player accompanied by the responsibility to bail out weaker states in times of economic crisis. And the more flexible (or hypocritical) approach to sovereignty might actually contribute to solving territorial disputes with China’s neighbors b­ ecause China’s leaders might place less emphasis on the sanctity of territorial bound­ aries: As Allen Carlson puts it, “In a reconstituted tianxia system, the territorial and jurisdictional concerns which have so preoccupied China’s leaders over the course of the last ­century could be re-­imagined as issues involving peripheral regions, not zero-­sum disputes over sovereign recognition. In this sense, a tianxia order might pave the way for novel solution of such controversies, and as such lead to greater stability in the region.” The prob­lem, as Carlson recognizes, is that states along China’s periphery are likely to construe an attempt to impose (or even articulate) a new normative hierarchical order in the East Asian region as a threat: “Within such a system it is clear that China

J u s t H i e r a r c h y b e t w e e n S t a t e s  137

is to occupy the paramount position, while ­those along its margins are expected to accept such dominance and show fealty to the center.”64 But China’s “peripheral” states need not show fealty in any official sense (hence the key difference with the tributary system), as long as they usually defer to the major power (China) on issues of global significance. Such arrangements seem unlikely now, but stranger t­ hings have happened: Who could have ­imagined the EU in the midst of World War II? And, unlike Eu­rope, the long history of an East Asian region with China as the center could provide a psychological basis for reestablishment and maintenance of attachment to an East Asian region led by China. Still, it must be recognized that China’s neighbors such as South ­Korea, Japan, and Vietnam—­the same countries that ­were tightly integrated in the China-­led tributary system—­seem distinctly worried by China’s growing economic and military might. So how could China regain the trust of its neighbors? Obviously a bellicose approach to solving regional disputes cannot be effective in the long term. At the end of the day, China must set a good model at home. As Yan Xuetong puts it, “For China to become a superpower modeled on humane authority, it must first become a model from which other states are willing to learn.”65 As regional leader, China would also try to provide neighboring states with mutual benefits that underpin weak reciprocity. At minimum, it means securing the peace. What­ever we think of China’s foreign policy, the fact that it has not launched any wars since 1979 should be a source of comfort. But China should aim for more. Ideally, it would provide the conditions for strong reciprocity by relying on such means as Ashoka-­style respectful and restrained speech and Xunzi-­style common rituals that generate a sense of community. Unlike the tributary system, which involved China teaching its supposed cultural and moral

138  c h a p t e r 3

inferiors, the learning curve would work both ways, with “peripheral” states learning from Chinese culture and China learning from neighboring states. The deepest ties between states in a hierarchical system are underpinned by the strongest pos­si­ble form of reciprocity. From a realpolitik point of view, the United States’ military hegemony in the East Asian region is perhaps the main obstacle to the development of an East Asian tianxia hierarchy led by China.66 But t­ hings could change. North K ­ orea is currently the major military threat in the East Asian region, but it is pos­si­ble that the divided Korean peninsula ­w ill unify over the next few de­cades in some form or another. At that point, ­there would be a weaker case for United States’ troops in the East Asian region, and a unified ­Korea would fall ­under the “natu­ral” influence of China due to its proximity and superior power in East Asia.67 China need not (and should not) send troops to K ­ orea to replace the Americans, but it could provide security guarantees to ­Korea, such as protection against invasion by neighboring countries. This kind of scenario may not appeal to Koreans in ­favor of full sovereignty, but sometimes less power­f ul countries need to make the best of less-­than-­ ideal solutions. Canada, for example, was invaded twice by its more power­ful southern neighbor (in 1775 and 1812, before Canada became an in­d e­p en­d ent country), and still t­ oday many Canadians take pride in being dif­fer­ent from Americans. But Canadians know they are a small country (in terms of population and global influence), and the government usually refrains f­ rom doing ­things that antagonize the bigger and more power­ful southern neighbor. Canada can occasionally object to U.S. foreign policy (for example, the Canadian parliament objected to the 2003 invasion of Iraq), but Canadians would never dream t­ oday of, say, inviting the British (or the Chinese)

J u s t H i e r a r c h y b e t w e e n S t a t e s  139

to build military bases in Canada as a buffer against the United States. Such arrangements also benefit the weaker party: Good ties with the Americans are valuable for Canadians ­because Canada does not have to spend much on the military, with the consequence that the Canadian government can devote more resources to improving the welfare of the Canadian ­people. So, yes, Canadians are not the equals of Americans on the international stage, but what’s the prob­lem if a bit of in­equality ­under the umbrella of an American-­led regional tianxia arrangement benefits the Canadian ­people?68 Still, it could be argued that American military bases in the East Asian region are ­really meant to check China’s rise. China may well become the biggest economic power in the world over the next few de­cades, with more demands for status and global influence, and perhaps the United States has no intention of reducing its military influence in the East Asian region. This kind of attitude can lead to a disastrous war between two major powers. Jonathan Renshon demonstrates empirically that states attributed less status than they are due based on material capabilities are overwhelmingly more likely (than “satisfied” states) to initiate militarized disputes.69 The policy implication should be obvious: “conflict may be avoided through status concessions before the escalation to violent conflict occurs.”70 Renshon has Rus­sia in mind, but exactly the same point applies in the case of China: If the United States genuinely wants to avoid war in the East Asian region, it should try to accommodate and make concessions to China’s desire to establish a regional hierarchy with itself at the head of the t­ able. In the 1970s, the United States did courageously cut its official diplomatic ties to Taiwan in order to recognize the preeminent role of China in the East Asian region, and it should make more such concessions in the ­future.

140  c h a p t e r 3

In short, the most ­viable path ­toward global peace in the region involves a bipolar world with the United States and China as heads of two regional hierarchies of states that also benefit the weaker states in the hierarchical relationships. Both China and the United States recognize each other’s leadership in their respective regions, and they work together to solve common global prob­lems such as climate change. But why should other major regional powers such as the EU and Rus­sia accept such an arrangement? The most impor­tant reason is that too many global leaders would make it more difficult to coordinate peaceful relations and work on joint global proj­ects. It’s fine if Rus­sia and the EU are recognized as less-­than-­major powers with more say in their own neighborhoods, but they c­ an’t be equals with China and the United States on the world stage. ­There must be a hierarchy of regional hierarchical systems. Perhaps the biggest challenge ­will be to accommodate India. The country’s rate of economic growth has recently overtaken China’s, and India may well achieve rough parity with China in terms of population and global clout over the next few de­cades.71 So how can the two countries work together? The situation may not look promising now (the two countries w ­ ere on the brink of another border war in 2017 and China’s closest partner in South Asia is Pakistan, India’s less-­than-­friendly neighbor). Again, we need to invoke Kautilya’s insight that two countries with contiguous borders often regard each other as natu­ral enemies. China and India went to war in 1962, and they have yet to resolve their territorial conflicts (in contrast, China has peacefully resolved territorial conflicts with eleven of its other neighboring countries). But the two countries w ­ ere both members of the nonaligned movement during the Cold War, and ­today China is India’s biggest trading partner, thus underpinning mutually beneficial relations of weak reciprocity. Ties between India and

J u s t H i e r a r c h y b e t w e e n S t a t e s  141

China have been improving since early 2018—China’s President Xi Jinping suggested that “shared Asian values” should trump the geopo­liti­cal differences between the two countries—­and India has emerged as the biggest beneficiary of the Chinese-­led Asian Infrastructure Investment Bank.72 Past history also points the way to a stronger form of reciprocity that underpins lasting peace.73 Buddhism spread peacefully from India to China, to the point that it has become far more influential in China. In the 1920s, the poet Tagore deeply marked Chinese intellectual culture when he visited China.74 The ­great Chinese intellectual Liang Shuming regarded Indian spiritual culture as the apex of ­human moral growth.75 And the learning was mutual: India benefited from China’s paper, gunpowder, and silk. Perhaps China’s greatest gift to India, according to Amitav Acharya, was the preservation of Buddhist texts. Chinese and Indian translators lived and worked in China and translated and preserved Buddhist texts. ­After Buddhism dis­appeared in India and original Indian texts ­were lost or destroyed by invaders, ­these Chinese translations preserved Buddhist sutras, which could then be retranslated for Indians.76 Buddhism would have been lost to Indians without Chinese help, just as Arabs preserved Greek texts in ­science and philosophy that would other­wise have been lost. Of course, some differences between China and India, such as dif­fer­ent ways of selecting po­liti­cal leaders, need to be respected. But such differences pale compared to what ­ought to be deep mutual re­spect between two countries with thousands of years of history and such glorious and diverse civilizations. Given that India and China had ties of strong reciprocity in the past, how might it be pos­si­ble to reestablish such ties in the ­future? Once again, we need to turn to the insights of ancient thinkers. If the leaders of the two g­ reat Asian powers follow Ashoka’s guidelines for respectful and restrained speech and

142  c h a p t e r 3

implement Xunzi’s ideas for rituals that generate a sense of community, their diplomatic, cultural, and people-­to-­people interactions might well (re)generate a strong sense of reciprocity. It is not impossible to image a ­future world with an Asian hierarchical system jointly led and managed by India and China, to the benefit of both countries, surrounding smaller states, and perhaps even the w ­ hole world.77 But the ­whole world includes not just ­humans: We also need to consider our relations with the animal kingdom. In the following chapter, we argue that hierarchical relations with animals can be morally justified if they are characterized by the princi­ ple of subordination without cruelty, though this princi­ple can and should be interpreted differently based on dif­fer­ent kinds of animals and the relation we (­humans) have with them.

4 Just Hierarchy between ­Humans and Animals s u b or di n at ion w i t hou t c ru e lt y When the stables burned down, Confucius was at court. On his return, he asked “Was anyone hurt?” He did not ask about the ­horses. —­t h e a na l ects of con f uci us, 10.1 4

we spend most of our time thinking about the obligations we owe to other h­ uman beings. Our news outlets report mainly on the horrors committed by some h­ uman beings against other ­human beings and few shed tears over, say, the number of innocent animals killed in devastating earthquakes. But we live in a natu­ral world that sustains our way of life, and we interact with animals that enrich our lives and make them meaningful. Try to imagine a world without animals: no pets, no birds in the sky, fish in the sea, or safaris in Zambia. Even if it ­were feasible to sustain ­human life without animals, it would be an infinitely 143

144  c h a p t e r 4

depressing place! So we need to ask about our proper relation with animals. The traditional view—in Western socie­ties—is that ­humans are on top of a moral hierarchy, with the power to dominate animals. But this view has been increasingly rejected of late. Many con­temporary thinkers defend the princi­ple that animals are our equals and should be treated as such. We defend the traditional view. But affirming a hierarchy between animals and h­ umans d­ oesn’t justify cruelty against animals. When it comes to the treatment of animals, we defend the princi­ple of subordination without cruelty. But that princi­ple is not sufficient to spell out the kinds of obligations we owe to animals. We have dif­fer­ent relations with dif­fer­ent kinds of animals, and we owe stronger obligations of care to animals that share human-­like traits and contribute most to our well-­being. At the other end of the spectrum, we owe least to ugly animals that harm ­humans, but the princi­ple of subordination without cruelty applies even in the case of nasty cockroaches.

1. Are Animals Our Equals? The modernization and industrialization of Western socie­ties has been largely beneficial for ­human members of ­those socie­ties. Karl Marx’s contemptuous dismissal of the “rural idiocy” of village life in India, where the natives worshiped cows and monkeys rather than dedicating themselves to self-­emancipation,1 may be an expression of racism and Eurocentrism, but not many of us would want to return to the days of grinding poverty, ­limited horizons, and outdoor toilets.2 The flip side of modernization, especially as it has spread to the rest of the world, is the toll it has taken on the animal kingdom. The ­human population has doubled since the 1960s, while wild animal populations have dropped by nearly a third. Over 56 billion animals are killed per year for

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  145

food (not including aquatic animals), and meat production is expected to double again by 2050.3 Marx himself would prob­ably blame capitalism—­the relentless quest for profit comes at the cost of habitats lost and animals consumed for the sake of satisfying endless ­human desires. If corporations need to sacrifice cows and monkeys for the sake of selling t­ hings to greedy ­human beings, then that’s the price of capitalism. But cows have been largely spared in India ­because they are viewed as sacred by the majority of Hindu worshipers, even in modernized parts of the country. So capitalism per se is not sufficient to explain the annihilation of animal species in recent times: Lack of concern for animal welfare in the economic realm must be supported and reinforced by cultural and religious ideas that view animals as mere means for ­human well-­being. In Western socie­ties, arguably, the main ideological culprit has been the Christian religion. The dominant official Christian doctrine has excluded animals from consideration as having souls and has stressed the Old Testament pronouncements that gave ­humans the power to rule over the animal kingdom: “Then God said, “Let us make mankind in his image, in our likeness, so that they may rule over the fish in the sea and the birds in the sky, over the livestock and all the wild animals, and over all the creatures that move along the ground.”4 The fate of animals took a turn for the worse in the seventeenth ­century with the theory of “animal-­ machines” put forward by René Descartes, one of the pioneers of the scientific revolution and the mechanistic vision of the natu­ral world. According to Descartes, not only do animals exist for the benefit of ­humans, but, in addition, they feel nothing: Animals are mere machines, automatons. They feel neither plea­sure, nor pain, nor anything ­else. Although they can emit cries when we cut them with knives, or contort themselves in

146  c h a p t e r 4

their efforts to avoid contact with a hot-­iron, it ­doesn’t mean that they feel pain in ­those situations. They are governed by the same princi­ples as a clock, and if their actions are more complex than ­those of a clock, it’s ­because ­these are machines built by ­humans, while animals are infinitely more complex machines, made by God.5 Such wrong-­headed views justified acts of unimaginable cruelty carried out in the name of the pursuit of scientific truth. As Gary Francione explains, “Descartes and his followers performed experiments in which they nailed animals by their paws onto boards and cut them open to reveal their beating hearts. They burned, scalded, and mutilated animals in ­every conceivable manner. When the animals reacted as though they ­were suffering pain, Descartes dismissed the reaction as no dif­fer­ent from the sound of a machine that was functioning improperly. A crying dog, Descartes maintained, is no dif­fer­ent from a whining gear that needs oil.”6 Not surprisingly, ­there was a counterreaction to ­these views. A few de­cades ­later, Voltaire expressed outrage at such practices: “Barbarians seize the dog, who wins so prodigiously over man in terms of friendship; they nail him on a t­ able, and dissect it alive to show its mezaraic veins. You discover in him all the same organs of feeling that are in yourself. Listen to me, mechanist: did nature arrange all the springs of feeling in that animal so that he ­doesn’t feel anything? Does he have nerves so he is impassive?”7 Such disputes came to a head in the nineteenth ­century. Claude Bernard, the ­great French physiologist, proclaimed that total disregard of distress and pain in his unanaesthetized animals was the attitude proper for scientists, but his own wife and ­daughters founded the first antivivisection society in Eu­rope ­after they came home to find that he had vivisected the domestic dog.8

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d A n i m a l s  147

Supporters of animal welfare argued against religious and secular outlooks that seemed to justify cruel treatment of animals. Most famously, the utilitarian thinker Jeremy Bentham argued that we should stop treating the species barrier as crucial, and instead treat all sentient beings as inside the moral community. Just as the French had recently abolished slavery in their colonies: The day may come when the rest of the animal creation may acquire ­those rights which never could have been withholden from them but by the hand of tyranny. The French have already discovered that the blackness of the skin is not reason why a ­human being should be abandoned without redress to the caprice of a tormentor. It may one day come to be recognized that the number of the legs, the villosity of the skin, or the termination of the os sacrum are reasons equally insufficient for abandoning a sensitive being to the same fate. What ­else is it that should trace the insuperable line? Is it the faculty of reason, or perhaps the faculty of discourse? But a full-­ grown ­horse or dog is beyond comparison a more rational, as well as more conversable animal, than an infant of a day or a week, or even of a month, old. But suppose they w ­ ere other­ wise, what would it avail? The question is not, Can they reason? Nor Can they talk? but Can they suffer.9 The con­temporary utilitarian thinker Peter Singer elaborated Bentham’s insights into a theory of animal liberation: “All animals are equal [to ­humans].” Since “the taking into account of the interests of the being, what­ever ­those interests may be—­ must, according to the princi­ple of equality, be extended to all beings, black or white, masculine or feminine, ­human or nonhuman” and “the capacity for suffering and enjoyment is a prerequisite for having any interests at all,” therefore “if a being suffers

148  c h a p t e r 4

t­ here can be no moral justification for refusing to take that suffering into consideration. No ­matter what the nature of the being, the princi­ple of equality requires that its suffering be counted equally with the like suffering—in so far as rough comparisons can be made—of any other being.”10 Singer’s theory has been hugely influential—­his book Animal Liberation has sold over half a million copies, and it has become the bible of the animal liberation movement—­but it has also been hugely controversial ­because he denies the value of sacred or inviolable h­ uman rights. His view that animals are the equals of ­humans—­combined with the utilitarian maxim that an act or policy is justified if it can maximize the happiness and reduce the suffering of the largest number—­would seem to justify the killing of babies in a per­sis­tent vegetative state or of elderly people in the advanced stages of a degenerative disease such as Alzheimer’s if it promotes happiness and reduces overall suffering in the ­human world and the animal kingdom. Singer does accept that the lives of higher beings—­“persons” that have rationality or self-­consciousness—­are more impor­tant than mere sentient beings, with the implication that if we came across a child and a dog drowning and we could only save one, we would be ­under a moral obligation to save the child. But for Singer, not all persons are ­humans, and some ­humans are not persons. An adult chimpanzee can exhibit more self-­consciousness, more personhood, than a newborn h­ uman infant. So if we came across a newborn infant with severe disabilities who had no ­family and a mature chimp with complex ­family and social ties and we could only save one of them, we might be u­ nder an obligation to save the chimp: “Killing them [infants], therefore, cannot be equated with killing normal ­human beings, or any other self-­conscious beings. No infant—­disabled or not—­has a strong claim to life as beings capable of seeing themselves as distinct entities existing

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  149

over time.”11 Such views radically conflict with the considered intuitions of (most) p­ eople in modern socie­ties. Infanticide may have been common in ancient Greece, but it is repulsive ­today. For one ­thing, advanced medical care can cure (or improve) some of the disabilities of babies. And we can grow into p­ eople with rich social ties: H ­ uman beings are not just beings, we are ­human becomings.12 Singer is working with a strangely static view of what it means to be a fulfilled and flourishing person. More surprisingly, perhaps, Singer’s utilitarian theory may also be inimical to the interests of the billions of animals raised and killed for ­human consumption in industrial animal farms. Between 1975 and 2000, the number of animals kept in captivity grew from 8 billion to 17 billion. As Torbjorn Tannsjo explains, “the probable explanation of this fact is that ­there is a causal connection: We hold more animals b­ ecause we eat them, and b­ ecause we who eat them become more numerous.” If we provide animals in captivity with good living and ­dying conditions, and they generally live lives worth experiencing, then from the point of view of total hedonistic utilitarianism “they have our interest in raising and eating them to thank for their lives.”13 With utilitarian friends like that, defenders of animal welfare do not need enemies. To c­ ounter views that justify sacrificing the interests of par­ tic­u­lar animals for some greater good, Sue Donaldson and ­Will Kymlicka propose a theory of animal rights that is grounded in basic rights. Animals, like h­ umans, should be seen as possessing inviolable rights. ­There are some t­ hings that cannot be done to individual animals—­like torture, killing, imprisonment, medical experimentation, separation from families—no ­matter what the benefit to ­human beings, the majority of animals, or ecosystems. This rights-­based approach is a natu­ral extension of the conception of moral equality underpinning the doctrine of ­human rights, but it is extended to animals as well.14 It’s worth

150  c h a p t e r 4

asking, however, if animals ­really possess human-­like inviolable rights? If so, we should feel the same sense of outrage when the basic rights of animals are ­v iolated as when the basic rights of ­humans are ­v iolated. It’s true that some defenders of ­human rights write as though they feel the same sense of outrage when the basic rights of animals are v­ iolated. For example, Donaldson and Kymlicka compare the initial pro­cess of domestication of animals to the “importation of slaves from Africa.”15 ­Others compare the modern-­day attitudes and methods ­behind society’s treatment of animals to the Nazi Holocaust in World War II.16 What­ever the validity of ­these comparisons, it would be somewhat odd if they translated into the same sense of outrage when animals and ­humans are enslaved and massacred. Sometimes we comfort victims of bad luck or evil deeds by the thought that ­things could have been even worse. But it would be breathtakingly insensitive, if not immoral, to comfort relatives of ­those murdered during the Holocaust with the thought that millions more chickens are murdered each year for h­ uman consumption. But let’s concede that defenders of animal rights ­really do feel the same sense of outrage when the basic rights of animals and ­humans are ­violated. If ­humans and animals possess equal rights, then animal rights advocates should also argue that the same penalties should be meted out to perpetrators of rights violations against h­ umans and animals. Raimond Gaita remarks that he has never met animal rights activists who acted as if they r­ eally believed that the slaughter of animals for meat is equivalent to murder.17 Donaldson and Kymlicka argue for a w ­ hole host of radical remedies and policies to protect the rights of animals, but they do not call for life-­long imprisonment of animal meat eaters or ­people who kill insects. They claim that it would be more just and po­liti­cally effective to connect the treatment of animals more directly to the language of h­ uman rights, but it would be neither

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  151

just nor po­liti­cally effective if they pushed their argument to its logical conclusion. That’s not to say it ­doesn’t make sense to use the language of rights to make the case for animals’ welfare, but it definitely d­ oesn’t make sense to say that animals and ­humans have equal rights. Put differently, at some basic level even animal rights activists agree that ­there is a hierarchy of moral concern between ­humans and animals, with h­ umans at the top of the hierarchy. ­These arguments are not merely theoretical. In one recent case, it was deemed necessary to kill a gorilla named Harambe to save a child who had entered its enclosure at the Cincinnati Zoo. Darting Harambe with a tranquilizer gun was judged too risky as the drugs ­don’t take immediate effect, and the act might have enraged the gorilla before sedating him. Peter Singer and Karen Dawn write that “as animal advocates, we ­don’t automatically deem the life of a boy as exponentially more impor­tant than that of a fellow primate, and we might have been inclined to risk the tranquilizer in order to save Harambe’s life,” but even they recognize that “it is easy for us to say, given that it ­wasn’t our child being dragged by a 400-­pound gorilla, so we are not e­ ager to blame zoo officials for their choice.”18 It is tempting to argue that such highly unusual cases should not inform everyday ethics about how to treat animals, but policy makers often need to decide w ­ hether to value ­human lives over animal lives. Should electric cars be programmed to swerve away from animals even if it endangers the life of the driver? Should scientists be forced to end all animal experimentation even if, as the Buddhist phi­ los­o­pher and animal advocate Matthieu Ricard recognizes, almost all the medicine we use has been tested on animals?19 Should governments legislate against massacres of disease-­ carrying animals? In one recent case, Australian scientists successfully wiped out more than 80 ­percent of disease-­carrying

152  c h a p t e r 4

mosquitoes which spread deadly diseases such as dengue fever and Zika.20 Again, it would seem extremely odd if animal advocates took their rights rhe­toric seriously and proposed to charge ­these scientists with murder. So we are back to the traditional Christian view that it may be necessary to affirm a moral hierarchy between ­humans and animals, with ­humans prioritized in cases of conflict. But this view is not distinctly Christian. The Analects of Confucius reports that “when the stables burned down, Confucius was at court. On his return, he asked, ‘Was anyone hurt?’ He did not ask about the ­horses” (10.14). Clearly Confucius valued the life of a h­ uman being over that of an animal.21 Such views, in fact, are almost universal. As Mary Midgley puts it, “the natu­ral preference for one’s species does exist. It is not, like race-­prejudice, a product of culture. It is found in all h­ uman cultures, and in cases of real competition it tends to operate very strongly.”22 Still, we can ask if “speciesism” is a good ­thing. Patriarchy was a near universal social phenomenon in the past, but progressive ­people in modern socie­ties reject the idea that men can and should dominate over ­women. In the same vein, domination over animals may be objectionable even if it’s widespread. For example, Fred Besthorn objects to a hierarchy of value between “higher” and “lower” life forms ­because it “establishes justification for the preferential treatment of some species while o­ thers may be ignored or, in the case of insects, annihilated.”23 Even insects, he suggests, should be treated as equals. But domination ­doesn’t necessarily justify indifference, much less annihilation. Ethical traditions that prioritize h­ uman life over animal life in cases of conflict do not seek to justify cruelty to animals. Quite the opposite. The Christian cosmic vision that grants “high dominion” of h­ uman beings created specially in God’s image “is a call to bring our highest aspirations for the cosmos in line with God’s, a call to exercise

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d A n i m a l s  153

the love, power, and creativity of God’s image within us ­toward the end of enabling the total flourishing of God’s world. . . . ​God’s call to high dominion is fundamentally incompatible with cruelty to animals, indifference to their suffering, and the conceit that they are h­ ere for us to do with as we please.”24 Confucians who justify a hierarchy between humans and animals also object to the suffering of animals: We owe less compassion to animals than to ­humans in cases of conflict, but we still need to extend compassion to all living creatures (see section 3 of this chapter for more details). It’s true that Confucians often compared unrestrained and intemperate “barbarians” to wild animals, but just as “barbarians” can be civilized by moral education, so the Confucian sage can tame wild beasts by ­music.25 In princi­ple, wild beasts, like uneducated p­ eople, can become less aggressive and more civilized, so we have moral obligations to them as well. Buddhism affirms a cosmological hierarchy between ­humans and animals—­sensible beings are reincarnated according to karma accumulated from former lives, and ­those with less virtue w ­ ill be reincarnated as animals rather than h­ umans26—­but all beings ultimately have the same Buddha nature, and Buddhists strongly affirm a commitment to compassion and against cruelty to animals. Stories from the Buddhist tradition suggest that ­humans with the highest level of compassion might be prepared to sacrifice their own lives to save the lives of animals.27 In short, the prob­lem is not that ­there is a special species bond between h­ uman beings. The prob­lem is when that bond translates into cruelty to animals. H ­ umans have a duty to avoid cruelty to animals, both b­ ecause animals can suffer and b­ ecause ­those who are cruel to animals are more likely to be cruel to ­humans. That said, the need to avoid cruelty is merely the bottom line, and we may also have more positive obligations to some animals. A pet owner should not be indifferent to a pet dog who

154  c h a p t e r 4

craves for her attention and affection, even if she does not mean to be cruel. H ­ ere we seek inspiration from the Donaldson and Kymlicka theory of animal rights. We reject their idea that ­humans and animals have equal negative rights (not to be killed, tortured, confined, and so on). ­There is a moral hierarchy between ­humans and animals, and if it’s a question of life and death, ­humans should have priority. But we accept their idea that we need to specify what positive obligations we owe to animals, such as an obligation to re­spect animal habitats, design our roads and buildings in a way that takes into account of animals’ needs, and obligations to care for animals who have become dependent on us. ­These positive relational duties arise out of our specific relationships with animals, and we owe dif­fer­ent kinds of duties to dif­fer­ent kinds of animals. Just as we owe dif­fer­ent ­duties to fellow citizens (e.g., ­free medical care in Canada) as compared to duties to tourists, so we owe dif­fer­ent duties to our pets than to visiting rats in our basement. ­There is, so to speak, a hierarchy of positive duties, with more duties to animals dependent on our care. Let’s say a bit more about why exactly we should value our pets, and what special duties may arise from that care.

2. Domesticated Animals: Subordination with Care It seems intuitively obvious that we owe dif­fer­ent duties to dif­ fer­ent kinds of animals. We owe more to animals with human-­ like traits such as intelligence, empathy, self-­consciousness, and the ability to be aware of themselves as distinct entities with a past and a ­future, such as chimpanzees and gorillas.28 If we have to test the efficacy of a new medicine that promises to cure a ­human illness on an animal, obviously we should choose a rat

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  155

over an ape. We also owe more to pets that share our lives, even if they are less intelligent and human-­like than ­great apes. But why do we owe more to pets compared to, say, wolves? Wolves are just as intelligent as dogs and suffer just as much from pain. Donaldson and Kymlicka use the language of citizenship to justify the extra positive duties we owe to animals that partake of our social lives. We owe special obligations to fellow citizens ­because we should re­spect ­people’s capacity to form morally significant attachments and relationships, including attachments to par­tic­u­lar communities. So relations between citizens are stronger than between citizens and noncitizens: For example, the interests of citizens (but not tourists) should decide w ­ hether to build more subways or homes. Donaldson and Kymlicka argue that the citizenship logic is equally compelling and applicable in the case of both ­humans and animals. Animals, like ­people, have a degree of “agency”; they have preferences and can “vote with their feet” if they are unhappy.29 Similar to ­humans, domesticated animals that live with us and share our lives are best viewed as equal co-­citizens in our po­liti­cal community whose interests count in determining our collective good. Some animals should be viewed as temporary visitors or noncitizen denizens (e.g., squirrels in our gardens) equivalent to visiting tourists, ­others should be viewed as residents of their own po­liti­cal communities whose sovereignty and territory we should re­spect (e.g., wolves) similar to the way we re­spect citizens of other countries, and domesticated animals should be viewed as the equivalent of full and equal co-­citizens b­ ecause of the way they have been bred over generations for interdependence with ­humans. While domesticated animals cannot actively exercise their demo­cratic self-­agency ­because they lack the capacity of speech and for rational reflection, they can be compared to c­ hildren and mentally disabled persons who have a right to have their

156  c h a p t e r 4

interests included in determining the public good of the community. Concretely, according to Donaldson and Kymlicka, it means that representatives can and should be appointed to represent the interests of domesticated animals in the po­liti­cal pro­ cess. Their representatives need to ensure that we re­spect the basic citizenship rights of animal co-­citizens: They should have freedom of movement (e.g., dogs should have the right to go to restaurants, as in France); ­there should be less use of animal ­labor (e.g., dogs should have more f­ ree time to play); domesticated animals should have the equal right to medical care (­there should be some form of animal insurance); they should have the right to a ­family (we should not take newborns away from their parents); they should have the right to reside in their own territory and should not be moved away without their consent (i.e., it’s okay if they leave on their own but they c­ an’t be forced to do so); they should not be hit or physically punished; they should have the right to sex (no castration); and the punishment for intentional killing of domesticated animals should be the same as for the killing of ­humans. But does it ­really make sense to think of domesticated animals as “full citizens of the polity . . . ​ with the full benefits and responsibilities of membership”?30 Do dogs and cats ­really have the same rights as fellow citizens? Consider the following (true) story, told from Daniel’s point of view.

Didi the Cat In 1997, I made a trip to Shenzhen, then known as the “wild West of China” (now it’s an ultramodern, high-tech city of over 12 million ­people). I was offered cat meat, and partook of the feast so as not to offend my hosts. But I felt guilty afterward. By some stroke of luck, my then head of department at the University of Hong Kong offered me a small kitten named Xiao Bao (­“little

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  157

trea­sure”) a few weeks l­ater. I gladly accepted the cat, and thought it would also make my two-­year-­old son Julien happy. Xiao Bao, however, was antisocial, always hiding from ­people and getting into mischief when he had the chance to do so, such as scratching furniture with his sharp claws. The only t­ hing that would keep Xiao Bao from scratching furniture was to spray him with a bit of ­water from a ­water ­bottle. I told my son to do the same during the day if he was alone with the cat. One time, I returned home to find that Julien was very upset about something. And the cat was missing. I tried to get my son to explain what happened, but Julien was not very expressive. He pointed to the ­water ­bottle, the furniture, and an open win­dow, and made movements to suggest that the cat had jumped out the win­dow. We lived on the twenty-­second floor, and I did not ask for more details. Julien was obviously traumatized by the turn of events, and I felt horrible (not to mention that I had to report the news to my head of department). I worried that Julien would be permanently scarred by this tragic accident, and I took Julien to the local SPCA to choose another cat, with the hope that he would become attached to a new cat and develop a healthy relationship with animals. We ­were shown one cage with three beautiful kittens—­siblings—­and Julien selected the cat who seemed most sociable and curious. We named him Didi, meaning “younger b­ rother” in Mandarin Chinese. He lived for fifteen years, sleeping most of the time in Julien’s bed. Didi was deeply attached to our ­family, ate with us at meal times, and spent time around us whenever he was awake. Didi had a lovely sense of humor, playing chase around the home (once I pulled a leg muscle during the game and had to miss work for a week, though my new head of department understandably remained skeptical about the reason for my absence). Julien always sought out Didi when he returned from school, and they ­were nearly

158  c h a p t e r 4

inseparable. Didi felt terrible when Julien was ill with a fever or cold, and he would cuddle up to Julien to bring him comfort. Didi himself was in terrible pain for the last ­couple of weeks of his life, and he remained immobile on the staircase, without any appetite. Julien brought him comfort, but I had to go away for ten days. When I returned, Didi held out his paw to greet me. Didi died a few hours l­ater, moving ­under the staircase to a peaceful resting place. We loved him and still mourn his death. Didi’s life was not entirely smooth, however. He was castrated at a young age. He hated other animals and ­either fled from them (in the case of a rabbit) or fought with them (in the case of a cat). He had to endure a disruptive move from Hong Kong to Beijing, including one month in quarantine ­after arrival in Beijing. He escaped from our Beijing apartment for a week, only to return severely bloodied and on the verge of death (we nursed him back to health, and he never tried to escape again). He was diabetic for the last two years of his life, and he had to be injected with insulin twice a day. And I feel somewhat guilty about an episode of physical punishment. Didi was used to his regular sleeping hours, but he was disrupted one eve­ning when Julien and I w ­ ere watching a World Cup final in the ­middle of the night. Didi retaliated by biting me on the arm, causing a deep bloody gash. Furious, I chased him around the home and whacked him. But it worked: Didi never bit me again. If we agree with Donaldson and Kymlicka’s theory of equal citizenship with domesticated animals, then Daniel is in trou­ble. On the plus side, Didi was provided with decent medical care, even if it was costly and not subsidized by the state (in contrast to subsidized medical care for ­human citizens in China). But the negative side should put Daniel in prison for life. He willingly ate cat meat. He (unwillingly) participated in the suicide of a young cat. He took Didi at a young age away from his cat f­ amily.

