Whose Tradition? Which Dao?

SUNY series in Chinese Philosophy and Culture

Roger T. Ames, editor

Whose Tradition? Which Dao?

Confucius and Wittgenstein
on Moral Learning and Reflection

JAMES F. PETERMAN

Published by State University of New York Press, Albany

2015 State University of New York
All rights reserved
Printed in the United States of America
No part of this book may be used or reproduced in any manner whatsoever
without written permission. No part of this book may be stored in a retrieval system
or transmitted in any form or by any means including electronic, electrostatic,
magnetic tape, mechanical, photocopying, recording, or otherwise without the prior
permission in writing of the publisher.
For information, contact State University of New York Press, Albany, NY
www.sunypress.edu
Production, Eileen Nizer
Marketing, Kate R. Seburyamo
Library of Congress Cataloging-in-Publication Data
Peterman, James F.
Whose tradition? Which Dao? : Confucius and Wittgenstein on moral learning and
reflection / James F. Peterman.
pages cm. (SUNY series in Chinese philosophy and culture)
Includes bibliographical references and index.
ISBN 978-1-4384-5419-1 (hardcover : alk. paper)
ISBN 978-1-4384-5421-4 (ebook)
1. Ethics. 2. Confucius. 3. Confucius. Lun yu. 4. Confucian ethics.
5. Wittgenstein, Ludwig, 18891951. I. Title.
BJ1012.P438 2015
170.92'2dc23

2014002776
10987654321

To All of My Teachers
., . . .
The Master said, If there are several people walking on the road, surely
there will be my guiding exemplars among them. I would choose [from
among] them whoever is adept [at complying with the Way] and then
follow them. In addition, I would choose whoever is not adept [at complying with the Way] and use their [examples] to rectify my conduct.
Analects 7.22

Contents

Preface ix
Acknowledgments
1 Introduction: A Prologue to an Unlikely Project

xv
1

2 Confucius, Wittgenstein, and the Problem of Moral Disagreement 39

3 Confucius, History, and the Problem of Meaning

69

4 Wittgenstein and the Problem of Understanding at a Distance

95

5 How to Be a Confucian Pragmatist without Losing the Truth

121

6 Saving Confucius from the Confucians

167

7 The Dilemmas of Contemporary Confucianism

185

8 Fingarette on Handshaking

219

9 Acknowledging the Given: Our Complicated Form of

Ritual Life

251

Afterword: The Way Backward or Forward:

Wittgenstein or Confucius?

271

Notes 275
Bibliography

307

Index 315

Preface

This book offers the first full-length comparative study of the ethics of
ancient Chinese ethicist Confucius and the moral aspects of the later
therapeutic approach to philosophy of twentieth-century philosopher Ludwig Wittgenstein. The title, Whose Tradition? Which Dao?: Confucius and
Wittgenstein on Moral Learning and Reflection, which alludes to Alasdair
MacIntyres book, Whose Justice? Which Rationality? (1989), takes seriously
a key claim of MacIntyres: Any sustainable version of moral inquiry must
not be committed to basic claims and principles that make that inquiry
impossible. To offer an example from MacIntyres playbook: If liberalism
claims that all moral traditions make arbitrary assumptions about moral
truth and it turns out that liberalism is itself a moral tradition, then liberalism makes claims that undermine its very possibility. I will refer to
this requirement as the requirement that moral traditions and their related
versions of moral inquiry may not be self-undermining. This principle of
evaluation of traditions, or what MacIntyre calls versions of moral inquiry,
can be traced back to the Socratic requirement that ethical judgments be
accounted for in a way that is coherent with the rest of the persons considered judgments. This book seeks to defend an interpretation of Confuciuss
project, depicted in the centrally important early Confucian text, Analects,
as operating in what Wittgenstein scholar Cora Diamond, taking a phrase
from Wittgenstein, refers to as the realistic spirit. The realistic spirit, as
distinct from the philosophical realist, seeks, as she puts it, to clarify our
life with concepts, including ethical life, in all its complexity, suspicious
of the simplification and nonsense bound up with traditional metaphysics.
Although the Socratic requirement that versions of moral inquiry not
be self-undermining is a basic principle for evaluation of competing versions
of moral inquiry, MacIntyres use of it to challenge the Confucian moral
tradition is unsuccessful. Although I explicitly take up MacIntyres challenge
to Confucianism in Chapter 7, the whole project of the book can be seen

ix

Preface

as offering an account of three key aspects of the version of moral inquiry

found in the centrally important Confucian text, the Analects, which offers
a distinctive, credible version of moral inquiry. This approach to moral
inquiry, like Wittgensteins quite similar approach to philosophical inquiry in
the realistic spirit, gives central place to moral practices and to reflection on
the meaning and significance of those practices by practitioners. Central to
Confucian moral practices is the practice of ritual ( li). Confucian moral
inquiry requires training in ritual, as well as reflection on the practice of
ritual guided by a master of such ritual practice and reflection.
Confucius approaches moral inquiry in a way that avoids abstract,
theoretical reflection on questions of moral epistemology and ontology. As
a result, the presentation of the Analects approach to moral inquiry is not
as fully developed as is required for systematic assessment. To solve this
problem, I turn to the later writings of Ludwig Wittgenstein, who offers an
approach to the relation between practice and reflection that is remarkably
similar to Confuciuss. Drawing on Wittgenstein to develop an account of
the early Confucian version of moral inquiry in Chapter 1, Introduction:
A Prologue to an Unlikely Project, I use this version of moral inquiry
later in the book to address a range of potential problems facing Confucian
moral inquiry, which, if not adequately addressed, threaten to undermine it.
In Chapter 2, Confucius, Wittgenstein, and the Problem of Moral
Disagreement, I take up the question of whether this version of moral
inquiry has adequate resources to address the problem of moral disagreement. The problem arises inevitably from the way in which this version
of moral inquiry avoids appeals to foundational moral epistemology and
ontology. Any account of moral inquiry that offers no account of how to
address the problem of moral disagreement is possibly self-undermining. I
argue that Confuciuss appeal to inherited practices learned by novices under
the guidance of a master offers a possible solution to the problem of moral
disagreement, one based on its commitment to the authority of a master
who transmits traditional norms to novices.
Another serious problem for Confucian moral inquiry, with its appeal
to founding texts, like the Analects, is the problem of meaning of texts written more than two thousand years ago in a non-Western culture. Appealing
to accounts of the meanings of the sentences in the Analects, which I refer
to as semantic nihilism and skepticism, John Makeham and Daniel Gardner have argued that the substance of the Analects has no meaning or no
knowable meaning of its own. For a version of moral inquiry that makes
essential appeal to its founding texts as exhibiting norms of conduct, this
result would be undermining, and this version of moral inquiry would be

Preface

xi

self-undermining. If Gardner and Makeham are correct, the meaning of

the founding texts of the Confucian tradition would not be available to
provide guidance to the tradition. Instead, each interpreter of these texts
creates merely his or her own personal meaning.
I discuss both of these views of meaning in Chapter 3, Confucius,
History, and the Problem of Meaning. By appealing to Wittgensteins view
of meaning as use and the view of the principle of charity implicit in that
account, in Chapter 4, Wittgenstein and the Problem of Understanding
at a Distance, I argue against these views of semantic nihilism and skepticism. I argue that the meanings of the sentences of the Analects themselves
are internally related to practices of interpretation within a community of
trained readers of early Chinese texts who are committed to making maximum sense of these sentences in light of historical evidence and to their
having learned the form of life that gives these sentences their meanings.
Specifically, I argue that making sense of some unfamiliar texts embedded in
an unfamiliar form of life requires learning the basic practices and original
language of that culture. I refer to this version of interpretive charity as the
Principle of Insider Competency.
Like other recent interpreters of early Chinese philosophy, I offer an
account of Confucian moral inquiry that gives a central place to practice
and reflection on the meanings of learned practices. We can refer to this
strain of interpretations as pragmatic. But the best known versions of such
interpretations, Donald Munros, Chad Hansens, David Halls, and Roger
Amess, tend to offer a pragmatic version of early Chinese philosophy that
holds that it operates without a concept of or interest in truth. The versions
offered of this basic view are, indeed, subtle, and I cannot do them justice
in this Preface. But I can say that if these accounts of early Confucianism
were true, the early Confucian version of moral inquiry would be selfundermining. Any putatively true claims Confucius would be makingand
he makes and implies many such claimswould be self-undermining. In
Chapter 5, How to Be a Confucian Pragmatist without Losing the Truth,
using a suggestion of Hall and Ames, I argue that the focus on pragmatics
need not come at the price of truth. I appeal to the philosophy of Wittgenstein to support my argument. I develop the view that truth claims depend
on background norms as their basis and that Confuciuss appeal to dao is
just such a norm. I also develop an argument of philosopher Xiao Yang,
which offers a way to understand Confuciuss speech acts that invoke dao
and presupposes the truth of Confuciuss spoken utterances.
Even if my arguments are plausible up to this point, they should
make readers familiar with the Confucian tradition uncomfortable. For early

xii

Preface

on, the post-Analects Confucian tradition tended to develop metaphysical

views of human nature and human psychology to underwrite Confuciuss
moral claims and practices. Furthermore, the formation of the Confucian
tradition as handed down to us is informed even today by the great Song
Dynasty Confucian Zhu Xis decidedly metaphysical interpretation of the
Analects. As a result of its importance, a reason must be given for rejecting
such an influential account. I take up this problem in Chapter 6, Saving
Confucius from the Confucians. There, I defend early, commonsensical
commentators on the Analects, such as Zheng Xuan and He Yan, who tended
at times to avoid metaphysical interpretations of key passages, like Analects
12.1. My criticism of Zhu Xis metaphysical interpretation of Analects 12.1
proceeds through the development of a trilemma that Zhu Xis metaphysical interpretation of original mind ( benxin) faces. The trilemma arises
from the following three possibilities concerning the criteria for use of the
concepts, like original mind ( benxin), in his commentary: (1) his
criteria for applying the concept of self-control are not different from our
ordinary criteria, or (2) they are different, or (3) they are not specified. In
each of these three cases, his commentary suffers by unnecessarily attributing a questionable account to Confucius. Developing the notion of depth
found in Wittgensteins later therapeutic approach to philosophy, moreover,
I argue that the apparent metaphysical depth and significance that Zhu Xis
interpretation seems to offer can be captured better without invoking his
metaphysics.
Another serious contemporary challenge to Confucianism as a modern
version of moral inquiry, not just an historical relic, rests on twin problems
that arise from the history and context of Confucianism. By virtue of its history in East Asia, Confucianism can seem to be so embedded in that history
and culture of East Asia that it appears incapable of justifying any objective, ethical claims of its own. Moreover, whatever claims it makes seem to
conflict with central moral intuitions common to modern Western (liberal)
cultures. I take up two forms of this challenge in Chapter 7, The Dilemmas of Contemporary Confucianism. According to philosopher Jiwei Ci,
contemporary Confucians must either embrace the essential commitments
of historical Confucianism and admit that we Western moderns cannot be
Confucians today, or we must remove problematic features of historical
Confucianism, in which case the position becomes formalistic, not distinct
from other more updated versions of, for example, communitarianism and
its variants. I also discuss Alasdair MacIntyres dilemma that Confucianism
must embed itself in a defensible account of human nature and in a related
moral ontology, if it is to avoid being nothing more than an account of

Preface

xiii

a local set of practices. Both of these dilemmas suppose that in order to

offer objective, justified moral claims, Confucianism must be embedded
in a theory of human nature and a related moral ontology. I resolve these
dilemmas by arguing that Confucianism needs neither to commit itself to a
theory of human nature or a related moral ontology. It needs merely to be
able to make ordinary truth claims about living well and what individuals
ought to do in particular circumstances. But as I argue, early Confucianism can make sense of its own truth claims, just as we all do every day, by
appealing to norms embedded in its local practices, what Wittgenstein calls
language-games, without appealing to moral ontology.
The distinctively Wittgensteinian approach that I bring to early Confucianism begs the question of what this approach to early Confucian moral
inquiry can offer us today. What can such a deflationary view of philosophy
contribute to a defense of early Confucianism? In the last two chapters, I
take up the philosophical need for acknowledgment of given aspects of
our human forms of life. The need for acknowledgment, avoided in most
traditional philosophy, which prefers to think of philosophy as seeking to
supply us with knowledge of basic principles, is central to Wittgensteins
later philosophy. Philosophical confusions often arise from a failure to
acknowledge basic features of our human form of life. In this way, acknowledgment of those basic features has by itself philosophical importance. In
Chapter 8, Fingarette on Handshaking, I argue that Herbert Fingarettes
groundbreaking book, Confucius: Secular as Sacred, offers a flawed theory
of ritual and a related interpretation of the Analects. His example of ritual
handshaking, does succeed, however, in offering an arresting example of a
ritual, which makes it easier than it would otherwise be for Westerners not
schooled in Confucianism to acknowledge ritual as a part of their moral
form of life. In Chapter 9, Acknowledging the Given: Our Complicated
Form of Ritual Life, I turn to the work of sociologist Erving Goffman on
contemporary, Western interaction rituals as a way to sketch out the range
and complexity of our Western face-to-face rituals, which Goffmans work
can help us to acknowledge. This range, suitably clarified, can provide us
with what Wittgenstein calls a perspicuous representation of ritual, which
can help clarify the role of ritual in what Wittgenstein calls our complicated form of life.
In making the arguments in this book, which I have sketched out
briefly in this Preface, I make no claim that that early Analects-style Confucian version of moral inquiry is reducible to Wittgensteinian modes of
reflection. Throughout, I argue for their important similarities. Moreover, I
look for ways in which so-called Wittgensteinian accounts of moral judg-

xiv

Preface

ment and critique can benefit from Confucian inquiry, just as Confucian
practice of moral inquiry can benefit from Wittgensteinian philosophical
investigations. They are, in fact, members of a family. From Confucius, we
know that family relations can be mutually supportive, and from Confucius
and Wittgenstein both, we know that family membership does not require
reduction of all to a single form. The use of Wittgenstein to articulate and
defend early Confucian versions of moral inquiry leaves us with the question
of how we can advance an early Confucian mode of inquiry today without
losing its distinctive character. In the Afterword, I take up this question and
offer a view of moral inquiry that appeals both to the Confucian tradition
for insights and to the Wittgensteinian quest for perspicuous presentations
of otherwise philosophically confusing aspects of our human form of life.1

Acknowledgments

Even though the actual writing of this book started in 2008, the idea for
the project first emerged in 1998 during a National Endowment for the
Humanities Institute, Chinese Philosophical and Religious Texts in Context, directed under the able leadership of Professor Henry Rosemont, who
was assisted by Professor Roger Ames. Never having had the opportunity to
study Eastern philosophies in a formal academic context and believing that
it would be good if I could, on occasion, add some Chinese philosophical
texts to my lower-level courses, I set out for Hawaii to attend this institute.
Before doing so, I began for the first time to read Confuciuss Analects and
was horrified at its approach to ethics and politics. I doubted my ability
to teach a text that is so dramatically disorganized and so fundamentally
committed to magical powers of ritual. By that time, however, I had already
signed up, and five weeks in Hawaii seemed attractive.
What happened at the institute was surprising, to say the least.
Although I started out with my typical skeptical method toward all things
philosophical, I found myself drawn into what felt like a different mode of
academic interaction. Later I realized that the culture of this institute reflected the key commitments of the Confucian texts under discussion. Central
to the interpretations offered of the Analects, especially Roger Amess, were
a focus on the practice of ritual and its creative and constructive character.
The relation between learning rituals and the ideals they embodied struck
me as importantly similar to Wittgensteins later approach to the relation
between ideals and language-games. I recall asking Roger Ames if anyone
had written on this relationship, and he said, No. I thought to myself,
Here is a project. On further reflection, it struck me that the Confucian
tradition offers us a glimpse of what it would mean to live in a tradition
committed to clarifying, rather than calling into question from the ground
up, its practices, so as to help those in that tradition to live in agreement
with their fundamental modes of living, an ideal central to Wittgensteins

xv

xvi

Acknowledgments

later philosophy. This was, I thought, a reflective tradition that Wittgenstein could warm up to. I returned to Tennessee jokingly claiming to have
converted to Confucianism, claiming to be the last living Confucian in
Tennessee. For his insights, advice, and ongoing encouragement over the
years, I am indebted to Roger Ames.
The academic year following this institute, I taught my first course
on Chinese Philosophy. I felt myself a fraud. Without reading the texts
in Chinese, given the variety of translations, I wondered, how could I
represent to students their real meanings? The next year I took advantage
of a follow-up East-West Center academic tour of sacred sites in China. I
took lots of photos of different types of temples, but the best thing that
happened on that tour happened on an overnight train trip: tour leader
Ned Davis, University of Hawaii professor of history, persuaded me that
it should be possible for me to learn Chinese and use it in my academic
work. I took the bait and the next summer traveled to Beijing Language
University to take part in its immersion course in Mandarin. Later that
fall, I sat in on University of CaliforniaBerkeley Professor Phil Riegels
Introduction to Classical Chinese. I spent the next semester poring over dictionaries, grammar books, and translations of the Analects, trying to develop
my own translation. Every summer since then, except 2008, I have spent
the summer in China studying Mandarin or classical Chinese and trying
to get more familiar with all things Chinese. In one form or another, this
travel was supported by Dean of the College John Gatta and the Research
Grants Committee at my home institution, Sewanee: The University of the
South. For the many opportunities given to me by Sewanee, including the
opportunity to teach at the most culturally Chinese academic institution in
the United States, I will be forever grateful.
In 2002, under the auspices of ASIANetwork, four of my students and
I traveled to Zhongshan University in Guangzhou, China, and interviewed
some forty students on their attitudes to parental authority. The results of this
study were published as The Fate of Confucianism in Contemporary China
in Asian Studies in America: Newsletter of the Asian Studies Development Program, Fall 2003. I wish to thank my Zhongshan University friends Wang
Kun, Ai Xiaoming, and Ke Qianting for their assistance and conviviality. I
am especially indebted to Ke Qianting, who spent hours with me in conversation later, when she was on a teaching-research fellowship at Sewanee.
Much of my early thinking about how to understand Confucianism under
a Wittgensteinian lens came out of this research and these conversations.
I am especially indebted to my mainland Chinese Mandarin teachers and friends: in Beijing, Xue Er; in Kunming, Xu Peng, Xu Feng, and