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  159

He castrated Didi. He confined Didi to the home and ­limited his freedom of movement. He forced Didi to move from one po­liti­cal community to another. He ­didn’t make any effort to represent Didi’s interests in the po­liti­cal pro­cess.31 And he engaged in physical abuse. E ­ ither Daniel is a monster, or t­ here is a prob­lem with Donaldson and Kymlicka’s theory. Of course, we hope that the reader ­w ill agree that the prob­lem lies with the theory.32 Yes, we owe special care to domesticated animals ­because they depend on us for their well-­being. But the idea that they should be treated as equal citizens is, to be blunt, absurd. That said, we owe special obligations of care to domesticated animals such as cats and dogs not just ­because they depend on us (­human caretakers) for their well-­being. We also owe special obligations to domesticated animals b­ ecause our relations with them enhance ­human well-­being. Consider what Immanuel Kant, writing in 1780, said about our relation with dogs: So far as animals are concerned, we have no direct duties. Animals are not self-­conscious, and are ­there merely as a means to an end. That end is man. We can ask “Why do animals exist?” But to ask “Why does man exist?” is a meaningless question. Thus, if a dog has served his master long and faithfully, his ser­vice, on the analogy of h­ uman ser­vice, deserves reward, and when the dog has grown too old to serve, his master ­ought to keep him ­until he dies. Such action helps to support us in our duties t­ owards ­human beings, where they are bounded duties. . . . ​If a man shoots his dog ­because the animal is no longer capable of ser­vice, he does not fail in his duty to the dog, for the dog cannot judge, but his act is inhuman and damages in himself that humanity which it is his duty to show ­towards mankind. . . . ​He who is cruel to animals becomes hard in his dealings with men.33

160  c h a p t e r 4

We can and should reject Kant’s view that animals are mere means for ­human welfare.34 They are ends in themselves who have their own goals and can suffer, and we have a duty to minimize their suffering. And we have an extra duty to care for pets who share our lives and depend on us for their well-­being. But we can endorse Kant’s argument that we should not be cruel to animals ­because such cruelty morally damages the perpetrator and is likely to spill over into the h­ uman world. Th ­ ere are countless stories of mass murderers and serial killers who started off with cruelty to animals.35 More commonly, social workers look to the treatment of pets as an indication of a ­house­hold that is cruel to its ­human members, especially ­children. It’s not just a ­matter of looking for evidence of abused animals. Functional and dysfunctional families have animals at the same rate but with one significant difference: the age of the animals. As Lynn Loar explains, “If you walk into a home with a six-­year-­old dog lounging on the carpet (or couch) or a nine-­year-­old cat sunning itself in the win­dow, your sense of risk should go down—­these p­ eople are stable enough to have maintained ­these animals over time. On the other hand, if you walk into a home with young animals, your sense of risk should go up—­not just ­because of the greater demands young creatures make, but also b­ ecause they tend to come and go quickly in troubled families.”36 If social workers observe that t­ here are new puppies and kittens (and no older animals) ­every time they visit a home, it is a good indication that cruelty has been inflicted not just against the animals, but also against the ­humans in the ­family.37 It’s not just a question of avoiding cruelty. Loving care for companion animals can also promote positive virtues such as friendship in the ­human world. Consider the relational ethics widespread in Sub-­Saharan Africa known as Ubuntu. The Ubuntu way of life values social harmony above other goods. As

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  161

Desmond Tutu put it, “Harmony, friendliness, community are ­great social goods. Social harmony is for us the summum bonum—­ the greatest good. Anything that subverts or undermines this sought-­after good is to be avoided like the plague.”38 In order to live a morally good life one o­ ught to have, first and foremost, rich social relations. Friendship is an impor­tant contributor to social harmony, which implies that animals who have the highest capacity to promote the virtue of friendship should have a higher moral status than other animals.39 Since dogs have special capacity to realize the virtue of friendship with h­ umans,40 they should be accorded more value compared to, say, snakes. That said, it’s worth keeping in mind that dogs (like other animals) can and should be subordinate to ­humans. It may well be that 40 ­percent of ­people would save their pet dog over a foreign tourist.41 But as a ­matter of policy, the priority (say, for firefighters) should be on saving h­ umans (including foreign tourists) over dogs in cases of conflict. In the same vein, t­ here may be limits to friendship with dogs. Dogs have special value both ­because they help to realize the virtue of friendship between dogs and ­humans and ­because ­humans can learn to extend that virtue to other h­ umans. But friendship with dogs should not displace friendship between ­humans to the point that adults prefer pets over ­children, especially if this leads to a substantial (­human) population decline and endangers a society’s capacity to reproduce itself.42 It’s not a big deal if a few grand­mothers prefer pet poodles over their own grandchildren, but public authorities may take preventive mea­sures if it appears likely that large numbers of ­people are turning into “antipeople” dog lovers. Yet another human-­centric reason to value domesticated animals is that they can help to promote concern for justice in society at large. More precisely, pets can encourage the motivation to support distributive schemes that benefit vulnerable and

162  c h a p t e r 4

marginalized members of the po­liti­cal community.43 ­Here we can draw on Elaine Scarry’s argument that a concern for beauty inspires a concern for justice.44 Scarry’s book On Beauty and Being Just—­p erhaps the most beautifully written work in con­ temporary Anglophone po­liti­cal theory—­puts forward several arguments on behalf of her thesis. For one ­thing, the single most enduringly recognized attribute of beauty—­symmetry—­also remains key in accounts of distributive justice and fairness “as a symmetry of every­one’s relation to one another.” The symmetry of beauty leads us, or assists us in discovering, the symmetry that eventually comes into place in the realm of justice. Second, “the symmetry, equality, and self-­sameness of the sky are pre­sent to the senses, whereas the symmetry, equality, and self-­sameness of the just social arrangements are not.” We engage with symmetry and equality of beautiful ­things via our senses, and we learn to extend the love of symmetry and equality to a concern with more abstract forms of justice. Second, “at the moment we see something beautiful, we undergo a radical decentering”: We cease to stand at the center of our own world, an attitude that can also underpin concern for helping needy members of society. More than that: We hope that a world of beauty ­will survive us, or even survive the ­whole ­human species. In that sense, the love of beauty is purely unselfish, without a need for reciprocity: “­people wish ­there to be beauty in the world even when their own interest is not served by it,” just as it’s appropriate to wish for justice in the world even if we d­ on’t personally benefit from (and may need to pay a price for) more just distributional arrangements. And fi­nally, “the fact that something is perceived as beautiful is bound up with an urge to protect it, or act on its behalf,” just as ­those concerned with justice have the motivation to protect and act on behalf of victims of injustice. Scarry does not mention domesticated animals—­her examples of beauty

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d A n i m a l s  163

include gods, birds, plants, “the tiny-­mauve-­orange-­blue moth on the brick, Augustine’s cake, a sentence about innocence in Hampshire”45—­but beautiful pets should have the same effects on ­human concern for justice. The same sorts of attitudes and emotions that lead h­ umans to love beautiful pets can also underpin and motivate support for just social arrangements that benefit the marginalized and needy members of the po­liti­cal community. But ­there’s a worry that love of beauty can also motivate injustice. Consider Nietz­sche’s view that what makes life on earth worth living are ­things like “virtue, art, ­music, dance, reason, intellect—­something that transfigures, something refined, fantastic, and divine.”46 ­These kinds of ­human excellences are threatened in a culture devoted to hedonistic satisfaction and obsessed with eliminating all forms of suffering. If ­people are committed to happiness and the elimination of suffering as a goal, nascent creative geniuses such as Beethoven (and Nietz­ sche himself) ­will waste their time in pursuit of ­those aims rather than create works of outstanding beauty. We do not need to endorse Nietz­sche’s account of what makes life worth living, but it’s hard to entirely disagree with his view that the untrammeled pursuit of beauty may conflict with concern for a morality of equality and pity for the suffering. A society that strongly encourages a morality of equality and pity for the suffering is more likely to support just distribution of material goods that benefit the marginalized (and often less beautiful) members of society, whereas a society that strongly values beauty is more likely to be characterized by lack of concern (if not contempt) for, say, mentally disabled ­people and the sick el­derly who need our help. Perhaps that’s why more works of beauty are created in dysfunctional Italy than in (relatively) just Sweden. Put differently, if the aim is to motivate concern for suffering, it may require active

164  c h a p t e r 4

mea­sures to withdraw from the world of beauty. Consider Buddha’s life story. He was brought up in the luxury of a palace, and his ­father decreed that he should be exposed only to ­things beautiful and ­wholesome. One day, however, he “came upon a peasant who was groaning by the roadside, wracked with pain from some excruciating illness. All his life Siddhartha had been surrounded by strapping body guards and healthy ladies of the court; the sound of groans and the slight of a disease-­wracked body ­were shocking to him. Witnessing the vulnerability of the ­human body impressed him deeply, and he returned to the palace with a heavy heart.”47 The glimpse of ugliness, old age, and death instilled in Siddhartha a longing to be exposed to the truth in its entirety. One remarkable night, a mysterious spell swept through the court, overpowering all but Siddhartha: “the slack-­ jawed courtesans snored, their limps akimbo, their jeweled fin­ gers dropped in their curries. Like crushed flowers, they had lost their beauty. Siddhartha did not rush to make order as we might have done; this sight only strengthened his determination. The loss of their beauty was just more evidence of impermanence.”48 As they slumbered, Siddhartha was fi­nally able to escape from the palace of beauty, and he embarked on his quest for full enlightenment. He shed his royal responsibilities and his ­family and set off alone into the forest and de­cided to remain in meditation ­until he knew the mind’s true nature and could benefit all beings. What­ever the truth of Buddhist religious tenets, it’s hard to disagree with the empirical point that being only exposed to beauty can make one indifferent to the world of suffering. So beauty can work both ways: It can motivate us to seek justice, but it can also motivate injustice. In the case of pet ­owners with beautiful (and often relatively expensive) pets, the love of beauty in the animal kingdom ­w ill not necessarily (or even

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  165

usually) extend to love of justice in the h­ uman world. If pet ­owners, say, spend large amounts of money on grooming loved dogs so they look even more symmetrical, they are not likely to develop concern for the poor and marginalized members of the ­human po­liti­cal community.49 But t­ here’s a dif­fer­ent kind of pet owner who may be more attuned to justice in the h­ uman world: ­those with cute pets. In Beijing and Shanghai, small brown pet poodles are all the rage. They seem completely dependent on their o­ wners, from el­derly grand­fathers who parade them in bicycle baskets to teenage girls who cuddle them in name-­ brand handbags. When the poodles walk, they seem like awkward and silly babies, and they arouse the same sentiments of care we have for the vulnerable and not-­so-­intelligent members of the ­human species. Yes, they are beautiful, but ugly pets can also be cute. The 2018 World’s Ugliest Dog context was won by a 125-­pound dog named Zsa Zsa “with red eyes, uncontrollable drool and baggy skin.”50 Zsa Zsa is so ugly that she can be regarded as cute and endearing. And she arouses the same sentiment that informs concern for justice in the h­ uman world: the desire to help far from perfect p­ eople in need. Our hypothesis—­ that a society with widespread concern for cute pets is more likely to be concerned with poor and vulnerable h­ umans—­needs to be empirically tested. At minimum, however, we can agree that cute and cuddly animals should not be eaten. Venezuelan president Nicolas Maduro recently urged citizens to breed rabbits and eat them as a source of protein and launched publicity campaigns to persuade the public “that rabbits ­aren’t pets but two and a half kilos of meat.” Although the country ­faces per­ sis­tent food shortages, the pi­lot proj­ect failed ­because “­people had put ­little bows on their rabbits and ­were keeping them as pets . . . ​a lot of ­p eople gave names to their rabbits, they took

166  c h a p t e r 4

them to bed.”51 ­There may be a case for eating animal meat— as we ­w ill argue in the next section—­but we have a special obligation to care for domesticated animals ­because they have the capacity to suffer and ­because they can do good in the ­human world by minimizing cruelty to fellow h­ umans, promoting the virtue of friendship, and motivating ­people to help weak and vulnerable members in society at large.52

3. Eating Animals: Subordination with Humane Exploitation Meat lovers who engage with the lit­er­a­ture on food ethics are bound to be depressed. The moral case against eating animal meat is very strong. It’s bad for the environment. In the United States, 99 ­percent of all animals eaten come from factory farms.53 Animals on industrial farms need to be fed, and as the number of animals on ­these farms rises, so does the amount of food that needs to be grown: “This leads to forest and other lands being turned into farmland. That, in turn leads to loss of habitat for wildlife, loss of biodiversity, and, in the case of forests that are clear cut, loss of carbon sinks that remove green­house gases from the atmosphere, and the replacement of them with sources of green­house gases, as agricultural fields are net emitters.”54 According to a 2006 UN report, “globally, green­house gas emissions from all livestock operations account for 18 ­percent of all anthropogenic green­house gas emissions, exceeding ­those from the transportation sector.”55 It’s bad for h­ uman health. Fifty years of medical research strongly supports the health benefits of the traditional Mediterranean diet: “a calorie-­limited diet high in fresh fruits and vegetables, ­whole grains, and olive oil and low in animal protein, particularly red meat, could lower the risk of

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d A n i m a l s  167

heart attacks and strokes, decrease chronic disease and extend life.”56 And it’s particularly terrible for the billions of animals born and bred and killed for ­human consumption. The cap­i­tal­ist imperative is to produce the most meat for the least money, with the result that animals on industrial farms suffer from almost permanently cruel conditions: “Intensive confinement (e.g., gestation crates for swine, battery cages for laying hens) often so severely restricts movement and natu­ral be­hav­iors, such as the ability to walk or lie on natu­ral materials, having enough floor space to move with some freedom, and rooting for pigs, that it increases the likelihood that animals suffer severe distress.”57 The pain is both physical and psychological, b­ ecause crowding is distressing to animals. And the ­whole ­thing often ends in cruel final moments. Th ­ ere are few regulations governing the killing of animals, and animals in slaughter­houses are often struck by panic at the moment of death.58 The way we treat animals killed for ­human consumption is the most widespread and systematic infringement of the ethical imperative to minimize cruelty to animals. So what can be done? In the long term, technological developments may take care of the prob­lem. For example, 3D printing can transform low-­value cuts of meat and by-product wastes into gourmet-­style meat cuts.59 This technology relies on real meat, but it promises to substantially reduce the need to breed (and kill) ­cattle in factory agricultural farms ­because much more meat could be used (and reproduced) for ­human consumption. Other technologies promise to entirely do away with the need to breed animals for ­human consumption. For example, we may be able to consume meat grown in labs from stem cells. No sentient being is created, just tissue, hence no living animal is directly harmed by the creation of this meat. However,

168  c h a p t e r 4

Donaldson and Kymlicka worry that “such a development would have spillover effects in terms of re­spect for the living. If animal stem cells, but not h­ uman ones, are used to grow frankenmeat, does this not mark a crucial difference in terms of the dignity of persons? It seems unlikely we would grow meat for ­human consumption from ­human stem cells. This would violate the taboo against cannibalism that says h­ umans a­ ren’t for eating. But in that case, would it not be a similar violation to eat flesh grown from animal stem cells?”60 Our response is that it’s not a similar violation ­because ­there is, and should be, a hierarchy of moral concern, with h­ umans on top (see section 1 of this chapter). A moral line should be drawn between killing animals and killing ­humans. Donaldson and Kymlicka only express concern ­because they hold the assumption that animals are the equals of ­people. But if we reject that assumption, then t­ here is good reason to encourage new technologies that create simulated meats without breeding animals in cruel conditions for ­human consumption. Still, we should recognize that for the foreseeable f­ uture most meat for h­ uman consumption w ­ ill come from animal factory farms. And with China “learning” from Western-­style meat-­eating habits, t­ hings are likely to get worse on a global level before they get better. So what can be done to minimize cruelty to animals bred for h­ uman consumption in the short to medium term? At a minimum, we should agree, regardless of our ethical and religious differences, that it’s bad to eat intelligent sentient beings, such as ­great apes. We may be morally superior to animals, but the line is fine with some animals, and it comes perilously close to cannibalism if we eat animals that look, think, and act like us in so many ways. Secondly, we should not eat domesticated and/or cute animals. We have an obligation to care for them both b­ ecause they share our lives

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  169

(in the case of pets) and ­because of the virtues they promote in the ­human world (see section 2 of this chapter). Beyond such moral bottom lines, we may need to allow for dif­fer­ent eating practices based on dif­fer­e nt religions and ethical systems. Religious visions of real­ity engage one’s deepest commitments and religious-­inspired ethics may have the best potential to motivate ethical food habits. But if the ethical imperative is to minimize the suffering of animals bred in cruel conditions for h­ uman consumption, some religions are better than o­ thers. In India, it is estimated that 35 ­percent of the population—450 million ­people—­are vegetarians, largely due to the influence of Hindu interpretations of ahimsa (nonviolence) that became pop­u­lar­ ized starting from the eighth ­century or so.61 The ideal of nonviolence ­toward living creatures “was reinforced by the idea of reincarnation and its implication that ­humans and animals ­were part of a single system of the recycling of souls: do not kill an animal, for it might be your grand­mother, or your grand­child, or you.”62 The command not to injure or kill any living creature is shared by Buddhism, which may explain why predominantly Buddhist countries such as Bhutan enact policies such as bans against hunting and fishing and restrictions on breeding animals for food.63 Other major religions do not directly (or even indirectly) justify vegetarianism, but they can and should be (re)interpreted so as to emphasize the need to avoid cruelty to animals raised for h­ uman consumption. Muslims should abstain from pork, and the Qur­an prohibits eating “carrion, blood, the flesh of swine, and that which has been dedicated to other than Allah, and ­those animals killed by strangling or by a violent blow or by a head-­long fall or by the goring of horns, and ­those from which a wild animal has eaten, except what you slaughter, and ­those

170  c h a p t e r 4

which are sacrificed on stone alters” (5.3).64 The prohibition against eating animals killed by strangling or by a violent blow can be interpreted as an expression of the princi­ple that we should not be cruel to animals. Judaism also bans pork, and the Torah says that “it is forbidden to inflict pain on any living creature. On the contrary, it is our duty to relieve the suffering of all living creatures.”65 The Christian religion is perhaps least friendly to vegetarianism—­St. Paul went so far as to associate vegetarianism with lack of faith—­but recent interpretations oppose the cruel treatment of animals for h­ uman consumption. Pope Benedict XVI described industrial farm animal production as the “degrading of living creatures as a commodity.”66 The commitment to a theocentric universe in which h­ uman beings have been dignified with a special calling is compatible with—­ and may require—­the view that God appointed stewards of creation who should take g­ reat care to avoid inflicting suffering on animals. Concretely, it means eating less meat, supporting less-­intensive farming methods, and adopting a greener diet.67 The large majority of Chinese p­ eople are not formally committed to a religious ethical system, but Confucian ethics often influences everyday life habits.68 Confucians, like Christians, posit a hierarchy between ­humans and animals, with ­humans on top. But it’s not a divinely ordained hierarchy. The hierarchy is an expression of the ideal of graded love: we love t­ hose closest to us, and we have an obligation to, say, save our ­mother over a stranger in cases of conflict. We have an obligation to extend that love, but the more we extend, the less the love, and the fewer the obligations. By the same logic, we are closer to h­ umans than to animals, hence we owe more to h­ umans than to animals. As the g­ reat ­Confucian scholar-­official Wang Yangming (1472–1528) put it: Someone asked, “A ­great man and an object are one, but why does [the Confucian classic] The ­Great Learning also say that

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  171

something is favored and something is not?” The master [Wang Yangming] said, “in princi­ple, ­there is naturally something favored and something not. For example, body is one, but [if ­there is a danger,] hands and feet are used to protect head and face. Does this mean that hands and feet are not favored? This is how it should be. We love both beasts and plants, but the heart can bear to use plants to feed beasts. We love both ­human beings and beasts, but the heart can bear to slaughter beasts to feed ­family, to make sacrifices, and to treat guests. We love both the closest kin and p­ eople in the street [i.e., strangers]. But if ­there is ­little food and soup, one can survive if one gets it and w ­ ill die if one does not, and if the food is not enough to save two, the heart can bear to save the closest kin and not the person in the street. This is how it should be. When it gets to my body and the closest kin, we cannot make distinctions anymore. For to treat p­ eople humanely and to treat ­things lovingly comes from this [love of one’s own body and closest kin]. If one can bear [to do anything] ­here, one can bear [to do anything] anywhere.69 Put differently, ­there is a hierarchy of compassion: We owe less compassion to animals than to ­humans, but we still owe them some compassion. So ­under what conditions might it be morally permissible to eat animals? According to the Analects, Confucius himself did not object to eating meat. Quite the opposite, he regarded meat as a ­great plea­sure only slightly below the plea­ sure of listening to other-­worldly ­music: “When the Master was in the state of Qi he heard the shao ­music, and for several months he did not know the taste of meat. He said, ‘I had no idea that ­music could reach such heights!’ ” (7.14).70 But he also expressed care for animals: When fishing, Confucius “used a fishing line but not a net,” and when hunting “he did not shoot at roosting birds” (7.27). He did not object to fishing and hunting per se, but

172  c h a p t e r 4

he objected to wasteful and cruel killings. A Confucian would therefore support efforts to minimize the suffering of animals bred for ­human consumption. Equally impor­tant, eating meat was a rare and special event in Confucius’s time; he referred to students who would bring him a “gift of a small bundle of dried meat” (7.7). Bai Tongdong draws the implication for con­ temporary food ethics: “we [should] take meats as rare delicacies, as we did before the modern age of mass-­manufactured meat.”71 Meat eating should be viewed as something to be enjoyed (only) on special occasions, similar to champagne on New Year’s Eve. Translated into real­ity, Bai’s suggestion would radically reduce if not eliminate the need for cruel exploitation of animals bred and killed for ­human consumption in industrial animal agriculture. ­There would still be a need to raise some animals for ­human consumption (­until the day we can grow delicious meat from stem cells), but the animals could be reared on free-­range farms that allow them to lead flourishing lives. The po­liti­cal community would not entirely do away with exploitation of animals for ­human use (­because ­humans still eat some animals), but it would be a humane form of exploitation.72 But why eat any meat at all, one may ask? Some communities—­ say, devout Hindus73—­entirely do away with the need to eat meat, so why c­ an’t the rest of the world? ­There may be a case for eating meat. It’s a weak case, but it’s still a case. One reason is individual-­centered. Some ­people ­really do love meat, and asking them to entirely eliminate meat from their diets would severely curtail their happiness, perhaps even to the point of depriving their lives of meaning. The proportion of meat lovers is particularly high in communities with rich and diverse meat-­ based culinary traditions. Surely it’s no coincidence that ­there are substantially fewer vegetarians (as a proportion of the ­population) in China, France, and Italy compared to the

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  173

United Kingdom and Germany.74 It is a much bigger sacrifice to become a vegetarian in a community that treats meat eating as an impor­tant part of the good life compared to communities that view meat (and food in general) as a necessity to fill the stomach. The second reason is relational. Ethical traditions such as Ubuntu and Confucianism value harmonious relations between ­people as the ­mother of all goods. If eating a certain animal can maximize harmonious social relations—­say, a turkey over Thanksgiving can help to nourish ­family ties—­then it may be justified. Of course, ­things may change. Killing animals in ritual ceremonies was once viewed as impor­tant for community bonding, and ­today most ­people recognize that “fake animals” (e.g., wooden depictions of sheep in sacrifices for Confucius’s birthday) can serve the same purpose. It is abstractly conceivable that broccoli can play the role of the turkey on Thanksgiving Day. U ­ ntil that day, however, we should allow for the possibility that turkeys brought up (and killed) u­ nder conditions of humane exploitation are impor­tant for ­family bonding once a year. What­ever the justice of meat eating, it is more po­liti­cally effective to campaign for substantial reduction of meat eating rather than the complete elimination of meat eating in meat-­ loving countries such as China and the United States. But even this desideratum may be pushing the limits of po­liti­cal feasibility. In China, the trend is ­toward more meat eating, not less.75 Perhaps the main cause of enthusiasm for meat eating—or lack of enthusiasm for curtailing the consumption of meat—is the invisibility of the cruel conditions u­ nder which animals are exploited for ­human consumption. Paul McCartney claimed that if slaughter­houses had glass walls, every­one would be vegetarian.76 If we knew the truth about the cruelty we inflict on animals for our benefit, we’d be opposed to the system of meat production. At some level, however, we do know the truth. In France, only

174  c h a p t e r 4

14 ­percent of ­people polled opposed the proposition “It is normal that ­people raise animals for their meat” while 65 ­percent answered yes to the question “Would it bother you to witness the slaughter of animals?”77 It’s as though we know the dirty and evil stuff that underpins the production of meat, but we choose to block it out. As Melanie Joy puts it, “the primary tool of the system is psychic numbing. Psychic numbing is a psychological pro­cess by which we disconnect, mentally and emotionally, from our experience; we ‘numb’ ourselves.”78 The very language we use helps in this pro­cess. In En­glish, we usually say “it” to refer to animals rather than “he” or “she,” as though the animal is a t­ hing rather than a living being. We use the term “sheep” to refer to the seemingly peaceful animal in pasture and “mutton” to the meat we eat, as though they are two dif­fer­ent animals.79 In French, we use beautiful sounding words such as “filet mignon” to cover up the fact we are eating a piece of flesh.80 The menus in elegant Chinese restaurants often use poetic language to refer to meat-­based dishes, sometimes even appropriating irreverent references to Buddhism that might be viewed as sacrilegious by devout members of that religious community: One of the most famous dishes of Fujian cuisine is termed 佛跳墙 [“Buddha’s Temptation”; more literally, “Buddha Jumps Over a Wall”], with obscure ingredients such as shark’s fin that even meat lovers might other­w ise find hard to stomach for ethical reasons. Would we relish our meat-­based meals if the local supermarket sold meat with warning labels against eating chopped up animals killed in cruel conditions for our benefit? So how should we deal with “psychic numbing”? One response is simply to accept it. Mencius famously said, “The attitude of an exemplary person ­towards animals is this: Once having seen them alive, he cannot bear to see them die, and once having seeing them cry, he cannot bear to eat their flesh.

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d A n i m a l s  175

That is why the exemplary person keeps his distance from the kitchen.”81 But perhaps Mencius—­other­w ise brilliant and humane—­advocated hy­poc­risy ­because he thought cruelty to animals was an unavoidable necessity to feed ­humans and exposing ­humans to the suffering of animals might inure them to the suffering of ­humans. ­Today we know better. We can and should reduce meat production for ­human consumption—­it’s better for the environment, for h­ uman health, and for the animals themselves—­and to the extent it’s necessary, it should be done ­under conditions of humane exploitation. Thus, public officials and educators should be encouraging exemplary persons and every­o ne e­ lse to enter modern-­d ay kitchens—­ industrial animal farms—to face the awful truth. At the national or regional level, it might mean regulation against cruel practices in industrial animal farms and government subsidies for ­free range farming,82 as well as publicly funded media campaigns that show the downside of meat production. At the city level, it might mean the promotion of meat-­free days: The city of Gand in Belgium set a good example by inaugurating a “weekly day without meat,” with public officials ordered not to eat meat and posters and educational campaigns urging citizens to do the same.83 At the local level, it might mean bringing impressionable school c­ hildren to the neighborhood slaughter­ house, then serving ­those same animals for lunch in the school cafeteria, along with the choice of a vegetarian option. To conclude, it is morally justifiable to care more for h­ umans than for animals, as well as to distinguish between dif­fer­ent levels of moral concern for dif­fer­ent kinds of animals, depending on their capacity to suffer and their relations with ­human beings. Ugly insects such as mosquitoes and cockroaches that reproduce in the billions and carry diseases that harm ­human well-­being should be at the bottom of the hierarchy. But the injunction

176  c h a p t e r 4

against cruelty applies even to the lowest forms of animal life. We might kill them (en masse) if they promote disease or (individually) to avoid painful bites,84 but we cannot torture them or use them as mere means for our benefit.85 Animals are not machines that we design for the sake of h­ uman welfare. As we w ­ ill see in the next chapter, dif­fer­ent hierarchical norms need to govern our dealings with machines.

5 Just Hierarchy between ­Humans and Machines on t h e n e e d for a m a s t e r-­s l av e r e l at ion Once ­adopted into the production pro­cess of capital, the means of l­ abour passes through dif­fer­ent metamorphoses, whose culmination is the machine, or rather, an automatic system of machinery . . . ​a moving power that moves itself; this automaton consisting of numerous mechanical and intellectual organs, so that the workers themselves are cast merely as its conscious linkages. . . . Rather, it is the machine which possesses skill and strength in place of the worker, is itself the virtuoso, with a soul of its own in the mechanical laws acting through it; and it consumes coal, oil ­etc. (matières instrumentales), just as the worker consumes food, to keep up its perpetual motion. The worker’s activity, reduced to a mere abstraction of activity, is determined and regulated on all sides by the movement of the machinery, and not the opposite. The science which compels the inanimate limbs of the machinery, by their construction, to act purposefully, as an automaton, does not exist in the worker’s consciousness, but rather acts upon him through the machine as an alien power, as the power of the machine 177

178  c h a p t e r 5

itself. . . . ​Hence the workers’ strug­gle against machinery. . . . ​ What capital adds is that it increases the surplus ­labour time of the mass by all the means of art and science, ­because its wealth consists directly in the appropriation of surplus ­labour time. . . . ​It is thus, despite itself, instrumental in creating the means of social disposable time, in order to reduce ­labour time for the w ­ hole society to a diminishing minimum, and thus to ­free every­one’s time for their own development. —­k a r l m a r x, t h e f r agm e n t on m ach i n e s

t­ here is a universal consensus in the modern world that slavery is a moral evil. No sensible person ­will question this value. Rightly so. ­Human beings need substantial freedom to realize their goals, what­ever ­those goals happen to be. And slaves, by definition, have no freedom. But machines without consciousness or feelings do not need freedom. Washing machines wash our clothes, and nobody worries about “exploiting” them for that purpose. They are supposed to serve h­ umans: To be blunt, we are their masters, and they are like slaves, but without the ability to understand their state of unfreedom and hope for something better. And that’s a good ­thing. We would understandably feel guilty if washing machines suddenly stood up for their rights and demanded fewer working hours and more meaningful work (though we’d laugh it off if we w ­ ere told they ­were programmed to do so). So we ­humans need to maintain our master-­slave relation with machines that lack consciousness. Unfortunately, perhaps, the development of artificial intelligence threatens to upend our master-­s lave relation with

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  179

machines. Th ­ ere is no immediate threat—­even the most advanced machines t­ oday lack consciousness and/or the ability to turn on their masters in the pursuit of other goals—­but technological optimists, including many of the brilliant minds in the field, predict the development of “super-­human” machines with computational abilities that go way beyond what ­humans can dream about and that share h­ uman traits such as the ability to understand, experience, and feel. And that threatens to invert our master-­slave relation with machines. If we are lucky, the superhuman machines ­will be on our side, and treat us the way we treat cute animals. But if we are unlucky, they w ­ ill treat us like slaves, or, if they ­don’t need (or like) us, treat us the way we treat cockroaches. The solution seems to be obvious: to stop the development of advanced machines with consciousness and/or the ability to destroy ­humans or turn us into their slaves. But that’s easier said than done. This chapter ­will discuss Karl Marx’s ideal of higher communism, perhaps the most influential account of a morally justifiable society with ­humans as masters and machines as slaves.1 But the Confucian tradition has more to offer for dealing with some of the short- to medium-­term ethical challenges of artificial intelligence, particularly in the Chinese context, and Confucian insights ­will be discussed in the second part of the chapter. Then we ­will return to the long-­term, existential threat posed by the development of artificial intelligence, and argue that a strong state, led by a communist party, is best placed to deal with that threat.

1. Marx on Machines In 2008, Daniel published a book proclaiming the end of Marxist ideology in China, and predicted that the CCP would be renamed the Chinese Confucian Party.2 This prediction was not

180  c h a p t e r 5

entirely off-­base. The revival of Confucianism has only accelerated since then. President Xi Jinping visited Qufu, the home ground of Confucianism, promising that he would seriously study two books about Confucian history and philosophy that ­were handed to him. The official “core socialist values,” plastered on walls and overpasses around China, include values such as “harmony” that owe more to the Confucian tradition than to Marxism. Primary and secondary schools now teach more of the Chinese classics, including Confucianism, than at any time since the communist revolution in 1949. And t­ here is widespread ac­ cep­tance in Chinese society that traditional values, with Confucianism as the core, can and should shape China’s ­future. What is more surprising is that the Marxist tradition has also made a comeback, in both official and unofficial circles. It started ­after the 2007–2008 global financial crisis, when scholars began to look to Marx’s critique of capitalism to understand what went wrong with market socie­ties. President Xi, while promoting the revival of “superior” traditional values, has si­mul­ta­neously reaffirmed the Marxist essence of the Chinese Communist Party and increased funding for the study of Marxism in universities. Most surprising, perhaps, is the revival of theorizing about “higher communism,” Karl Marx’s end point of ­human history. This ideal is beautiful in theory—­machines would do most, if not all, of the dirty work, the state would “wither away,” and h­ umans would be ­free to develop their “creative essences” in a harmonious society without any coercive power—­but it played a dubious role in postrevolutionary Chinese history. It inspired Mao the poet-­philosopher and helped to legitimize the ­Great Leap Forward—­the attempt to move straight to a higher communist society on the basis of ­w ill power alone—­and it ended with the deaths of tens of millions of Chinese p­ eople.3 The basic prob­lem with the ­Great Leap Forward, as Deng Xiaoping well

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d M a c h i n e s  181

understood, is that it ignored Marx’s insight that poor countries need to go through capitalism on the way to higher communism. The cap­i­tal­ist mode of production treats workers as mere tools in the productive pro­cess and puts technology to use for the purpose of enriching a small minority of cap­i­tal­ists. But it does have an impor­tant virtue: It has the consequence of developing the productive forces—­technology and the knowledge to make use of it—­more than any previous economic system. The reason is that cap­i­tal­ists compete with one another to make a profit; hence they have an incentive to develop ever more efficient means to produce goods, creating a large material surplus without which communism would not be feasible. If communism is implemented without the developed productive forces that underpin material abundance—­similar to the ­Great Leap Forward—­then it ­won’t last long. Without an “absolutely essential material premise,” as Marx put it in The German Ideology—­“want is merely made general, and with want the strug­gle for necessities would begin again, and the old filthy business would necessarily be restored.”4 That’s why Marx justified British imperialism in India: Yes, it would be exploitative and miserable for Indian workers, but the foundations would be laid for socialist rule. The CCP’s defense of no-­holds-­barred capitalism in China—as Deng Xiaoping famously put it, “to get rich is glorious”—­has roots in a similar logic. In the Marxist framework, the moral point of the w ­ hole ugly pro­cess is to ­free the large mass of humankind from the need to engage in drudge ­labor. Technology ­w ill be highly developed, and at a certain point—­the moment of revolution—­private property ­will be abolished, and machines will be made to work for the betterment of humanity instead of the interests of one small class. Technology ­will do the dirty work needed to meet ­people’s physical needs, and ­people ­w ill fi­nally be f­ ree to go fishing, read books, design and create works of beauty, and so

182  c h a p t e r 5

on. Unpleasant ­labor w ­ ill be l­ imited to the maintenance of machinery and other tasks necessary to keep the system ­going, but this “realm of necessity” would not take up most of the working day. But when is China supposed to implement communism? And how w ­ ill the transition come about? Such questions w ­ ere not openly discussed in the period of market-­based reforms, even by Marxist theorists. One reason may be an aversion to utopian theorizing that followed the disasters of the ­Great Leap Forward and the Cultural Revolution. But the commitment to “higher communism” was not completely shelved. This ideal informed the work of Wang Huning, China’s most influential po­liti­cal theorist—­adviser to presidents Jiang Zemin and Hu Jintao, and now member of the ruling Standing Committee of the Politburo.5 It is no small ­matter that “private” owner­ship of homes in China is ­limited to a seventy-­year lease.6 The implication is that private property can (and should?) be expropriated at some point in the ­future, if required to bring about a truly equal communist society. But talking about communism in the midst of cap­i­tal­ist reforms reduces the likelihood of achieving it: If cap­i­tal­ists are made aware of the possibility that communism might require expropriation of the cap­i­tal­ist class, they might think that their property rights are not stable and hence might not be willing to invest in ways that are necessary to develop the productive forces now. And some Chinese workers may not be willing to sacrifice in the interests of f­ uture generations:7 If workers are made aware of the plan to implement communism in the ­future, they might not be willing to undergo the sacrifices that are required to get t­ here. Over the past few years, however, the ideal of higher communism has been revived as a topic of theoretical interest among po­liti­cal thinkers in China.8 The main reason is that higher communism—­a society where machines do the ­labor

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d M a c h i n e s  183

required to meet our physical and material needs, and we are ­free to develop as we see fit—­has once again come to be seen as a realistic possibility: not now, but a few de­cades from now. Artificial intelligence, of course, is the magic pill. In the short to medium term, it may put millions of workers out of work and exacerbate in­equality.9 But, as a leading advocate of “beneficial AI” puts it, “although ­people often discuss the disappearance of jobs with doom-­and-­gloom connotations, it ­doesn’t have to be a bad ­thing! Luddites obsessed about par­tic­u­lar jobs, neglecting the possibility that other jobs might provide the same social value. Analogously, perhaps ­those who obsess about jobs ­today are being too narrow-­minded: we want jobs b­ ecause they can provide us with income and purpose, but given the opulence of resources produced by machines, it should be pos­si­ble to find alternative ways of providing both the income and the purpose without jobs.” And no need to expropriate cap­i­tal­ists who own the machines: “If machines can one day produce all current goods and ser­vices at minimal cost, then t­ here’s clearly enough wealth to make every­one better off. In other words, even relatively modest taxes could then allow governments to pay for basic income and ­free ser­vices.”10 It is tempting to dismiss the technological optimism of a scientist who may feel more at home writing about “gaining resources though cosmic settlement”11 than about the suffering of workers who lose their jobs on the way to a high-­tech utopia. But Karl Marx himself was similarly prone to technological optimism. In chapter XV of Capital, “Machinery and Modern Industry,” Marx movingly depicts the suffering and mind-­dulling work of laborers compelled to serve machines in modern factories: The life-­long specialty of h­ andling one and the same tool, now becomes the life-­long specialty of serving one and the same

184  c h a p t e r 5

machine. Machinery is put to the wrong use [from a moral point of view, ­because ­humans are supposed to be the masters of machines], with the object of transforming the workman, from his very childhood, into a part of the detail-­machine. In this way, not only are the expenses of his reproduction considerably lessened, but at the same time his helpless dependence upon the factory as a w ­ hole, and therefore upon the cap­i­tal­ist, is rendered complete. . . . ​That factory work exhausts the ner­vous system to the uttermost, it does away with the many-­sided play of the muscles, and confiscates ­every atom of freedom, both in bodily and intellectual activity. The lightening of the ­labor, even, becomes a sort of torture, since the machine does not ­free the laborer from work, but deprives the work of all interest. . . . ​It is not the workman that employs the instruments of ­labor, but the instruments of ­labor that employ the workman.12 But the cap­i­tal­ist cannot just rely on exploitation of ­labor to squeeze profits. He (or, more rarely, she) must constantly revolutionize the means of production to maintain a competitive edge against other cap­i­tal­ists. And this leads to the development of machines that require more than workers “bound hand and foot for life to a single detail operation.” Machines become more complex, and the cap­i­tal­ist must find and make use of workers with the ability to engage in more vari­ous and complex forms of work: Modern Industry . . . ​through its catastrophes imposes the necessity of recognizing, as a fundamental law of production, variation of work, consequently fitness of the laborer for varied work, consequently the greatest pos­si­ble development of varied aptitudes. . . . ​Modern Industry, indeed, compels society, u­ nder penalty of death, to replace the detail-­worker of

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  185

t­ oday, crippled by life-­long repetition of one and the same trivial operation, and thus reduced to the mere fragment of a man, by the fully developed individual, fit for a variety of ­labors, ready to face any change of production, and to whom the dif­fer­ent social functions he performs, are but so many modes of giving ­free scope to his own natu­ral and acquired powers.13 Once the workers seize owner­ship of the means of production, they ­will be equipped to lead a many-­sided life that combines the joys of dif­fer­ent kinds of work, and they can freely choose what they do as opposed to being forced to work for cap­i­tal­ists who exploit their l­abor for profit. The same sorts of schools that train workers for varied work in a cap­i­tal­ist mode of production w ­ ill also become more widespread in communist society: “­there can be no doubt that when the working-­class comes into power, as inevitably it must, technical instruction, both theoretical and practical, ­w ill take its proper place in working-­class schools.”14 Marx ­didn’t specify when the transition would happen, but in some sense he was even more optimistic than Silicon Valley billionaires who seek to prolong their lives so they can witness the age of “Singularity” when AI can solve all ­human prob­lems: He rushed to write and publish Capital ­because he thought the communist revolution was about to occur in his day and worried his writings would soon be overtaken by events. We can infer that Marx would have been thrilled by the prospect that artificial intelligence gradually replaces the menial jobs that require “single-­detail operation,” though he would likely have called for expropriation of the means of production to ensure that the benefits of increased production are distributed fairly to workers. He would also have called for a “dictatorship

186  c h a p t e r 5

of the proletariat” to put down the “remnants of the bourgeoisie” and to ensure that ­people are indeed trained in both practical and theoretical skills, thus allowing for the flourishing of the “fully-­developed individual.” So far so good: The Chinese Communist Party, what­ever its faults, is a modern instantiation of the “dictatorship of the proletariat.” But Marx predicted that the state would eventually “wither away” in higher communism: Developed machinery would produce all the goods h­ uman beings need, and t­ here would be no need for a coercive state to protect the interests of a ruling class.15 We’d all be truly ­free and equal, with the opportunity to develop our many-­sided talents as we see fit. ­Here’s a prob­lem. Marx ­didn’t foresee the possibility that machines might eventually outsmart even “fully-­developed individuals,” with the risk that h­ uman beings could be enslaved by machine-­masters. Given this possibility, it would be foolhardy to even hope that the state ­will wither away. ­There is a need for a strong and capable pro-­human organ­ization that represents the interests of the large majority of p­ eople, not just to facilitate the transition to a society composed of “fully-­developed individuals,” but also to ensure that h­ uman beings maintain their dominance over machines. Machines are supposed to serve us, not the other way around, and only a strong state can guarantee our dominance, or at least prolong it as much as pos­si­ble. One hopes that such worries w ­ ill only become live po­liti­cal issues in the long term, several de­cades if not hundreds of years from now. In the short to medium term, however, the development of artificial intelligence w ­ ill lead to ethical dilemmas, and we must look beyond the Marxist tradition for insights, especially in the Chinese context.

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d M a c h i n e s  187

2. The Role of Confucian Role Ethics First, a bit more on Marx’s vision of communism. He envisioned a ­future where machines would do most, if not all, of the socially necessary ­labor. We ­will all be provided with our basic material needs, and we w ­ ill be ­free to lead our lives as we see fit. But what would we do in a higher communist society, with all that f­ ree time? Karl Marx himself ­doesn’t provide much guidance. He was a famously prolific writer, and his collected works amount to forty volumes. Yet he wrote hardly anything about the topic that made him most (in)famous: the ideal of a communist society. In retrospect, that was a ­mistake: It would have been more difficult to misuse this ideal in his name had he said more about what communism is—­and ­isn’t. But Marx was notoriously averse to utopian theorizing, and he d­ idn’t want to speculate in any detail about the ­future. Still, ­there are few tantalizing lines. In the Critique of the Gotha Program—­his last major work—he wrote: “In a higher phase of communist society, ­after the enslaving subordination of the individual to the division of ­labor, and therewith also the antithesis between m ­ ental and physical ­labor, has vanished; ­after ­labor has become not only a means of life but life’s prime want.”16 This line helps to answer some of the worries of AI thinkers ­today. With all that ­free time, how can we ensure that ­people w ­ on’t sink into a life of mindless hedonism? Might we not prefer to play video games all day, particularly if AI can provide as-­yet-­undreamt of levels of excitement? Not necessarily. Marx reminds us that if we get the right training—an education in the practical skills and theoretical knowledge provided by the administrators of the higher communist state—we ­will be motivated to realize our creative essences by inventing and producing works of beauty. We ­w ill be motivated to work b­ ecause we love our work, not ­because we need to work to make a living. The urge

188  c h a p t e r 5

to create ­will be our prime want, and we are likely to get satisfaction from both manual work and work of the mind. Marx thinks that the capacity for creative work is what makes ­humans distinctive and we ­will fi­nally all be able to realize that capacity in a higher communist society. In Capital, he offers the following example: “A spider conducts operations that resemble ­those of a weaver, and a bee puts to shame many an architect in the construction of her cells. But what distinguishes the worst architect from the best of bees is this, that the architect raises his structure in imagination before he erects it in real­ity.”17 The example is somewhat misleading—­because architects d­ on’t actually build their own buildings (creative cooking may be a better example)—­but the point is that we’d all have the opportunity to realize both our bodily and ­mental talents. In his ­earlier work The German Ideology, Marx offers a somewhat more detailed (and frequently cited) example of life “in communist society, where nobody has one exclusive sphere of activity but each can become accomplished in any branch he wishes, society regulates the general production and thus makes it pos­si­ble for me to do one ­thing ­today and another tomorrow, to hunt in the morning, fish in the after­noon, rear c­ attle in the eve­ning, criticise ­after dinner, just as I have a mind, without ever becoming hunter, fisherman, herdsman or critic.”18 This example may also seem a bit odd ­because hunting, fishing, and rearing c­ attle seem to be more typical activities of life in low-­technology socie­ties. But the point is not that we’d revert back to a “primitive” existence. Nor is the point that we ­will spread out talents all over the place without ever specializing in one area of work. Rather, Marx suggests that what we do in a higher communist society would not be a socially defined role determined by the need to find a job in a par­ tic­u­lar area. We ­will do what we want to do, and that involves realizing our capacity for creative work.

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  189

So perhaps we ­don’t have to worry so much that ­people in a higher communist society with slavelike AIs w ­ ill waste their time rather than seek to develop their talent for creative work. From a Confucian perspective, however, Marx’s view is an overly narrow account of the good life for h­ uman beings. Of course the capacity for creative work is an impor­tant feature of what it means to be ­human, but is it always life’s prime want? Is it even necessary for all h­ umans? Is not pos­si­ble to, say, live a fulfilling life of ser­vice to other p­ eople, even if the work is not particularly creative? Confucians, in contrast, emphasize that a good life is characterized first and foremost by rich and diverse social relations. This is not just a descriptive banality about how our identities are ­shaped by our communities, but is rather a normative claim that ­human flourishing is constituted by social relations of certain kinds, with the implication that we have an obligation to nourish ­those relations. Confucianism prizes social ways of life in the physical world above all ­else. The Analects of Confucius—­ the key text in the (diverse) Confucian tradition—is mainly about how we should relate to other ­people. On the one hand, ­people are the main sources of our plea­sure: As Confucius famously put it in the opening passage of the Analects, “Is it not delightful to have friends coming from afar?”19 On the other hand, the constraints on our plea­sure are mainly a function of responsibilities we owe to other ­people, not to anything otherworldly (such as God or superhuman AI) or to animals. We owe the most to the people who have done t­ hings for us. Other-­regarding morality begins via interaction with f­amily members, and t­ hose moral duties are extended to other h­ uman beings by way of other communal forms of life. Confucianism does not have much to say about our obligations to animals and the natu­ral world (compared to ethical traditions such as Daoism), but Mencius

190  c h a p t e r 5

(the second-­most-­influential Confucian thinker ­after Confucius himself) did point to the dangers of an ecologically unsustainable way of life.20 In any case, any reasonable interpretation (or extension) of Confucianism t­oday would recognize that ­humans can flourish only in the context of an ecologically sustainable way of life, including healthy relations with animals. In other words, the key relations for ­human well-­being are t­ hose within the f­ amily, the society (or country), and the international world (meaning, t­ oday, among countries), and between ­humans and the natu­ral world.21 But which social relations ­matter most? In a Confucian ethical framework, the key social relations do not m ­ atter equally. That is, Confucianism defends the value of partiality: Our ethical obligations are strongest to ­those with whom we have personal relationships, and they diminish in intensity the farther we go from ­those relationships. We do have an obligation to extend love beyond intimates, but it is not expected that the same degree of emotions and responsibilities ­will extend to strangers, and even less so to nonhuman forms of life.22 The contrast with the Marxist conception of the good life should be clear. Marxists value the capacity to realize diverse forms of creative work, and Confucians value the capacity to nourish diverse forms of social relations, especially relations with intimates. In one sense, both traditions are right: In a higher communist society, we’d seek to realize our capacity for creative work as well as nourish our valued social relations. So we’d have even less time to waste on mind-­dulling and compassion-­killing video games. But the traditions do conflict in impor­tant ways. From a Confucian perspective, the Marxist tradition places excessive reliance on the value of ­free choice. Yes, it’s good that we’d be ­free to choose the work we want to do, as opposed to being tied to forms of work de­cided by cap­i­tal­ist masters. And we

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  191

should do our best to stop AI from replacing cap­i­tal­ists as our masters.23 But some of our most valued attachments are not a ­matter of f­ ree choice. We owe obligations to our parents w ­ hether we like it or not, just as we owe obligations to our po­liti­cal community, including re­spect for its history and culture. We simply find ourselves in certain social roles that constitute our identity, and we need to do our best to nourish commitments that make ­those roles flourish. As Roger Ames puts it, the Confucian sensibility is “to promote a model of interdependent relationships that ­will best accomplish the goal of getting the most out of your constitutive relations.”24 That’s not to say ­there is no room for individual choice—we still need to think about which communal attachments are constitutive and which ones more contingent, and how to balance the obligations to the roles assigned by our constitutive communities25—­but what’s clear (from a Confucian perspective) is that some commitments are not up for choice. Such ideas about commitments that cannot be freely unchosen are not mere theory: They also inform the everyday, deeply held beliefs of ­people in East Asian socie­ties ­shaped by Confucian values. In the West, by contrast, a standard view—­shaped by liberal/Marxist perspectives that prioritize the value of individual autonomy—is that I should have the choice to stick to, or withdraw from, par­tic­u­lar roles once I reach the age of reason (around eigh­teen years old). I can choose to be a filial ­daughter or son, but it is no g­ reat mortal sin if I choose not to. In East Asian socie­ties, such constitutive relations are not m ­ atters of individual choice: I cannot, and should not, withdraw from the role of filial ­daughter or son. I have a lifelong obligation to care for my parents, and in a fundamental sense, it is not up to me, no ­matter how I am treated by my parents. It’s not a coincidence (nor is it particularly controversial) that several East Asian socie­ties

192  c h a p t e r 5

enforce filial piety by means of ­legal punishment.26 Similar to the obligation to care for ­children in Western socie­t ies, such obligations do not depend on individual choice, and the “deviants” can and should be punished. In short, constitutive relations set constraints on what we’d do even in higher communist socie­ties.27 We can choose our work, but we c­ an’t choose all the social relations that inform who we are and bestow obligations upon us. So from a Confucian perspective, our po­liti­cal communities can and should encourage the development of AI to the point that it helps us to realize our constitutive commitments. If AI ­frees us from socially necessary work and makes it easier to devote time to care for our parents with love and compassion, for example, then such developments should be welcomed. As Jason Borenstein and Yvette Pearson put it, “the removal of some burdensome aspects of care-­giving might lessen existing tendencies to detach oneself from ­those in need of care. The intervention of robot caregivers could improve ­family unity and other interpersonal relationships ­because they would not be tainted by our aversion to unpleasant tasks.”28 But the development of AI should be restricted if it undermines constitutive attachments: For example, if a cute robot renders loving care for el­derly parents unnecessary ­because it relieves adult c­ hildren of all caring obligations and/or el­derly parents are fooled into thinking the robot sincerely cares about their well-­being to the point that the parents care more for the robots than their own c­ hildren.29 The film Her persuasively suggests that even healthy adults can fall in love with an unusually clever, witty, and attentive AI, so such worries about the effects of cute caring robots on physically or psychologically impaired el­derly parents are not so ­far-­fetched. Another example: AI-­aided teaching should be encouraged if it helps with learning, but not if it undermines the

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  193

personal ties between teacher and student that underpin life-­ long commitments to learning. The general point is that Confucian values, modernized and defensible in the con­temporary era, can and should set the standards that help us to distinguish between desirable and malevolent forms of AI. But let us turn to a more detailed discussion of how Confucian ethics might shape thinking in one par­tic­u­lar area: the development of automated cars powered by AI.

Confucian Cars In a c­ ouple of de­cades’ time, our cities and highways may be filled with driverless cars. Already cities such as Toronto are launching experiments with special zones reserved mainly for driverless cars, with restrictions on private cars.30 Liberal individualists may complain that driverless cars can undermine the sense of autonomy that comes from driving cars. But the advantages of driverless cars powered by AI are too overwhelming: They promise to get us where we want to go in more efficient ways, and with far less risk of accidents. And the autonomy provided by driving may not be a form of autonomy worth defending. From a Marxist perspective, driving is not an expression of our capacity for creative work, except perhaps in the case of professional race car ­drivers. Better to leave the dreary task of driving to the machine, so the h­ uman inside the car can spend her time developing creative talents. And from a Confucian perspective, it’s better to fully devote one’s attention to nourishing harmonious relations with intimates (for example, by Skyping with one’s parents) than waste time driving a car. Still, many ­people might prefer to have the choice of driving cars (think of the liberation that w ­ omen car d­ rivers must feel in Saudi Arabia, which has just legalized car driving for ­women!), and

194  c h a p t e r 5

it’s impor­tant not to entirely foreclose that option, at least for the foreseeable ­future. It’s also impor­tant to consider what ethical values to program into cars. Most of the theoretical debate about the ethics of driverless cars involves highly unlikely scenarios, such as ­whether the car should decide to sacrifice the ­human in the car by crashing into a tree if the only other option is plowing into a bunch of school c­ hildren. But how many of us have faced ethical dilemmas about which persons to kill in our driving ­careers? The answer is prob­ably zero.31 Such scenarios are variations of the trolley narratives—­whether to pull a switch that leads to the death of one person if the passive option involves killing a few ­people. But we d­ on’t normally set policy according to the constraints of highly improbable scenarios (­shall we prevent the building of win­dows big enough to fit h­ uman bodies ­because of the risk that mafia hoodlums might throw bodies outside of it?). And from a Confucian perspective, such questions need to be answered with more contextual knowledge about the relevant ­human relations (for example, we’d have more qualms about killing one’s m ­ other than a stranger if it’s required to save two strangers). In any case, it would be controversial to assert that Confucian ethics should be the overriding source of ethical decision making if real lives are at stake, even in socie­ties thoroughly ­shaped by Confucian ethics.32 ­There is a strong case for also programming driverless cars with deontological concerns for unqualified re­spect for life and utilitarian concerns that value happiness for the greatest number. Perhaps the altruistic driver should have the right to program a car that puts his or her life at risk rather than the life of pedestrians or other ­drivers in cases of conflict, and for the relatively selfish ­others, life-anddeath decisions can be left to random decision making among dif­fer­ent ethical theories. Fortunately, such extreme outlier

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  195

scenarios w ­ ill be of more concern to phi­los­o­phers than to policy makers. In more common everyday scenarios, Confucian ethics might have a more impor­tant role to play, especially in socie­ties ­shaped by a Confucian heritage. Two U.S.-­based thinkers—­Amitai Etzioni and Oren Etzioni—­ penned a thought-­provoking article titled “Incorporating Ethics into Artificial Intelligence” that draws implications for programming the ethics of driverless cars.33 They begin with the assumption “that cars and other machines are not emotional beings that experience pain or shame, but unfeeling tools made to serve ­humans.” Hence, the relevant social question is how to make t­ hose tools (or slaves without consciousness) operate in ways constrained by morality. At a minimum, they argue that driverless cars must be programmed to abide by the law of the land: “for instance, that a car must stop completely at places marked by a stop sign.” Beyond that, however, the authors argue that the ethical decisions of driverless cars should be left to individual choice. They envisage: a new AI program that ­will “read” the own­er’s moral preferences and then instruct t­ hese machines to heed them. We call it an ethics bot. An ethics bot is an AI program that analyzes many thousands of items of information (not only information publicly available on Internet but also information gleaned from a person’s local computer storage and that of other devices) about the acts of a par­tic­u­lar individual in order to determine that person’s moral preferences. . . . ​For instance, an ethics bot may conclude that a person places high value on environmental protection if it finds that said person purchases recycled paper, drives a Prius, contributes to the Sierra Club, prefers local food, and never buys Styrofoam cups. It would then instruct that person’s driverless car to refuel using only environmentally friendly gas,

196  c h a p t e r 5

to run on the air conditioning only if the temperature is high, and to turn off the engine at stops.34 But this proposal is based on questionable individualist assumptions.35 For one ­thing, some (all?) driverless cars in the ­future may be owned by the community rather than the individual. Driverless cars may be embedded in a web-­network of driverless cars and programmed as a group rather than individually.36 From a moral point of view, the proposal is problematic as well: Leaving driving habits up to individuals’ “ethic bots” is only likely to reproduce, if not exacerbate, the unfairness of driving in cities populated by rude and impolite d­ rivers. If Daniel’s “ethics bot” determines that he consistently drives at the speed limit and never gives way to other cars and pedestrians in cases of ambiguity, and Pei’s “ethics bot” determines a moral preference for politely giving way to other cars and pedestrians, the machines ­w ill force Pei’s car to give way to Daniel’s car if they happen to meet at intersections without clear rules of who should go first. Is that fair? Obviously not. For deference and civility (礼让) to work, it has to be reciprocal. If only one person defers to the other on a systematic basis, the polite person w ­ ill experience a sense of unfairness and w ­ ill change his or her habits for the worse, and the society overall ­will become less civil. More fundamentally, perhaps, such an individualist approach to ethical decision making may be appropriate in a po­liti­cal community with a strong libertarian outlook, but it seems odd in a context ­shaped by Confucian ethics that places strong emphasis on constant moral self-­improvement guided by moral exemplars and the occasional paternalistic hand of the government.37 Is it wrong if the government “forces” cars to be programmed so that

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d M a c h i n e s  197

they are more civil and polite than what some ­drivers would other­w ise prefer if ­every non-­legally-­binding decision ­were left up to the individual? From a Confucian perspective, the default option should be in f­ avor of the value of “rang” (让)—­ sometimes translated as “let” or “deference” in En­glish (it sounds much more positive in Chinese). The very fact of being in civil cars might be a good teaching moment for ­those other­wise inclined to be rude and selfish ­drivers. Perhaps ­drivers should be allowed to override the “rang” command in urgent cases—­say, if a car needs to rush a person to a hospital—­but the cars can also be programmed to let other d­ rivers know (e.g., by flashing lights) that it’s an exceptional situation, not meant to override the general commitment to “rang.” Such arguments might seem odd in an American context with a strong libertarian tradition, but they ­will have more purchase in socie­ties ­shaped by Confucian values. That said, we do not mean to imply that China as a w ­ hole should program cars with a uniform code of Confucian-­ influenced values. ­There may be a strong case for Confucian cars in, say, Qufu, a city with a strong Confucian ethos.38 But perhaps cars in Lhasa should be programmed with Buddhist values that honor the sanctity of animal life (e.g., cars can be programmed to drive at slow speeds so as to minimize the killing of all forms of animals, including insects). The general point is that social norms widely endorsed in par­tic­u­lar communities should shape the task of incorporating ethics into artificial intelligence, that t­ hose norms should aim to morally elevate be­ hav­ior rather than pander to the interests of the least-­civil citizens (even if they other­wise obey the law), and that ­there can be morally legitimate variation of ­those social norms in dif­fer­ ent contexts.39

198  c h a p t e r 5

­ ese are issues to worry about in the short to medium term. Th In the medium to long term, we need to worry about existential threats posed by the pos­si­ble development of “superhuman” AI. Let us now (re)turn to this topic.

3. Silicon Valley vs. the Chinese Communist Party The Holy Grail of research on artificial intelligence is the development of artificial superintelligence (ASI), defined by Nick Bostrom as “an intellect that is much smarter than the best human brains in practically ­every field, including scientific creativity, general wisdom and social skills.”40 It is tempting to use the word “God” as a replacement for “superintelligence,” but t­ here is no guarantee that ASI ­will be benevolent and serve ­human purposes. The worry is not just that ASI ­will develop consciousness and decide to invert our master-­slave relation with machines. Even without consciousness, ASI can pose an existential threat to humanity: Bostrom posits the example of an ASI exclusively devoted to the making of paper clips that slowly but surely ends up converting the w ­ hole world, and eventually the w ­ hole universe, for the single-(un)minded aim of making paper clips. Along the way, ­humans would eventually be converted into raw material that helps make paper clips in a more efficient way.41 Fortunately, we are a long way from ASI. In fact, we are still a long way from artificial general intelligence (AGI), or the ability of a machine to perform any intellectual task that a ­human can. At the moment, even the most intellectually impressive AI can do no more than perform efficiently the narrow task assigned to it. If the task involves computational ability or the ability to pro­ cess information, the AI can surpass h­ umans in that task, but it

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  199

c­ an’t transfer that ability to other tasks. The computer program AlphaGo can beat the best player in the world at Go—­a game more mathematically challenging than chess—­but it ­can’t beat a five-­year-­old child at hide-­and-­seek. And that’s the way it should be. Why take the risk of inventing something that threatens to turn us into slaves, or to exterminate the ­whole h­ uman species?42 It seems obvious that it is desirable to regulate the development of AI so that it ­doesn’t evolve into something that threatens our well-­being. It is of course fine—­indeed, desirable—to promote research that can serve ­human interests, such as AI that can do boring but socially necessary ­labor. It is also fine to do research on AI that can supplement, if not merge with, our bodies and brains if the aim is to cure diseases and prevent cognitive malfunctions.43 But the research should be ­limited if it aims to develop AIs that could end up undermining the well-­being of ­humans, if not wholly inverting our master-­slave relation with machines.44 The prob­lem, however, is that researchers on AI, particularly in Silicon Valley, may not agree that it is desirable to curb work on AI. Many AI researchers did endorse the “Asilomar AI Princi­ ples” meant to keep AI beneficial. The Asilomar AI Princi­ples include a commitment to deal with long-­term risks—­“Risks posed by AI systems, especially catastrophic or existential risks, must be subject to planning and mitigation efforts commensurate with the expected impact”—­and the common good—­ “Superintelligence should only be developed in the ser­vice of widely shared ethical ideals, and for the benefit of all humanity rather than one state or organ­ization.”45 But such princi­ples have no teeth and the libertarian culture of Silicon Valley militates against any serious attempt to curb research that threatens to develop conscious or (intentionally or not) malevolent AI. A survey of wealthy technologists by po­liti­cal scientists at Stanford

200  c h a p t e r 5

University found that while many technologists support left-­ liberal policies on social and economic issues, “they may strive to push Demo­crats away from regulation on business—­including the growing calls for greater rules around the tech industry.”46 To be charitable, the Silicon Valley tech elites, especially the researchers, may be motivated (at least partly) by the love of the pursuit of scientific truth, and consequently they oppose any curbs designed to limit the pursuit of truth. They may interpret the “widely shared ethical ideals” mentioned in the Asilomar AI Princi­ples as the quest for truth, and many scientists may indeed have a quasi-­Platonic belief that Truth and Goodness intersect at some deep, mystical level,47 hence no need to regulate the search for truth.48 Consider the case of Google. The com­pany’s found­ers seem to be motivated by the idealistic aim of pushing the quest for scientific truth as far as it can go: “At the epicenter of Google’s bulging portfolio is one master proj­ ect: The com­pany wants to create machines that replicate the ­human brain, and then advance beyond. This is the essence of its attempts to build an unabridged database of global knowledge and its efforts to train algorithms to become a­ dept at finding patterns, teaching them to discern images and understand language.”49 And ­there seems ­little concern, or even awareness, that ­things can go wrong: As co-­founder of Google Sergey Brin put it, “Certainly if you had all the world’s information directly attached to your brain, or an artificial brain that was smarter than your brain, you’d be better off. . . . ​Perhaps in the ­future, we can attach a l­ittle version of Google that you just plug into your brain.”50 Blind optimism of this sort may help to explain Google’s hostility to any attempts at regulation of its activities. But huge economic interests are also at stake. Google owes its ultimate allegiance to its shareholders, not to humanity at large, and it is not likely to pursue aims that conflict with its

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  201

profit-­making imperative. The com­pany resorts to dubious means for the sake of profit such as trampling on copyright law and using data from users without their permission for the purpose of targeted advertising, including the use of data from countries where Google does not pay any taxes.51 More worryingly, Google is plunging full steam ahead with an “Apollolike program for reaching artificial general intelligence: a proj­ect called Google Brain.”52 The com­pany spent nearly $12.5 billion on research and development in 2016, and it has hired some of the best minds in the world, including the strategy of poaching leading professors from academia (and depriving students of mentoring by the best minds in the AI field). What happens if Google reaches its aim of developing AGI ahead of its competitors? If its shareholders ­favor the use of technology for short-­term profit, how can the rest of us not worry about its potential misuses, including the pos­si­ble inversion of the master-­slave relation we are supposed to have with machines? The billionaire technologists of Silicon Valley are hedging their bets by planning for the colonization of other planets and fantastical cures for mortality, but that provides l­ ittle comfort for the rest of us.53 It is ironic that the biggest threat to our planet comes from a com­pany with the motto “­Don’t be evil.”54 So what can be done? The best case scenario is a global consensus that emerges from extensive deliberation between dif­fer­ ent countries and cultures about the need to regulate AI that poses an existential threat to humanity, leading to a feasible and desirable global regime that regulates dangerous forms of AI.55 At the very least, ­there should be global support for research that investigates ways of preventing or mitigating potential harm.56 It is certainly worth supporting such initiatives, ­whether they come from governments or from civil society.57 ­There are reasons to be optimistic. The Montreal Protocol on Substances that Deplete the Ozone Layer is an example of successful global

202  c h a p t e r 5

action designed to combat a common threat to humanity (and the natu­ral world).58 Notwithstanding some loopholes and close-­calls, we have managed to regulate the development of nuclear weapons and ­human cloning. And notwithstanding recent setbacks, we are making some pro­gress on climate change. But it may be more of a challenge to regulate research on AI. For one t­ hing, we are far from a global consensus on the need to regulate AI: In contrast to the scientific consensus on climate change,59 experts differ widely in terms of their estimates of when computers ­will be able to simulate ­every aspect of ­human intelligence, with some “deniers” plausibly dismissing the ­whole ­thing.60 It’s hard to imagine any global action on regulation without a global consensus among experts. More worrisome, it’s more difficult to monitor research on AI compared to, say, research on nuclear weapons. ­There has been an explosion of research on AI in dif­fer­ent countries, some small scale, some large scale, and it would require a sustained, massive-­scale international effort to regulate. Even if efforts succeed in one country, ­others may not join b­ ecause t­ here is a po­liti­cal race for advantage. As Rus­sian president Vladimir Putin put it, AI is “the f­ uture, not only for Rus­sia, but for all humankind. . . . ​It comes with colossal opportunities, but also threats that are difficult to predict. . . . ​W hoever becomes the leader in this sphere ­will become the ruler of the world.”61 So it’s r­ eally a question of who gets ­there first. In short, the best case scenario may not be the most likely scenario, and we also need to consider feasible second (or third) best scenarios. What seems more likely is that an organ­ization ­will take the global lead. If w ­ e’re not cheering for Google to become the ruler of the world, then we should cheer for a nonprofit-­oriented organ­ization that is committed to the well-­ being of humanity at large. The United States government, what­ ever we think of its foreign policy, is not a likely candidate.