Acknowledgments

xvii

Zhu Lan; and to Keats School teacher and comrade in the dao, Pan Siyi,
without whom my grasp of classical Chinese language and culture would
be all the poorer.
In 2008, under the auspices of the Fulbright Foundation, I spent ten
months in Taipei, Taiwan, as a senior fellow, living and working at Academia
Sinica and traveling from one coffee spot to the next, writing the first draft
of this book. My application for this fellowship was improved by expert
advice from my friends and colleagues, Professor Charles Brockett, Professor
Richard OConnor, and Dean Rita Kipp, and from Fulbright Foundation
Senior Officer David Adams.
Although my work in Taipei was supported and sustained by the
professional services offered by Fulbright Taiwan, some of the most important conversations and opportunities were made possible by good friend
Professor Kirill Thompson of Taiwan National University, who gave of his
precious time to give me feedback on my work and to make sure I was in
contact with other Taiwan National University philosophers. To him, I am
profoundly indebted. In addition to introducing me to colleagues, Kirill
also arranged for me to spend four hours a week with my teacher, Yang
Youwei Laoshi, in what has turned out to be his last chance to teach a long
list of Western scholars: Roger Ames, Carine DeFoort, Christian Joachim,
John Makeham, Randy Peerenbohm, Lisa Raphaels, and Kirill Thompson,
to name a few. I feel humbled to have been the last student among such a
distinguished group. Having published nothing of his own, Yang Laoshis
lifetime study of the Chinese classics lives on in the work of these scholars.
Yang Laoshi grew up in Beijing in a Confucian family, studying the
classics with the help of his uncle and mother, herself a committed Confucian from a family of scholar officials. Yang Laoshis linguistic insights
were, to say the least, profound. He was nothing if not opinionated, fully
confident of the correctness of every one of his judgments, even when they
changed from day to day, able to riff from the meaning of a character, to
the bankruptcy of contemporary culture, to the problem with American
foreign policy, as if he were discussing one topic. Despite his stubbornness
and the ferocity of his judgments and arguments, he was willing to listen
when I was prepared, in my poor Chinese, to press my own points. From
time to time, I had the feeling that I was in the presence of a Chinese
Wittgenstein, someone who represented an earlier culture, out of tune with
his own times, uncompromising in his intellectual and moral commitments,
but also in his commitment to me as his student.
I traveled two times a week to Yang Laoshis apartment, where we
crawled line by line through the Analects. On occasion, I was able to share

xviii

Acknowledgments

with him some of my work on this text. Although his academic work in the
United States in the 1960s had been in psychology, he was deeply interested
in philosophy, but he was not familiar with the details of Wittgensteins
writings. He was especially attracted to my characterization and use of
Wittgensteins view of nonsense. Later in our times together, he seemed to
relish characterizing certain problematic translations of Analects passages as
pihua, nonsense, in this Wittgensteinian sense. I would like to think that
on a few occasions such as these, I managed to become a partner with
him in inquiry.
In the summer of 2010, I spent a week with Yang Laoshi when we
looked over the translations of passages I use in this book. I am grateful
to him for his help and for the time we spent together. Two days before I
wrote the initial draft of the Acknowledgments for this book, Yang Laoshi
passed away. The books dedication, To all my teachers, although not
solely for him, finds in my own (xin, heart and mind) a large place for
him and for the kinds of thinking and writing my time with him in Taipei
opened up for me.
In addition to my weekly visits to Yang Laoshi while in Taipei, I was
invited by my friend Professor Sato Masayuki of Taiwan National Universitys Philosophy Department to participate in their departments lively
conference series, to give a lecture to the Philosophy Department during
that fellowship year, to give another lecture the next fall, and to deliver
three lectures on parts of this book in the spring of 2009. On these occasions, I benefited from conversations with Professor Masayuki and his colleague, Professor Christian Wenzel. This book would be poorer without
those opportunities and without their efforts to welcome me and to discuss
issues that arose in the early version of my manuscript.
I am indebted to my friend Professor P. J. Ivanhoe of City University
of Hong Kong, who read the penultimate version of this manuscript and
provided a range of helpful suggestions about how to connect my arguments with the aspects of the Confucian tradition that I have not had an
opportunity to investigate and with recent discussions of which I was not
fully aware. My thanks also go to my friend and colleague at Sewanee,
Professor Andrew Moser, who listened and gave feedback during hours of
my thinking out loud about the issues I take up here, and to my copy editor, Kathy Hamman, whose perceptive linguistic sense made the language
of the text clearer than I was able to make it by myself.
I have to express my deepest feelings of gratitude to my wife, Merissa
Tobler, who resolutely, without a hint of complaint, let me reserve ten years
of much of my spare time and attention to this project.

Acknowledgments

xix

Earlier versions of portions of this book this book were published

in the following journals: Chapter 6: Just the Details: A Wittgensteinian
Defense of Lunyu Early Commentarial Tradition, 2008
Conference Proceedings. Chapter 3:
? (Do the Sentences of the Analects Have Any Meaning?). The Frontiers
of Social Sciences (), 160 (10), 2008, 5458.
Earlier versions of chapters were presented in the following forums:
Chapter 1: Master and Novice in Confucius, The International Society
for Comparative Studies of Chinese and Western Philosophy. Eastern
American Philosophical Association Meeting, New York City, December 28, 2009.
Chapter 2: Confucius and Wittgenstein on Moral Disagreement, Wittgenstein Workshop. Virginia Military Academy, Lexington, Virginia,
September 2008.
Chapter 2: Invited Conference Presentation, Ludwig Wittgenstein Meets
Confucius. Contemporary Research on Chinese Ethical Thought.
Philosophy Department, Taiwan National University, Taipei, Taiwan,
November 1315, 2008.
Chapter 3: Are the Sentences of Lunyu Meaningful?: Kongzi, Wittgenstein,
and the Problem of Meaning, Taiwan National University Philosophy
Department, Taipei, Taiwan, December 10, 2007.
Chapter 3: ? (Are the Sentences of Lunyu Meaningful?). Presented in Chinese at Academia Sinica, Taipei, Taiwan
(ROC), June 26, 2008.
Chapter 3: Are the Sentences of Lunyu Meaningful?: Kongzi, Wittgenstein,
and the Problem of Meaning, International Society for Comparative
Studies of Chinese and Western Philosophy, The 3rd ISCWP Constructive-Engagement International Conference on The Methodology
of Comparative Philosophy, Department of Philosophy & Institute
of Foreign Philosophy, co-sponsored by the Peking University Center
for Comparative Philosophy, Beijing, China, and San Jose State University, San Jose, California, USA, June 2008.
Chapter 6: Just the Details: A Wittgensteinian Defense of Lunyu Early
Commentarial Practice, Program for East Asian Classics, 2008 EastAsian Lunyu Studies Conference. Taiwan National University, March
89, 2008.

xx

Acknowledgments

Chapters 1, 2, and 6: Invited Lecturer. Three Lectures on Kongzi and

Wittgenstein: The Way of Ethics without Philosophy, Department of
Philosophy, Taiwan National University, May 2527, 2009.
Chapter 8: Fingarette on Handshaking, Southeastern Early Chinese
Roundtable. University of Georgia, Athens, Georgia, October 2325,
2009, and ASDP National Conference, Seattle, Washington, March
2931, 2012.
Sewanee, Tennessee
April 1, 2013

Introduction
A Prologue to an Unlikely Project

In the beginning was the deed. Goethe, Faust I.

Quoted by Wittgenstein, On Certainty, section 396
Among the disciples, there was one who, on his own, had written down
the Masters teachings. Hearing of this, the Master said, Sages and
worthies teach in the same way that physicians prescribe medicine. They
always match the treatment to the ailment, taking into consideration
the various symptoms and, whenever appropriate, adjusting the dosage.
Their sole aim is to eliminate the ailment. They have no predetermined
course of action. Were they indiscriminately to stick to a predetermined
course [of treatment], rarely would they avoid killing their patients.
Now with you gentlemen, I do nothing more than diagnose and polish
away each of your particular prejudices or obsessions. As soon as you
manage to make these changes, my words become nothing but useless
tumors. If, subsequently, you preserve my words and regard them as
dogma, you will one day mislead yourselves and others. Could I ever
atone for such an offense?1
Xilu, quoting Wang Yangming

Introduction
In the course of talking to philosophers about this book, though many
expressed some interest in the project, just as many expressed reservations.
What possibly could be the connection between Confucius and Wittgenstein? Others seemed to think that even if there were a substantial, important
connection, it would be very difficult to establish.
1

Whose Tradition? Which Dao?