J u s t H i e r a r c h y b e t w e e n ­H u m a n s   a n d M a c h i n e s  203

In October 2016, the Obama White House issued a document titled “Preparing for the ­Future of Artificial Intelligence” that mentions the possibility that super-­intelligent machines could cause “havoc, with h­ umans no longer in control of their destiny at best and extinct at worst,” but it concludes “that long-­term concerns about super-­intelligent General AI should have ­little impact on current policy.”62 The report does discuss the regulatory challenges of AI-­enabled products such as self-­driving cars, but not surprisingly the Trump administration, which has made a princi­ple of rolling back government regulations from the Obama era, has not taken any action to further regulate AI research.63 Nor do ­things look promising at lower levels of government. In July 2017, Elon Musk, then CEO of Tesla, addressed a group of governors and called for a regulatory agency to be formed to gain insight into and regulate fast-­moving AI development, adding that “right now the government d­ oesn’t even have insight. . . . ​Once t­ here is awareness p­ eople w ­ ill be extremely afraid, as they should be.” But he was rebuffed by Arizona governor Doug Ducey, a Republican who said, “I was surprised by your suggestion to bring regulations before we know what we are dealing with.”64 Even if U.S. po­liti­cal leaders in the ­future are more open to the idea of regulating research on AI, we can be sure that Silicon Valley companies ­will fight tooth and nail against regulation, further diminishing the prospects of po­liti­cal success. So what’s the alternative? A recent report by Sinovation Ventures notes that “­today’s consensus view is that the United States and China have begun a two-­way race for AI dominance. . . . ​ China has put the full power of the state ­behind its drive for AI dominance.”65 And China may win the race. Like the United States, China has a culture that venerates entrepreneurs, and it is taking a ­giant leap forward in global technology innovation.66 But China has two extra advantages. It “has an abundance of data to

204  c h a p t e r 5

train AI-­learning algorithms b­ ecause of its huge population of Internet users—­more than 700 million. China’s thriving mobile Internet ecosystem also provides a test bed for AI researchers to collect and analyze demographics and transactional and behavioral big data and to conduct large-­scale experiments at a much higher level than foreign counter­parts. This combination places Beijing in a unique position to dominate AI in just a de­ cade.”67 Second, the country is led by the Chinese Communist Party. The CCP can and does take a long-­term view: It “supports AI as a strategic area supported by high-­level policies with ambitious and quantifiable targets, inter-­ministry coordination, government funding for research and development, support for workplace development, and suggestions for international collaboration and expansion.”68 Chinese cities and public officials compete to promote AI innovation and design AI friendly infrastructure proj­ects.69 The CCP supports (or refrains from interfering with) companies that use AI-­powered tools designed to bring top-­flight education and health care to remote regions of the country.70 But the CCP w ­ ill also ensure that no one private com­ pany has more power than the government. The CCP has recently taken small stakes in internet ­giants Baidu, Alibaba Group, and Tencent Holdings (collectively known as BAT), and free-­market critics complain that the government ­will have the power to monitor and constrain BAT’s activities, but that is precisely the point. At the end of the day, the CCP is supposed to serve the ­people, and it aims to steer research and economic activity so that they do not threaten ­human well-being.71 This includes dealing with ­potential long-­term challenges to h­ uman dominance. In March 2017, the Chinese government announced the establishment of a state-­sponsored AI lab focusing on deep learning, widely regarded as the forerunner to superintelligence. In July 2017, the CCP released a three-­step roadmap to take the global lead in artificial intelligence by 2030.72

J u s t H i e r a r c h y b e t w e e n ­H u m a n s  a n d M a c h i n e s  205

What would it mean for China to win the race to dominate AI?73 Liberal critics worry that the CCP w ­ ill harness big data and AI “to perfect the police state.”74 Technologies such as facial recognition software ­will indeed enable the police and state-­security agencies to keep track of “destabilizing ele­ments” in society. It may be more difficult to or­ga­nize challenges to the dominance of the CCP. But the big picture is that the large majority of Chinese ­people have more substantial freedom than in any period in Chinese history—­the freedoms to marry, find jobs, travel abroad, and so on—­and ­there’s no reason to expect that t­ hese freedoms ­will be curtailed. More pertinently, it may not be a bad ­thing if the new technologies facilitate the monitoring of “destabilizing ele­ ments” in the form of AI researchers who threatens to invert our master-­slave relation with machines. As committed Communists, the leaders of the CCP can and should do their best to maintain ­human dominance, all the while encouraging the use of machines that ­free ­humans from the need to engage in drudge ­labor. From the rest of the world’s perspective, perhaps the worry is that China ­will be able to rule the world with its AI-­powered economic, po­liti­cal, and military might. But ­there’s a difference between ruling the world and helping the world to deal with global challenges. If the Trump administration has relinquished global leadership in such areas as climate change, United Nations peacekeeping, and win-­w in trade accords, why ­shouldn’t the CCP step up to the plate? Why s­ houldn’t the CCP also take the lead in international efforts to regulate AI?75 In November 2017, the CCP hosted the largest-­ever meeting of po­liti­cal parties from around the world, with representatives from over 120 countries and 400 po­liti­cal parties (Daniel was an invitee, as an in­de­pen­ dent academic). Most of the representatives ­were from communist and progressive parties from around the world, but the gathering included representatives from the Republican Party in the United States and the Conservative Party in the United

206  c h a p t e r 5

Kingdom. President Xi addressed the delegates and urged them to focus on the common destiny of humankind, invoking ancient ideals such as tianxia (“world making”) and datong (“­grand harmony”), all the while stressing the need to re­spect cultural and po­liti­cal diversity, and he criticized t­ hose who urge China to promote its “model” to the rest of the world. He also invoked the challenge of AI, and in an other­wise serious speech, managed a wry smile when he mentioned that some thinkers propose to colonize other planets. No, he said, the earth is our home, and we need to make it a better place. Let’s assume that China wins the race for AI supremacy by 2030. It ­will be a Pyrrhic victory if the CCP regulates AI research at home while Google and others continue with their reckless research into super AI that threatens to invert our master-­slave relation with machines. The CCP w ­ ill need to regulate AI research not just in China, but in the rest of the world as well. It ­doesn’t mean acting like a “global human-­led Orwellian surveillance state where certain kinds of AI research are banned.”76 It just means ensuring that AI research ­doesn’t threaten to invert our master-­slave relation with machines. And the formula should be “先礼后兵,” usually translated as “first ritual, then force,” but meaning ­here “first soft power, then hard power.” The first step should be to regulate AI research in China, and to ­inspire the rest of the world by positive example. Then China should strive to persuade other countries by such means as economic incentives to join in an international accord on AI regulation. If such efforts fail ­because they are blocked by Google or others, then China would need to consider economic sanctions and, as a last resort, military force. If we are unlucky, the last war involving humans w ­ ill be a clash between the Chinese Communist Party and Google’s unfriendly creation, and for the sake of humanity we need to pray for the victory of the CCP.77

No t e s

Introduction 1. Liu Yaodong. “不喝酒，如何在山东的酒桌上活下来” [If you ­don’t drink alcohol, how can you survive in Shandong’s drinking t­ ables?], June 28, 2018, https://­ baijiahao​.­baidu​.­com​/­s​?­id​=­1604692088818172035&wfr​=­spider&for​=­pc. We translated from Chinese (and French) ­unless other­wise indicated. 2. Patriarchal norms may also be weakening in parts of rural Shandong province: Pei and Daniel ­were once invited to the home of a rural resident near Shandong University, and Pei was asked to occupy the principal honored guest seat next to their host. Also, we w ­ ere told of an episode that suggests Shandong-­style patriarchal norms may not always work in the interests of the supposedly powerfully men: In one rural ­family, the grand­mother served chicken to her female relatives in the kitchen and then served sweet potatoes to the men in the dining hall who w ­ ere blithely unaware that they ­were being treated as second-­class members of the ­family. The famous Qing dynasty novel Dream of the Red Chambers has many such stories of “weapons of the weak” exercised by ­women and servants of the ­house­hold. 3. The closest analog in Western culture is King Arthur’s round t­ able, which was meant to be an expression of equality (see Wace, Roman de Brut: A History of the British [Text and Tradition], trans. Judith Weiss [Exeter: University of Exeter Press, 2002]). 4. It is similar in other academic settings: One of the most challenging jobs in Chinese academia is to prepare the seating arrangements at conferences, with seats being assigned according to a professor’s age, status, and/or level of achievement. The conferences are almost always accompanied by a formal picture-taking ceremony with all conference participants. For photos, the seating/standing positions are usually not assigned beforehand, but the hierarchies are well known to participants: The closer to the center of the first row, the more prestigious the social position, and ­there ­will usually be mock “fights” to avoid being closer to the center. The visiting foreigners who are ignorant of ­these rituals are often “pushed” to the front and center, and they w ­ ill occupy ­those positions without any re­sis­tance, but ­here too they are forgiven for their 207

208  N o t e s t o I n t r o du c t i on moral transgressions b­ ecause foreign “barbarians” are not expected to know the norms of civility in Chinese academia. 5. The earliest justification for drinking rituals in the Chinese tradition—­from the Book of Documents (尚书)—is an argument for restraint from drinking alcohol. This Western Zhou (1122–771 BCE) work is critical of the unrestrained drinking in the ­earlier Shang dynasty on the grounds that it led to extravagance and po­liti­cal chaos. Drinking rituals are thus meant to regulate and limit drinking (Yu Zhong, 喻中读《尚书》[Reading the Book of Documents with Yu Zhong], Beijing: Peking University Press, 2011, 203). This justification for ritualized drinking, unfortunately, is often ­violated in con­temporary practice. Thanks to Kong Xinfeng for this reference. 6. Liu Yaodong, “不喝酒，如何在山东的酒桌上活下来” [If you d­ on’t drink alcohol, how can you survive in Shandong’s drinking ­tables?], June 28, 2018, https://­ baijiahao​.­baidu​.­com​/­s​?­id​=­1604692088818172035&wfr​=­spider&for​=­pc. 7. The “penalty” for refusing a toast, fortunately, is not as consequential as in the past. According to a famous story in A New Account of the Tales of the World (世说新 语, Shishuo Xinyu), written by Liu Yiqing (刘义庆; 403–444) during the Southern and Northern dynasties (420–589), one of the richest officials in Western Jin dynasty (266–420), Shi Chong (石崇, 249–300) invited General Wang Dun (王敦, 266–324) and Chancellor Wang Dao (王导, 276–339) for drinking: “Whenever Shi Chong invited guests to a banquet, he always had beautiful ­women serve the wine. If a guest failed to drain his cup, Shi would have a servant decapitate the w ­ oman who served him. . . . ​The chancellor [Wang Dao] was not much of a drinker, but he forced himself to drain ­every toast ­until he was dead drunk. However, each time it came to the generalissimo (Wang Dun) he deliberately refused to drink in order to observe what happened. Even ­after three w ­ omen had been beheaded his expression remained ­unchanged, and he still refused to drink. When the chancellor chided him, he said. ‘If he wants to kill p­ eople from his own h­ ouse­hold, what is that to you?’ ” Yiqing Liu, Jun Liu, and Richard B. Mather, A New Account of Tales of the World (Ann Arbor: University of Michigan Press, 2002), 458. 8. One study of five Chinese provinces, including Shandong, found that men drank 13.4 times more than w ­ omen. See World Health Organ­ization, “WHO Global Status Report on Alcohol 2004,” http://­www​.­who​.­int​/­substance​_­abuse​/­publications​/­en​ /­china​.­pdf. Such facts have impor­tant consequences, and may help to explain why ­women find it more difficult to rise to impor­tant posts in the business and po­liti­cal worlds in provinces with greater per capita alcohol consumption (­because much of the trust that greases the wheels of social interaction is formed over meals with ritualized alcohol consumption). Of course, ­there are more direct and consequential ­factors that explain discrimination against ­women in China. Most worrisome, men do not typically help much with child rearing and h­ ouse­work. Hence, w ­ omen are often

N o t e s t o I n t r o du c t i o n   209 denied equal opportunities in job applications ­because it is assumed they ­will be burdened by f­ amily responsibilities and hence cannot fully devote themselves to paid work in the “public” realm (this kind of prejudice has only worsened since the abolishment of the one child policy, b­ ecause it’s now thought that most w ­ omen w ­ ill have even more ­family responsibilities). Unfortunately, ­there is as yet ­little social pressure to provide state-­funded nurseries for ­women and to encourage men to take on more ­house­work and ­family responsibilities (Shanghai is perhaps the most gender-­equal part of China ­because men often help with ­house­work). 9. Joe C. Magee and Adam Galinksy, “Social Hierarchy: The Self Reinforcing Nature of Power and Status,” The Acad­emy of Management Annals 2 (2008): 354. 10. We are grateful to David Wong for his email on evolution and hierarchy. 11. Henok Mengistu, Joost Huizinga, Jean-­Baptiste Mouret, and Jeff Clune, “The Evolutionary Origins of Hierarchy,” Computational Biology 12, no. 6 (2016): 1–23. 12. Peter Turchin, “The Evolution of Hierarchy,” December 20, 2014, https://­ evolution​-­institute​.­org​/­blog​/­the​-­evolution​-­of​-­hierarchy​/­. For an empirically informed argument that shows how status hierarchies more efficiently or­ga­nize patterns of be­hav­ior to produce larger public goods, see Brent Simpson, Robb Willer, and Cecilia L. Ridgeway, “Status Hierarchies and the Organ­ization of Collective Action,” So­cio­log­i­cal Theory 30, no. 3 (2012): 149–166. 13. Emily M. Zitek and Larissa Z. Tiedens, “The Fluency of Social Hierarchy: The Ease with Which Hierarchical Relationships Are Seen, Remembered, Learned, and Liked,” Journal of Personality and Social Psy­chol­ogy 102, no. 1 (2012): 98–115. 14. The use of psychedelic drugs, however, has the power to overturn (unconscious) hierarchies in the mind and sponsor unconventional thinking. See Michael Pollan, How to Change Your Mind: What the New Science of Psychedelics Teaches Us About Consciousness, ­Dying, Addiction, Depression, and Transcendence (New York: Random House, 2018), ch. 5). But overturning of hierarchies is not always a good ­thing. Ego dissolution and the feeling of oneness with the universe experienced ­under the influence of psychedelic drugs may undermine attachment to morally bad hierarchies such as racism and sexism, but such experiences may also have the effect of abolishing attachment to good forms of hierarchy. In the 1960s, LSD may have played a positive role in the United States by casting doubt on the virtues of military adventurism in Vietnam, but it also played a negative role by leading teen­agers to shed attachment to their parents. 15. Robert Wright, Why Buddhism Is True: The Science and Philosophy of Meditation and Enlightenment (New York: Simon and Schuster, 2017), 7. 16. Ibid., 177, 229. 17. Yuval Noah Harari, Sapiens: A Brief History of Humankind (New York: Harper, 2015), 136.

210  N o t e s t o I n t r o du c t i on 18. To be more precise, most ­people consciously condemn racism. Racism still rears its ugly head at the unconscious level. See American Psychological Association, “Speaking of Psy­chol­ogy: Understanding Your Racial Biases,” http://­www​.­apa​.­org​ /­research​/­action​/­speaking​-­of​-­psychology​/­understanding​-­biases​.­aspx. 19. Robert Bellah et al., Habits of the Heart: Individualism and Commitment in American Life (Berkeley: University of California Press, 1985). 20. Harold J. Leavitt, Top Down: Why Hierarchies Are H ­ ere to Stay and How to Manage Them More Effectively (Cambridge, MA: Harvard Business School Press, 2004). 21. One apparent advantage of traditional hierarchies with roles set at birth, and where every­one knows their place in the social pecking order, is that such arrangements seem to provide social stability and po­liti­cal order. As the case of France’s ancien régime shows, however, seemingly fixed roles often had gray areas that allowed for flexibility, and occupants of power­ful positions had to defend their positions against upstarts with new interpretations of the past. Hence, ­those with power had to stay ever vigilant, and the seemingly stable social order masked tensions that eventually exploded into revolution. See Fanny Cosendey, Le rang: Préséance et hiérarchies dans la France d’Ancien Régime (Paris: Éditions Gallimard, 2016), esp. 453–469. A similar story can be told about early Chinese history: The seeming stability of Legalist-­inspired social and po­liti­cal order provided by self-­proclaimed First Emperor of China Qin Shi Huang, with fixed roles for dif­fer­ent groups of ­people, masked instability that led to the collapse of the short-­lived Qin dynasty. 22. Mary Beard, SPQR: A History of Ancient Rome (New York: Liveright, 2015), 144. 23. The ­great Warring States Confucian thinker Xunzi argued that “from scholar officials (士) upwards, all must be moderated through ritual and m ­ usic; the commoners and the ‘hundred clans’ (百姓) must be restricted through law and methods [of rule]” (富国). The Han dynasty text Book of Rituals expressed a similar view that “common ­people and downwards should not be restricted by ritual and scholar-­ officials and upwards should not be subject to punishment” (礼记· 曲礼上). In the Song ­dynasty an implicit norm spared scholar-­officials from humiliating bodily ­punishments and execution, whereas common ­people ­were subject to such ­legal punishments. Ding Chuanjing, ed., 宋人轶事汇编 [Edited collection of anecdotes from the Song Dynasty] (Beijing: Zhonghua Shuju, 1981), 7–8. Thanks to Yuri Pines for references. 24. Michael Walzer, Thick and Thin: Moral Argument at Home and Abroad (Notre Dame, IN: University of Notre Dame Press, 1994). 25. Edmund Burke, Select Works of Edmund Burke, Vol. 2: Reflections on the Revolution in France (1790) (Indianapolis, IN: Liberty Fund, 1999). A few years l­ ater, Napoleon did precisely what Burke feared. Thanks to Jay Boggis for the reference.

N o t e s t o I n t r o du c t i o n   211 26. More precisely, it is not controversial to reward acts of “moral merit” with reactions of praise and admiration and symbolic trophies, but it is much more controversial to reward them with money. See Tim Scanlon, Why Does In­equality M ­ atter (Oxford: Oxford University Press, 2018), ch. 8. 27. Jean-­Jacques Rousseau, The Confessions, trans. J. M. Cohen (London: Penguin Books, 1953), 476. 28. In Chinese, by contrast, young c­ hildren and teen­agers rarely argue for equal treatment. 29. For a defense of the “doctrine of sufficiency” and a critique of economic egalitarianism, see Harry J. Frankfurt, On In­equality (Prince­ton, NJ: Prince­ton University Press, 2015). 30. Elizabeth S. Anderson, “What Is the Point of Equality?” Ethics 109 ( January 1999): 288. 31. Elizabeth S. Anderson, “Equality,” in The Oxford Handbook of Po­liti­cal Philosophy, ed. David Estlund (Oxford: Oxford University Press, 2012), 40. 32. For an argument that “progressive Confucians” also have good reasons for resisting oppression, see Stephen C. A ­ ngle, “Vicious Oppression, Valuable Deference, and Provisional Hierarchy: A Con­temporary Confucian Perspective,” paper presented at the Hierarchy and Equality Workshop, sponsored by the Berggruen Institute held at Stanford University’s Center for Advanced Study in the Behavioral Sciences, March 2016. 33. Joseph Chan, “Demo­cratic Equality or Confucian Hierarchy?,” paper presented at the Hierarchy and Equality Workshop sponsored by the Berggruen Institute held at Stanford University’s Center for Advanced Study in the Behavioral Sciences, March 2016. This paragraph draws on Chan’s paper. 34. Michael Walzer, Spheres of Justice: A Defense of Pluralism & Equality (Oxford: Basil Blackwell, 1983). For a similar method, see David Miller, Princi­ples of Social Justice (Cambridge, MA: Harvard University Press, 2001). 35. Canada may be an exception: the Progressive Conservative Party of Canada was a leading po­liti­cal party from 1942 u­ ntil its dissolution in 2003. Our use of the term is not meant to endorse or overlap with the center-­right po­liti­cal outlook of that po­liti­cal party. 36. One impor­tant exception is Liang Shuming (1893–1988), who expressed commitment to both traditional and socialist values. 37. For an attempt to define the key characteristics of Chinese culture, see Ge Zhaoguang, What Is China? Territory, Ethnicity, Culture, and History, trans. Michael Gibbs Hill (Cambridge, MA: Harvard University Press, 2018), 96–98. 38. One impor­tant counter-­current is virtue politics in the po­liti­cal thought of Re­ nais­sance Italy, which has striking parallels to Confucian-­style po­liti­cal meritocracy.

212  N o t e s t o I n t r o du c t i on See James Hankins, Virtue Politics: Po­liti­cal Thought in Re­nais­sance Italy from Petrarch to Machiavelli (Cambridge, MA: Harvard University Press, 2019), conclusion. 39. Thanks to Mattias Ottervik for a helpful email. 40. http://­www​.­12371​.­cn​/­2018​/­11​/­25​/­ARTI1543146320637564​.­shtml (the fourth article). 41. The dif­fer­ent ways airlines reward frequent flyers is another manifestation of the cultural difference between Western-­style preference for material hierarchies and East Asian–­style preference for social hierarchies. Gold card holders of Air Canada’s Aeroplan frequent flyer program receive material benefits in the form of f­ ree upgrades. In contrast, Platinum card holders of Air China’s Phoenix frequent flyer program receive social benefits in the form of personal attention by stewards or stewardesses who provide information about flight details (the material benefit tends to be l­ imited to a ­bottle of mineral ­water). 42. The theoretical exception might be Marx’s ideal of communism: a society with abundant resources that are fairly distributed and (thus) t­ here is no need for p­ eople to divide into dif­fer­ent classes or for the state to enforce a property system that benefits the ruling class. The closest modern approximation of Marx’s ideal might be super-­ wealthy Norway, which does indeed have a strong bias in f­ avor of both social and material equality. 43. Yan Yunxiang “Un­balanced Reciprocity: Asymmetrical Gift Giving and Social Hierarchy in Rural China,” in The Question of the Gift: Essays Across Disciplines, ed. M. Osteen (London: Routledge Press, 2002), 77–94. The modern Chinese word “suzhi” (素质), meaning “quality” in a literal sense, is often used to distinguish between categories of ­people in ways that may be shocking to the Western ear. “Suzhi” rankings usually refer to hierarchies of morality and civility (e.g., low “suzhi” ­people spit in public), but they can sometimes refer to somewhat vague hierarchical rankings of p­ eople (a Chinese friend at a leading American university once told Daniel that he could not get along well with his wife ­because of her lower “suzhi”: he considered himself a 9 out of 10, but she was a mere 7). 44. Li Jin, Cultural Foundations of Learning: East and West (Cambridge: Cambridge University Press, 2012). 45. Richard E. Nisbett, The Geography of Thought: How Asians and Westerners Think Differently . . . ​and Why (New York: ­Free Press, 2003). 46. Erin Meyer, The Culture Map: Decoding How ­People Think, Lead, and Get ­Things Done Across Cultures (New York: Public Affairs, 2014). 47. Stephen C. A ­ ngle, “Building Bridges to Distant Shores,” in Appreciating the Chinese Difference: Engaging Roger T. Ames on Methods, Issues, and Roles, ed. Jim Behuniak (Albany: SUNY Press, 2018), 165.

N o t e s t o C h a p t e r   1   213

Chapter 1. Just Hierarchy between Intimates: On the Importance of Shifting Roles 1. More precisely, they are fixed for one’s lifetime b­ ecause they can change over the course of several lives depending on one’s karma. 2. Zhang Xianglong, 家与孝： 从中西间视野看 [­Family and Filial Piety: From the Perspective of Chinese and Western Thought] (Beijing: SDX Joint Publishing Com­pany, 2017). 3. He Xinyin. 何心隐文集 [Collected Works of He Xinyin] (Beijing: Zhonghua Shuju, 1960), 27–28. 4. Ibid., 38. 5. Ibid., 65. 6. The Tang dynasty Confucian thinker Han Yu did argue for an interactive relation between teacher and student—­“弟子不必不如师， 师不必贤于弟子” [The students d­ on’t have to be inferior to the teachers, and the teachers d­ on’t have to be more virtuous and capable than the students], but he ­didn’t go so far as to claim that students and teachers can learn equally from each other or that the relation between teacher and friends, as modeled on the ideal of equal friendship, is the highest form of social relation. 7. Tian Wenjun, “何心隐新论” [A new analy­sis of He Xinyin’s thought], Wuhan Daxue Xuebao (Shehui Kexue Ban) 5 (1986): 15–21. 8. Quoted in Dennis C. Rasmussen, The Infidel and the Professor: David Hume, Adam Smith, and the Friendship that S­ haped Modern Thought (Prince­ton, NJ: Prince­ ton University Press, 2017), 5. 9. Ibid., 6. 10. Aristotle, Nicomachean Ethics Book VIII, 12, 7, 9. 11. Ibid. For a critique of Aristotle’s view, see Alexander Nehamas, On Friendship (New York: Basic Books, 2016). According to Nehamas, “Experience shows that morality is often irrelevant to, or even in conflict with, our love for our friends . . . ​not only do we love our friends despite their shortcomings but, sometimes, we love them ­because of them: think of the self-­importance or the forgetfulness that makes your friend so dear to you and so irritating to every­one ­else” (p. 27). He also points to the example of the movie Thelma and Louise “whose uncomfortable, well, moral is that friendship and immorality are often bound together” (p. 189). In China, t­ here is a rich tradition—­both literary and real—of criminals bound together in brotherly friendship. But the fact that immoral ­people can be good friends does not detract from Aristotle’s point that two equal friends bound by the common pursuit of virtue is a morally higher or more desirable form of friendship. 12. He’s and Aristotle’s conceptions of ideal friendship, however, are not identical. For He, the Confucian-­inspired view of the good life is a quest for constant

214  N o t e s t o C h a p t e r   1 self-­improvement, and friends learn from each other for the purpose of moral and intellectual improvement. For Aristotle, ideal friends must already be perfectly virtuous to befriend each other, so friends do not learn or improve from befriending each other. 13. Michel de Montaigne, On Friendship, trans. M. A. Screech (London: Penguin Books, 2004), 3. 14. Aristotle, Nicomachaen Ethics, Book VII, 1158a10. 15. Ibid., 6. 16. Quoted in Nehamas, On Friendship, 46. New research provides some evidence for Montaigne’s subjective experience of merging with his friend: “Scientists have found that the brains of close friends respond in remarkably similar ways as they watch a series of short videos: the same ebbs and swells of attention and distraction, the same peaking of reward pro­cessing ­here, boredom alerts t­ here. The neural response evoked by the videos . . . ​proved so congruent among friends, compared to patterns seen among ­people who ­were not friends, that the researchers could predict the strength of two p­ eople’s social bond based on their brain scans alone.” See Natalie Angier, “You Share Every­thing With Your Bestie: Even Brain Waves,” New York Times, April 16, 2018. 17. Emmanuel Levinas, Eros, Littérature, et Philosophie (Paris: Éditions Grasset, 2013): 173–174. 18. Ibid., 192. 19. Emmanuel Levinas, Totalité et Infini. Essai sur l’extériorité (Paris: Le livre de poche, 1990), 22. For more extensive discussion, see Wang Pei, A Study of Eros and Individuation in Levinas’s Early Philosophy, Ph.D thesis [in Chinese], Tsing­hua University, 2016. 20. Christopher Boehm, Hierarchy in the Forest: The Evolution of Egalitarian Be­hav­ ior (Cambridge, MA: Harvard University Press, 1999). Boehm draws the po­liti­cal implication that successful and long-­lasting h­ uman socie­ties involve some form of hierarchy along with mechanisms that limit abuses of power. In the modern world, national democracies where the rank and file band together to dominate their potential master along with institutionalized checks on power are most “in tune with ­human po­liti­cal and social nature” (ibid., 255). Boehm dismisses Marx’s ideal of communism ­because ­human hierarchical tendencies are too strong to allow dominant competition to evaporate and the state to wither away, but he does not consider the possibility that hierarchical po­liti­cal meritocracies with checks on power may also be desirable and realistic in modern large-­scale socie­ties (see chapter 2). 21. It’s worth noting that neither Hume nor Smith married—­and it’s pos­si­ble that neither experienced a physically intimate relation with a lover—so they w ­ ere not in a position to compare the ideals of friendship and love from their own experience. 22. We do not mean to imply that lovers cannot have intimate relations in the daytime. In ancient China, however, ­there was a taboo against daytime physical interaction.

N o t e s t o C h a p t e r   1   215 23. One of the negative outcomes of the period of economic reform in China is that ­women’s rights have often taken a turn for the worse. In Maoist days, with communal living and workers who ­were not expected to excel and get more in return, men and w ­ omen had relatively equal opportunities (more precisely, equally ­limited opportunities). But ­today, with family-­based homes where w ­ omen do most of the work, employers often prefer men who can (supposedly) devote themselves ­wholeheartedly to the job. 24. It was codified into a single text by the first c­ entury BCE and thereafter defined as one of the “Five Classics” that helped shape the educational curriculum throughout East Asia for much of the subsequent two millennia. 25. Quoted in (and translated by) Michael Puett, “Equality, Hierarchy and the Discontents: Theories from Classical China,” paper presented at Berggruen Workshop on Hierarchy and Equality, Center for Advanced Study in the Behavioral Sciences, Stanford University, March 13, 2016, 1. 26. Ibid. 27. Wendy Doniger and Sudhir Kakar, “Introduction,” in Kamasutra: A New, Complete En­glish Translation of the Sanskrit Text, translated and edited by Wendy Doniger and Sudhir Kakar (Oxford: Oxford University Press, 2009), xi. 28. Ibid., 8 29. Wendy Doniger, The Mare’s Trap: Nature and Culture in The Kamasutra (New Delhi: Speaking Tiger Publishing Pvt. Ltd., 2015), 151. 30. Doniger and Kakar, Kamasutra, 13, 14. 31. Ibid., 56. See Doniger, The Mare’s Trap, ch. 2, for an extensive discussion of sex and power in the Kamasutra. 32. Doniger and Kakar, Kamasutra, 146–147. 33. See also Doniger, The Mare’s Trap, ch. 5. 34. Ibid., 34, 58. 35. Ibid., 58. 36. Ibid., 60–61. Note that we do not condone vio­lence between intimate lovers, but if it’s voluntary on both sides (what we’d call sadomasochism t­ oday), the moral and ­legal implications are not so straightforward. 37. Ibid., 50. 38. Ibid., 64. 39. Ibid., 47. China’s earliest erotic texts put forward a similar view: see Richard Shusterman, “Sex and Somaesthetics: Appreciating the Chinese Difference,” in Appreciating the Chinese Difference, 104–105. 40. Doniger, The Mare’s Trap, 51. We do not mean to imply that role reversals should equalize in the long term. ­There may be a case for “extra” dominance by the female partner if it helps to challenge deeply held patriarchal norms and practices. 41. Ibid., 10.

216  N o t e s t o C h a p t e r   1 42. Ibid., 78. 43. Ibid., 42. 44. Mencius 长幼有序 (Teng Wen Gong I); Xunzi 长幼有差 (4.14–4.15). 45. In 1915–1919, however, the New Culture movement inverted the traditional veneration of the elderly—­“young ­people ­were perceived to be relatively uncorrupted by old traditions and hopefully not yet infected by a diseased culture and a sick society.” See Maurice Meisner, Mao’s China and ­After: A History of the P ­ eople’s Republic, 3rd ed. (New York: The ­Free Press, 1999), 15. This reversal culminated in the Cultural Revolution’s violent effort to destroy the Four Olds. With the rise of China’s economic and po­liti­cal strength and the revival of re­spect for tradition, such views are not widely held ­today. 46. Mencius (Liang Hui Wang I). 47. See Zhang Xianglong, “孝意识的时间分析” [A temporal analy­sis of the consciousness of filial piety], Journal of Peking University 43, no. 1, (2006): 14–24. 48. As it turns out, crows ­were praised for expressing the virtue of filial piety in traditional Chinese culture on the grounds that they brought food to el­derly parents who could not provide for themselves (Alan Chan and Tan Sor-­Hoon, Filial Piety in Chinese Thought and History [London: Routledge, 2004], 59). For Confucian thinkers, however, filial piety is not just about providing for material needs: It must be accompanied by re­spect for el­derly parents and emotions of love and care. 49. Tang Wenming, “人伦理念的普世意义及其现代调适” [The universal significance of the “Ren Lun” concept and its modern adjustment: On the defense and reconstruction of the Wu Lun princi­ples by modern Confucian scholars], Daode yu wenming 2015(6): 5–12. 50. It’s similar in other countries with a Confucian heritage: In Singapore parents can take their adult ­children to court if they fail to provide material support, a law that is justified with reference to the value of filial piety (see “Maintenance of Parents Act,” singaporeinfopedia, http://­eresources​.­nlb​.­gov​.­sg​/­infopedia​/­articles​/­SIP​_­1614​_­2009​ -­11​-­30​.­html). 51. Peter Wohlleben, The Inner Life of Animals (Vancouver/Berkeley: Greystone Books, 2017), 114. 52. The opening chapter of the Xunzi, “An Exhortation to Learning” makes a similar point. The line “Learning must never stop” is memorized by all Chinese students preparing for the national university entrance examinations. 53. This paragraph draws on Daniel A. Bell, The China Model (Prince­ton, NJ: Prince­ton University Press, 2016), 95–96. 54. Daniel Goleman, “What Makes a Leader?” in On Leadership (Boston: Harvard Business Review Press, 2011), 8. 55. Stephen S. Hall, Wisdom: From Philosophy to Neuroscience (New York: Vintage Books, 2010), 228–229.

N o t e s t o C h a p t e r   1   217 56. Ibid., 255. 57. This paragraph draws on Daniel A. Bell, China’s New Confucianism (Prince­ton, NJ: Prince­ton University Press, 2008), 153. 58. Zhang Taisu, “Social Hierarchies and the Formation of Customary Property Law in Pre-­Industrial China and ­England,” The American Journal of Comparative Law 62 (2014): 71. 59. On the other hand, the el­derly often seem to act in “selfish” ways in demo­cratic socie­ties, supporting policies that benefit the el­derly at the cost of ­future generations, and providing more formal and informal power to the el­derly may exacerbate ­these tendencies. 60. See Fang Zhaohui, “三纲真是糟粕吗?” [Are the “Three Cardinal Guides” ­really so worthless? Re-­examining the historical and practical significance of the Three Cardinal Guides], Tianjin Social Sciences 2 (2011): 48–49; Tang Wenming, “人伦理念的 普世意义及其现代调适” [The universal significance of the “Ren Lun” concept and its modern adjustment], Morality and Civilization 6 (2015): 9. 61. It would also be absurd to subject f­ amily elders to formal mea­sures designed to minimize nepotism and corruption. In imperial China (as well as t­ oday), public officials ­were assigned to posts outside of their home communities so as to decrease the likelihood of nepotism and corruption, but it would be inhumane to assign power­ful ­family elders to dif­fer­ent families. 62. Li Yuxiu, “Students’ Rules,” http://­tsoidug​.­org​/­dizigui​_­trans​_­simp​.­php, 24–27. 63. ­There may be exceptions to the rule: For example, a parent may be a permanent caretaker of a severely disabled child with a relatively short life span. Such arrangements can be morally justified if they are informed by love and care and ­free of oppression and domination. But hierarchies between intimates who, at some point in time, develop into adults with the capacity to reason typically involve shifting hierarchies. 64. It could be argued that role reversals induced by Alzheimer’s disease are not morally justified ­because they are not voluntary on the side of the parent. But an el­ derly parent at an early stage of Alzheimer’s can “sign on” to this kind of role reversal in advance. In any case, we do not mean to imply that all role reversals need to be voluntary: In the case of shifting “nighttime hierarchies,” for example, the role reversals may not be voluntary at the time they take place, even if both parties might need to endorse them ­after the fact if pressed to do so. 65. In traditional China, however, the dominance of husbands over wives was accompanied by high status in the ­family for ­mothers: The value of filial piety served to increase w ­ omen’s status when she become a m ­ other b­ ecause a son would typically defer to his ­mother. 66. Aristotle, The Politics and The Constitution of Athens, ed. Stephen Everson, trans. Benjamin Jowett (Cambridge: Cambridge University Press, 1996), 1253b35–1254a1.

218  N o t e s t o C h a p t e r   1 67. Ibid., 1255a5–10. 68. Ibid., 1255a22–25. 69. Ibid., 1255a25–28. 70. Ibid., 1254a21–23. 71. Mencius, Teng Wen Gong I, http://­ctext​.­org​/­mengzi​/­teng​-­wen​-­gong​-­i (translation has been slightly modified). 72. The context of the passage is a discussion of rulers who spend time cooking their own meals and cultivating the fields along with the ­people. Mencius suggests that wise and virtuous rulers should not waste time on such activities and should instead focus on ruling. 73. Mencius also defends the well-­field system, an equal distribution of land at the community level with a mixture of plots for public and private (family-­based) use and that aims to promote a sense of mutual caring among villa­gers (Teng Wen Gong I). 74. Ibid., 1254a34–36, 1254b3–10, 1254b16–24. 75. Ibid., 1254b38–1255a1. 76. Christopher Shields, Aristotle (London and New York: Routledge, 2007), 14. 77. Aristotle, The Politics and The Constitution of Athens, 1255b27–28. 78. This section draws on Daniel A. Bell, China’s New Confucianism (Prince­ton, NJ: Prince­ton University Press, 2008), ch. 5. 79. The Analects of Confucius, 2.8 [translation modified]. 80. Confucius may have highlighted the need to serve el­derly parents with love precisely ­because it’s less “natu­ral” than serving ­children with love. 81. Mencius, Liang Hui Wang I [translation modified]. 82. In the famous Chinese novel A Dream of Red Mansions (红楼梦) by Cao Xueqin (a descendant from a declining aristocratic ­family in the Qing dynasty), the servants in the Jia Mansion often played an active role in ­family life. The old nannies with a long history of serving the Jia f­ amily ­were treated with special re­spect. In ancient China, nannies often breastfed the ­children of aristocratic families and remained physically close to the ­children, and the ­children ­were supposed to show gratitude to the nannies during the course of their education as virtuous aristocrats. In this novel, some young “masters” disliked their imperious nannies, but they did not dare to be openly rude to them. At f­ amily gatherings, the old servants who served the first generation of the ­family had the privilege of sitting down, while the young masters could only stand by the side. 83. This paragraph draws on Bell, China’s New Confucianism, ch. 5. 84. ­There may be additional forms of exploitation of personal servants who work outside the home, such as when Don Giovanni (in Mozart’s opera) demands that his servant Leporello help or­ga­nize his sexual escapades and cover up his (resulting) crimes of rape and murder. But Leporello was not entirely innocent: He knew he was helping with immoral deeds, and he accepted ­because he was given more money.

N o t e s t o C h a p t e r   2   219 85. In other cultures, dif­fer­ent norms may help to morally regulate the relation between h­ ouse­keeper and employer. In E ­ ngland’s aristocratic culture, butlers concerned themselves with the moral status of the employer and the most respected professional butlers selected employers who made impor­tant and morally desirable contributions to public affairs (for a fictitious account, see Kazuo Ishiguro, The Remains of the Day [London: Faber and Faber, 1989]). 86. More realistic, perhaps, is the possibility of role exchanges between teacher and student over the course of a lifetime. The Confucian view of the student-­ teacher relationship is that the teacher needs to be prepared that the student ­w ill eventually surpass the teacher, and serve as the former teacher’s teacher, at least in some domains of learning (see Joseph Chan, “Demo­cratic Equality or Confucian Hierarchy?” paper presented at Berggruen Workshop on Hierarchy and Equality, Center for Advanced Study in the Behavioral Sciences, Stanford University, March 13, 2016, 4–5). 87. Given that wealth and opportunities tend to get transmitted from generation to generation in cap­it­ al­ist socie­ties, however, such generational shifts of roles are highly unlikely.