While working on this book, I have become increasingly convinced

that the connections I somewhat dimly sensed at the outset are real but
that clarifying these connections is considerably more difficult. Some of the
difficulties stem from the inherent problems in comparative philosophy, and
some have to do with the targets of my comparison: Confucius and Wittgenstein. The difficulties of comparative philosophy concern the nature and
possibility and value of comparison across cultural, historical, and linguistic
divides; these divides or differences are so extensive that finding sufficient
similarities to compare can be nearly impossible. The difficulty of using
Confucius and Wittgenstein as objects of comparison has to do with the
complex question of which versions of each thinkers positions to select.
Even if we could specify only three different versions of each position, we
would have nine different possible targets of comparison.
This chapter is a prolegomena to the study I have undertaken. The
procedure will be to specify which Confucius and whose Wittgenstein will
be compared and then, given these choices, the points of the comparison.
I take the goal of comparative philosophy not only to make comparisons
between philosophical views from different parts of the world, but also to
articulate for each side of the comparison possible problems and resources
more readily grasped in the other side. The justification of the selection of
versions will come from the textual support for those selections and from
the reflective, critical fruit of those versions.
However, if the argument I develop in Chapter 1 is reasonable, interpretive justification is not separable from evaluative argument. For, as I
argue, the principle of charity rests on the fact of under-determination of
interpretation by textual and historical evidence. This means we are required
to adopt those interpretations of texts and authors that maximize the reasonableness of those positions.2 If that is true, it is inevitable that we acknowledge different versions of Wittgenstein and of Confucius. This is not to say
that all versions are equally well defended, but it does mean that in many
cases the sort of gotcha appeals to textual evidence, designed to refute an
alternative interpretation, are not likely to do that by themselves. For even
gotcha passages will be subject to competing interpretations and differing
interpretive weights. What tips the balance in these interpretive arguments
is often an unstated appeal to the claim that one version of the position is
stronger than the other, that it is thought to be philosophically superior.
But one philosophers thinking that an interpretive argument is superior
does not make it so. Simple appeal to textual evidence, with one possible
interpretation, will not necessarily refute alternatives. The superiority of an
interpretation will be borne out in some way by appeal to reasoning and

Introduction

explication, making it clear that one version of the position is superior to

the other.
Although it is impossible to state the whole range of reasons I have
for adopting a particular version of the Wittgenstein position, it is possible
to specify that version and to explain why it sheds light on and shows the
strength of key aspects of Confuciuss project. Nonetheless, my approach
contains a few possible problems that I will spell out in this introduction
and address in the comparative exposition.
If we draw a standard distinction between normative ethics and metaethics, we might wonder how it can be possible to compare Confuciuss
teachings, which appear normative, with Wittgensteins later philosophy,
which says next to nothing about ethics and would appear, if it were used
to think about ethics, to be meta-ethical.3 After all, Wittgensteins discussion of language-games is designed more to clarify how to resolve conceptual confusions about language and concepts rather than to show which,
of the range of possible language-games, including ethical language-games,
is the correct one (whatever that would mean). In contrast, Confucius is
deeply embedded in a moral tradition, which he takes a stand on, studies,
and passes on to others. Thus, an attempt to compare Wittgensteins later
philosophy with Confuciuss teachings would appear to be a meta-ethical
normative ethical divide if ever there were one.
A second issue I need to address is that even if I can show that
Wittgensteins later philosophical project and Confuciuss teachings share
sufficient similarities to overcome this initial objection, I still may not have
shown that the comparisons I am making bear any fruit for either position.
After all, with sufficient cleverness, any position can be said to be similar to
another in some respects. The similarities need, however, to be illuminating and bear some fruit. For, at least in my conception of the comparative
philosophical enterprise, the point of comparison should be an examination
of the ways in which the comparisons and contrasts bear dialectical fruit,
both in terms of the issues they raise and the resources they make available
to the positions under discussion.
My task in this chapter will be to provide a general sketch of the ways
in which these two positions can be benefited by joining forces. In Chapter
2, I raise fundamental dilemmas facing Confuciuss teaching and Wittgensteins later philosophy. Both projects forego foundational theories, and in
doing so, arguably face problemseven if differentof not being able to
justify key claims they must make in the course of carrying themselves
forward. For, as I will argue, both seek to embody a form of spirit that
they leave unjustified by theory. By embodying a theoretically ungrounded

Whose Tradition? Which Dao?

spirit, it can be argued, they both are subject to the charge of arbitrariness, and both require some theoretical grounding if they are to be carried
forward. I will argue that by combining forces, the two forms of spirit can
develop even better strategies for addressing these criticisms than they are
able to muster alone. In the course of this comparative study, I discovered
that the unusual juxtaposition of Wittgensteins and Confuciuss philosophies, including their ungrounded spirits, produces not only a different,
stronger spirit than either one embodies alone, but in so doing, addresses
outstanding issues, such as the meaning and truth of the sentences of the
Analects, as well as its contemporary relevance.
The way I develop these arguments depends on which versions of
Wittgenstein and Confucius I choose to discuss. My version of Wittgenstein shares some similarities with the so-called New Wittgenstein,4 who
is suspicious of metaphysical/epistemological theories designed to provide
an explanation of justification for ordinary linguistic practices. Wittgenstein
holds that once our ordinary criteria for something being true or real are
applied to a situation to justify the correctness of a sentence, there are no
further epistemological, metaphysical, foundational questions to raise about
the correctness of the language-game in which the utterance takes place. As
Wittgenstein says of any particular language, This is the language-game that
is being played.5 The fate of Confuciuss self-cultivationist approach to ethics and related eschewal of metaphysics, I argue, is wedded to the success of
Wittgensteins very similar project. However, Wittgensteins later avoidance
of sustained discussion of ethics benefits from the sort of supplementation
offered by Confuciuss Analects.
Despite these affinities between my Wittgenstein and the New Wittgenstein, following recent work of Meredith Williams and Nigel Pleasants,
I argue that Wittgenstein would not agree with the New Wittgensteinian
tendency to be suspicious of traditional moral authority. I take these issues
up in some detail in Chapter 2.
In this chapter, I will provide an overview of Confuciuss and Wittgensteins teachings. The overviews goal is to persuade readers that the two
philosophers share a roughly similar account of the norms embedded in
human life and language, despite the large differences between their projects.
I will call these accounts or commitments their shared basic insight that
our primary relationship to norms, one necessary for understanding them, is
through learning. This approach constitutes a key stumbling block for interpretation and evaluation of both thinkers projects. As a result of this insight,
both projects address the problems they facefor Confucius, how to restore
a life lived in conformity with dao, meaning the set of norms governing

Introduction

human life and, for Wittgenstein, how to resolve conceptual confusion in

philosophy and protect shared moral insights of humanitythrough appeal
to contingent practices and the sorts of self-cultivation designed to bring
us in agreement with them, not through the construction of foundational,
justificatory theories.

Bedrock Practices
An important caveat to any comparative presentation of Wittgensteins later
philosophy alongside Confucian teaching in the Analects is that whereas
Wittgenstein offers a view of language that would play a role in investigating
and resolving the problems of philosophy, Confuciuss teaching is designed
to clarify dao with the goal of fostering self-cultivation and culture-wide
recovery of dao, which had been left in disrepair with the gradual dissolution of the political and cultural power of the Zhou Dynasty. Despite these
differences, Wittgensteins project centers on the key issue of understanding the role of norms within human life with an emphasis on language.
Confuciuss project is formed out of a sense of the basic character of dao,
the norms governing human life and, primarily, the care of human relationships. It is true that some later passages in the Analects take up the political
project of rectifying names, but Confucius does not develop an interest in
language and rectification of names in general, beyond his limited interest
in functional terms, like father and ruler. These terms imply norms,
which, when strictly applied to people, require applying them to people
who must live up to those norms.6
One additional noteworthy similarity between the texts is their preoccupation with teaching and learning. Confucius describes the need for
a devotion to learning (haoxue), especially of ritual (li), as the
principal first step in self-cultivation and also offers a phenomenology of
what it is like to move along the path from novice to master of dao.
Moreover, one of Confuciuss disciples, Master You (Youzi), supplements
Confuciuss focus on learning ritual with an account of how performing
relatively concrete practices, like keeping ones word (xin), provides a
basic practice that prepares one for the more complex range of practices
that make up righteousness (yi). Youzis focus includes an emphasis on
being filial or obedience to parents, which he generalizes to obedience to
those in authority. Both Confucius and Youzi emphasize that basic, bedrock
practices are the first steps and constitutive features of understanding and
practicing norms.7

Whose Tradition? Which Dao?

Although there is reason to think that as an adult, learning dao from

a teacher is different from learning language and related basic concepts for
the first time from ones parents; if we analyze this apparent difference using
Wittgensteins approach to clarifying the content of concepts and norms
by appeal to learning contexts, this difference does not matter for purposes
of conceptual clarification. Wittgensteins learning contexts in the opening
passages of Philosophical Investigations, for example, focus on the learning
situation of those who lacked the key concepts being taught. Learning contexts in the Analects, however, focus on the learning of young adults and
adults instructed by Confucius. But Wittgenstein in other writings, such
as Remarks on the Foundation of Mathematics, focuses on the learning situations of a range of learners. What a concept or norm consists of is shown
in the way it is learned and taught. What is bedrock will be displayed in
the learning context. In discussing paradoxes in mathematics, Wittgenstein
claimed, All the puzzles I will discuss...can be exemplified by the most
elementary mathematicsin calculations which we learn from ages six to
fifteen.8 But his discussion of learning in Philosophical Investigations also
takes up teaching how to help adults (presumably) to weigh and balance
imponderable evidence of whether another persons feelings are genuine.9
Although Confucius does not discuss the ritual of teaching children
and their learning, Youzis account could apply to children even if Youzi, like
Confucius, is concerned only with moral cultivation in adults. In contrast,
Wittgenstein focuses on language-games, which he describes as the sorts of
simple language usages learned by children that adults use to clarify and
critique philosophical views. Those same usages of language, he says, also
apply to the process of learning concepts and related words as constitutive
of their meanings. Although Wittgenstein does not say much about moral
language and concepts, he does say something; thus, it is possible to imagine fruitful use of his mode of clarification to reflect on the meanings of
moral concepts. Wittgensteins views tend to blur this distinction between
children and adult learning while offering enough resources for the reader
to discover both similarities and differences.10
In the rest of this section of the Introduction, I will explore some of
the connections between Wittgensteins and Confuciuss views of learning
and teaching as they apply to what Wittgenstein takes as bedrock practices, those practices, which, as the inherited background of our actions and
beliefs, we convey in teaching-learning contexts, and which, as constitutive
of those beliefs and actions, are not justifiable by us. He says of them:
Once I have exhausted the justifications, I have reached bedrock, and my
spade is turned. Then I am inclined to say: This is simply what I do.11