Chapter 2. Just Hierarchy between Citizens: On the Importance of Ser­vice 1. See, e.g., Mogens Herman Hansen, The Athenian Democracy in the Age of De­mos­ the­nes: Structure, Princi­ples, and Ideology (Oxford: Blackwell, 1991). 2. Sarah Conley draws on extensive evidence from social psy­chol­ogy and behavioral economics to show that our irrational choices often undercut the achievement of our own goals, and she draws the po­liti­cal implication that we could advance our goals more effectively if government ­were to prevent us from acting in accordance with our decisions by means of such policies as banning cigarettes and forced savings. But she inexplicably asserts that “what we need is a demo­cratically elected government” to implement “coercive paternalism.” See Sarah Conley, Against Autonomy: Justifying Coercive Paternalism (Cambridge: Cambridge University Press, 2013), 2. For discussion that focuses more explic­itly on the irrationality of voters, see Bryan Caplan, The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Prince­ton, NJ: Prince­ton University Press, 2007); and Christopher H. Aden and Larry M. Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (Prince­ ton, NJ: Prince­ton University Press, 2016). Jason Brennan argues that voter irrationality is a good reason to rethink the value of electoral democracy in his two books, The Ethics of Voting (Prince­ton, NJ: Prince­ton University Press, 2011) and Against Democracy (Prince­ton, NJ: Prince­ton University Press, 2017). Chapter 1 of Daniel’s book The China Model (Prince­ton, NJ: Prince­ton University Press, 2015) asks w ­ hether

220  N o t e s t o C h a p t e r   2 the theory and practice of po­liti­cal meritocracy can help to remedy four key flaws of electoral democracy. Samuel Bagg recognizes that it is difficult if not impossible to substantially upgrade the po­liti­cal competence of citizens in modern-­day electoral democracies but that electoral democracy should instead be justified on the grounds that demo­cratic states are more resistant to dangerous forms of capture than nondemo­ cratic alternatives. See Samuel Bagg, “The Power of the Multitude: Answering Epistemic Challenges to Democracy,” American Po­liti­cal Science Review 12, no. 4 (2018): 891–904. The rest of this chapter attempts to show that Chinese-­style po­liti­cal meritocracy can be improved so as to avoid po­liti­cal ossification and allow for more demo­cratic participation without electoral democracy at the top. 3. Yves Sintomer, “From Deliberative to Radical Democracy? Sortition and Politics in the Twenty-­First ­Century,” Politics and Society 46, no. 3 (2018): 337–357; Wang Shaoguang, 民主, 共和与抽签: 从雅典到威尼斯 (Democracy, republic and sortition: From Athens to Venice) (Beijing: CITIC Press, 2018). 4. Niall Ferguson, The Square and the Tower: Networks and Power, from the Freemasons to Facebook (New York: Penguin Press, 2017), 21. 5. Ibid., 62. 6. Zhao Dingxin, The Confucian-­Legalist State: A New Theory of Chinese History (Oxford: Oxford University Press, 2015), 13. 7. Yuri Pines, “Between Merit and Pedigree: Evolution of the Concept of ‘Elevating the Worthy’ in Pre-­imperial China,” in The East Asian Challenge for Democracy: Po­liti­cal Meritocracy in a Comparative Context, ed. Daniel A. Bell and Chenyang Li (New York: Cambridge University Press, 2013), 161–202. 8. This section draws on Wang Pei, “Debates on Po­liti­cal Meritocracy in China: A Historical Perspective,” Philosophy and Public Issues (new series) 7, no. 1 (2017): 63–71. 9. Zhao, The Confucian-­Legalist State, 14. 10. A more historically precise terminology might be the “Mohist-­Confucian” ideal of po­liti­cal meritocracy, since the “Legalist” idea that public officials should be selected mainly according to administrative capability (rather than moral qualities) can be traced to Mozi. See Yuri Pines, “Pitfalls of Meritocracy: Elevating the Worthy in Early Chinese Thought,” paper presented at workshop on po­liti­cal meritocracy, Harvard University, November 2018. 11. See Yao Yang, “An Anatomy of the Chinese Selectocracy,” https://­www​ .­research​gate​.­net​/­publication​/­294883730​_­An​_­Anatomy​_­of​_­the​_­Chinese​_­Select​ ocracy; Zheng Yongnian, The Chinese Communist Party as an Orga­nizational Emperor: Culture, Reproduction, and Transformation (London: Routledge, 2010); and Pan Wei, ed. 中国模式: 解读人民共和国的60年 [China model: Explaining sixty years of the ­People’s Republic] (Beijing: Central Compilation and Translation Press, 2009).

N o t e s t o C h a p t e r   2   221 12. Francis Fukuyama argues that Denmark is the country that comes closest to realizing the ideal of liberal democracy. See Francis Fukuyama, Po­liti­cal Order and Po­ liti­cal Decay: From the Industrial Revolution to the Globalization of Democracy (New York: Farrar, Straus and Giroux, 2015). But it seems absurd to suggest that the po­liti­ cal system of a relatively homogenous, well-­off country of 5.7 million ­people surrounded by small, friendly neighbors should be used as the benchmark for assessing po­liti­cal success in large-­scale po­liti­cal communities with hundreds of millions of ­people (if Denmark ­were to be compared with the most well-­off sections of Shanghai, the differences might not seem so stark). 13. Liang Shuming, 乡村理论建设 [Rural construction theory] (Beijing: Shiji Chubanshe, 2006; orig. pub. 1937), 124. According to Liang, the tradition of respecting the sage w ­ ill continue to be influential b­ ecause the development of science and technology would lead to a new tradition (in the West as well) with every­one taking the advice of sages and the decision by the minority would be willingly accepted by the majority. Other influential early twentieth-­century intellectuals also rejected “one person, one vote” for modern China: Yang Du, Kang Youwei, and Liang Qichao argued that national se­lection by means of one person, one vote would lead to po­liti­cal chaos and division by power­ful countries (Wang Hui, 世纪的诞生 [The birth of the ­century], Kaifeng Shidai, 4 [2017]), http://­www​.­opentimes​.­cn​/­Abstract​/­2365​.­html. 14. See Bell, The China Model, 147. 15. It is hard to imagine a po­liti­cally realistic case for Western socie­ties given the almost universal consensus in ­favor of one person, one vote as a way of selecting leaders, not to mention the dramatic rise of antielitist (aka “populist”) sentiment of late. But it ­hasn’t always been the case: Plato’s Republic is a defense of po­liti­cal meritocracy and the nineteenth-­century liberal thinker John Stuart Mill favored extra votes for educated ­people. Such proposals may become ­viable again if Western-­style democracies fail to perform well in comparison to Chinese-­style po­liti­cal meritocracies. For a normative case in ­favor of injecting po­liti­cal meritocracy in the demo­cratic politics of Western socie­ties, see Elena Ziliotti, “The Meritocratic Challenge to Demo­cratic Theories,” Ph.D. thesis, National University of Singapore and King’s College, London, 2018. 16. Think of the challenge of AI (see chapter 5 for discussion). 17. Po­liti­cal surveys, however, show that most Chinese interpret democracy to mean “politics for the ­people” rather than “politics by the p­ eople.” See Shi Tianjian, The Cultural Logic of Politics in Mainland China and Taiwan (New York: Cambridge University Press, 2015), ch. 7. We work with the definition of politics by the p­ eople ­because nondemo­cratic po­liti­cal systems can (theoretically speaking) also implement the princi­ple of politics for the ­people. 18. This discussion draws on Daniel Bell, “Why China Guards Its Se­lection Secrets so Jealously,” Financial Times, October 30, 2017.

222  N o t e s t o C h a p t e r   2 19. We do not mean to imply that t­ hese prob­lems are unique to po­liti­c al meritocracies—­the abuse of power in the form of corruption, for example, may be just as bad in electoral democracies (think of Malaysia)—­but they pose more of an existential threat in a system without the safety valve of competitive elections (see Bell, The China Model, ch. 3 for discussion). 20. See Liu Zehua, “王权主义：中国思想文化的历史定位” [Monarchy: A historical orientation of Chinese intellectual culture], Tianjin shehui kexue 3 (1998): 59–62. 21. Discussion with Wang Shaoguang at Shandong University, April 27, 2018. 22. This discussion draws on Daniel Bell, “China’s Corruption Clampdown Risks Policy Paralysis,” Financial Times, May 2, 2017. 23. Wang Qian and Zhang Yan, “Drunken Driving Crashes, Injuries Declining,” China Daily, October 10, 2014. 24. This example suggests that the Chinese expression “先礼后兵” (literally, first ritual, then force, but meaning something like, first Confucian-­style soft power, then Legalist-­style hard power) may have it backward. But the fear of harsh punishment in the short term can only help to transform inner morality in the long term if the initial fear of punishment goes along with a commonly held social value (­people knew that drinking and driving is bad, but they only paid attention to this value once it was backed up by harsh punishment for violations). 25. The government ­doesn’t openly acknowledge the use of Legalist means, but public officials are more frank in private conversation. 26. See Gong Ting and Wu Muluan, “我国 2000–2009 年腐败案例研究报告：基于 2800 余个报道案例的分析” [Research report on China’s corruption cases between 2000 and 2009: An empirical analy­sis of 2800 cases], Shehuixue yanjiu 4 (2012): 204– 220. That’s not to say the w ­ hole system was corrupt prior to the anticorruption drive. The national university entrance examination system (高考) has been perceived as relatively fair in the reform era and helps to underpin po­liti­cal stability (put negatively, if the examination system becomes widely seen as corrupt and unfair, it may endanger the ­whole po­liti­cal system). 27. “China’s Anti-­Corruption Campaign Expands with New Agency,” BBC News, March 20, 2018, https://­www​.­bbc​.­com​/­news​/­world​-­asia​-­china​-­43453769. 28. For an empirically informed argument that the campaign is indeed an attempt to root out systemic corruption prob­lems and strengthen party-­led meritocracy, see Xi Lu and Peter Lorentzen, “Rescuing Autocracy from Itself: China’s Anti-­Corruption Campaign,” November 6, 2016​.­ https://­ssrn​.­com​/­abstract​=­2835841 or http://­dx​.­doi​ .­org​/­10​.­2139​/­ssrn​.­2835841. In a ­later paper, however, the same authors also found that individuals with personal ties to Xi Jinping appear to be exempt from investigation. See Peter Lorentzen and Xi Lu, “Personal Ties, Meritocracy, and China’s Anti-­ Corruption Campaign,” November 21, 2018, https://­papers​.­ssrn​.­com​/­sol3​/­papers​ .­cfm​?­abstract​_­id​=­2835841.

N o t e s t o C h a p t e r   2   223 29. See Henny Sender, “Xi’s Crackdown on Corruption Is a Boon to Corporate China,” Financial Times, April 4, 2017. 30. Sebastian Heilmann, “Policy Experimentation in China’s Economic Rise,” Studies in Comparative International Development 43, no. 1 (2008): 1–26. 31. That’s not to say that the campaign has been completely effective. U ­ ntil the day China becomes so wealthy that the country can afford Singapore-­style salaries for all its public officials, a certain degree of corruption is likely to remain a permanent feature of the po­liti­cal landscape. In the Tang dynasty, corruption was even viewed as a necessary feature of the afterlife: Relatives of the dead would burn paper that could be transformed “into ethereal ­matter that the dead could employ in the afterlife, especially to reduce their suffering by bribing officials in the underworld.” Charles Benn, China’s Golden Age: Everyday Life in the Tang Dynasty (Oxford: Oxford University Press, 2002), 271. 32. The enemies include external enemies. The Trump administration’s “trade war” against China has only exacerbated the sense of paranoia b­ ecause of the sense (widely held among Chinese po­liti­cal elites) that the U.S. government aims not just to establish a more fair trading system, but also to curb China’s development and international influence, if not overthrow the w ­ hole Chinese po­liti­cal system. When Stephen Bannon, former chief strategist for Donald Trump, openly calls for a fight with the CCP—­“the greatest existential threat ever faced by the United States,” https://­www​ .­washingtonpost​.­com​/­opinions​/­steve​-­bannon​-­were​-­in​-­an​-­economic​-­war​-­w ith​ -­china​-­its​-­futile​-­to​-­compromise​/­2019​/­05​/­06​/­0055af36​-­7014​-­11e9​-­9eb4​-­0828f5389013​ _­story​.­html​?­utm​_­term​=­​.­604f157e7ed0—the Chinese government reacts by mobilizing the p­ eople for some sort of lasting conflict with the United States, with less room for dissent and alternative viewpoints. A more cooperative relationship with the United States would also allow for a more open society in China. 33. Li Yongzhong of the China Acad­emy of Supervision and Discipline Inspection proposed that amnesty be granted to officials who come clean. Zhang Weiying of the Guanghua School of Management also argued that amnesty might rescue the party. 34. Ray Huang, 1587, A Year of No Significance: The Ming Dynasty in Decline (New Haven: Yale University Press, 1981), 76. 35. As of late 2019, the anticorruption drive does indeed seem to be winding down. 36. Communist Party schools including the Central Party School increasingly teach Confucian classics to cadres. 37. The current policy in Xinjiang may be another manifestation of the “First Legalism, then Confucianism” approach to dealing with pressing social and po­liti­cal challenges. To address the threat of extremist religion and foreign-­funded separatist movements, the government embarked on a Legalist-­inspired policy of imprisoning hundreds of thousands of Uyghurs in “re-­education centers.” Just as the heavy-­handed

224  N o t e s t o C h a p t e r   2 anticorruption campaign has implicated several hundred thousand public officials without always clearly distinguishing between “good” and “bad” officials, so too the current “pre-­emptive” campaign against religious extremism in Xinjiang does not always distinguish between “good” and “bad” ­people, and the longer it lasts, the more resentment it ­will cause. ­There may be a good case for job training, Mandarin Chinese language teaching, education in Chinese law, and re-­interpretation of history that shows the positive contributions of the Uyghur ­people and the Islamic religion to the idea of “Chineseness” so as to increase attachment of the Uyghur ­people to the po­ liti­cal community, but such mea­sures are not likely to be effective if they rely on heavy-­ handed coercion for extended periods. The most optimistic outcome is that the government ­will soon move from “Legalist” coercion to “Confucian” soft power in Xinjiang. 38. Xunzi, 9.1 (see also 9.4) [translations of Xunzi draw on John Knoblock’s translation of Xunzi (Changsha: Hunan Publishing House, 1999), but they have been modified]. 39. This section draws on Wang Pei, “Debates on Po­liti­cal Meritocracy in China: A Historical Perspective,” Philosophy and Public Issues (new series) 7, no. 1 (2017): 53–71. 40. Lao Gan, “汉代察举制度考” [On the recommendation system in the Han Dynasty], “中研院历史语言研究所集刊,” 17 (April 1948): 79–129. 41. Yan Buke, 阎步克，波峰与波谷：秦汉魏晋南北朝的政治文明 [Peaks and valleys: Po­liti­cal cultures of the Qin, Han, Wei, Jin, and Northern and Southern Dynasties] (Beijing: Peking University Press, 2018), 91. 42. Filial piety became even more significant in the Eastern Han dynasty. 43. Fan Ye, 后汉书· 荀爽列传 [Book of the L ­ ater Han: Biographies of Xunshuang] (Beijing: Zhonghua Shuju, 2007). 44. Hu Pingsheng, 孝经译注 [On The Classic of Filial Piety] (Beijing: Zhonghua Shuju, 1996). 45. Bai Xiaoyong, “易学与东汉政治初探” [On the relation between the study of the Book of Changes and politics in the Eastern Han Dynasty], Zhejiang Academic Journal 1 (2013): 42–47. 46. Tang Changru, 九品中正制度试释 [On the nine-­rank system] (Beijing Zhonghua Shuju, 2011); Chen Lingguo, “两晋九品中正制与选官制度” [The nine-­rank system in the Western and Eastern Jin Dynasty], Lishi yanjiu 3 (1987): 101–115; Hu Baoguo, “关于九品中正制的几点意见” [A few points on the nine-­rank system], Lishi yanjiu 1 (1988): 191–192; Hu Shuyun, “九品官人法性质辨析” [The nature of the nine-­rank system], Dongbei shifan daxue bao 6 (2003): 78–84. 47. Shen Yue, “宋书· 列传· 卷九十四” [Song Shu, Biographies, vol. 94] (Beijing, Zhonghua Shuju, 1974).

N o t e s t o C h a p t e r   2   225 48. Fang Xuanling, “晋书· 列传· 第十五章” [ Jin Shu, biographies, vol. 15] (Beijing, Zhonghua Shuju, 1974). 49. Li Cunshan, “朱子‘学校贡举私议’ 述评” [On Zhu Xi’s comments about chief examiners in academic schools], Journal of the Gradu­ate School of the Chinese Acad­emy of Social Sciences 2 (2011): 19–27. 50. Pan Fuen and Xu Yuqing, “论二程的人才观” [On the two Chengs’ view of talents], Lanzhou daxue xuebao 1 (1987): 35–41. 51. But for an empirically informed argument that examination success in the Tang dynasty did successfully test for virtue, see Bernard Yeung, “Notes on Meritocracy: Insights from Tang’s Civil Ser­vice Exam and Poetry,” paper presented at “Meritocracy in China and India Workshop,” Harvard Center Shanghai, May 16–17, 2018. 52. More precisely, a distorted form of po­liti­cal meritocracy became an unjust po­ liti­cal hierarchy that needed to be remedied by a new interpretation and institutionalization of po­liti­cal meritocracy. 53. See Bell, The China Model, ch. 3, sections  1 and 2 for more extensive discussion. 54. Joseph Chan, Confucian Perfectionism: A Po­liti­cal Philosophy for Modern Times (Prince­ton, NJ: Prince­ton University Press, 2014), 36. 55. Quoted in ibid., 37. 56. See Bell, The China Model, chapter 3, section 3, for more extensive discussion. 57. The need for a more participatory hierarchy is not unique to China’s po­liti­cal system: Terence Nichols argues that the Roman Catholic church also needs to adopt a more participatory hierarchy. Terence Nichols, “Who’s Afraid of Hierarchy? The ­People of God Have to Be Or­ga­nized: H ­ ere’s How,” Commonweal (April 7, 2000): 16–18. Singapore may be a positive example for both China and the Roman Catholic Church: The government relied on rule by an elite that excluded the rest of the population from po­liti­cal participation and relied on heavy-­handed repression and censorship to maintain its rule, but it has peacefully evolved into a much more open and participatory po­liti­cal community over the past few de­cades, all the while maintaining its commitment to po­liti­cal meritocracy. That said, po­liti­cal reform of this sort may be easier to successfully accomplish in a small po­liti­c al community such as Singapore. 58. Media report mentioned in Liu Yawei, “Preface,” China Elections and Governance Review 1, no. 1 (February 2009): 1. 59. He Baogang, “A New Interpretation of the China Model: In Search of Mixed Governance in Con­temporary China,” paper presented at the 2019 IPP International Conference on Civilization and Governance, Guangzhou, August 17–18, 2019. 60. See http://­www​.­bjreview​.­com​.­cn​/­quotes​/­t xt​/­2007​-­02​/­28​/­content​_­57047​ .­htm.

226  N o t e s t o C h a p t e r   2 61. This is not to imply that China should adopt ways of protecting individual rights typical of litigious socie­ties such as the United States: Joseph Chan has argued that Confucian ethics widely shared in China justifies more reliance on informal mechanisms such as mediation designed to restore (or not worsen) harmonious relations, with rights as a fallback apparatus when other mechanisms fail. Chan, “A Confucian Perspective on H ­ uman Rights,” in The East Asian Challenge for ­Human Rights, ed. Joanne R. Bauer and Daniel A. Bell (Cambridge: Cambridge University Press, 1999), 212–240. 62. In The China Model, Daniel proposes the idea of a referendum on “vertical demo­cratic meritocracy” that would be an explicit form of consent by the p­ eople. This proposal assumes that the electorate tends to be unusually well informed when they vote in referenda on major constitutional issues compared to voting patterns in regular demo­cratic elections, as seems to have been the case with the two referenda on the question of Quebec in­de­pen­dence. Since then, however, Brexit has shaken Daniel’s faith in referenda. If the electorate in the world’s most mature democracy can vote in less-­than-­rational ways—­actual interaction with Eu­ro­pean mi­grants in the UK inversely correlated with votes in ­favor of Brexit—­why should we expect voters to be more sensible in a relatively poor country without a long history of democracy? So if t­ here is a referendum on vertical demo­cratic meritocracy in China, perhaps it should be accompanied by meritocratic checks, such as a ­simple multiple choice test on po­ liti­cal options set by in­de­pen­dent experts. In any case, this proposal for a referendum seems far-­fetched in the con­temporary Chinese context. 63. Zhang Yongle, Journal of Chinese Humanities 4, trans. Colleen Howe, no. 1 ­(August 2018): 58. 64. Ibid., 58–59, 59. 65. Ibid., 61. 66. In the Cultural Revolution, millions of educated urbanites had to spend extended periods in the countryside to be “educated” by farmers. It was a miserable experience for many intellectuals (especially ­because they had no indication of when they could return to the cities), but it did have the positive consequence that public officials in charge of the initial period of economic reform had experience with the countryside and a good feel for the needs of the farmers (we thank Wang Hui for this insight). In more recent years, cadres lost touch with ordinary ­people and ­were more likely to implement policies insensitive to their needs. 67. We use scare quotes to cast doubt on the term “nonpo­liti­cal” ­because Confucian thinkers also point to the po­liti­cal significance of family-­based virtues (see, e.g., The Analects of Confucius, 2.21) that ancient Greek and liberal thinkers traditionally regarded as private and nonpo­liti­cal family-­based virtues. But Confucian thinkers still tend to think that the best life involves serving the public as a government official, so

N o t e s t o C h a p t e r   2   227 work outside the government has implicitly been downgraded in terms of its social value. 68. This section draws on Cao Feng, “Pre-­Qin Daoist Reflections on the Xianneng [Po­liti­cal Meritocracy],” Journal of Chinese Humanities 4, trans. Caterina Weber, no. 1 (August 2018): 65–90. 69. Quoted in ibid., 69. 70. Quoted in ibid., 69. 71. Quoted in ibid., 74, 72, 75, 75, 76, 76. 72. Ibid., 77. 73. Quoted in ibid., 77, 81, 85. 74. This section draws on Daniel A. Bell, “Vertical Demo­cratic Meritocracy in China: Response to Comments,” Journal of Chinese Humanities 4, no. 1 (August 2018): 111–123. 75. Simon May traces the origin of the culture of cuteness to Walt Disney’s Mickey Mouse in the 1920s, but the East Asian (more intense) variation took off from Japan in the 1970s. See Simon May, The Power of Cute (Prince­ton, NJ: Prince­ton University Press, 2019), chs. 4–5. See also Big Think, “Why Do Japa­nese ­People Love, Cuteness? Learn the Science of ‘Kawaii’ .” http://­bigthink​.­com​/­paul​-­ratner​/­why​-­do​-­the​-­japanese​ -­love​-­cute​-­things. The culture of cuteness, especially the use of animated images in digital conversation, may be more widespread and long lasting in Confucian-­influenced East Asian countries that prioritize politeness and indirect talk b­ ecause online communication is inherently straightforward and p­ eople like to use funny images to relax the communication atmosphere and minimize the risk of hurting other ­people’s feeling. See Serenitie Wang and Joseph Kleinhenz, “How an Eight-­Year-­Old American Boy Became a Viral Sensation in China, CNN, November 9, 2018. 76. In Shanghai’s not-­so-­cold winters, it has become almost compulsory to clothe cute small dogs with colorful jackets, to the point that it becomes somewhat jarring to spot a “naked” dog walking the streets. 77. This is not meant to be a criticism. One of the prob­lems with traditional email is that messages cannot convey emotions and hence could be easily misunderstood. But now we can add a smiley face or other such symbols to convey the emotions meant to accompany our messages, which reduces the risk of misunderstandings. 78. Hiroshi Nittono, Michiko Fukushima, Akihiro Yano, and Hiroki Moriya, “The Power of Kawaii: Viewing Cute Images Promotes a Careful Be­hav­ior and ­Narrows Attentional Focus,” PlosOne, September  26, 2012, http://­j ournals​ .­p los​ .­o rg​ /­plosone /­article​?­id​=­10​.­1371​/­journal​.­pone​.­0046362. Thanks to Julien Bell for a helpful email. 79. May, The Power of Cute, 9. It’s worth noting, however, that the culture of cuteness in its extreme form entails brutality ­toward animals: Some cat ­owners in China

228  N o t e s t o C h a p t e r   3 make their pets undergo a painful operation to cut single eye lids into larger (and cuter) double eyelids. “有人竟花1万给猫拉了双眼皮” [Believe it or not, somebody spent 10,000 rmb on double eye-­lids for cats], Guancha Syndicate, February 27, 2019. 80. If the culture of cuteness is (at least partly) a reaction against an ultracompetitive meritocratic po­liti­cal system underpinned by an ultracompetitive educational system, one might expect the culture of cuteness not to have substantial social impact in more easygoing, less competitive socie­ties. This hypothesis is supported by the fact that the culture of cuteness has had ­little social impact in the world’s happiest countries such as Norway and Denmark.

Chapter 3. Just Hierarchy between States: On the Need for Reciprocity 1. David A. Lake, Hierarchy in International Relations (Ithaca and London: Cornell University Press, 2009), 3. See also Jonathan Renshon, Fighting for Status: Hierarchy and Conflict in World Politics (Prince­ton, NJ: Prince­ton University Press, 2017); and John M. Hobson and J. C. Sharman, “The Enduring Place of Hierarchy in World Politics: Tracing the Social Logics of Hierarchy and Po­liti­cal Change,” Eu­ro­pean Journal of International Relations 11 (1) (2005): 63–98; and Alexander Wendt and Daniel Friedheim, “Hierarchy u­ nder Anarchy: Informal Empire and the East German State,” International Organ­ization 49, no. 4 (1995): 689–721. 2. According to “hegemonic stability theory,” the international system is more likely to remain stable when a single state is the dominant power. See Joshua Goldstein, International Relations (New York: Pearson-­Longman, 2005), 107. What­ever the truth of this theory in the past, it is no longer an option given the rise of China. But perhaps the international system is more likely to be stable if two rather than several states are dominant global powers. 3. Yan Xuetong, Ancient Chinese Thought, Modern Chinese Power (Prince­ton, NJ: Prince­ton University Press, 2011), 105. 4. Lake, Hierarchy in International Relations, 138. 5. Yan, Ancient Chinese Thought, Modern Chinese Power, 105. 6. Lane, Hierarchy in International Relations, 178. 7. See Kautilya, King, Governance, and Law in Ancient India: Kautilya’s Arthasastra, trans. Patrick Olivelle (New Delhi: Oxford University Press, 2013). 8. Ibid., 401–402. 9. Olivelle, “Introduction,” in Kautilya, King, Governance, and Law in Ancient India: Kautilya’s Arthasastra, 48. 10. Mao Zedong, “Quotations,” https://­www​.­marxists​.­org​/­reference​/­archive​/­mao​ /­works​/­red​-­book​/­index​.­htm.

N o t e s t o C h a p t e r   3   229 11. Kautilya, King, Governance, and Law in Ancient India: Kautilya’s Arthasastra, 302. 12. Ibid., 283. 13. Quoted in Patrick Olivelle, “Relations between States and Rulers in Ancient India: Asoka [aka Ashoka], Kautilya, and Manu,” 13. Paper presented at the workshop on “Classical Indian & Chinese World Views on Global Order: A Comparison,” Berggruen Institute, January 28–31, Bangkok 2018. 14. Ibid., 9. 15. Mark McClish and Patrick Olivelle (eds. and trans.), “Introduction,” Kautilya, The Arthasastra: Se­lections from the Classic Indian Work on Statecraft (Indianapolis: Hackett, 2012), xlix. 16. Quoted in Patrick Olivelle, “Relations between States and Rulers in Ancient India: Asoka, Kautilya, and Manu,” 13, 18. 17. Quoted in Theodore de Bary et al., eds. Sources of Indian Tradition (Delhi: Motilal Banardidass, 1963), 153. 18. Quoted in Olivelle, “Relations between States and Rulers in Ancient India,” 18, 19. 19. Edict XII, quoted in Rajeev Bhargava, “Asoka’s Dhamma as Civic Religion: Toleration, Civility, Communal Harmony,” paper presented at the workshop on “Classical Indian & Chinese World Views on Global Order: A Comparison,” Berggruen Institute, Bangkok, January 28–31, 2018, 58–59. 20. Ibid., 47. 21. Rajeev Bhargava draws a parallel with the Confucian ideal of harmony (see Bhargava, “Asoka’s [aka Ashoka] Dhamma as Civic Religion”). 22. Arguably, Xunzi’s student Han Feizi, who systematized the “Legalist” tradition in ancient China, is a more consistent amoral realist than Kautilya. See Daniel A. Bell, Beyond Liberal Democracy: Po­liti­cal Thinking for an East Asian Context (Prince­ton, NJ: Prince­ton University Press, 2006), ch. 8. Han Feizi, however, went rogue. He systematically rejected key assumptions of his teacher’s theory—­that ­human nature can be improved, that ­legal punishment should come second to rituals, and that “humane authority” should set the moral standard for evaluating rulers—­and proposed cruel policies, such as killing Confucian thinkers, that would have resulted in the execution of his own teacher. 23. This section draws on Daniel A. Bell, China’s New Confucianism: Politics and Everyday Life in a Changing Society (Prince­ton, NJ: Prince­ton University Press, 2008), ch. 3. 24. Ritual per se is not sufficient. As we w ­ ill see, rituals can engage and thus transform the emotions, but Xunzi also argues for extensive and life-­long reading of ­great works to enhance the mind (the Xunzi leads with chapter 1, “An Exhortation to Learning,” a work well known to Chinese university students ­because they need to memorize passages for the national university examinations).

230  N o t e s t o C h a p t e r   3 25. Eric L. Hutton, “Introduction,” in Xunzi, Xunzi: The Complete Text, trans. with an introduction by Eric L. Hutton (Prince­ton, NJ: Prince­ton University Press, 2014), xxvii. 26. Patricia Buckley Ebrey, Confucianism and F ­ amily Rituals in Imperial China (Prince­ton, NJ: Prince­ton University Press, 2014), 228. 27. The one exception, of course, is the hierarchical relationship between the living and dead: The dead cannot be conscious of the common fate with the living, and they do not have the ability to reciprocate in any way (assuming that ghosts are not real). 28. The next section draws on Yan, Ancient Chinese Thought, Modern Chinese Power, ch. 2, as well as Daniel’s introduction to this book. 29. Xunzi, 10, “Enriching the State” (quoted in Yan, Ancient Chinese Thought, Modern Chinese Power, 81). 30. Xunzi, 11, “Humane Authority and Hegemony” (quoted in Yan, Ancient Chinese Thought, Modern Chinese Power, 88–89). 31. Xunzi, “On the Regulations of a Humane Authority” (9.5). ­Here we mark an area of disagreement with Yan Xuetong, who criticizes Xunzi’s notion of humane authority on the grounds that he neglects to mention that it also needs a foundation in hard power. Xunzi does have an extensive discussion of domestic policies that the humane authority should try to implement, including the need for a complex bureaucracy (see esp. books 9 and 13) designed to benefit the p­ eople and strengthen the state. 32. Xunzi, “Enriching the State.” 33. Quoted in Yan, Ancient Chinese Thought, Modern Chinese Power, 96. 34. Xunzi 18, “Correcting: A Discussion.” Quoted in Yan, Ancient Chinese Thought, Modern Chinese Power, 96–97. 35. Yan, Ancient Chinese Thought, Modern Chinese Power, 98. 36. John K. Fairbank and Ssu-­Yu Teng, “On the Ch’ing Tributary System,” Harvard Journal of Asiatic Studies 6 (1941): 138–139. 37. D. C. Kang, East Asia before the West: Five Centuries of Trade and Tribute (New York: Columbia University Press, 2010), 105. 38. Ibid., 63, 114. See also Ge, What Is China? 138. 39. Wang Gungwu, “Ming Foreign Relations: Southeast Asia,” in The Cambridge History of China, Vol. 8: The Ming Dynasty, eds. D. Twitchett and F. W. Mote, part 2: 1368–1644 (Cambridge: Cambridge University Press, 1998), 320. 40. Babones, American Tianxia, 23. 41. Immanuel C. Y. Hsu, China’s Entrance into the F ­ amily of Nations (Cambridge: Harvard University Press, 1960), 8–9. Con­temporary research shows that common Confucian values still inform the former “tributary states” of East Asia such as ­Korea and Vietnam. See Nisbett, The Geography of Thought: The Cultural Foundations of

N o t e s t o C h a p t e r   3   231 Learning; and Bell, Beyond Liberal Democracy. The Confucian values that w ­ ere transmitted to Vietnam also helped the Viet­nam­ese to resist foreign invaders in subsequent history (see Neil Sheehan, A Bright Shining Lie: John Paul Vann and Amer­i­ca in Vietnam [New York: Vintage, 1989]). 42. Kang, East Asia before the West, 8–9, chs. 3–4. 43. Zhang Feng, Chinese Hegemony: G ­ rand Strategy and International Institutions in East Asian History (Palo Alto, CA: Stanford University Press, 2015), 7, 177. 44. Alasdair Ian Johnston, Cultural Realism: Strategic Culture and G ­ rand Strategy in Chinese History (Prince­ton, NJ: Prince­ton University Press, 1995), xi. As David Kang notes, however, Johnston’s study focuses mainly on nomads on China’s northern and Western frontier; Korean, Japan, and Vietnam shared China’s “Confucian” worldviews and had far more stable and peaceful relations with China (Kang, East Asia before the West, 10). 45. Zhang Feng, “Rethinking the ‘Tribute System’: Broadening the Conception Horizon of Historical East Asian Politics,” Chinese Journal of International Politics 2, no. 4 (2009): 545–574. 46. Zhuang Guotu, “ 略论朝贡制度的虚幻: 一古代中国与东南亚的朝贡关系为列” [An account of the unreality of the tributary system: An example of the relationships in the tributary system in ancient China and East Asia], Nanyang wenti yanjiu [Research on prob­lems in Nanyang] 3 (2005): 1–8. 47. Yan, Ancient Chinese Thought, Modern Chinese Power, 104. 48. For an argument that “untruths” can be useful despite knowing they are false, see Anthony Appiah, As If: Idealization and Ideals (Cambridge, MA: Harvard University Press, 2017). 49. Rasmussen, The Infidel and the Professor, 174–175. 50. Hutton, “Introduction,” in Xunzi, Xunzi: The Complete Text, xxix. 51. See Bell, Beyond Liberal Democracy, ch. 2. 52. Courtney J. Fung, “China and the Responsibility to Protect: From Opposition to Advocacy,” United States Institute of Peace, June 8, 2016, https://­www​.­usip.org​ /­publications/2016/06/china-­and-­responsibility-­protect-­opposition-­advocacy. 53. James Kynge, China Shakes the World: A Titan’s Rise and Troubled ­Future—­And the Challenge for Amer­ic­ a (New York: Mari­ner Books, 2017). 54. See President Trump’s speech at Davos on January 26, 2018, https://­www​ .­weforum​.­org​/­agenda​/­2018​/­01​/­president​-­donald​-­trumps​-­davos​-­address​-­in​-­f ull​ -­8e14ebc1–79bb​-­4134​-­8203​-­95efca182e94​/­. 55. “Making the world,” to convey an active, nonstatic ideal, may be a more appropriate of tianxia (thanks to Roger Ames for this point). 56. Li Yangfan, “天下”观念” [On the concept of Tianxia], Guoji zhengzhi yanjiu no. 1 (2002): 107–111.