Introduction

The Confucian project of self-cultivation and understanding dao as

grounded in the fundamental master-novice relationship is the means by
which norms, which are embodied, are transmitted and understood. In
certain contexts, emphasis is given to the parent-child (father-son) relationship and in other contexts, to the teacher-disciple relationship. Nevertheless,
this project provides room for reflection on these norms but only as an
outgrowth of learning practices.
Confucius offers, however, nothing in the way of a metaphysics of
norms and little or nothing of what we would call normative theory. Some
commentators have found this feature of Confuciuss project problematic.
We can distinguish three basic interpretive approaches to Confuciuss project:

A. Embodying a version of some Western philosophical theory

B. Being insufficiently philosophical

C. Presenting an alternative to Western philosophy.

MacIntyre12 and Slingerland,13 who read the Analects as presenting a view

similar to Aristotles virtue ethics, as well as Hall and Ames,14 who understand
the text in terms of Whiteheads process philosophy or Deweys pragmatism,
exemplify type A. Fingarette,15 Graham,16 Hansen,17 and Schwartz18 exemplify
type B by treating the Analects as philosophically unsophisticated and in need
of some theoretical foundation. Both of these types arise from a fundamental principle held by many philosophers, that a non-Western philosophical
text is philosophically significant only if it contains a theory that provides a
principled justification for actions or beliefs. In contrast, I join with those
of type C, the pluralists (Eno19 and Nivison20), who treat the Analects as an
alternative to Western philosophy but do so by arguing that the Analects offers
not theory, but self-cultivation practices and reflections (si).
Nonetheless, despite his departure from standard forms of philosophical theory, Confucius, I argue, is still some sort of realist about norms.
Although he offers no metaphysics of norms and is more concerned to guide
others self-cultivation than to offer abstract clarifications of dao, Confucius
says that the dao, which set of norms governing human life, is something
a society either has or fails to have.21 He also says that dao is something
a culture can get closer to and possibly even embody through a life selfcultivation.22 These are formulations nearly all realists would be happy with
even if they were to think more needed to be said. Yet, we might wonder,
how can Confucius proceed with any confidence about the reality of
norms if he lacks any philosophical account to give of its reality?

Whose Tradition? Which Dao?

This problem seems all the more pressing given that when Confucius
discusses daos constitutive ideals, such as goodness or humanity ( ren),
filiality ( xiao), and trust ( xin), he often characterizes them in various
ways.23 And in some cases, his characterizations seem designed specifically
for the person with whom he is talking. Lacking any comprehensive definition and any account of the reality of dao, one might argue that it is hard
to see what basis Confucius might have for treating dao as real. Without
such an account, we might think we need to accuse him of ethical provincialism if not wholesale bias.24
There are several aspects to an adequate account of Confuciuss socalled realism. If he is a realist, then he is a realist without a theoretical
elaboration of his realism. Confucian realism would have to be understood
as one would understand it from the vantage point of his fundamental
project of self-cultivation in those practices, attitudes, and reflective understandings and forms of sensitivity and responsiveness that constitute dao
(the norms of living well).
The burden of my argument will be to show what Confuciuss form
of realism involves and why philosophers might want to take it seriously as
an alternative to theoretical, metaphysical accounts of the reality of norms.
To that end, I examine Wittgensteins approach to realism in Chapter 2.
Unlike Confuciuss realism, Wittgenstein intentionally seeks to place limits
on the meaningfulness of metaphysical language as a way of protecting our
complicated forms of life and those ways of thinking and speaking central to
it. Before that argument is made, it is important to understand Confuciuss
and Wittgensteins shared basic insight, which is central to the form of realism that Confucius adopts. I borrow a term from Wittgenstein scholar and
philosopher Cora Diamond; she uses realistic spirit to capture Wittgensteins realism. Confucius also embraces the realism of the realistic spirit.
At this point, I wish to capture one aspect of that spirit, specifically
the way in which Confucius appears to have no interest in questions of
metaphysics and definitions of key normative concepts; nevertheless, he
thinks of dao and its constitutive norms as real. In the account I offer, the
reality of dao together with its constitutive ideals are embodied in Confucian
practices of learning; these practices constitute and illustrate real instances of
dao and its ideals. The primary way to understand those ideals is to learn the
ritual practices that embody them. If this is true, then all we need to know
about Confucian ideals comes from learning and acquiring them with the
assistance of someone who has mastered them. Confuciuss down-to-earth
instructions plus any extensions of the ideals taught by master practitioners,

Introduction

such as Youzis modifications of Confuciuss teachings, will provide us with

all there is to know about these concepts and ideals.
The best way to make sense of this approach to dao and its constitutive ideals, in terms of Wittgensteins and Confuciuss shared notion, is that
as novices we learn the basics of how to embody norms. Later, after years
of study and practice, we may become masters of how to embody those
norms in their full complexity. It is possible to understand Confuciuss
project of inquiry into morality along the lines of inquiry into the nature
and meaning of the game of chess. Someone who wants to understand chess
first needs to learn the basics of chess. After extended practice mastering
the basics, the novice will, perhaps by being offered examples of exemplary
chess moves from the history of chess, be able to operate at higher levels
of mastery, and eventually, if lucky, be able to create his own exemplary
chess moves. We can imagine such a level of mastery as being accompanied
by reflection on the meaning of chess as playing a role in the human form
of life. This reflection depends on the persons earlier levels of mastery and
would not be possible without them. But this project would be a form of
realism about chess, its basic rules, its history, its exemplary moments,
and so forth. Confuciuss similar realism lies in his acknowledgment that
central normative practices can be taught by those who have mastered them
and that learning from this teaching establishes students basic competency.
His acknowledgment that the basic norms can be mastered and taught to
others limits what those norms are.
Given this basic approach, like Wittgenstein, we can view Confucius
as proceeding in a realistic spirit in his self-cultivation teaching project. He
can insist that students be trained in basic practices of propriety. He can
also embrace a certain depth of dao that escapes simple clarification in terms
of definitions yet allows for clarification as learners move to higher levels
of mastery of basic practices or mastery of even higher-level practices.25
Moreover, he can allow that some folks exhibit a higher degree of mastery
of how to conform to dao than others. In short, dao can be understood
beyond clear formulas, and it can be considered real precisely because its
norms can be taught and students can choose to act on this learning either
correctly or incorrectly. From the vantage point of someone engaged in a
self-cultivationist project, this is realistic enough.26
My basic approach to this realistic spirit of Confucianism derives
from the way that Wittgenstein investigates how norms are learned and how
they work in the context of various practices. In his later investigations,
Wittgenstein presents a sustained examination of rule-following and how

10

Whose Tradition? Which Dao?

understanding concepts is constituted by learning bedrock, norm-governed

practices. He offers a way to understand these phenomena without offering abstract, overly simple views of them, which gives rise to irresolvable
paradoxes. For example, an intellectualistic view of rule-following might
allow that to understand a formula involves grasping the rule that allows
an individual to carry it out, but because rules can be interpreted in various
ways, rule-following rests on grasping an acceptable interpretation of the
rule. But even deciding whether the interpretation of the rule is acceptable
or correct can be interpreted variously.27 Wittgenstein concludes that this
account makes it unclear what rule-following consists of and, indeed, makes
rule-following impossible due to its required distinction between getting the
rule right or wrong. In place of a view that gives central place to interpretation, Wittgenstein argues that there has to be a way of understanding a rule
that is not an interpretation. Instead, Wittgenstein claims, rule-following is
a practice.28 In addition, in On Certainty, he maintains that some empirical
propositions function as rules that are part of the bedrock practices that lie
at the bottom of our ways of thinking and talking:
94. But I did not get my picture of the world by satisfying myself
of its correctness; nor do I have it because I am satisfied of its
correctness. No: it is the inherited background against which I
distinguish between true and false.
95. The propositions describing this world-picture might be part
of a kind of mythology. And their role is like that of rules of a
game; and the game can be learned purely practically, without
learning any explicit rules.29
In these passages, Wittgenstein makes the point that to learn the rules
and the ways of thinking and talking connected to them is to learn those
practices. And those practices are learned from master teachers who, when
they can, instill in novices the basic competency required to operate with
basic beliefs and concepts within the target area of belief and language.
But learning bedrock practices does not exhaust how we learn concepts. In
Philosophical Investigations, Wittgenstein also discusses whether a person can
learn to become an expert in judging the genuineness of others feelings.
Such judgment requires being influenced by imponderable evidence. A
master can teach a novice to use such evidence not by teaching a system,
but through teaching correct judgments and giving the right hints.30 Witt-

Introduction

11

gensteins account of how concepts are learned finds important parallels in

Confuciuss account of moral learning. Next, I will investigate these parallels.

Bedrock Learning
A key feature of bedrock learning is that a novice is required to follow
blindly the teaching of the master. What distinguishes bedrock learning
from other forms of learning is the fact that we do it without justification.
It is just what we do. As Wittgenstein says:
How am I able to obey a rule?If this is not a question about
causes, then it is about the justification for my acting this way
in complying with the rule.
Once I have exhausted the justifications I have reached
bedrock, and my spade is turned. Then I am inclined to say,
This is simply what I do.31
My act of obeying a rule rests on bedrock practices, which, while
they supply me with a way to justify proceeding as I do, are not themselves
justified. They are what I do. But Wittgenstein is at pains to point out that
what I do is the by-product of learning from someone who is already a
master of the practice I am engaged in. So at this bedrock level, a person
could just as well say, This is what I learned. Or, This is what we accomplished practitioners do. Or, This is what a master of this practice does.32
The novice lacks a grasp of the conceptual terrain he is being initiated
into, and only later, after mastering a bedrock practice, can he come to
have the basic concepts of that terrain. Therefore, Wittgenstein distinguishes
between ostensive teaching and definition: the latter requires concepts
in the terrain of the definition, but the former does not. So, if I am to ask
what the meaning of the word tree is and understand possible answers, I
already need to have mastered the concept of plants, height, longevity, and
so on. In contrast, ostensive teaching simply involves pointing and naming,
as a basic form of learning required before a person has understood the
concepts of tree, plants, and so forth.
Meredith Williams discusses this feature of Wittgensteins view of
concepts in a way that is instructive. She argues that for Wittgenstein,
how a person learns a concept is constitutive of that concept. She quotes
Wittgenstein in support of her claim:

12

Whose Tradition? Which Dao?

It may now be said: The way the formula is meant determines

which steps are to be taken. What is the criterion for the way
the formula is meant? It is, for example, the kind of way we
always use it, the way we are taught to use it.33
For this reason, Wittgenstein often turns to the question of how a concept
gets learned in the course of clarifying the concept. But what is distinctive
and important about Williamss account is her clarification of the masternovice relationship. This relationship is crucial, for unless we understand this
relationship, we will mischaracterize the character and context of bedrock
learning.
A master cannot function to teach a novice about a concept, for
example, the concept of expressing appreciation, without a good deal of
stage setting. The context of teaching presupposes the practice of saying
thank you in various contexts, done with the right expression of gratitude.
The master herself must have mastered those practices. Based on her prior
learning, she intends to teach the child and through a pattern of imitative
behavior expects the child eventually to learn to engage in this behavior
without prompting. The novice begins learning this language without the
concept of gratitude, which she will only later come to understand. But until
the novices thank you utterances come as matter of course, are suitably
expressive, and she feels that she must utter them in the right contexts, she
will not yet have adopted the concept of gratitude. Her developed sense of
what she must do in situations where we typically feel a need to express
gratitude constitutes her grasp of a new concept. As Wittgenstein indicates
in his discussions of learning mathematical concepts: This must shews that
he has adopted a concept.34 Prior to this, the child will have a feeling of
pleasure at those things that please her teacher. Her first efforts to imitate
her teacher are only courtesy attributions of understanding of the concepts
to be learned.35 Until she has a mastery of the basics of the concepts and
related judgments and knows when and how she must employ them, she
lacks those concepts. The sort of bedrock learning, say, of the concept of
table, requires blind obedience on the part of the novice that finally issues
in this sense on the part of the learner that she must say table to describe
certain objects.36 This requirement is fully clear in Wittgensteins discussions
of the problem of skepticism, especially in a passage on learning in which
he seems to liken the adult skeptic to the difficult child who doesnt believe
his teacher or schoolbooks:37
314. Imagine that the schoolboy really did ask and is there a
table there even when I turn around, and even when no one

Introduction

13

is there to see it? Is the teacher to reassure himand say of

course there is!? Perhaps the teacher will get a bit impatient,
but think that the boy will grow out of asking such questions.
315. That is to say, the teacher will feel that this is not really a
legitimate question at all. And it would be just the same if the
pupil cast doubt on the uniformity of nature, that is to say, on
the justification of inductive arguments.The teacher would
feel that this was only holding them up, that this way the pupil
would only get stuck and make no progress.And he would be
right. It would be as if someone were looking for some object
in a room; he opens a drawer and doesnt see it there; then he
closes it again, waits, and opens it once more to see if perhaps
it isnt there now, and keeps on like that. He has not learned to
look for things. And in the same way this pupil has not learned
how to ask questions. He has not learned the game that we are
trying to teach him.
316. And isnt it the same as if the pupil were to hold up his
history lesson with doubts as to whether the earth really...?
317. This doubt isnt one of the doubts in our game. (But not
as if we chose this game!)38
At the level of bedrock learning, there is no room for creative projection of a concept onto novel items that the ordinary concept does not
include. Indeed, bedrock teaching that tolerated such free-wheeling application of a concept would be irresponsible. Williamss account sounds strikingly Confucian:
For the novice...as part of the process of training itself, an
indispensable courtesy is extended to his behavior and utterances.
They are accorded the status of actions and judgments before
they really are such.39
This makes the novice doubly dependent on the community.
Like the master, his action is what it is only against the background of its historical and social setting; but unlike the master,
this status is not ensured by his own competency, but by that of
the master. For many performances of a novice, there is simply
no fact of the matter as to whether he understands correctly
or not. This is because it is not enough to go on correctly; the

14

Whose Tradition? Which Dao?

correct performances must be the exercise of the right kind of

disposition. Acting from ones competency, understanding, and
acting correctly all go hand in hand.40
The novice, like his master, depends on the historical setting for the
practices they eventually share in, but the novices competency as someone
who understands is dependent on the competency of his master. Without
having the proper disposition to go on correctly, that is, to do as the master
does or would do, the novice cannot understand what the master is teaching. His novice-level understanding is dependent on the full understanding
of his master. And those gray-area performances of the novice, not yet
clearly arising from the right disposition, are called instances of understanding only because of their derivation from his learning from a master, who
has that settled disposition.
Although Confucius does not address the question of the constitutive
relation between learning and concepts or the relation between that constitutive relation and the dependence of the novices understanding on the masters
understanding, he does show a profound sense of the need for bedrock learning of ritual, as this learning forms a persons moral sensibility. Confuciuss
disciple, Master You (Youzi) shows an understanding of the importance of
children learning filial obedience and other simple virtues as constituents
of more comprehensive virtues.41 The Analects, then, makes, among others,
two important claims about (ren), moral goodness.42 Youzi claims that
the basis of being morally good is xiao, being filial, and the route to being
morally good is through practicing ritual. Consider the following passage:
. . . .
.. . . .
.
Youzi said, There seldom is one as a man who, being filial
and fraternal, is strongly inclined to go against superiors. There
has never been one [who was] not inclined to go against his
superiors [who] is strongly inclined to foment rebellion. The
ruler should undertake the fundamentals. After fundamentals
have been undertaken, the Way (the standard for appropriate
conduct in interpersonal relationships) is established. Being filial
and fraternal, arent they perhaps the root of being morally good
(in respect of conduct in interpersonal relationships)?43

Introduction

15

In contrast, Confucius claims that the practice of ritual brings about

(ren), moral goodness:
. . . . .
. . . .. .
. . . . . .
Yan Yuan asked how to become morally good. Our Master said,
Controlling oneself and returning to ritual practice is the way to
become morally good. On a single day, if a person has controlled
himself and returned to practicing ritual, then the whole empire
would categorize him as being morally good. Becoming morally
good comes from oneself; how could it come from others? Yan
Yuan said, May I hear the details? The Master said, Dont
look if it does not comply with ritual action. Dont speak if
it does not comply with ritual action. Dont act if it does not
comply with ritual action. Yan Yuan said, Although I am not
intelligent enough, please let me devote myself to these words
[instructions].44
Due to the fragmentary character of the Analects and Confuciuss tendency to make hints and suggestions about how to engage in self-cultivation,
we are left wondering how these claims about xiao and ren might be related.
I suggest the following relationship. Filial piety and ritual propriety have a
fairly clear relationship. A child first learns forms of ritual propriety within
the family. The successful transmission of ritual propriety requires that the
child shows appropriate obedience to and so filial respect toward his parents
and teachers in the course of learning bedrock ritual practices. At the same
time, those ritual practices provide the child with ways to express his respect
and love for his parents and elder brothers. These forms of expression also
have appropriate correlates outside the family.
In whatever ways we clarify the relationship between parents as teachers and their children, it is clear that Confucius considers the learning
of ritual practices as the basis for his self-cultivation project. Indeed, he
emphasizes devotion to learning (haoxue) as the key to self-cultivation.
And although reflection, that is, thinking and questioning, is important,
reflection cannot make headwayand, in fact, will get us into troubleif
it is disconnected from bedrock learning:
. , . , .

16

Whose Tradition? Which Dao?