232  N o t e s t o C h a p t e r   3 57. Li Fang, “试论唐朝的“中国”与“天下” [ An analy­sis of “China” and “tianxia” in the Tang dynasty], Research on the History and Geography of China’s Borderlands 17, no. 2 (2007): 10–20. 58. Yuri Pines, “Changing Views of ‘Tianxia’ in Pre-­Imperial Discourse,” Oriens Extremus 43, nos. 1/2 (2002): 108, http://www​.­oriens​-­extremus​.­de​/­inhalt​/­pdf​/­43​ /­OE43​-­09​.­pdf. 59. Zhao Tingyang, “All-­under-­Heaven and Methodological Relationism: An Old Story and New World Peace,” in Con­temporary Chinese Po­liti­cal Thought: Debates and Perspectives, eds. Fred Dallmayr and Zhao Tingyang (Lexington: University of Kentucky Press, 2012), 46–66. For Zhao’s latest formulation, see Zhao, 天下的当代性 [The possibility of “All-­under-­Heaven] (Beijing: Citic Press, 2016). 60. William A. Callaghan, “Chinese Visions of World Order: Post-­hegemonic or a New Hegemony?” International Studies Review 10 (2008): 749–761. 61. Zhao, “All-­under-­Heaven and Methodological Relationism,” 73–74. 62. Ibid., 79. 63. Zhang Feng, “The Tianxia System: World Order in a Chinese Utopia,” Chinese Heritage Quarterly 21 (March 2010): 108–112. 64. Allen Carlson, “Moving beyond Sovereignty? A Brief Consideration of Recent Changes in China’s Approach to International Order and the Emergence of the Tianxia Concept,” Journal of Con­temporary China 20, no. 68 (2011): 101–102. 65. Yan, Ancient Chinese Thought, Modern Chinese Power, 99. 66. See Yan Xuetong, Leadership and the Rise of ­Great Powers (Prince­ton, NJ: Prince­ton University Press, 2019), 198–199. 67. For an argument that China and ­Korea should reestablish a Ming-­like system underpinned by the princi­ple that “Ritual lies in the deference of the small to the big and the caring of the big for the small,” but without adherence to the formal tributary system, see Orun Kihyup Kim, “­Korea’s Experiences with Big Neighbors,” paper presented at the Berggruen Institute workshop “What Is Tianxia?,” Peking University, Beijing, China, June 16–17, 2018. 68. Salvatore Babones puts forward the idea of an “American tianxia” appropriate for the modern world, with Canada and other Anglo-­Saxon allies in the zone of “internal barbarians.” See Salvatore Babones, American Tianxia: Chinese Money, American Power, and the End of History (Bristol: Policy Press, 2017), 22. This proposal may work as a defense of an American-­led tianxia hierarchical system in North Amer­i­ca and Eu­rope, but it is a complete nonstarter if China and other “wild” barbarians are meant to endorse such an order. 69. Renshon, Fighting for Status, ch. 5. 70. Ibid., 270. 71. See Gideon Rachman, Easternization: Asia’s Rise and Amer­i­ca’s Decline from Obama to Trump and Beyond (New York: Other Press, 2017).

N o t e s t o C h a p t e r   4   233 72. Amy Kazmin and Ben Bland, “China and India Use Summit to Push for Improved Ties,” Financial Times, April 28, 2018; Kiran Stacey, Simon Mundy, and Emily Feng, “India Benefits from AIIB Loans despite China Tensions,” Financial Times, May 18, 2018. More generally, China has been seeking better ties with its neighbors (at least partly) ­because the United States has been more aggressively working to ­counter China’s rise. 73. The point ­here is not that strong reciprocity should replace weak reciprocity: Strong reciprocity that is not founded on common economic interests may not be very stable; put negatively, if strong reciprocity imposes (or coexists with) severe financial constraints on one country in a relation between other­wise friendly states, it may not be long lasting (thanks to Zhang Feng for this point). In other words, the most stable form of reciprocity between states would be founded on both forms of reciprocity. 74. http://­en­glish.cri.cn/12954/2016/05/05/2743s926586.htm. Tagore also visited the Forbidden City for a meeting with the “last” emperor. See Reginald Fleming ­Johnston, Twilight in the Forbidden City (Vancouver: Soul Care Publishing, 2008), 335 [orig. pub. 1934]. We ­will know that India-­China relations are on track for ties of strong reciprocity when leading poets and writers from India are invited to meet members of the Standing Committee of the Politburo. 75. Liang Shuming, 东西文化及其哲学 [Eastern and Western cultures and their philosophies] (Beijing: The Commercial Press, 2009) [orig. pub. 1921]. 76. Email sent to Daniel on February 24, 2017. 77. We do not mean to imply that the ideal of an Asian regional order led by the two most populous countries in Asia is the only po­liti­cally realistic and morally ­desirable scenario for Asia’s f­ uture. One can imagine other scenarios, for example, a security alliance between China and Rus­sia, or joint patrols of Pacific sea lanes by the United States and China, that also benefit surrounding smaller countries.

Chapter 4. Just Hierarchy between H ­ umans and Animals: Subordination without Cruelty 1. See Chris Brown, “Our Side? Critical Theory and International Relations,” in Critical Theory and World Politics, ed. Richard Wyn Jones (Boulder: Lynne Rienner, 2001), 197. 2. The same holds true in India: the film Toilet (based on a true story) is an inspiring tale that links female emancipation with the right to indoor toilets. 3. Sue Donaldson and ­Will Kymlicka, Zoopolis: A Po­liti­cal Theory of Animal Rights (Oxford: Oxford University Press, 2011), 2. 4. Genesis 1:26, https://­www​.­biblegateway​.­com​/­passage​/­​?­search​=­Genesis+ 1percent3A25–26&version​=­NIV.

234  N o t e s t o C h a p t e r   4 5. Quoted in Matthieu Ricard, Plaidoyer pour les animaux: Vers une bienveillance pour tous (Paris: Allary Editions, 2014), 27. 6. Gary Francione, Introduction to Animals Rights: Your Child or the Dog? (Philadelphia: ­Temple University Press, 2010). 7. Quoted in Ricard, Plaidoyer pour les animaux, 27. 8. Mary Midgley, Animals and Why They ­Matter (Athens: The University of Georgia Press, 1983), 28. 9. Quoted in Peter Singer, Animal Liberation, 40th anniversary edition (New York: Open Road, 2016), 45. 10. Singer, Animal Liberation, 28, 34, 37, 38. 11. Quoted in Kevin Toolis, “The Most Dangerous Man in the World,” Guardian, November 6, 1999, https://­w ww​.­theguardian​.­com​/­lifeandstyle​/­1999​/­nov​/­0 6​ /­weekend​.­kevintoolis. The first part of this paragraph draws on Toolis’s article. 12. Roger T. Ames, “Achieving Personal Identity in Confucian Role Ethics: Tang Junyi on ­Human Nature as Conduct,” Oriens Extremus 49 (2010): 143–166. 13. Thorbjorn Tannsjo, “It’s Getting Better All the Time,” excerpted in Food, Ethics, and Society: An Introductory Text with Readings, edited by Anne Barnhill, Mark Budolfson, and Tyler Doggett (New York and Oxford: Oxford University Press, 2017), 364–365. 14. Donaldson and Kymlicka, Zoopolis, ch. 2. 15. Ibid., 74. 16. Charles Patterson, Eternal Treblinka: Our Treatment of Animals and the Holocaust (New York: Lantern Books, 2002). 17. Raimond Gaita, The Phi­los­o­pher’s Dog (New York: Routledge, 2016). 18. Peter Singer and Karen Dawn, “Harambe the Gorilla Dies, Meat Eaters Grieve,” Los Angeles Times, June 5, 2016, http://­www.latimes.com/opinion/op-­ed/la-­oe-­singer​ -­dawn-­harambe-­death-­zoo-20160605-­snap-­story.html#. 19. Ricard, Plaidoyer pour les animaux, 197–198. 20. Jessie Yeung, “Australian Experiment Wipes Out over 80 ­Percent of Disease-­ Carrying Mosquitoes,” CNN, July 10, 2018. 21. Confucius, The Analects, http://­w ww.indiana.edu/~p374/Analects_of​ _­Confucius_(Eno-2015).pdf (translation modified). As the ­great Song-­dynasty Confucian Zhu Xi put it, “非不爱马, 然恐伤人之意多, 故未暇问. 盖贵人贱畜, 理当如此.” [It ­isn’t ­because the Master ­didn’t love the ­horses, but his concern for ­human beings was overwhelming. Since h­ uman beings are more precious than animals, the Master followed the right princi­ple.] (Zhu Xi, “四书章句集注” [Commentaries on Four Books] [Beijing: Zhonghua Shuju, 2011], ch. 10.13). It does not follow that Confucius was indifferent to animals. Quite the opposite: According to one account, Confucius cried and mourned publicly for a Kirin [a semimythical creature with features of dif­fer­ent

N o t e s t o C h a p t e r   4   235 animals] that was captured and died. The story is recorded at the end of the Spring and Autumn Annals, and Confucius died two years l­ater (Yang Bojun, “春秋左传注· 哀公十四年” [Commentary on Zuozhuan [Beijing: Zhonghua Shuju, 1990]). The Kirin (or Jilin) became known in China and other East Asian cultures as a creature said to appear with the imminent arrival or passing of a sage or illustrious ruler. 22. Midgley, Animals and Why They ­Matter, 104. In the Chinese context, Daoism may be viewed as a counter-­example: animals ­were viewed as (equal) embodiments of the Dao (Way) and even as potential teachers of ­human beings. See Louis Komjathy, “Animals and Daoism,” Encyclopaedia Britannica, September 11, 2011, http://­ advocacy.britannica.com/blog/advocacy/2011/09/daoism-­and-­animals/. Daoist concern for animal welfare often translates into vegetarianism for priests and community leaders. In ancient China, animals ­were sometimes worshiped as gods on a higher plane than h­ uman beings, even without a doctrinal connection with Daoism. Buddha, gods, and fairies ­were symbols of virtue and worshiped for public goods, but animal gods w ­ ere easier to access for common folk, and they w ­ ere often worshiped for private and personal reasons, such as requests for money or a beautiful wife. See Kang Xiaofei, The Cult of the Fox: Power, Gender, and Popu­lar Religion in Late Imperial and Modern China (New York: Columbia University Press, 2006), 132–133. The animal gods could also be used to fight against unjust hierarchies: For example, oppressed ­people could pretend to be animal gods in order to criticize officials who lacked virtue, and young w ­ omen who did not dare to publicly reject arranged ­marriages sometimes pretended to be possessed by animal gods to safely escape unwanted marriages. 23. Fred H. Besthorn, “Deep Ecological ‘Insectification’: Integrating Small Friends with Social Work,” in Animals in Social Work: Why and How They ­Matter, ed. Thomas Ryan (Houndsmills: Palgrave Macmillan, 2014), 11. 24. Matthew C. Halteman, “Compassionate Eating as Care of Creation,” in Food, Ethics, and Society, 295. 25. Reol Strerckx, The Animal and the Daemon in Early China (Albany: State University of New York Press, 2002), 123–163. 26. More precisely, sensible beings can be reincarnated in one of six ways or worlds depending on the karma accumulated from previous lives: (1) the world of gods, fairies, or other celestial beings; (2) the world of demigods; (3) the world of ­human beings; (4) the world of animals; (5) the world of the famished; (6) the world of hell. The first three worlds are benevolent, and the beings are more or less virtuous. The last three worlds are not benevolent, and beings lack virtue and suffer more. Evidently, ­human beings are in a higher position than animals. See Robert E. Buswell, Encyclopedia of Buddhism, vol. 2 (New York: Macmillan Library Reference, 2003), 711–712.

236  N o t e s t o C h a p t e r   4 27. H ­ ere’s one story. Siddhartha was a king in one of his previous lives (before he became the Buddha). One day he came upon a pigeon chased by a huge hawk. The pigeon pleaded with the king to save its life. The hawk counterargued: “I have been starving for days, and ­will be dead without food. So long as you are aiming at the salvation of all beings, how could you save me if you stop me from eating this pigeon? Am I not one of ­those beings?” The king asked the hawk: “What kind of food do you need? I ­will try my best to satisfy you so long as you do not take the pigeon as your food.” The hawk asked for meat that must weigh as much as the pigeon. So the king cut off his own flesh to substitute for the pigeon. For some reason, he had nearly cut off all his flesh and still did not reach the required weight. He then asked his minister to kill him so that his marrow could add more weight. At that time, the hawk revealed himself as Sakka (the ruler of Heaven) and paid his re­spects (Liu Luming, “金光明经” [Golden Light Sutra]. Beijing: Zhonghua Shuju, 2010). 28. See Ferris Jabr, “The Person in the Ape,” Lapham’s Quarterly, https://­www​ .­laphamsquarterly​.­org​/­states​-­mind​/­person​-­ape. 29. One recent study shows that wild dogs in Botswana “vote” by means of sneezes before deciding on collective displacement, with more “votes” increasing the likelihood of displacement. Paul Seabright, “De la démocratie chez les chiens sauvages,” Le Monde, May 27–28, 2018. 30. Donaldson and Kymlicka, Zoopolis, 14, 155. See chapters 3–5. 31. To be fair (to Daniel), it would have been impossible to represent his interests in China, where even citizens do not have the equal right to participate in the po­liti­cal pro­cess (which is not necessarily wrong: see chapter 2). But it would be equally difficult to represent the (equal) interests of animals in electoral democracies that represent the (equal) interests of ­human citizens in the legislature since ­there is no mechanism to do so. And what would it mean for domesticated animals to be treated as equals in the legislature? Imagine a po­liti­cal community with twice as many cats as ­people. According to Donaldson and Kymlicka’s theory, ­there should be twice as many representatives of cat interests as representatives of h­ uman interests in the legislature. 32. The reader may not endorse all of Daniel’s actions, such as the use of physical punishment. But t­ here may be a stronger case against physical punishment of ­people than against animals. One recent book draws on a wide range of empirical studies to argue that corporal punishment of ­children by parents or guardians is always harmful and should be banned everywhere. See Patrick Lenta, Corporal Punishment: A Philosophical Assessment (New York: Routledge, 2018). But even animal advocates sometimes concede that physical punishment of domesticated animals may not always be morally wrong to the point that it should be banned: For example, Peter Wohlleben argues that it should be up to the h­ uman caretaker to decide if it’s acceptable to put a special collar on a dog or cat that delivers an electric shock when the animal crosses a

N o t e s t o C h a p t e r   4   237 certain boundary. See Peter Wohlleben, The Inner Life of Animals (Vancouver/Berkeley: Greystone Books, 2017), 170. In any case, it’s quite obvious that animals without the capacity for complex language may not respond as well to moral persuasion compared to ­human beings, and hence the argument against physical punishment for animals is not so straightforward. 33. Quoted in Midgley, Animals and Why They ­Matter, 51. 34. See Christine Korsgaard’s critique of her “other­wise favorite phi­los­o­pher” in “Getting Animals in View,” in Food, Ethics, and Society, 366–372. 35. Lynn Loar, “ ‘How Is Fido?’: What the F ­ amily’s Companion Animal Can Tell You about Risk Assessment and Effective Interventions—­If Only You Would Ask!,” in Midgley, Animals in Social Work, 136. For a recent case, see Kevin ­Sullivan, William Wan, and Julie Tate, “Florida Shooting Suspect Had a History of Explosive Anger, Depression, Killing Animals,” Washington Post, February 15, 2018. 36. Loar, “How Is Fido?” p. 138. 37. The causal arrow may work both ways: It is also likely that he or she who is cruel to ­humans becomes hard in his dealings with animals. 38. Quoted in Daniel A. Bell and Thaddeus Metz, “Confucianism and Ubuntu: Reflections on a Dialogue between Chinese and African Traditions,” Journal of Chinese Philosophy 38, Supplement (2011): 88. 39. We borrow this insight from Cordeiro Rodrigues, pre­sen­ta­tion at Sun Yat-­Sen University (Zhuhai), June 21, 2018. 40. Dogs w ­ ere the first animals to be domesticated by p­ eople, more than 15,000 years ago. See Brian Fagan, “Nous ne serons jamais tous vegetariens,” Le Figaro, April 9, 2017, which helps to explain why they became “man’s best friend.” 41. Nicholas Kristof, “Choosing Animals Over P ­ eople?,” New York Times, April 7, 2018. 42. At the moment, higher education, economic fluctuation, and urbanization may be more impor­tant ­causes of subreplacement fertility, but it is conceivable that excessive love of pets (over ­children) w ­ ill play an impor­tant role in the ­future. 43. John Rawls famously defended a princi­ple of justice that prioritizes the worst off (see A Theory of Justice), but similar views w ­ ere also defended by early Confucian thinkers (see Chan, Confucian Perfectionism, ch. 7). 44. Elaine Scarry, On Beauty and Being Just (Prince­ton, NJ: Prince­ton University Press, 1999), part 2. 45. Ibid., 97, 101, 110, 123, 80, 111–112. 46. Quoted in Brian Leiter, “Friedrich Nietz­sche: The Truth Is Terrible,” TLS Online. https://­www​.­the​-­tls​.­co​.­uk​/­articles​/­public​/­friedrich​-­nietzsche​-­truth​-­terrible​/­. Accessed July 20, 2018. 47. Dzongsar Jamyang Khyentse, What Makes You Not A Buddhist (Boulder: Shambhala, 2008), 8.

238  N o t e s t o C h a p t e r   4 48. Ibid., 8, 11. 49. Consider the case of the owner in Beijing who bought his beautiful dog a $500,000 pet mansion. See Danielle Paquette and Luna Lin, “­After Making His Owner Rich, This Border Collie Gets to Live in a $500,000 Pet Mansion in Beijing,” Washington Post, July 22, 2018. 50. Sarah Mervosh, “Congratulations? Zsa Zsa, an En­glish Bulldog, Wins the World’s Ugliest Dog Contest,” New York Times, June 24, 2018. 51. “Venuzuela’s ‘Plan Rabbit’ Encounters ‘Cultural Prob­lem,’ ” BBC News, September 14, 2017. 52. The argument that domesticated animals can help motivate h­ umans to care for the weak and the vulnerable in the h­ uman world may also apply to wild but cute animals such as pandas and koala bears that resemble ­human babies and similarly trigger the desire to care for vulnerable and dependent ­humans. 53. Jonathan Safran Foer, Eating Animals (London: Penguin, 2010), 5. 54. Anne Barnhill, Mark Budolfson, and Tyler Doggett, eds., “Industrial Animal Agriculture,” in Food, Ethics, and Society, 326. 55. “Pew Commission Says Industrial Scale Farm Animal Production Poses ‘Unacceptable’ Risks to Public Health, Environment,” in Food, Ethics, and Society, 343. 56. Paul Greenberg, “How to Get Amer­i­ca on the Mediterranean Diet,” New York Times, July 19, 2018. This article r­ ose to the top of the “most emailed” New York Times list of emailed articles. 57. “Industrial Animal Agriculture,” in Food, Ethics and Society, 344. 58. One friend from France became a vegetarian ­because he observed the panic of cows at the moment before execution in a slaughter­house. 59. Sue Neales, “3D Printed Meat Makes the Cut,” The Weekend Australian, May 3, 2017. 60. Donaldson and Kymlicka, Zoopolis, 152. 61. Ricard, Plaidoyer pour les animaux, 36–37. 62. Wendy Doniger, “Compassion ­toward Animals, and Vegetarianism,” in Food, Ethics, and Society, 303. 63. Ricard, Plaidoyer pour les animaux, 45. That said, only 15 ­percent of Bhutanese are vegetarians, and animal welfare advocates face big challenges in Bhutan. See ­Michael Tobias and Jane Gray Morrison, “Animal Rights in Bhutan,” http://­www​ .dancingstarfoundation.org/articles_Animal_Rights_in_Bhutan.php. 64. Quoted in “Food and Religion,” in Food, Ethics, and Society, 276. 65. Quoted in Ricard, Plaidoyer pour les animaux, 33. 66. Quoted in “Compassionate Eating as Care of Creation,” in Food, Ethics, and Society, 293. 67. Ibid., 296.

N o t e s t o C h a p t e r   4   239 68. See Anna Sun, Confucianism as a World Religion: Contested Histories and Con­ temporary Realities (Prince­ton, NJ: Prince­ton University Press, 2013). 69. Quoted in (and translated by) Bai Tongdong, Against Po­liti­cal Equality: The Confucian Case (Prince­ton, NJ: Prince­ton University Press, 2019), 134. 70. Confucius, The Analects of Confucius: A Philosophical Translation, trans. Roger T. Ames and Henry Rosemont, Jr. (New York: Ballantine Books, 1998). 71. Bai Tongdong, Against Po­liti­cal Equality, 274. 72. On the permissibility of killing (and eating) f­ ree range animals, see Roger Scruton, “Eating Our Friends,” in Food, Ethics, and Society, 391–395. 73. Devout Buddhists, in contrast, do not object to meat if it’s given to them as offerings. 74. Vegetarians are 5 ­percent of the population in China and France compared to 10 ­percent in Germany and 14 ­percent in the United Kingdom, https://­en​.­wikipedia​ .­org​/­wiki​/­Vegetarianism​_­by​_­country. 75. But the Chinese government has recently unveiled a plan to cut meat consumption by 50 ­percent in 2030. See Oliver Milman and Stuart Leavenworth, “China’s Plan to Cut Meat Consumption by 50% Cheered by Climate Campaigners,” https://­www​ .­theguardian​.­com​/­world​/­2016​/­jun​/­20​/­chinas​-­meat​-­consumption​-­climate​-­change. 76. Quoted in Melanie Joy, Why We Love Dogs, Eat Pigs, and Wear Cows (San Francisco: Conari Press, 2010), 71. 77. Ricard, Plaidoyer pour les animaux, 110. 78. Joy, Why We Love Dogs, Eat Pigs, and Wear Cows, 18. 79. H.-­S. Afeissa and J.-­B. Jeangene Vilmer, “Preface,” in Philosophie animale: Différence, responsabilité, et communauté, eds. H.-­S. Afeissa and J.-­B. Jeangene Vilmer (Paris: VRIN, 2015), 10. 80. Ricard, Plaidoyer pour les animaux, 64. 81. Mencius, Liang Hui Wang I, https://­ctext​.­org​/­mengzi​/­liang​-­hui​-­wang​-­i, translation modified. 82. In the United States, unfortunately, t­ hings took a turn for the worse in the 1980s when new legislation shifted the burden of quality control in slaughter­houses from the government to the plants themselves ( Joy, Why We Love Dogs, Eat Pigs, and Wear Cows, 76). 83. Ricard, Plaidoyer pour les animaux, 84. 84. Followers of the Jain religion who push the commitment to nonviolence to an extreme end would disagree with the view that it’s acceptable to purposefully kill insects. More than that, they take special precautions not to accidentally kill insects, such as being careful not to crush insects when they walk and putting a cloth on the mouth to avoid swallowing insects while breathing the air (Ricard, Plaidoyer pour les animaux, 40). For a (depressing) fictitious account of a follower of the Jain religion

240  N o t e s t o C h a p t e r   5 in the con­temporary United States, see Philip Roth, American Pastoral (New York: Houghton Mifflin, 1997). 85. An artist in the Philippines recently showed a live cockroach being tortured in an electric chair. See Connor Boyd, “Cockroach Is Executed in an Electric Chair: Artist Is Condemned for Animal Cruelty a­ fter Killing Bug That Flew into His Home,” Daily Mail, July 5, 2018. The point was to make a statement against capital punishment, but the means do not justify the ends.

Chapter 5. Just Hierarchy between ­Humans and Machines: On the Need for a Master-­Slave Relation 1. The Marxist-­inspired master-­slave meta­phor refers to the objective relation between the (machine) slave that serves the (­human) master, rather than the other way around. The historical real­ity of the master-­slave relation between two h­ umans has a subjective dimension: According to Hegel’s argument in The Phenomenology of Spirit, the ­human master needs the ­human slave to recognize his status as a slave in order for the master to feel like a master. In the case of the relation between h­ uman and machine, the machine (without consciousness or feeling) cannot feel like a slave, and the master need not feel like a master. For an argument that h­ umans should not feel like a masters when they engage with machines (­because thinking of ourselves as masters makes us more vulnerable and ironically makes our own be­hav­ior alienated), see Mark Coeckelbergh, “The Tragedy of the Master: Automation, Vulnerability, and Master,” Ethics and Information Technology 17 (2015): 219–229. 2. Daniel A. Bell, China’s New Confucianism (Prince­ton, NJ: Prince­ton University Press, 2008), ch. 1. 3. See Maurice Meisner, Mao’s China and ­After: A History of the ­People’s Republic, 3rd ed. (New York: The ­Free Press, 1999), chs. 11, 13. 4. See Karl Marx, The German Ideology, https://­www​.­marxists​.­org​/­archive​/­marx​ /­works​/­1845​/­german​-­ideology​/­ch01a​.­htm. 5. See Haig Patapan and Wang Yi, “The Hidden Ruler: Wang Huning and the Making of Con­temporary China,” Journal of Con­temporary China (October 2017): 9. See Wang Huning, 政治的逻辑 [The logic of politics] (Shanghai: Shanghai ­People Press, 2004). 6. In princi­ple, however, the leases can be extended or renewed a­ fter 70 years. See http://­www.xinhuanet.com/fortune/2017–03/08/c_129504739.htm. 7. It depends on the time horizon for the implementation of communism: Many Chinese workers seem willing to be exploited in cap­i­tal­ist enterprises if it benefits their ­children and grandchildren, but t­ here may be less willingness to be exploited for the sake of ­people in a communist society that may still be de­cades away. 8. See Feng Xiang, “我是阿尔法—­论人机伦理” [On human-­computer ethics], ­Wenhua zongheng 12 (2017): 128–139.

N o t e s t o C h a p t e r   5   241 9. See “A Study Finds That Nearly Half of Jobs Are Vulnerable to Automation,” The Economist, April 24, 2018, https://­www​.­economist​.­com​/­graphic​-­detail​/­2018​/­04​/­24​ /­a​-­study​-­finds​-­nearly​-­half​-­of​-­jobs​-­are​-­vulnerable​-­to​-­automation. Researchers on AI believe ­there is a 50 ­percent chance of AI outperforming h­ umans in all tasks in 45 years and of automating all h­ uman jobs in 120 years. See Katja Grace, et al., “When ­Will AI Exceed H ­ uman Per­for­mance? Evidence from AI Experts,” May 2017, https://­ www​.­researchgate​.­net​/­publication​/­317100941​_­W hen​_­Will​_­AI​_­Exceed​_­Human​ _­Performance​_­Evidence​_­from​_­AI​_­Experts. In China, however, what was once viewed as a prob­lem—­a rapidly aging society and a contracting working-­age workforce (largely as a result of the one-­child policy)—­may actually be an advantage, ­because AI can replace ­humans in the factories while also providing some form of care for the el­derly. See John Markoff, “Artificial Intelligence Could Improve How We Age,” WorldPost, October 18, 2017, https://­www​.­washingtonpost​.­com​/­news​/­theworldpost​ /­w p​/­2017​/­10​/­18​/­artificial​-­intelligence​-­could​-­improve​-­how​-­we​-­age​/­​?­utm​_­term​=­​ .­e05a276779d8. 10. Max Tegmark, Life 3.0: Being H ­ uman in the Age of Artificial Intelligence (New York: Alfred A. Knopf, 2017), 126, 128. Nathan Gardels and Nicholas Berggruen, however, argue that such proposals meant to equalize income are less likely to address the ­causes of economic in­equality rooted in owner­ship. To remedy this prob­lem, they argue for a scheme in which all citizens own an equity share in the robots that produce wealth in an AI-­driven world. See Gardels and Berggruen, Renovating Democracy: Governing in the Age of Globalization and Capital (Berkeley: University of California Press, 2019), ch. 3. Proposals meant to equalize owner­ship are more likely to be taken seriously in a country such as China that is openly committed to communism. 11. Ibid., 218–245. To be fair, such statements are preceded by an extensive discussion of the way technology drives in­equality in the United States and an observation that the government is not implementing mea­sures to deal with this trend. 12. Karl Marx, Capital, vol. 1, in The Marx-­Engels Reader, 2nd ed., edited by R ­ obert C. Tucker (New York: Norton and Com­pany, 1978), 408, 409. 13. Ibid., 412. 14. Ibid., 414. 15. Marx recognized ­there might still be a need for an administrative apparatus to fulfill tasks such as coordination and rec­ord keeping. See Bertell Ollman, “Marx’s Vision of Communism,” https://­www​.­nyu​.­edu​/­projects​/­ollman​/­docs​/­v ision​_­of​ _­communism​.­php. But t­ hese tasks could conceivably be done by advanced machines. In ancient Greece, ­there was a strict dichotomy between the citizens who de­cided on policy and the “neutral” public servants who implemented their policies. But the public servants ­were slaves—­owned by the po­liti­cal community and without the freedom to choose their own way of life—­w ho did the administrative and bureaucratic

242  N o t e s t o C h a p t e r   5 “neutral” work of serving the state and implementing the decisions of the “po­liti­cal” citizens who deliberated about and de­cided on public policy. See Paulin Ismard, La démocratie contre les experts: Les esclaves publics en Grèce ancienne (Paris: Éditions du Seuil, 2015). Of course, what made the po­liti­cal system unjust is that it relied on a foundation of both private and public slaves who did the socially necessary work. In higher communism, ­there would be no need for a coercive apparatus, and advanced machines would be both like “private” slaves who help with domestic work and like literate and intelligent “public” slaves who help with the relatively mundane and repetitive bureaucratic and administrative tasks necessary for any well-­ coordinated community. But the system would not be morally justified if (once) the machines develop consciousness and/or the capacity to suffer. 16. Karl Marx, Critique of the Gotha Programme, https://­www​.­marxists​.­org​/­archive​ /­marx​/­works​/­1875​/­gotha​/­ch01​.­htm. 17. Karl Marx, Capital, vol. 1, part III, “The Production of Absolute Surplus Value,” ch. 7, “The L ­ abour Pro­cess and the Pro­cess of Producing Surplus Value,” https://­www​ .­marxists​.­org​/­archive​/­marx​/­works​/­1867​-­c1​/­ch07​.­htm. 18. Karl Marx, The German Ideology, https://­www​.­marxists​.­org​/­archive​/­marx​ /­works​/­1845​/­german​-­ideology​/­ch01a​.­htm. 19. Confucius, The Analects, http://­classics.mit.edu/Confucius/analects.1.1.html (modified). 20. Mencius, The Works of Mencius, http://­ctext.org/mengzi. 21. See chapter 4, section 3, for a more extensive discussion of Confucianism and animal ethics. 22. See chapter 3, section 3, for a more extensive discussion of the Confucian ideal of graded love. 23. It’s worth noting that the Confucian tradition recognizes the permanent need for a state, unlike the Marxist tradition. The classic Confucian text the Rec­ord of Rites, from the Han dynasty, offers an account of a communist-­like ideal society known as the ­Great Harmony (大同), but that society is still characterized by a po­liti­cal system that aims to select and promote rulers with superior ability and virtue (选贤与能). 24. Roger T. Ames, Confucian Role Ethics: A Vocabulary (Honolulu: University of Hawai’i Press, 2011), 261. 25. See Daniel A. Bell, “Roles, Community, and Morality: Comment on Confucian Role Ethics,” in Appreciating the Chinese Difference, 203–212. 26. See Daniel A. Bell, Beyond Liberal Democracy: Po­liti­cal Thinking for an East Asian Context (Prince­ton, NJ: Prince­ton University Press, 2007), ch. 9. In China’s new social credit system, commitment (or lack thereof) to filial piety affects one’s score (­needless to say, such a policy would be hugely controversial in Western socie­ties that do not prioritize filial piety to the same extent).

N o t e s t o C h a p t e r   5   243 27. The idea of constraint suggests closure rather than openness of possibilities, but the opposite may be true: Just as constraints on language allow for the creative use of language (other­wise ­human sounds would be random, with no meaning), so constraints on roles allow us to express our creativity. See Henry Rosemont, Jr., Against Individualism: A Confucian Rethinking of the Foundations of Morality, Politics, ­Family, and Religion (Lanham, MD: Lexington Books, 2015), 16, 103–104. 28. Jason Borenstein and Yvette Pearson, “Robot Caregivers: Ethical Issues across the ­Human Lifespan,” in Robot Ethics: The Ethical and Social Implications of Robotics, eds. Patrick Lin, Keith Abney, and George A. Bekey (Cambridge, MA: The MIT Press, 2014), 261. 29. See Luo Dingsheng and Wu Shihong, “浅谈智能护理机器人的伦理问题” [A discussion about the ethical prob­lems of intelligent nursing robots], Technology and Society 8, no. 1 (2018): 25–39. Perhaps robot caregivers should be designed to be ugly (but not so horrifying as to scare ­people) so as to minimize the risk that el­derly parents love them more than real ­people. 30. Ian Austen, “City of the F ­ uture? H ­ umans, Not Technology, Are the Challenge in Toronto,” New York Times, December 29, 2017. 31. This is not to deny that many ­drivers engage in maneuvers designed to avoid killing, such as swerving suddenly to avoid a child or dog, but t­ hese actions typically are instinctive (without the deliberation that typically accompanies trolley scenarios) and do not usually involve decisions about which person(s) to kill. Thanks to Jay Boggis for this point. 32. Researchers from MIT’s Media Lab found that p­ eople in socie­ties ­shaped by Confucian values such as China, Japan, and South ­Korea that emphasize re­spect for older members of society ­were more likely to prefer saving older ­people over saving the young. See Echo Huang, “The East and West Have Very Dif­fer­ent Ideas on How to Save in a Self-­Driving Car Accident,” Quartz, November 1, 2018. It ­doesn’t follow, however, that policy makers in East Asia should program self-­driving cars with a preference to kill young c­ hildren over the el­der­ly; surely informed deliberation is necessary before such decisions are implemented. 33. Amitai Etzioni and Oren Etzioni, “Incorporating Ethics into Artificial Intelligence,” Journal of Ethics 21, no. 4 (2017): 403–418, https://­philpapers​.­org​/­archive​ /­ETZIEI​.­pdf. 34. Ibid. 35. We leave aside the difficulty of programming an algorithm that can effectively distinguish between personal habits and moral outlooks and adjudicate between conflicting habits and conflicting moral outlooks. 36. Li Yanhong, “Zhineng Geming” [Intelligent revolution] (Beijing: Citic Press, 2017), 182.

244  N o t e s t o C h a p t e r   5 37. It’s also somewhat odd that Amitai Etzioni should put forth an argument for leaving moral decision making outside the realm of the law and strictly to the individuals b­ ecause the self-­described communitarian theorist has written persuasive critiques of libertarianism in his ­earlier works. See Etzioni, The Spirit of Community: Rights, Responsibilities, and the Communitarian Agenda (New York: Crown Publishers, 1993). S­ houldn’t a communitarian ­favor “nudging” ­toward more responsible social be­hav­ior (so long as no basic rights are ­violated)? 38. See Kong Xinfeng and Daniel A. Bell, “曲阜: 儒家文化之城” [Qufu: The city of Confucian culture], in 城市的精神II: 包容与认同 [The spirit of cities II: Inclusion and identity], eds. Daniel A. Bell and Avner de Shalit (Chongqing: Chongqing Press, 2017), 52–85. 39. While countries need to strive for relatively impartial treatment between citizens, it may be more justifiable for cities to promote a conception of the good life widely endorsed by residents of cities (aka city-­zens) for the following reasons: (1) Most countries have several cities with dif­fer­ent dominant values, so citizens can choose among cities that more closely approximate their own conceptions of the good life; (2) cities (with the exception of city-­states) ­don’t have armies, so they are less likely to coerce nonmainstream city-­zens to adhere to dominant values and ways of life (this is not to deny that cities have police forces that can also exercise coercion, but police forces do not have the power to go to war); and (3) cities, compared to rural areas and small towns, tend to be more open and tolerant, hence nonmainstream city-­zens are less likely to be subject to informal social pressure. See Daniel A. Bell and Avner de-­Shalit, The Spirit of Cities (Prince­ton, NJ: Prince­ton University Press, 2011). Hence, it may be more justifiable for cities, rather than countries, to program dominant values in self-­driving cars. We can foresee a f­ uture where cars in Jerusalem (but not Tel Aviv) would be programmed to re­spect religious ways of life, cars in Montreal (but not Toronto) would be programmed to communicate in French (with En­glish as an option), ­etc. 40. See Nick Bostrom, “How Long before Superintelligence,” https://­nickbostrom​ .­com​/­superintelligence​.­html. 41. Nick Bostrom, Superintelligence: Paths, Dangers, Strategies (Oxford: Oxford University Press, 2014), 123. 42. The “slave” meta­phor refers only to the idea that machines should be designed to serve h­ uman beings. It does not mean that h­ uman “masters” can do what­ever they want to machines. For example, ­there is a good case to treat machines with care, both ­because bad care reflects badly on the character of the ­human (somebody who takes plea­sure from destroying beautifully designed AIs is likely to be violent to real p­ eople too) and b­ ecause AIs can learn immoral be­hav­ior and may act on it if they have the ability to do so (thanks to Song Bing for this point). Th ­ ese caveats imply that h­ umans can lose the moral authority to dominate machines if they act in immoral ways: Think of the HBO show Westworld that portrayed ­human “masters” in such a despicable

N o t e s t o C h a p t e r   5   245 light that many h­ uman viewers w ­ ere cheering for the machine rebels t­ oward the end of season one. 43. It is more controversial to promote research designed to improve bodily and ­mental functions beyond pre­sent biological limits, mainly ­because such improvements are most likely to benefit an already privileged minority and to exacerbate opportunity and income gaps in society and the world at large. 44. This is easier said than done: See Murray Shanahan’s discussion of the difficulty of engineering artificial intelligence designed to prevent harm to humanity, such as Isaac Asimov’s First Law of Robotics that a robot cannot allow a ­human being to come to harm. See Shanahan, The Technological Singularity (Cambridge, MA: The MIT Press, 2015), 173–177. Shanahan concludes that the prob­lems are not necessarily insurmountable and ­because so much is at stake, “it’s worth thinking about the prob­ lem now” (ibid., 177). 45. Articles 21 and 23 (quoted in Tegmark, Life 3.0, 331). In the same vein, the “Open AI Charter” aims for “Broadly Distributed Benefits: We commit to use any influence we obtain over AGI’s deployment to ensure it is used for the benefit of all, and to avoid enabling uses of AI or AGI that harm humanity or unduly concentrate power” (https://­blog​.­openai​.­com​/­openai​-­charter​/­), but it too lacks any enforcement mechanism. 46. Farhad Manjoo, “Silicon Valley’s Politics: Liberal, With One Big Exception,” New York Times, September 6, 2017. 47. Buddhists thought also holds the view that metaphysical truth and goodness are linked at some deep, mystical level. See Wright, Why Buddhism Is True, 229. Not coincidentally, Buddhism is quite influential among Silicon Valley techies. 48. To be fair, one’s assessment of the need to regulate AI may also depend on one’s assessment of certain risks. One scientist in China told Daniel that we need to allow for the un­regu­la­ted development of AI ­because we ­will need the help of AI to ward off superintelligent invaders from outer space. 49. Franklin Foer, World without Mind: The Existential Threat of Big Tech (New York: Penguin Press, 2007), 33. 50. Quoted in ibid., 38. 51. The co-­founders of Google argued against advertising-­funded search engines while they ­were students at Stanford on the grounds that they would be biased ­toward the advertisers and away from the needs of the consumers but “they discarded their own wisdom long ago.” See Foer, World Without Mind, 212. 52. Ibid., 52. See also Jonathan Taplin, Move Fast and Break Th ­ ings: How Facebook, Google, and Amazon Have Cornered Culture and Undermined Democracy (London: Pan Books, 2017). 53. At a small workshop with leading academics and tech elites at Stanford University a few years ago, Daniel had a discussion with a billionaire founder of a tech

246  N o t e s t o C h a p t e r   5 com­pany. His interlocutor did worry that “bad” AI may turn on h­ umans, but he was optimistic that some ­people would be able to escape planet earth by colonizing other planets. In the long term, however, the billionaire recognized that a supersmart AI ­will catch up with ­humans eventually, no ­matter where we go. So he has committed himself to vegetarianism, and he explained that we should treat animals well to set a good model for AIs once they have more power over us. Daniel told him about his plan to write a book that would assert the need for maintain our master-­slave relation with machines, and the concerned billionaire warned against this proj­ect on the grounds that Daniel would be one of the first targets of the power­ful AIs. Daniel’s interlocutor thought all t­ hese scenarios could happen in our lifetime (if the billionaire is right, Daniel would like to emphasize that although this book is co-­ authored, he is primarily responsible for this chapter, so please, supersmart AI, leave Pei out of this). 54. For some reason, the unofficial motto was erased from Google’s code of conduct in May 2018. In the film Ex Machina—­perhaps the most realistic of the “malevolent” AI science fiction films (the scientific adviser was Murray Shenahan, author of The Technological Singularity)—­the “bad” AI was created by a socially deprived and sex-­starved billionaire leader of a com­pany that seems loosely modeled on Google. 55. Thanks to Sean Oheigeartaigh for helpful discussion at a Berggruen Institute workshop on “AI, Robotics and Society,” Peking University, March 17–18, 2018. 56. Thanks to Billy Wheeler for helpful comments. 57. See Allison Duettmann et al., “Artificial General Intelligence: Coordination & ­Great Powers,” Foresight Institute Strategy Meeting on AGI, 2018, https://­foresight​.­org​ /­wp​-­content​/­uploads​/­2018​/­11​/­AGI​-­Coordination​-­Geat​-­Powers​-­Report​.­pdf. 58. EPA, “International Actions: The Montreal Protocol on Substances that Deplete the Ozone Layer,” https://­www​.­epa​.­gov​/­ozone​-­layer​-­protection​/­international​ -­actions​-­montreal​-­protocol​-­substances​-­deplete​-­ozone​-­layer. 59. Scientific consensus, unfortunately, is not sufficient to motivate global po­liti­ cal action if the perceived cure is too costly. In the case of the Montreal protocol, the threat from chlorofluorocarbons was effectively handled ­because ­there was a scientific consensus about the cause of the prob­lem and the cost of solving it was not perceived to be too high. 60. Meg Houston Maker, “AI@50,” https://web.archive.org/web/20140513052243/ http://­www.megmaker.com/2006/07/ai50_first_poll.html. 61. Quoted in David Meyer, “Vladimir Putin Says Whoever Leads in Artificial Intelligence ­Will Lead the World,” Fortune, September  4, 2017, http://­fortune​ .com/2017/09/04/ai-­artificial-­intelligence-­putin-­rule-­world/. 62. Executive Office of the President, “Preparing for the ­Future of Artificial Intelligence,” https://­obamawhitehouse​.­archives​.­gov​/­sites​/­default​/­files​/­whitehouse​ _­files​/­microsites​/­ostp​/­NSTC​/­preparing​_­for​_­the​_­future​_­of​_­ai​.­pdf, 8.

N o t e s t o C h a p t e r   5   247 63. The Trump administration did release an executive order on “Maintaining American Leadership in Artificial Intelligence,” but it is meant to promote (rather than regulate) AI research, and it lacks enforcement mechanisms. See Caleb Watney, “Trump’s Artificial Intelligence Strategy: Aspirations without Teeth,” Lawfare, February 20, 2019. 64. Quoted in Tim Higgins, “Elon Musk Lays Out Worst-­Case Scenario for AI Threat,” Dow Jones Institutional News, July 16, 2017. 65. Ian Bremmer, “China Embraces AI: A Close Look and a Long View,” Sinovation Ventures, December 2017. 66. Rebecca Fannin, “China’s Secret Goal Is to Crush Silicon Valley,” CNBC News, May 22, 2018, https://­www​.­cnbc​.­com​/­2018​/­05​/­22​/­chinas​-­secret​-­goal​-­is​-­to​-­crush​ -­silicon​-­valley​.­html. 67. Edward Tse, “Inside China’s Quest to Become the Global Leader in AI,” The Worldpost, October 19, 2017. See also Kai-­Fu Lee, AI Superpowers: China, Silicon Valley, and the New World Order (Boston: Houghton Mifflin Harcourt, 2018), 18. 68. Quoted from Yujia He, “How China Is Preparing for an AI-­powered F ­ uture,” Wilson Briefs, June 2017, 1. In response, ­there are increasing calls for United States policy makers to develop a strategy for large-­scale AI research to ­counter the supposed competitive threat from the Chinese government’s support for AI. See Gregory C. Allen, “Understanding China’s AI Strategy: Clues to Chinese Strategic Thinking on Artificial Intelligence and National Security,” Centre for a New American Security, February 2019. But it remains to be seen if U.S. policy makers can implement the required long-­term strategic vision consistent with rule by dif­fer­ent po­liti­cal parties over the next few de­cades. Meanwhile, smaller countries such as the United Kingdom are also positioning themselves for a new era of “AI nationalism.” Even thinkers who endorse the ideal of AI as a global public good are calling for more expansive national AI strategies to protect smaller countries’ economic, military, and technological interests in the short term. See Ian Hogarth, “AI Nationalism,” Blog, June 13, 2018 (thanks to W ­ ill Fairbairn for a helpful email). 69. Lee, AI Superpowers, 63–65, 84, 98–99, 133–134. 70. Ibid., 114, 123. 71. See Kevin Rudd, “Understanding China’s Rise ­Under Xi Jinping,” The Sinocism China Newsletter, March 18, 2018, https://­nb​.­sinocism​.­com​/­p​/­understanding​-­chinas​ -­rise​-­under​-­xi​-­jinping​-­by​-­the​-­honourable​-­kevin​-­rudd. 72. “China Recruits Baidu, Alibaba and Tencent to AI ‘National Team,’ ” South China Morning Post, November 22, 2017. 73. ­Needless to say, t­ here are other possibilities. For example, perhaps the EU ­w ill win the “race” and w ­ ill also take the lead in effective regulation of malevolent AI. At the moment, however, it seems to be a “race” between Silicon Valley and China.

248  N o t e s t o C h a p t e r   5 74. Willy Wo-­Lap Lam, “Beijing Harnesses Big Data & AI to Perfect the Police State,” Jamestown, July 21, 2017; Paul Mozur, “Inside China’s Dystopian Dreams: A.I., Shame and Lots of Cameras,” New York Times, July 8, 2018; George Soros, “The AI Threat to Open Socie­ties,” Proj­ect Syndicate, January 24, 2019. For a more optimistic take, see Louise Lucas, “China Leads Asian Neighbors on Data Privacy,” Financial Times, May 31, 2018. 75. Given that potentially dangerous and unethical gene-­editing of ­human embryos took place in China, ­there may be doubts about the willingness of the CCP to regulate potentially malevolent AI research. But the question is how the CCP reacts to rogue research of this kind: It has punished the rogue scientist and joined efforts at regulating or banning such research. See Pam Belluck, “How to Stop Rogue Gene-­ Editing of ­Human Embryos,” New York Times, January 23, 2019. 76. Tegmark, Life 3.0, 191. 77. This verdict assumes that the CCP is seriously committed to the promotion of communist and Confucian values and to ­human well-­being in general. If that’s not the case, then the CCP loses the moral authority to lead the global crusade against malevolent AI.

s e l e c t e d bi bl io g r a p h y

Achen, Christopher H., and Larry M. Bartels. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Prince­ton, NJ: Prince­ton University Press, 2016. Afeissa, Hicham-­Stéphane, and Jean-­Baptiste Vilmer, eds. Philosophie animale: Différence, responsabilité, et communauté. Paris: VRIN, 2015. Allen, Gregory C. “Understanding China’s AI Strategy: Clues to Chinese Strategic Thinking on Artificial Intelligence and National Security.” Center for a New American Security, 2019. American Psychological Association. “Speaking of Psy­chol­ogy: Understanding Your Racial Biases.” http://­www​.­apa​.­org​/­research​/­action​/­speaking​-­of​-­psychology​ /­understanding​-­biases​.­aspx. Accessed August 11, 2019. Ames, Roger T. “Achieving Personal Identity in Confucian Role Ethics: Tang Junyi on ­Human Nature as Conduct.” Oriens Extremus 49 (2010): 143–166. Ames, Roger T. Confucian Role Ethics: A Vocabulary. Honolulu: University of Hawai‘i Press, 2011. Anderson, Elizabeth S. “Equality.” In The Oxford Handbook of Po­liti­cal Philosophy, edited by David Estlund, 40–57. Oxford: Oxford University Press, 2012. Anderson, Elizabeth S. “What Is the Point of Equality?” Ethics 109, no. 2 (1999): 287–337. ­Angle, Stephen C. “Building Bridges to Distant Shores.” In Appreciating the Chinese Difference: Engaging Roger T. Ames on Methods, Issues, and Roles, edited by Jim Behuniak, 159–182. Albany: SUNY Press, 2018. ­Angle, Stephen C. “Vicious Oppression, Valuable Deference, and Provisional Hierarchy: A Con­temporary Confucian Perspective.” Paper presented at the Hierarchy and Equality Workshop, sponsored by the Berggruen Institute, held at Stanford University’s Center for Advanced Study in the Behavioral Sciences, Stanford University, March 2016. Appiah, Anthony. As If: Idealization and Ideals. Cambridge, MA: Harvard University Press, 2017. 249

250  s e l e c t e d b i b l io g r a p h y Aristotle. Nicomachean Ethics: Translation, Introduction, Commentary. Edited and translated by Sarah Broadie and Christopher Rowe. New York and Oxford: Oxford University Press, 2002. Aristotle. The Politics and the Constitution of Athens. Stephen Everson (ed.). Benjamin Jowett (trans.). Cambridge: Cambridge University Press, 1996. Babones, Salvatore. American Tianxia: Chinese Money, American Power and the End of History. Bristol: Policy Press, 2017. Bagg, Samuel. “The Power of the Multitude: Answering Epistemic Challenges to Democracy.” American Po­liti­cal Science Review 12, no. 4 (2018): 100–114. Bai, Tongdong. Against Po­liti­cal Equality: The Confucian Case. Prince­ton, NJ: Prince­ ton University Press, 2019. Bai, Xiaoyong. “易学与东汉政治初探” [On the relation between the study of the Book of Changes and politics in the Eastern Han Dynasty]. Zhengjiang xuekan 1 (2013): 42–47. Barnhill, Anne, Mark Budolfson, and Tyler Doggett. Food, Ethics, and Society: An Introductory Text with Readings. Oxford: Oxford University Press, 2016. Beard, Mary. SPQR: A History of Ancient Rome. New York: Liveright, 2015. Bell, Daniel A. Beyond Liberal Democracy: Po­liti­cal Thinking for an East Asian Context. Prince­ton, NJ: Prince­ton University Press, 2006. Bell, Daniel A. The China Model: Po­liti­cal Meritocracy and the Limits of Democracy. Prince­ton, NJ: Prince­ton University Press, 2016. Bell, Daniel A. China’s New Confucianism: Politics and Everyday Life in a Changing Society. Prince­ton, NJ: Prince­ton University Press, 2008. Bell, Daniel A. “Roles, Community, and Morality: Comment on Confucian Role Ethics.” In Appreciating the Chinese Difference: Engaging Roger T. Ames on Methods, Issues, and Roles, edited by Jim Behuniak, 203–212. Albany: SUNY Press, 2018. Bell, Daniel A. “Vertical Demo­cratic Meritocracy in China: Response to Comments.” Journal of Chinese Humanities 4, no. 1 (2018): 111–123. Bell, Daniel A., and Avner de-­Shalit. The Spirit of Cities: Why the Identity of a City ­Matters in a Global Age. Prince­ton, NJ: Prince­ton University Press, 2011. Bell, Daniel A., and Thaddeus Metz. “Confucianism and Ubuntu: Reflections on a Dialogue between Chinese and African Traditions.” Journal of Chinese Philosophy 38 (Supplement) (2011): 78–95. Bellah, Robert N., Richard Madsen, William M. ­Sullivan, Ann Swidler, and Steven M. Tipton. Habits of the Heart: Individualism and Commitment in American Life. Berkeley: University of California Press, 1985. Benn, Charles. China’s Golden Age: Everyday Life in the Tang Dynasty. Oxford: Oxford University Press, 2002.

s e l e c t e d b i b l io g r a p h y   251 Besthorn, Fred H. “Deep Ecological ‘Insectification’: Integrating Small Friends with Social Work.” In Animals in Social Work: Why and How They ­Matter, edited by Thomas Ryan, 3–17. Houndsmills: Palgrave Macmillan, 2014. Bhargava, Rajeev. “Asoka’s Dhamma as Civic Religion: Toleration, Civility, Communal Harmony.” Paper presented at the workshop on “Classical Indian & Chinese World Views on Global Order: A Comparison,” Berggruen Institute, Bangkok, January 2018. Boehm, Christopher. Hierarchy in the Forest: The Evolution of Egalitarian Behaviour. Cambridge, MA: Harvard University Press, 1999. Borenstein, Jason, and Yvette Pearson. “Robot Caregivers: Ethical Issues across the ­Human Lifespan.” In Robot Ethics: The Ethical and Social Implications of Robotics, edited by Patrick Lin, Keith Abney, and George A. Bekey, 251–266. Cambridge, MA: The MIT Press, 2014. Bostrom, Nick. Superintelligence: Paths, Dangers, Strategies. Oxford: Oxford University Press, 2014. Breeder, Ian. “China Embraces AI: A Close Look and a Long View.” Sinovation Ventures, December 2017. Brennan, Jason. Against Democracy. Prince­ton, NJ: Prince­ton University Press, 2017. Brennan, Jason. The Ethics of Voting. Prince­ton, NJ: Prince­ton University Press, 2011. Brown, Chris. “Our Side? Critical Theory and International Relations.” In Critical Theory and World Politics, edited by Richard Wyn Jones, 191–204. Boulder: Lynne Rienner, 2001. Burke, Edmund. Select Works of Edmund Burke. Vol. 2: Reflections on the Revolution in France (1790). Indianapolis, IN: Liberty Fund, 1999. Buswell, Robert E. Encyclopedia of Buddhism. New York: Macmillan Library Reference, 2003. Callaghan, William A. “Chinese Visions of World Order: Post-­Hegemonic or a New Hegemony?” International Studies Review 10, no. 4 (2008): 749–761 Cao, Feng. “Pre-­Qin Daoist Reflections on the Xianneng [Po­liti­cal Meritocracy].” Journal of Chinese Humanities 4, Caterina Weber (trans.), no. 1 (2018): 65–90. Cao, Xueqin. A Dream of Red Mansions (1791). Beijing: ­People’s Lit­er­a­ture Press, 2002. Caplan, Bryan. The Myth of the Rational Voter: Why Democracies Choose Bad Policies. Prince­ton, NJ: Prince­ton University Press, 2007. Carlson, Allen. “Moving beyond Sovereignty? A Brief Consideration of Recent Changes in China’s Approach to International Order and the Emergence of the Tianxia Concept.” Journal of Con­temporary China 20, no. 68 (2011): 89–102. Chan, Alan, and Tan Sor-­Hoon. Filial Piety in Chinese Thought and History. London: Routledge, 2004. Chan, Joseph. Confucian Perfectionism: A Po­liti­cal Philosophy for Modern Times. Prince­ ton, NJ: Prince­ton University Press, 2014.

252  s e l e c t e d b i b l io g r a p h y Chan, Joseph. “A Confucian Perspective on ­Human Rights.” In The East Asian Challenge for H ­ uman Rights, edited by Joanne R. Bauer and Daniel A. Bell, 212–240. Cambridge: Cambridge University Press, 1999. Chan, Joseph. “Demo­cratic Equality or Confucian Hierarchy?” Paper presented at Berggruen Workshop on Hierarchy and Equality, Center for Advanced Study in the Behavioral Sciences, Stanford University, March 13, 2016. Chen, Linguo. “两晋九品中正制与选官制度” [The nine-­rank system in the Western and Eastern Jin dynasty]. Lishi yanjiu 3 (1987): 101–115. Coeckelbergh, Mark. “The Tragedy of the Master: Automation, Vulnerability, and Master.” Ethics and Information Technology 17 (2015): 219–229. Confucius. The Analects of Confucius: A Philosophical Translation. Translated by Roger T. Ames and Henry Rosemont Jr. New York: Ballantine Books, 1998. Confucius. The Analects. The Internet Classics Archive, 2018. http://­classics.mit.edu​ /­Confucius/analects.1.1.html. Accessed June 30, 2018. Confucius. The Analects of Confucius. Translated by Arthur Waley. London: Routledge, 2012. Conley, Sarah. Against Autonomy: Justifying Coercive Paternalism. Cambridge: Cambridge University Press, 2013. Cosendey, Fanny. Le rang: Préséance et hiérarchies dans la France d’Ancien Régime. Paris: Éditions Gallimard, 2016. de Bary, William Theodore, Stephen Hay, Royal Weiler, and Andrew Yarrow, eds. Sources of Indian Tradition. Delhi: Motilal Banardidass, 1963. Ding, Chuanjing, ed. 宋人轶事汇编 [Edited Collection of Anecdotes from the Song Dynasty]. Beijing: Zhonghua Shuju, 1981. Donaldson, Sue, and W ­ ill Kymlicka. Zoopolis: A Po­liti­cal Theory of Animal Rights. Oxford: Oxford University Press, 2011. Doniger, Wendy. “Compassion ­toward Animals, and Vegetarianism.” Excerpted in Food, Ethics, and Society: An Introductory Text with Readings. Edited by Anne Barnhill, Mark Budolfson, and Tyler Doggett, 301–304. New York and Oxford: Oxford University Press, 2017. Doniger, Wendy. The Mare’s Trap: Nature and Culture in the Kamasutra. New Delhi: Speaking Tiger Publishing Pvt. Ltd., 2015. Doniger, Wendy, and Sudhir Kakar. “Introduction.” In Kamasutra: A New, Complete En­glish Translation of the Sanskrit Text. Oxford: Oxford University Press, 2009. Duettmann, Allison, Olga Afanasjeva, Stuart Armstrong, Ryan Braley, Jessica Cussins, Jeffrey Ding, Peter Eckersley, Melody Guan, Alyssa Vance, and Roman Yampolskiy. “Artificial General Intelligence: Coordination & ­Great Powers.” Foresight Institute Strategy Meeting on AGI, 2018. https://­foresight​.­org​/­wp​-­content​/­uploads​ /­2018​/­11​/­AGI​-­Coordination​-­Geat​-­Powers​-­Report​.­pdf. 

s e l e c t e d b i b l io g r a p h y   253 Ebrey, Patricia B. Confucianism and ­Family Rituals in Imperial China: A Social History of Writing about Rites. Prince­ton, NJ: Prince­ton University Press, 2014. Etzioni, Amitai. The Spirit of Community: Rights, Responsibilities, and the Communitarian Agenda. New York: Crown Publishers, 1993. Etzioni, Amitai, and Oren Etzioni. “Incorporating Ethics into Artificial Intelligence.” Journal of Ethics 21, no. 4 (2017): 403–418. https://­philpapers​.­org​/­archive​/­ETZIEI​ .­pdf. Executive Office of the President. Preparing for the F ­ uture of Artificial Intelligence. White House, 2016. https://­obamawhitehouse​.­archives​.­gov​/­sites​/­default​/­files​ /­whitehouse​_­files​/­microsites​/­ostp​/­NSTC​/­preparing​_­for​_­the​_­future​_­of​_­ai​ .­pdf. Accessed June 30, 2018. Fairbank, John K., and Ssu-­Yu Teng. “On the Ch’ing Tributary System.” Harvard Journal of Asiatic Studies 6, no. 2 (1941): 135–246. Fan, Ye. 后汉书 [Book of the L ­ ater Han]. Beijing: Zhonghua Shuju, 2007. Fang, Xuanling. 晋书 [ Jin Shu]. Beijing: Zhonghua Shuju, 1974. Fang, Zhaohui. “三纲真是糟粕吗?” [Are the “Three Cardinal Guides” really so worthless? Re-­examining the Historical and Practical Significance of the Three Cardinal Guides]. Tianjin Social Sciences 2 (2011): 47–52. Feng, Xiang. “我是阿尔法-­论人机伦理” [I am Alpha: On human-­computer ethics]. Wenhua zongheng 12 (2017): 128–139. Ferguson, Niall. The Square and the Tower: Networks and Power, from the Freemasons to Facebook. New York: Penguin Press, 2017. Foer, Franklin. World without Mind: The Existential Threat of Big Tech. New York: Penguin Press, 2007. Foer, Jonathan S. Eating Animals. London: Penguin UK, 2010. Francione, Gary. Introduction to Animal Rights: Your Child or the Dog? Philadelphia: ­Temple University Press, 2010. Frankfurt, Harry J. On In­equality. Prince­ton, NJ: Prince­ton University Press, 2015. Fukuyama, Francis. Po­liti­cal Order and Po­liti­cal Decay: From the Industrial Revolution to the Globalization of Democracy. New York: Farrar, Straus and Giroux, 2015. Fung, Courtney J. “China and the Responsibility to Protect: From Opposition to Advocacy.” United States Institute of Peace, June 8, 2016. https://­w ww​.­usip.org​ /­publications/2016/06/china-­and-­responsibility-­protect-­opposition-­advocacy. Accessed July 20, 2018. Gaita, Raimond. The Phi­los­o­pher’s Dog. New York: Routledge, 2016. Gardels, Nathan, and Nicholas Berggruen. Renovating Democracy: Governing in the Age of Globalization and Capital. Berkeley: University of California Press, 2019. Ge, Zhaoguang. What Is China? Territory, Ethnicity, Culture, and History. Translated by Michael Gibbs Hill. Cambridge, MA: Harvard University Press, 2018.

254  s e l e c t e d b i b l io g r a p h y Goldstein, Joshua. International Relations. New York: Pearson-­Longman, 2005. Gong, Ting, and Wu Muluan. “我国2019年腐败案例报告：基于2800余个报道案例的 分析” [Research report on China’s corruption cases between 2000 and 2009: An empirical analy­sis of 2800 cases]. Shehuixue yanjiu 4 (2012): 204–220. Grace, Katja, John Salvatier, Allan Dafoe, Zhang Baobao, and Owain Evans. “When ­Will AI Exceed H ­ uman Per­for­mance?” AI Experts. 2017. https://­www​.­researchgate​ .­net​/­publication​/­317100941​_­W hen​_­Will​_­AI​_­Exceed​_­Human​_­Performance​ _­Evidence​_­from​_­AI​_­Experts. Accessed March 14, 2018. Hall, Stephen S. Wisdom: From Philosophy to Neuroscience. New York: Vintage Books, 2010. Halteman, Mattew C. “Compassionate Eating as Care of Creation.” Excerpted in Food, Ethics, and Society: An Introductory Text with Readings. Edited by Anne Barnhill, Mark Budolfson, and Tyler Doggett, 292–300. New York and Oxford: Oxford University Press, 2017. Hankins, James. Virtue Politics: Po­liti­cal Thought in Re­nais­sance Italy from Petrarch to Machiavelli. Cambridge, MA: Harvard University Press, 2019. Hansen, Mogens Herman. The Athenian Democracy in the Age of De­mos­the­nes: Structure, Princi­ples, and Ideology. Norman: University of Oklahoma Press, 1991. Harari, Yuval Noah. Sapiens: A Brief History of Humankind. New York: Harper, 2015. He, Baogang. “A New Interpretation of the China Model: In Search of Mixed Governance in Con­temporary China.” Paper presented at the 2019 IPP International Conference on Civilization and Governance, Guangzhou, August 17–18, 2019. He, Xinyin. 何心隐文集 [Collected works of He Xinyin]. Beijing: Zhonghua Shuju, 1960. He, Yujia. “How China Is Preparing for an AI-­Powered F ­ uture.” Wilson Briefs, June 2017. Heilmann, Sebastian. “Policy Experimentation in China’s Economic Rise.” Studies in Comparative International Development 43 (2008): 1–12. Hobson, John M., and, J. C. Sharman. “The Enduring Place of Hierarchy in World ­Politics: Tracing the Social Logics of Hierarchy and Po­liti­cal Change.” Eu­ro­pean Journal of International Relations 11 (2005): 63–98. Hsu, Immanuel C. Y. China’s Entrance into the F ­ amily of Nations. Cambridge, MA: Harvard University Press, 1960. Hu, Baoguo. “关于九品中正制的几点意见” [A few points on the nine-­rank system]. Lishi yanjiu 1 (1988): 191–192. Hu, Pingsheng. 孝经译注 [On The Classic of Filial Piety]. Beijing: Zhonghua Shuju, 1996. Hu, Shuyun. “九品官人法性质辨析” [The nature of the nine-­rank system]. Dongbei shifan daxue bao 6 (2003): 78–84. Huang, Echo. “The East and West Have Very Dif­fer­ent Ideas on How to Save in a Self-­ Driving Car Accident.” Quartz, November 1, 2018.

s e l e c t e d b i b l io g r a p h y   255 Huang, Ray. 1587, A Year of No Significance: The Ming Dynasty in Decline. New Haven: Yale University Press, 1981. Ishiguro, Kazuo. The Remains of the Day. London: Faber and Faber, 1989. Ismard, Paulin. La démocratie contre les experts: Les esclaves publics en Grèce ancienne. Paris: Éditions du Seuil, 2015. Jabr, Ferris. “The Person in the Ape: A History of H ­ umans Trying and Failing to Understand the Minds of Apes.” Lapham’s Quarterly. 2018. https://­w ww​ .­laphamsquarterly​.­org​/­states​-­mind​/­person​-­ape. Johnston, Alasdair I. Cultural Realism: Strategic Culture and ­Grand Strategy in Chinese History. Prince­ton, NJ: Prince­ton University Press, 1995. Johnston, F. R. Twilight in the Forbidden City. Vancouver: Soul Care Publishing, 2008 [orig. pub. 1934]. Joy, Melanie. Why We Love Dogs, Eat Pigs, and Wear Cows: An Introduction to Carnism. San Francisco: Conari Press, 2011. Kang, D.C. East Asia before the West: Five Centuries of Trade and Tribute. New York: Columbia University Press, 2010. Kang, Xiaofei. The Cult of the Fox: Power, Gender, and Popu­lar Religion in Late Imperial and Modern China. New York: Columbia University Press, 2006. Kautilya. King, Governance, and Law in Ancient India: Kautilya’s Arthasastra. Translated by Patrick Olivelle. New Delhi: Oxford University Press, 2013. Khyentse, Dzongsar Jamyang. What Makes You Not a Buddhist. Boulder: Shambhala Publications, 2008. Kim, Orun K. “­Korea’s Experiences with Big Neighbors.” Paper presented at the Berggruen Institute workshop “What Is Tianxia?,” Peking University, June 16–17, 2018. Komjathy, Louis. “Animals and Daoism.” Encyclopaedia Britannica. Advocacy for Animals. 2011. http://­advocacy.britannica.com/blog/advocacy/2011/09/daoism​ -­and-­animals. Accessed July 20, 2018. Kong, Xinfeng, and Daniel A. Bell. “曲阜: 儒家文化之城” [Qufu: The city of Con­ fucian culture]. In 城市精神2： 包容与认同 [The spirit of cities II: Inclusion and identity]. edited by Daniel A. Bell and Avner de Shalit, 52–85. Chongqing: Chongqing Press, 2017. Korsgaard, Christine M. “Getting Animals in View.” Excerpted in Food, Ethics, and Society: An Introductory Text with Readings. Edited by Anne Barnhill, Mark Budolfson, and Tyler Doggett, 366–372. New York and Oxford: Oxford University Press, 2017. Kynge, James. China Shakes the World: A Titan’s Rise and Troubled F ­ uture—­And the Challenge for Amer­i­ca. New York: Mari­ner Books, 2017. Lake, David A. Hierarchy in International Relations. Ithaca, NY, and London: Cornell University Press, 2009. Lao, Gan. “汉代察举制度考” [On the recommendation system in the Han Dynasty]. In 中研院历史语言研究所集刊 [Collected journals of the Institute of History and

256  s e l e c t e d b i b l io g r a p h y Philology]. Taiwan: Institute of History and Philology, Academia Sinica, vol. 17, April (1948): 79–129. Leavitt, Harold J. Top Down: Why Hierarchies Are ­Here to Stay and How to Manage Them More Effectively. Cambridge, MA: Harvard Business School Press, 2004. Lee, Kai-­Fu. AI Superpowers: China, Silicon Valley, and the New World Order. Boston: Houghton Mifflin Harcourt, 2018. Lenta, Patrick. Corporal Punishment: A Philosophical Assessment. New York: Routledge, 2018. Levinas, Emmanuel. Eros, littérature, et philosophie. Paris: Éditions Grasset, 2013. Levinas, Emmanuel. Totalité et infini. Essai sur l’extériorité. Paris: Le livre de poche, 1990 [orig. pub. La Haye: Martinus Nijhoff, 1961]. Li, Cunshan. “朱子‘学校贡举私议’ 述评” [On Zhu Xi’s comments of chief examiners in school]. Journal of the Gradu­ate School of the Chinese Acad­emy of Social Sciences 2 (2011): 19–27. Li, Fang. “试论唐朝的‘中国’与‘天下’” [An analy­sis of “China” and “Tianxia” in the Tang Dynasty]. Zhongguo bianjiang yanjiu 17 (2007): 10–20. Li, Jin. Cultural Foundations of Learning: East and West. Cambridge: Cambridge University Press, 2012. Li, Yangfan. “天下观念” [On the concept of Tianxia]. Guoji zhengzhi yanjiu, no. 1 (2002): 107–111. Li, Yanhong et al. Zhineng Geming [Intelligent revolution]. Beijing: Citic Press, 2017. Li, Yuxiu.弟子规 [Di Zi Gui]. Translated by Feng Xinming. http://­tsoidug.org/ dizigui_trans_simp.php. Accessed June 30, 2018. Liang, Shuming. 东西文化及其哲学 [Eastern and Western cultures and their philosophies]. Beijing: The Commercial Press, 2009 [orig. pub. 1921]. Liang, Shuming. 乡村理论建设 [Rural construction theory]. Beijing: Shiji Chubanshe, 2006 [orig. pub. 1937]. Liu, Luming. 金光明经 [Golden Light Sutra]. Beijing: Zhonghua Shuju, 2010. Liu, Yiqing, Liu Jun, and Richard B. Mather. A New Account of Tales of the World. Ann Arbor: University of Michigan Press, 2002. Liu, Zehua. “王权主义: 中国思想文化的历史定位” [Monarchy: A historical orientation of Chinese intellectual culture]. Tianjin shehui kexue 3 (1998): 59–62. Loar, Lynn. “ ‘How Is Fido?’: What the ­Family’s Companion Animal Can Tell You about Risk Assessment and Effective Interventions—­If Only You Would Ask!,” excerpted in Animals in Social Work: Why and How They ­Matter, edited by Thomas Ryan, 135–150. Basingstoke: Palgrave Macmillan, 2014. Lorentzen, Peter, and Xi Lu. Personal Ties, Meritocracy, and China’s Anti-­Corruption Campaign, 2018. https://­papers​.­ssrn​.­com​/­sol3​/­papers​.­cfm​?­abstract​_­id​=­2835841​. Accessed March 14, 2019.