If one learns something but does not successfully reflect on

its meaning, he will learn in vain. If one merely thinks about
something in the abstract but does not first learn it, he will
face trouble.45
This slogan, reminiscent of Kants concepts without intuitions are empty,
intuitions without concepts blind, also captures a key commitment of
Wittgenstein in his two methodological dicta: A philosophical problem has
the form: I dont know my way about. And, the philosophers aim is to
shew the fly the way out of the fly bottle.46 In both dicta, Wittgenstein
is concerned to combat a form of reflection that proceeds disconnected
from concrete linguistic usage and practice, embedded in specific forms of
life (or language-games). Yet he is also at pains to acknowledge that our
forms of language can give rise to misleading philosophical pictures, which
tend to bewitch us into thinking about our concepts as detached from
their role in the commerce of ordinary life and language. Philosophical
reflection that is guided by misleading pictures and detached from concrete linguistic usage leads to irresolvable conceptual puzzlement. Mastery
of language-games by itself does not protect us from becoming bewitched
by pictures. And reflection that is not guided by concrete linguistic usage
embedded in specific language-games and forms of life will encourage rather
than resolve this sort of bewitchment. The resolution of philosophical torment and the results of philosophical peace come from reflection guided
by concrete linguistic usage. Linguistic usage protected by such reflection
prevents philosophical confusion or resolves it if we fall into it. But this
sort of limitation on reflection is central to Analects 2.15. Our reflections
about dao must arise out of and be limited by prior learning, primarily,
but not exclusively, out of ritual.
A question, however, arises about whether this comparison can be
made between Wittgensteins master-novice relationship, which, for the most
part, concerns learning contexts in which the novice lacks the concepts that
he is being taught by the master, and Confuciuss Master-novice relationship, which concerns an adult novice who understands the concepts he is
being taught if only at an elementary level. In discussing a childs learning,
Wittgenstein distinguishes between ostensive definition and ostensive teaching. The former requires some level of understanding of the concepts being
used and the ability to ask the meanings of the words that express them. So,
a question such as, What is four? when asked by someone who already
knows what numbers are, asks for a definition. Pointing to four objects
would offer an ostensive definition of four. Ostensive teaching occurs when

Introduction

17

the novice has not yet learned basic mathematics and does not yet have a
concept of number. A parent or teacher saying, This is four, while pointing to four objects, is part of the training. Under normal circumstances,
this type of instruction leads the pupil to an understanding of the concept
of number, including the number four. Until that level of mastery occurs,
the novice can be said to be counting, adding, and knowing what four and
other numbers mean only because of the context of the learning and his
relationship to the master of counting.
In contrast, the Analects Master-novice relationships occur primarily
between adults and young adults who have already, presumably, mastered
the basics of ordinary language, including the moral language of the rituals
and virtues. As helpful as this comparison might be between a child learning numbers and an adult learning how to practice moral goodness, it is
not exact. Moreover, if bedrock beliefs and practices are understood only
as the beliefs and practices taught to child-novices who are being initiated
for the first time into a conceptual terrain and related language-games, this
simplistic understanding does not represent the situation of Confuciuss
novices, who are mostly young adults and are already trained in the basic
language and concepts of living morally good lives.47
My response to this concern is multifaceted. There is nothing in the
Analects to suggest that Confucius would be hostile to the claim that moral
understanding starts at a young age and that the training he offers adults is
secondary to that. In fact, in Analects 19.12, Zixia makes this very argument:
.., , .
.. ., . .. .
.. . . .
. . .
Youzi said, The young disciples of Zixia serve as sprinklers,
sweepers, dealing with guests, their coming and going, and these
they are worthy to do. But these are merely minor subjects; as
for regulation of the basics, they have none. How can I deal
with them? Zixia heard this, saying, Alas, Youzis words are
excessive. As for the way of the well-cultivated man, what first is
transmitted? Afterward what should last be transmitted? We can
think of it as similar to groups of plants and trees. We classify
them according to their differences. The way of the gentleman,
how could it be so falsely distorted? As for one who grasps both
the beginning and the end, wont he alone be a wise man?

18

Whose Tradition? Which Dao?

Although not Confuciuss argument, this passage shows an early appreciation of the need to begin in childhood with very basic practices. To this
passage, we might also add the famous passage in which Confucius offers
his spiritual autobiography, Analects 2.4, in which he says that he set his
mind on cultivation since he was fifteen years old. Presumably, his serious
level of commitment did not arise from nothing, but rather, from his own
appreciation at that early age of the benefits of learning he had gained
prior to this.48
Nonetheless, it is true that moral cultivation of children is not the
core focus of the Analects. I would venture to say that because his project
concerned moral cultivation of adults, it was not a topic Confucius felt a
strong need to discuss. It is also noteworthy that despite emphasizing the
importance of filiality, Youzi himself does not discuss filiality in children.
The demands of filiality that he and Confucius discuss are demands of
adult children toward parents. Children might seem to be on a moral
holiday.
Other texts in the early Confucian canon also tend to ignore the
problem of moral training in children. Two sections of The Record of Rituals
( Liji) address these issues, albeit in limited ways. These texts probably
postdate the Analects and reflect the efforts of authors to expound upon
aspects of ritual that go beyond discussions in the Analects. Some suggest a
very permissive approach toward children. Consider the seventh section of
the Inner Pattern ( neize) chapter of the Liji, which contrasts early
morning household requirements for children and adult family members:
.
The children go earlier to bed, and get up later, according to
their pleasure. There is no fixed time for their meals.49
However, sections 7680 specify a curriculum from early childhood
to adulthood:
. . .
When the
to use the
taught to]
[in a] low
of leather;

child was able to take its own food, it was taught

right hand. When it was able to speak, a boy [was
respond boldly and clearly; a girl, submissively and
[tone of voice]. The former was fitted with a girdle
the latter, with one of silk.

Introduction

..
. .
At six years, they were taught the numbers and the names of
the cardinal points; at the age of seven, boys and girls did not
occupy the same mat nor eat together; at eight, when going out
or coming in at a gate or door and going to their mats to eat
and drink, they were required to follow their elders: the teaching
of yielding to others was now begun; at nine, they were taught
how to number the days.

.
At ten, [the boy] went to a master outside and stayed with him
(even) overnight. He learned the [different classes of ] characters
and calculation; he did not wear his jacket or trousers of silk; in
his manners he followed his early lessons; morning and evening
he learned the behavior of a youth; he would ask to be exercised
in [reading] the tablets, and in the forms of polite conversation.

.
At thirteen, he learned music, and to repeat the odes, and to
dance the ko [of the duke of Zhou]. When a full-grown lad, he
danced the xiang [of King Wu]. He learned archery and chariot
driving. At twenty, he was capped, and first learned the [different classes of ] ceremonies, and might wear furs and silk. He
danced the da xia [of Yu] and attended sedulously to filial and
fraternal duties. He might become very learned, but did not teach
others[his object being still] to receive and not to give out.
.

...
At thirty, he had a wife and began to attend to the business
proper to a man. He extended his learning without confining it

19

20

Whose Tradition? Which Dao?

to particular subjects. He was deferential to his friends, having

regard to the aims [which they displayed]. At forty, he was first
appointed to office, and according to the business of it, brought
out his plans and communicated his thoughts. If the ways [which
he proposed] were suitable, he followed them out; if they were
not, he abandoned them. At fifty, he was appointed a great
officer, and labored in the administration of his department. At
seventy, he retired from his duties. In all salutations of males,
the upper place was given to his left hand.50
I suggest that this later text, from the early Han, represents a stage of
institutionalization of moral cultivation beyond the sort of adult training
Confucius gave to individuals. Although we cannot use these passages to
argue that Confucius himself held these views of early childhood moral
training, we can use these passages to show that early expositors of Confuciuss teachings for adults did not see any tension between his accounts
and related accounts of childhood training.
Perhaps the more difficult issue is how far we can understand Confuciuss project as a training of adult novices by a master on the model of
Wittgensteins initial training of children into a conceptual terrain they have
not yet encountered. I will now argue that there is no problem extending Wittgensteins view of training in bedrock practices from children to
adults. This argument requires two steps: First, I will argue that bedrock
practices are not essentially connected to the training of novices whose
learning requires equipping them with the most basic concepts and related
linguistic behavior. Second, I will argue that, in some contexts, young adults
function as novices, learning higher-level bedrock practices from a master
of such practices.
Wittgensteins discussions of master-novice relations are varied. We
know from his discussions of puzzles about mathematics that he sought to
understand how to resolve them by appealing to mathematical learning of
children from ages six to fifteen:
Knowing our everyday languagethis is one reason I can talk
about them. Another reason is that all of the puzzles I will discuss can be exemplified by the most elementary mathematicsin
calculations which we learn from ages six to fifteen, or what
we might easily have learned, for example, Cantors theorem.51
Wittgenstein is pointing out that the puzzles that concern him as a philosopher involve ordinary concepts, and the puzzles can be exhibited in elementary

Introduction

21

mathematics. In this context, the bedrock that he appeals to involves ordinary

ways of counting, speaking about proofs, operating with proofs (constructing them, understanding them, and drawing consequences from them), and
performing the elementary mathematical operations that exhibit these. Many
of the examples he uses in his discussions are of proofs in geometry, a more
advanced form of mathematics than arithmetic. Nevertheless, any discussion
of conceptual issues in geometry would rest on the bedrock of geometrical
practice, including the practice of speaking about lines, points, and circles,
contrasting proofs using compass and ruler, and so on.
For those who had not yet studied geometry, making headway on
clarifying the concept of proof might depend first on teaching them the
rudiments of geometrical practice and the sorts of proof constructions in
Euclid. It would not do, however, for the student to challenge the claim
that Euclids proofs are proofs. For example, it would not do for the student
of Euclidean geometry to challenge the proof procedure that supports the
Pythagorean theorem. The elements of the proof are basic to Euclidean
geometry. Bedrock for this proof would be drawing a square, drawing a
triangle, and so forth. The bedrock status of these practices and related
beliefs is that they must be accepted by the novice as given facts, as elements to be mastered if he is to do and understand geometry. The fact that
geometry is taught after basic arithmetic, taking up the concepts of line,
point, and circle, already part of the novices conceptual repertoire, does not
make basic claims and practices bedrock claims and practices for geometry.
Questioning bedrock claims betrays the questioners failure to understand
them. When someone questions our bedrock claims, rather than assume
we are required to provide a foundational defense, Wittgenstein says we
simply need to indicate that this is how we do it: If I have exhausted the
justifications I have reached bedrock, and my spade is turned. Then I am
inclined to say: This is simply what I do.52
The authority of the teacher-master of geometry rests on the persons
having mastered geometry, as learned from teachers who were masters. As a
result, when challenged to justify his authority in these matters, the teachers
response might very well be, I have learned geometry. Nothing else would
be required.
We might then liken Confuciuss young adult and more mature novices as akin to students of geometry. They bring to their instruction some
ethical practices and language that allows them to raise questions about the
meaning of ethical terms, such as (ren) goodness. Confucius tends to
respond to those questions by getting the questioner to return to a bedrock practice and especially focuses on ritual practice. Youzi also counsels
starting with simple practices that are constituents of more complex ideals.