s e l e c t e d b i b l io g r a p h y   257 Luo, Dingsheng, and Wu Shihong. “浅谈智能护理机器人的伦理问题” [A discussion about the ethical prob­lems of intelligent nursing robots]. Technology and Society, no. 1 (2018): 25–39. Magee, Joe C., and Adam Galinksy. “Social Hierarchy: The Self-­Reinforcing Nature of Power and Status.” The Acad­emy of Management Annals 2 (2008): 351–398. Mao, Zedong. Quotations from Mao Tse Tung. Translated by David Quentin and Brian Baggins. Marxists​.­org, 2018. https://­www​.­marxists​.­org​/­reference​/­archive​/­mao​ /­works​/­red​-­book​/­index​.­htm. Accessed June 30, 2018. Marx, Karl. Capital. Marxists​.­org, 2018. https://­www​.­marxists​.­org​/­archive​/­marx​ /­works​/­1867​-­c1​/­ch07​.­htm. Accessed June 30, 2018. Marx, Karl. Capital. In The Marx-­Engels Reader. Second edition. Edited by Robert C. Tucker. New York: Norton and Com­pany, 1978. Marx, Karl. Critique of the Gotha Programme. Marxists​.­org, 2018. https://­www​.­marxists​ .­org​/­archive​/­marx​/­works​/­1875​/­gotha​/­ch01​.­htm. Accessed June 30, 2018. Marx, Karl. The German Ideology. Marxists​.­org. https://www​.­marxists​.­org​/­archive​ /­marx​/­works​/­1845​/­german​-­ideology​/­ch01a​.­htm. Accessed June 30, 2018. May, Simon. The Power of Cute. Prince­ton, NJ: Prince­ton University Press, 2019. McClish, Mark, and Patrick Olivelle, eds. and trans. “Introduction.” The Arthasastra: Se­lections from the Classic Indian Work on Statecraft. Indianapolis: Hackett, 2012. Meisner, Maurice. Mao’s China and ­After: A History of the ­People’s Republic. 3rd ed. New York: The ­Free Press, 1999. Mencius. The Works of Mencius. http://­ctext.org/mengzi. Mengistu, Henok, Joost Huizinga, Jean-­Baptiste Mouret, and Jeff Clune. “The Evolutionary Origins of Hierarchy.” Computational Biology 12, no. 6 (2016): e1004829.​ https://­doi:10​.­1371​/­journal. pcbi.1004829. Meyer, Erin. The Culture Map: Decoding How P ­ eople Think, Lead, and Get Th ­ ings Done Across Cultures. New York: Public Affairs, 2014. Midgley, Mary. Animals and Why They M ­ atter. Athens: University of Georgia Press, 1998. Miller, David. Princi­ples of Social Justice. Cambridge, MA: Harvard University Press, 2001. Montaigne, Michel de. On Friendship. Translated by M. A. Screech. London: Penguin UK, 2004. Nehamas, Alexander. On Friendship. New York: Basic Books, 2016. Nichols, Terence L. “Who’s Afraid of Hierarchy? The P ­ eople of God Have to Be Or­ga­nized. ­Here’s How.” Commonweal 127, no. 7 (2000): 16–18. Nisbett, Richard E. The Geography of Thought: How Asians and Westerners Think Differently . . . ​and Why. New York: ­Free Press, 2003. Olivelle, Patrick. “Relations between States and Rulers in Ancient India: Asoka, Kautilya, and Manu,” 13. Paper presented at the workshop on “Classical Indian &

258  s e l e c t e d b i b l io g r a p h y Chinese World Views on Global Order: A Comparison,” Berggruen Institute, Bangkok, January 2018. Ollman, Bertell. Marx’s Vision of Communism. New York: New York University, 2018. https://­w ww​.­nyu​.­edu​/­projects​/­ollman​/­docs​/­v ision​_­of​_­communism​.­php. Accessed September 20, 2018. Pan, Fu’en, and Xu Yuqing. “论二程的人才观” [On the two Chengs’ view of talents]. Lanzhou daxue xuebao 1 (1987): 35–41. Pan, Wei, ed. 中国模式: 解读人民共和国的60年 [China model: Explaining sixty years of the ­People’s Republic]. Beijing: Central Compilation and Translation Press, 2009. Patapan, Haig, and Wang Yi. “The Hidden Ruler: Wang Huning and the Making of Con­temporary China.” Journal of Con­temporary China 27, no. 109 (2017): 1–14. Patterson, Charles. Eternal Treblinka: Our Treatment of Animals and the Holocaust. New York: Lantern Books, 2002. Pines, Yuri. “Between Merit and Pedigree: Evolution of the Concept of ‘Elevating the Worthy’ in Pre-­imperial China.” In The East Asian Challenge for Democracy: Po­ liti­cal Meritocracy in a Comparative Context, edited by Daniel A. Bell and Chenyang Li, 161–202. New York: Cambridge University Press, 2013. Pines, Yuri. “Changing Views of  ‘Tianxia’ in Pre-­imperial Discourse.” Oriens Extremus 43 (2002): 101–116. Pines, Yuri. “Pitfalls of Meritocracy: ‘Elevating the Worthy’ in Early Chinese Thought.” Paper presented at the workshop on Po­liti­cal Meritocracy. Harvard University, Cambridge, MA. November 2018. Pollan, Michael. How to Change Your Mind: What the New Science of Psychedelics Teaches Us About Consciousness, ­Dying, Addiction, Depression, and Transcendence. New York: Random House, 2018. Puett, Michael. “Equality, Hierarchy and the Discontents: Theories from Classical China.” Paper presented at Berggruen Workshop on Hierarchy and Equality. Center for Advanced Study in the Behavioral Sciences. Stanford University. March 13, 2016. Rachman, Gideon. Easternization: Asia’s Rise and Amer­i­ca’s Decline from Obama to Trump and Beyond. New York: Other Press, 2017. Rasmussen, Dennis C. The Infidel and the Professor: David Hume, Adam Smith, and the Friendship that S­ haped the Modern World. Prince­ton, NJ: Prince­ton University Press, 2017. Rawls, John. A Theory of Justice. Cambridge, MA: Harvard University Press, 2009. Renshon, Jonathan. Fighting for Status: Hierarchy and Conflict in World Politics. Prince­ ton, NJ: Prince­ton University Press, 2017. Ricard, Matthieu. Plaidoyer pour les animaux: Vers une bienveillance pour tous. Paris: Allary Editions, 2014.

s e l e c t e d b i b l io g r a p h y   259 Rosemont, Henry Jr. Against Individualism: A Confucian Rethinking of the Foundations of Morality, Politics, ­Family, and Religion. Lanham, MD: Lexington Books, 2015. Roth, Philip. American Pastoral. New York: Random House, 2016. Rousseau, Jean-­Jacques. The Confessions. Translated by J. M. Cohen. London: Penguin Books, 1953. Scanlon, Thomas M. Why Does In­equality M ­ atter. Oxford: Oxford University Press, 2018. Scarry, Elaine. On Beauty and Being Just. Prince­ton, NJ: Prince­ton University Press, 2013. Scruton, Roger. “Eating Our Friends.” Excerpted in Food, Ethics, and Society: An Introductory Text with Readings. Edited by Anne Barnhill, Mark Budolfson, and Tyler Doggett, 391–394. New York and Oxford: Oxford University Press, 2017. Shanahan, Murray. The Technological Singularity. Cambridge, MA: The MIT Press, 2015. Sheehan, Neil. A Bright Shining Lie: John Paul Vann and Amer­i­ca in Vietnam. New York: Vintage Books, 1989. Shen, Yue. 宋书 [Book of Song]. Beijing: Zhonghua Shuju, 1974. Shi, Tianjian. The Cultural Logic of Politics in Mainland China and Taiwan. New York: Cambridge University Press, 2015. Shields, Christopher. Aristotle. London and New York: Routledge, 2007. Shusterman, Richard. “Sex and Somaesthetics: Appreciating the Chinese Difference.” In Appreciating the Chinese Difference, edited by Jim Behuniak, 91–110. New York: State University of New York Press, 2018. Simpson, Brent, Robb Willer, and Cecilia L. Ridgeway. “Status Hierarchies and the Organ­ization of Collective Action.” So­c io­l og­i­cal Theory 30, no.  3 (2012): 149–166. Singer, Peter. Animal Liberation. 40th anniversary ed. New York: Open Road, 2016. Sintomer, Yves. “From Deliberative to Radical Democracy? Sortition and Politics in the Twenty-­First ­Century.” Politics and Society 46, no. 3 (2018): 337–357. Strerckx, Reol. The Animal and the Daemon in Early China. Albany: State University of New York Press, 2002. Sun, Anna. Confucianism as a World Religion: Contested Histories and Con­temporary Realities. Prince­ton, NJ: Prince­ton University Press, 2013. Tang, Changru. 九品中正制度试释 [On the nine-­rank system]. Beijing: Zhonghua Shuju, 2011. Tang, Wenming. “人伦理念的普世意义及其现代调适” [The universal significance of the “Ren Lun” concept and its modern adjustment: On the defense and reconstruction of the Wu Lun princi­ples by modern Confucian scholars]. Daode yu wenming 6 (2015): 5–12.

260  s e l e c t e d b i b l io g r a p h y Tannsjo, Thorbjorn. “It’s Getting Better All the Time.” Excerpted in Food, Ethics, and Society: An Introductory Text with Readings, 362–365. Edited by Anne Barnhill, Mark Budolfson, and Tyler Doggett. New York and Oxford: Oxford Univesity Press, 2017. Taplin, Jonathan. Move Fast and Break ­Things: How Facebook, Google, and Amazon Have Cornered Culture and Undermined Democracy. London: Pan Books, 2017. Tegmark, Max. Life 3.0: Being H ­ uman in the Age of Artificial Intelligence. New York: Alfred A. Knopf, 2017. Tian, Wenjun. “何心隐新论” [A new analy­sis of He Xinyin’s Thought]. Wuhan Daxue Xuebao (Shehui Kexue Ban), no. 5 (1986): 15–21. Turchin, Peter. “The Evolution of Hierarchy.” The Evolution Institute, 2014. https://­ evolution​-­institute​.­org​/­blog​/­the​-­evolution​-­of​-­hierarchy. Accessed July 20, 2018. Wace, Roman de Brut: A History of the British (Text and Tradition). Translated by Judith Weiss. Exeter, UK: University of Exeter Press, 2002. Walzer, Michael. Spheres of Justice: A Defense of Pluralism & Equality. Oxford: Basil Blackwell, 1983. Walzer, Michael. Thick and Thin: Moral Argument at Home and Abroad. Notre Dame: University of Notre Dame Press, 1994. Wang, Gungwu. “Ming Foreign Relations: Southeast Asia.” In The Cambridge History of China. Vol. 8: The Ming Dynasty, part 2: 1368–1644, 301–332. Edited by D. Twitchett and F. W. Mote. Cambridge: Cambridge University Press, 1998. Wang, Hui. “世纪的诞生” [The birth of the c­ entury]. Kaifang Shidai 4 (2017). http://­ www​.­opentimes​.­cn​/­Abstract​/­2365​.­html. Accessed June 30, 2018. Wang, Huning. 政治的逻辑 [The Logic of Politics]. Shanghai: Shanghai ­People Press, 2004. Wang, Pei. “Debates on Po­liti­cal Meritocracy in China: A Historical Perspective.” Philosophy and Public Issues (new series) 7, no. 1 (2017): 53–71. Wang, Pei. “A Study of Eros and Individuation in Levinas’s Early Philosophy.” Ph.D. thesis [in Chinese]. Tsinghua University, 2016. Wang, Shaoguang. 民主, 共和与抽签: 从雅典到威尼斯 [Democracy, republic and ­sortition: From Athens to Venice]. Beijing: CITIC Press, 2018. Wendt, Alexander, and Daniel Friedheim. “Hierarchy ­Under Anarchy: Informal Empire and the East German State.” International Organ­ization 49, no. 4 (1995): 689–721. Wohlleben, Peter. The Inner Life of Animals: Love, Grief, and Compassion: Surprising Observations of a Hidden World. Vancouver and Berkeley: Greystone Books, 2017. Wright, Robert. Why Buddhism Is True: The Science and Philosophy of Meditation and Enlightenment. New York: Simon and Schuster, 2017. Xi, Lu, and Peter Lorentzen. “Rescuing Autocracy from Itself: China’s Anti-­Corruption Campaign,” November 6, 2016. https://­ssrn​.­com​/­abstract​=­2835841. Accessed July 23, 2018.

s e l e c t e d b i b l io g r a p h y   261 Xunzi. Xunzi: The Complete Text. Translated by Eric L. Hutton. Prince­ton, NJ: Prince­ ton University Press, 2014. Xunzi. Xunzi: A Translation and Study of the Complete Works. Vol. 1. Translated by John Knoblock. Stanford, CA: Stanford University Press, 1988. Yan, Buke. 波峰与波谷：秦汉魏晋南北朝的政治文明 [Peaks and valleys: Po­liti­cal cultures of the Qin, Han, Wei, Jin, and Northern and Southern dynasties]. Beijing: Peking University Press, 2018. Yan, Xuetong. Ancient Chinese Thought, Modern Chinese Power. Prince­ton, NJ: Prince­ton University Press, 2011. Yan, Xuetong. Leadership and the Rise of G ­ reat Powers. Prince­ton, NJ: Prince­ton University Press, 2019. Yan, Yunxiang. “Un­balanced Reciprocity: Asymmetrical Gift Giving and Social Hierarchy in Rural China.” In the Question of the Gift: Essays Across Disciplines, edited by M. Osteen, 77–94. London: Routledge Press, 2002. Yang, Bojun. 春秋左传注 [Commentary on Zuozhuan]. Beijing: Zhonghua Shuju, 1990. Yao, Yang. “An Anatomy of the Chinese Selectocracy.” Researchgate 2016. https://­www​ .­researchgate​.­net​/­publication​/­294883730​_­A n​_­A natomy​_­of​_­the​_­Chinese​ _­Selectocracy. Accessed July 20, 2018. Yeung, Bernard. “Notes on Meritocracy: Insights from Tang’s Civil Ser­vice Exam and Poetry.” Paper presented at “Meritocracy and India Workshop,” Harvard Center Shanghai, May 16–17, 2018. Yu, Zhong. 喻中读《尚书》 [Yu Zhong reads Shangshu]. Beijing: Peking University Press, 2011. Zhang, Feng. Chinese Hegemony: ­Grand Strategy and International Institutions in East Asian History. Palo Alto, CA: Stanford University Press, 2015. Zhang, Feng. “Rethinking the ‘Tribute System’: Broadening the Conception Horizon of Historical East Asian Politics.” Chinese Journal of International Politics 2, no. 4 (2009): 545–574. Zhang, Feng. “The Tianxia System: World Order in a Chinese Utopia.” Chinese Heritage Quarterly 21 (2010): 108–112. Zhang, Taisu. “Social Hierarchies and the Formation of Customary Property Law in Pre-­Industrial China and ­England.” The American Journal of Comparative Law 62 (2014): 171–220. Zhang, Xianglong. 家与孝：从中西间视野看 [­Family and filial piety: From the perspective of Chinese and Western thought]. Beijing: SDX Joint Publishing Com­ pany, 2017. Zhang, Xianglong. “孝意识的时间分析” [A temporal analy­sis of the consciousness of filial piety]. Journal of Peking University (Philosophy and Social Science Edition) 43, no. 1 (2006): 14–24.

262  s e l e c t e d b i b l io g r a p h y Zhang, Yongle. “The ­Future of Meritocracy.” Translated by Colleen Howe. The Journal of Chinese Humanities 4, no. 1 (2018): 49–64. Zhao, Dingxin. The Confucian-­Legalist State: A New Theory of Chinese History. Oxford: Oxford University Press, 2015. Zhao, Tingyang. “All-­under-­Heaven and Methodological Relationism: An Old Story and New World Peace.” In Con­temporary Chinese Po­liti­cal Thought: Debates and Perspectives, edited by Fred Dallmayr and Zhao Tingyang, 46–66. Lexington: University of Kentucky Press, 2012. Zhao, Tingyang. 天下的当代性 [The possibility of “All-­under-­heaven”]. Beijing: Citic Press, 2016. Zheng, Yongnian. The Chinese Communist Party as an Orga­nizational Emperor: Culture, Reproduction, and Transformation. London: Routledge, 2010. Zhu, Xi. 四书章句集注 [Commentaries on Four Books]. Beijing: Zhonghua Shuju, 2011. Zhuang, Guotu. “略论朝贡制度的虚幻: 以古代中国与东南亚的朝贡关系为列” [An account of the unreality of the tributary system: An example of the relationships in the tributary system in ancient China and East Asia]. Nanyang wenti yanjiu 3 (2005): 1–49. Ziliotti, Elena. “The Meritocratic Challenge to Democratic Theories.” Ph.D. thesis. National University of Singapore and King’s College, London, 2018. Zitek, Emily M., and Larissa Z. Tiedens. “The Fluency of Social Hierarchy: The Ease with Which Hierarchical Relationships Are Seen, Remembered, Learned, and Liked.” Journal of Personality and Social Psy­chol­ogy 102, no. 1 (2012): 98–115.

I n de x

Acharya, Amitav, 141 alcohol consumption, 6–7, 208n5, 208n7–208n8 Alibaba Group, 204 Alpha Go, 199 Ames, Roger, 191 amoral realism, 111 Analects of Confucius, The, 49–50, 152, 171, 189 Anderson, Elizabeth, 15 ­Angle, Stephen, 27 Animal Liberation (Singer), 148 animals: Ashoka and, 115; beauty and, 164–65; Bentham and, 147; Buddhism and, 153; Chris­tian­ity and, 145, 147, 152–53, 170; as citizens, 155–56, 158–59; Confucius and, 152–53, 170–72, 190; cruelty to, 153–54, 160, 168, 173; cuteness and, 102–3, 103, 104, 227n75, 227n77–227n80; Descartes and, 145–46; Didi (cat), 156–59, 236n31–236n32; duties owed to, 154–55, 159–60; eating meat and, 166–76; filial piety and, 46, 216n48; food technology and, 167–68; Harambe (gorilla), 151; justice and, 161–62; Kant on dogs, 159–60; lack of concern for, 145; liberation of, 147–49; meat production and, 145;

number of in captivity, 149; population of, 144; psychic numbing and, 174–75; religion and animal ethics, 169–70; Singer and, 147–49; social harmony and, 160–61, 173; theory of rights for, 149–52, 154; Xiao Bao (cat), 156–57; Zsa Zsa (dog), 165 Aristotle, 32–34, 55–59, 63, 69, 213n12 Arthasastra (Kautilya), 111 artificial intelligence: artificial general intelligence (AGI), 198; artificial superintelligence (ASI) and, 198, 203; Asilomar AI Princi­ples and, 199–200; driverless cars and, 183–97, 244n39; Google and, 200–202, 206, 245n51; higher communism and, 183–86; race for dominance in, 203–6; regulation of research into, 202–4, 247n68; threat to humanity and, 178–79, 199, 201–4, 206, 245n53. See also machines Ashoka, 114–17, 132, 137, 141 Asian Infrastructure Investment Bank, 141 Asilomar AI Princi­ples, 199–200 Association of Southeast Asian Nations (ASEAN), 110 263

264 i n de x Athens, 67 Australia, 151–52 automated cars. See driverless cars autonomy, 191, 193 Babones, Salvatore, 127, 232n68 Baggini, Julian, 21 Baidu, 204 Bai Tongdong, 172 beauty, 162–64 Beauty and Being Just, On (Scarry), 162–63 Bentham, Jeremy, 147 Bergruen Institute, 21 Bernard, Claude, 146 Besthorn, Fred, 152 Bhargava, Rajeev, 116 Blanchard-­Field, Fredda, 49 Boehm, Christopher, 35, 214n20 Boétie, Étienne de la, 34 Book of Changes, The, 87 Book of Filial Piety, The, 87 Book of Han, The, 70 Borentstein, Jason, 192 Bostrom, Nick, 198 bowing, 23 Brin, Sergey, 200. See also Google Buddhism, 114–15, 117, 141, 153, 164, 236n27 Burke, Edmund, 13 Callahan, William, 133 Canada, 138–39 Cao Feng, 100 Capital (Marx), 183–85, 188 capitalism, 145, 167, 180–85 Carlson, Allen, 136–37 cars. See driverless cars Central Commission for Discipline Inspection, 81

Chan, Joseph, 16, 93 Cheng Hao, 90–91 Cheng Xu, 31 Cheng Yi, 90–91 ­children: equality and, 29; filial piety and, 45–47; Sweden and, 23 China: age based hierarchies and, 45–55; alcohol consumption and, 6–7, 80–81, 208n5, 208n7–208n8; artificial intelligence and, 203–6; ASEAN and, 110; ­belt and road initiative, 124; Confucianism and, 22, 179–80; corruption and, 81–84; Cultural Revolution and, 13, 22, 71–72, 182, 226n66; culture of cuteness and, 102–3, 103, 104, 227n75, 227n77–227n79, 228n80; democracy and, 72, 75, 92, 95, 225n57, 226n62; drunk driving and, 80–81; economic war with the US and, 223n32; filial piety and, 29–30, 45–46; friendship and, 18; ­Great Leap Forward and, 180, 182; higher communism and, 179–83; Legalism and, 69–72; Marxism and, 180; meat consumption and, 173; Ming, 126; organ­ization department of, 76–77; patriarchy and, 38–39; po­liti­cal meritocracy and, 18–19, 71–78; poodles and, 165; Property Law and, 95–96; reciprocity and, 124, 136–38, 140–41; social hierarchies and, 22–24; sovereignty and, 130–31; Tianxia hierarchy and, 135–42; tributary system in, 126–29, 131, 137 Chinese Communist Party, 23, 51, 81, 93–94, 179–81, 204–6 Chris­tian­ity, 145, 147, 152–53, 170 circle of kingdoms, the, 112

i n de x  

265

class: po­liti­cal meritocracy and, 85; rituals and, 120–21 cloning, 202 communism, 56, 179–88, 241n15 Confessions, The (Rousseau), 14 Confucianism: animals and, 152–53, 170–73, 190; Cheng Hao and, 90–91; Cheng Yi, 90–91; China and, 179–80; Dong Zhongshu and, 70; driverless cars and, 193–97; economic equality and, 51; emotional intelligence and, 48–49; filial piety and, 46, 54; formal seating arrangements and, 4; friendship and, 31–32; harmony and, 52; He Xinyin and, 31–33, 213n12; hierarchy and, 22; imperial examination system and, 90; learning and, 47–48; marriage and, 54; Mencius and, 45–46, 57, 60, 130, 132, 174–75, 189–90; moral growth and, 49–50; politics and, 70–71, 84, 93, 96, 98, 104, 226n67; recommendation system and, 86; rituals and, 121; social relations and, 189–92; tianxia and, 133–35; Xunzi and, 117; Zhu Xi and, 90, 134 Confucius: animals and, 152, 171, 234n21; authoritative rule and, 93; filial piety and, 59; friendship and, 31–32, 189; learning and, 48; moral growth and, 49–50 creative work, 188–91 Critique of the Gotha Program (Marx), 187 cuteness, 102–3, 103, 104, 227n75, 227n77–227n80

Dawn, Karen, 151 democracy: ancient Athens and, 65; China and, 72, 75, 92, 95–96, 225n57, 226n62; large-­scale po­liti­cal communities and, 67–68, 73, 221n12; limiting of po­liti­cal power and, 79; po­liti­cal meritocracy and, 75, 95–96 Deng Xiaoping, 180 Descartes, René, 145–46 dharma, 114–15, 117 distributive princi­ple, 15 domination, 35, 37–38 Donaldson, Sue, 149–50, 154–56, 158–59, 168 Dong Zhongshu, 70 Doniger, Wendy, 40–41 Dream of Red Mansions, A (Cao), 218n82 driverless cars, 193–97, 244n39 Ducey, Doug, 203

Daoism, 20, 98–102, 104, 235n22; Laozi, 98–99; Zhuangzi, 99–100 datong, 206

­family members. See relations with ­family members feminism, 37–38, 215n23

East Asia, 135–42 Ebrey, Patricia Buckley, 121 economic equality, 50–51 emotional intelligence, 48–49 Emperor Han Wu Di, 70, 86 Emperor Qin Shi Huang, 69–70 Emperor Wanli, 84 equality, 15–16, 26, 35 Erdogan, Tayyip, 68 Etzioni, Amitai, 195–96, 244n37 Etzioni, Oren, 195–96 Eurocentrism, 144 Eu­ro­pean imperialism, 127 Eu­ro­pean Union, 136–37

266 i n de x Ferguson, Niall, 68 filial piety, 29–30, 45, 87–88, 191–92, 217n59, 217n61, 217n63, 217n65 formal meals. See formal seating arrangements formal seating arrangements, 1–7, 207n4 France, 173–74 Francione, Gary, 146 friends. See relations with friends Gaita, Raimond, 150 gender: driving and, 193; formal seating arrangements and, 2–3; the Kamasutra and, 40–45; patriarchy and, 37–38; po­liti­cal meritocracy and, 85 German Ideology, The (Marx), 181, 188 global financial crisis (2007–2008), 180 Google, 200–202, 206, 245n51 ­Great Leap Forward, 180, 182 ­Great Learning, The, 133–34 Han dynasty, 70–71, 86–88 Han Yu, 213n6 Harari, Yuval Noah, 11 harmony, 52 He Baogang, 95 hedonistic utilitarianism, 149 hegemonic state, 123 Her (film), 192 He Xinyin, 31–33, 213n12 hierarchical po­liti­cal rule: collective leadership and, 101; Confucianism and, 70–72; corruption and, 81–84; democracy and, 75, 95–96; efficiency and, 68–70; imperial examination system and, 90–93; justification of, 93–105; Legalism and, 69–72, 80, 83, 105; limiting po­liti­cal power and, 78–81; nine-­rank system and,

88–89; ossification of, 67, 85–86, 88, 91, 104; po­liti­cal meritocracy and, 73–76, 92, 104–5, 221n15; recommendation system and, 86–88; se­lection pro­cesses and, 84–86; transparency and, 76–78. See also politics hierarchy: age based, 45–55; biology and, 8–9; of compassion, 171–72; cultural differences and, 24–25; Cultural Revolution and, 22; in defense of, 13–21, 25–28; definition of, 8; efficiency and, 9–11, 68; equality and, 7, 11–12, 15–16; ­human relations and, 35; power and, 6, 14, 68–69; progressive conservative values and, 12–13; psychedelic drugs and, 209n14; rituals and, 118–26, 128; skepticism and, 21–22; social relations and, 16–17; between states, 108–10. See also just hierarchies; power; relations between states; social hierarchies; unjust hierarchies higher communism. See communism History of the Jin Dynasty, 89 History of the ­Later Han Dynasty, 87 History of the Liu Song Dynasty, 89 house­keepers. See relations with ­house­keepers Hu Jintao, 182 ­human beings, 148–49 humane authority, 124–26 ­human rights, 12, 78, 96, 130, 149–50, 226n61 Hume, David, 32, 129 India: animals and, 169; Ashoka and, 114–17, 132; cows and, 145; Kautilya and, 111–13, 131, 140; Marx and, 144; Tianxia hierarchy and, 140–42 injustice, 163–64

i n de x   Japan, 23, 37–38, 102, 110 Jiang Zemin, 182 Jin Dynasty, 89 Johnston, Alasdair Ian, 128 Joy, Melanie, 174 just hierarchies: defense of, 13–21; definition of, 3; shifting roles and, 30, 45–46. See also hierarchy; shifting roles; unjust hierarchies justice, 16, 161–64 Kalinga war, 114 Kamasutra (Vatsyayana), 40–45 Kang, David, 127 Kant, Immanuel, 159–60 Kautilya, 111–13, 131, 140 kings, 110–13 ­Korea, 138 Kymlicka, ­Will, 149–50, 154–56, 158–59, 168 Lake, David A., 108, 110 Laozi, 98–99. See also Daoism learning, 47–48 Legalism, 22, 69–72, 80, 83, 105, 117, 223n37 Levinas, Emmanuel, 34–35 Liang Shuming, 74, 141 liberal demo­crats, 120 Libertarianism, 23 Liu Song Dynasty, 89 Loar, Lynn, 160 love, 33, 35, 134. See also relations with lovers lovers. See relations with lovers machines: artificial intelligence and, 178–79, 183, 192–95, 199, 201–2; artificial superintelligence and, 198, 203; Asilomar AI Princi­ples and,

267 199–200; Capital (Marx), 183–84; driverless cars and, 193–97, 244n39; Google and, 200–202, 206, 245n51; higher communism and, 179–88; race for AI dominance and, 203–6; as slaves, 178, 240n1, 244n42. See also artificial intelligence Maduro, Nicolas, 165 mandala, theory of the, 112–13 Maoism, 97, 104 Mao Zedong, 112 marriage, 53–54 Marxism: creative work and, 188–91; driving and, 193; hierarchical rituals and, 120; higher communism and, 20, 56, 181, 186, 241n15; machines and, 177–78, 183–85, 240n1; village life in India and, 144–45 mass line, 97–98, 105 Maurya dynasty, 111 May, Simon, 103 McCartney, Paul, 173 Mencius, 45–46, 57, 60, 130, 132, 174–75, 189–90 Midgley, Mary, 152 Ming dynasty, 31, 127–28 modernization, 144–45 Montaigne, Michel de, 33–35 Montreal Protocol, 201, 246n59 moral growth, 49–50 morality, 114–17, 120, 122, 131, 189, 195–96 Musk, Elon, 203 NATO, 109 Nehamas, Alexander, 213n11, 214n16 New Culture Movement, 216n45 Nietz­sche, Friedrich, 163 North ­Korea, 92, 136, 138 nuclear weapons, 202

268 i n de x Olivelle, Patrick, 112, 115 oppression, 15–16 Orban, Victor, 68 ozone layer, 201 parents: equality and, 29; filial piety and, 30, 45–47 patriarchy: formal seating arrangements and, 3, 207n2; the Kamasutra and, 42; social relations and, 37–38; wealth and, 51 Peace of Westphalia treaty (1648), 108 Pearson, Yvette, 192 philosophy, 28 Plato, 85, 92, 129 po­liti­cal meritocracy: China and, 18–19, 71–78; corruption and, 81–84; democracy and, 75; justifying a, 93–105; large scale po­liti­cal communities and, 73; limiting po­liti­cal power and, 78–81; se­lection systems and, 84–93; support for, 74; transparency and, 76–78. See also hierarchical po­liti­cal rule; politics po­liti­cal theory, 15, 68, 129–30, 162 politics: collective leadership and, 101; Confucianism and, 70–71, 84, 93, 96, 98, 104, 226n67; corruption and, 74, 76, 81–84; Daoism and, 98–102, 104; democracy and, 95–96, 221n12; hierarchical po­liti­cal rule and, 68–76; hy­poc­risy and, 129–31; international, 111–13, 116–17; justification for po­liti­cal rule and, 93–105; Kautilya and, 111–13; large-­scale po­liti­cal communities and, 67; limiting po­liti­cal power and, 78–81; Maoism, 97, 104; mass line and, 97–98, 105; po­liti­cal meritocracy and, 71–76, 92, 104–5; po­liti­cal se­lection pro­cesses and,

84–93; power and, 13, 19, 67, 78–84, 92–93, 96, 98, 105; small-­scale po­liti­cal communities and, 67–68; tianxia and, 132–34; transparency and, 76–78. See also hierarchical po­liti­cal rule power: animals and, 145; Ashoka and, 114; China and, 133, 136–39; Chinese Communist Party and, 204, 206; culture of cuteness and, 103–4; filial piety and, 46, 51–53; friends and, 34–35; gender and, 37–38, 41–43; hierarchy and, 6, 14, 68–69; lovers and, 30, 35–37, 41–43; moral, 128; politics and, 13, 19, 67, 78–84, 92–93, 96, 98, 105; rituals and, 119–22; sibling relationships and, 55; slavery and, 56–57; states and, 108–10, 112, 116, 124; tributary system and, 128–29; wealth and, 13 progressive conservatism, 12–13, 21 progressive social values, 7–8 psychic numbing, 174 Puett, Michael, 39 Putin, Vladimir, 202 Qin dynasty, 86 Qin empire, 69–70 Qing dynasty, 53, 127 racism, 144 Rawls, John, 15 reciprocity, 47, 106–8, 110, 113, 116, 122, 124, 126–27, 132, 136–37, 233n73 Rec­ords of Rites, The, 39 relations between states: equal sovereignty and, 129–30, 132, 136; global responsibilities and, 130–31; hierarchical rituals and, 122–29, 137; hierarchy between, 108–10, 228n2;

i n de x   humane authority and, 124–26; morality and, 131; reciprocity and, 106–8, 110, 113, 116–17, 123–24, 126–29, 132, 136–38, 140–41; sovereignty and, 108; tianxia hierarchy and, 132–42 relations with ­family members: arguments for age based hierarchies and, 47–52; filial piety and, 45–47, 191–92, 217n59, 217n61, 217n63, 217n65; shifting roles and, 52–55 relations with friends, 29–36 relations with ­house­keepers, 30, 60–65, 218n84, 219n85 relations with lovers, 30, 36–45 Renshon, Jonathan, 139 Republic, The (Plato), 85, 129 Ricard, Matthieu, 151 rituals, 118–26, 128–30, 137 Rock Edicts, 114 role reversals. See shifting roles Rousseau, Jean-­Jacques, 14 Rus­sia, 140 Scarry, Elaine, 162 security hierarchies, 109–10 sex, 50. See also Kamasutra (Vatsyayana) Shandong Province: alcohol consumption and, 6–7. See also formal seating arrangements Shandong University, 2–3 shifting roles: corporations and, 40; death of a ruler and, 39; Eu­ro­pean Carnivals and, 39–40; formal seating arrangements and, 6; between intimates, 40–45; just hierarchies and, 30; the Kamasutra and, 42–45; relations with ­family members and, 52–55; relations with ­house­keepers and, 64–65; relations

269 with lovers and, 37. See also just hierarchies; social hierarchies Silicon Valley, 199–206 Singer, Peter, 147–49, 151 Sintomer, Yves, 68 slavery, 55–59, 63 Smith, Adam, 32 social egalitarians, 15 social hierarchies: China and, 22–24, 212n41–212n42; defense of, 14; normative dimensions and, 8; rituals and, 119–22, 130. See also hierarchy social relations, 16–17, 189–92 social status: bowing and, 23; formal seating arrangements and, 1–7 Song Dynasty, 90, 133 soul, 34, 57–58 South ­Korea, 23, 37–38 sovereignty, 108, 136 Soviet Union, 109, 112 Stanford University, 21 states. See relations between states strong reciprocity. See reciprocity subordination without cruelty, 20, 144 Su Dongpo, 93 Sui Dynasty, 90 Sweden, 23 symmetry, 162, 165 Symposium (Plato), 34 Tagore, 141 Taiwan, 139 Tang Dynasty, 90 Tannsjo, Torbjorn, 149 Theory of Justice, A (Rawls), 15 theory of the mandala, 112–13 thin ­human rights, 12. See also ­human rights tianxia, 132–42, 206

270 i n de x transparency, 76–78 tributary system, 126–31, 137 Trudeau, Pierre, 38 Trump, Donald, 13, 26, 68, 75, 107–8, 131 Turchin, Peter, 9 Tutu, Desmond, 161 Ubuntu, 160–61, 173 United Kingdom, 107 United Nations, 108–9, 166 United States, 13, 23–25, 107–9, 112, 131, 138–40, 166, 202–6, 223n32, 247n68 unjust hierarchies: Athens and, 67; caste system and, 30; formal seating arrangements and, 2–3; society and, 14–16; static roles and, 11–12, 210n21. See also hierarchy Vatsyayana. See Kamasutra (Vatsyayana) vegetarianism, 169 Venezuela, 165 vio­lence, 30 Voltaire, 146 Walzer, Michael, 12, 16 Wang Huning, 182 Wang Qishan, 81, 84 Wang Shaoguang, 68

Wang Yingming, 170–71 war, 56–57, 111–14 Warring States period, 84 Warsaw Pact, 109 weak reciprocity. See reciprocity wealth, 23–24 Western Zhou dynasty, 125–26 ­women: formal meals and, 3; po­liti­cal meritocracy and, 85 Wright, Robert, 10 Xi Jinping, 141, 180, 206 Xunzi, 45, 85, 117–24, 129–30, 137, 142, 210n23 Yang Xiaodu, 81 Yan Xuetong, 109, 126, 129, 137, 230n31 Zhang Feng, 128, 135 Zhang Taisu, 50, 52 Zhang Yongle, 96–98 Zhao Dingxin, 70 Zhao Tingyang, 132–33, 135 Zhejiang Province, 7 Zhou dynasty, 133 Zhuang Guotu, 128 Zhuangzi, 99–100. See also Daoism Zhu Xi, 90, 134 Zika virus, 152

a no t e on t h e t y pe

This book has been composed in Arno, an Old-­style serif typeface in the classic Venetian tradition, designed by Robert Slimbach at Adobe.