22

Whose Tradition? Which Dao?

In these similarities, the Analects basic approach to moral cultivation and

understanding moral ideals is consistent with Wittgensteins appeal to bedrock practices. Wittgensteins view of bedrock practices is wide enough to
incorporate learning by beginners who completely lack the concepts, as well
as by those who have a rudimentary understanding of the concepts. The
latter are learning moral conduct, what he calls the inherited background
of our actions and beliefs,53 at a more advanced level.
This approach to moral cultivation is foreign to those of us who have
grown up under the ideals of Western liberalism, but I would think that
even liberals could admit that not just any personal understandings of moral
concepts are cogent. For example, even if we admit a plurality of competing
views of justice, not any action can be understood as just. In Chapter 2, I
argue that liberal conceptions of moral disagreement that accept this account
of a moral bedrock, with masters teaching novices, must admit limitations
on freedom of thought if they are to avoid lapsing into moral relativism.
In the next section, I investigate the ways in which this account of morality
is, nevertheless, consistent with the claim that judgments of how to act in
a particular case sometimes rest on imponderable evidence, which, though
assessable, leaves us with something less than moral certainty.

Imponderable Evidence
Beyond this bedrock level of learning, Wittgenstein discusses a second
level of mastery that concerns judging based on imponderable evidence.
He understands imponderable evidence as information containing one or
more factors that cannot be assessed or measured in exact terms, such as a
persons emotional state at a particular time. Such judging can be learned
through hints and suggestions. Even though his discussion focuses on the
specific question of whether or not there might be experts at judging the
genuineness of a persons feelings, we might consider this a discussion of an
example that would be relevant to other cases where judgments are based on
imponderable evidence. But even this level of mastery would rest on first
learning bedrock practices of the sort recently described, without which one
would lack basic competency, but would also later go beyond that mastery.
Consider what Wittgenstein says about judging others motives:
There is such a question as: Is this a reliable way of judging
peoples motives? But in order to be able to ask this we must
know what judging a motive means; and we do not learn this
by being told what motive is and what judging is.54

Introduction

23

Wittgensteins point here is that there are bedrock practices of judging a motive that must be learned as a condition of understanding what
judging a motive means. One already has to have mastered those basic
practices of attributing motives before the question of the reliability of the
method makes sense to the asker or to the listener of this question. But
then, he says we can distinguish two types of methodological investigations.
One is conceptual, having to do with the methods central to learning the
concept. The other is, let us say, psychological and has to do with what
would give us the most reliable result.55
Despite the bedrock practices we learn that make us competent to use
the term judging a motive, Wittgenstein says that there is in general no
agreement over the question of whether an expression of feeling is genuine
or not.56 When discussing agreement in mathematics, he tells us that
Mathematicians do not in general quarrel over the result of a calculation. (This is an important fact.) Were it otherwise . . . then
our concept of mathematical certainty would not exist.57
And because in the language-game of mathematics, quarrels in general
do not break out over the result of a calculation, mathematical certainty
is conceptually connected to complete agreement. However, certainty in
psychology, where there is generally no complete agreement, has its own
character.58 Certainty in psychologywhich, Wittgenstein says, we might
call subjective to mark the difference between the language-games of psychology and mathematicswould be dependent on expert judgment of
imponderable evidence, not on the learning of a technique that produces
the same results for all who have mastered the language-game.
Wittgensteins comments about whether such expert judgment can be
taught is crucial for his clarification of what the concept of expert judgment consists of. For how we teach it is conceptually constitutive of what
it is. He says,
Can someone be mans teacher in this? Certainly. From time
to time he gives him the right tip.This is what learning and
teaching are like here.What one acquires here is not a technique; one learns correct judgments. There are also rules, but
they do not form a system, and only experienced people can
apply them right. Unlike calculating rules.59
It is important to Wittgenstein that in the case of using imponderable evidence, we do not acquire a technique. But what does that mean?

24

Whose Tradition? Which Dao?

Presumably,
learning a technique would be more straightforward. We know
that to learn a technique, learners are required to get the hang of the correct
sorts of similarities they should pay attention to, say, in projecting the concept of table to what is for them novel instances of tables. However, Wittgenstein emphasizes that the level of agreement on how to project a concept
is almost complete. We can also clarify the concept of table by saying, This
is a table, while pointing to a table, thereby providing a single paradigm
to be used in ones projection. But when it comes to weighing imponderable evidence, the projection onto novel cases, as well as the learning of the
concept, requires attending to multiple aspects of a persons conduct and
demeanor: what he says in other contexts, his body language, his tone in
expressing his feeling, what we know about his general character, the specific
context of the conduct, etc. Presumably, such expertise would also eventuate
in weighted judgments, where the person making or assessing judgments
of others could distinguish between those judgments that are possible but
weakly supported, versus those that are possible but strongly supported. If
people wanted to improve their ability to make weighted judgments, they
would seek out a person with this ability who had demonstrated success at
transmitting it to others. Short of that, they might seek out such a person
for advice when they need to make such judgments.
Just as Confucius takes seriously learning bedrock practices, one
of his disciples also acknowledges that the skill of addressing questions
involving imponderable evidence requires a higher-level of learning beyond
understanding bedrock practices, one used to address questions involving
imponderable evidence that can be tackled only by those who have a special
knack. Yan Hui, Confuciuss favorite student, describes Confucius as having
such a knack.
In one passage of the Analects, we find Yan Huis perspective on what
it is like to be guided by his master. This description could function as an
account by a novice describing what it is like to be guided by a teacher
with a knack for expert judgment when the novice lacks that ability:
. . . . .
.. . . .
. . .
Yan Yuan sighed admiringly, saying, The more I look upward
toward it (Confuciuss dao), the higher it seems. The more I dig
deeper into it, the more impenetrable it seems. When I see it
ahead, suddenly it is behind. Our Master is adept at guiding

Introduction

25

others step by step: He broadens our knowledge by having us

practice cultural refinement. He restrains our conduct by having us practice ritual actions. Even if I intend to complete this
instruction, I cant. When I have exhausted my capabilities, it
seems like our Masters way still stands there, majestically. Even
though I desire to follow it, I have no way.60
Here Yan Hui describes a stage in which the novice has grasped
the reality of Confuciuss dao, struggles to follow or emulate the Masters
way, but cannot figure out how to complete the next steps and judgments
required for him to proceed by himself. He can be guided by Confucius
and can appreciate the correctness of his judgments in leading him on the
path, but Yan Hui cannot make those judgments himself. We might be
able to locate Yan Huis stage as somewhere in the middle, between ages
fifteen and thirty, in terms of Confuciuss own account of the stages of
development along the way:
. . . .
. ...
I have set up my mind in cultivation since I was fifteen. I have
had my stance since I was thirty. I have been no longer bewildered
since I was forty. I have understood the heavenly mandate since I
was fifty. I have thoroughly understood others words since I was
sixty. I have no longer surpassed the code rules while following
my heart and minds desires since I was seventy.61
Yan Huis incapacity to judge how to carry on would seem to reflect
a stage prior to the loss of bewilderment. This would be a stage in which
the Masters hints and suggestions have not yet brought about the capacity
in his student to make appropriate decisions and judgments autonomously.
Yan Hui acknowledges the correctness of his teachers judgments and authority as teacher, despite the complexity of learning how to follow dao. He
acknowledges Confuciuss status as having a kind of expertise that he
lacks, the expertise of knowing how to carry on, step-by-step, along the way.
Analects 9.11 is sometimes thought to indicate a mystical strain in
Confucian thought,62 but it need not indicate that.63 It also seems to point to
the ways in which any successful complex practice, most especially one that
does not form a system, is accessed and understood by its master in ways
that differ from the novices understanding. The master grasps the norms

26

Whose Tradition? Which Dao?

governing the practice and how to carry on with them, even if he cannot
articulate them as a system. The novice grasps the reality of the norms
governing the practices once removed, through his teacher. That being the
case, it is no wonder that a standard exchange between Confucius and his
interlocutor have to do with whether particular persons or actions exhibit
(ren) (goodness) and with how they might better understand and fulfill
the complex requirements of ren and its constitutive ideals.64
Indeed, the sort of perplexity that Yan Hui expresses would be natural
for the sort of teaching that Confucius engages in. His teaching, he claims,
requires a certain kind of student, one who can quickly figure out how to
carry on with practicing and mastering succeeding steps of Confuciuss way
by himself:
.. . . .
The Master said: If one isnt in the fen status (of seeking to
understand even if he hasnt yet understood), I wont guide
[open him, instruct] him. If one isnt in the status of fei [trying his best to express his intentions in words, even if he hasnt
yet expressed any single word], I wont guide him. If I mention
one corner and he doesnt infer to the other three same types
of corners, I wont repeat [my former instruction].65
This confirms the claim that Confucius teaches by hints and suggestions,
and the novice needs to grasp how to carry on to the next corners, that is,
intuitively go on to the next steps by himself.
There is, however, a possible problem with the distinction I have
drawn between learning bedrock practices and using imponderable evidence to make a judgment. We might wonder the following: why should
we think of Yan Huis confession in Analects 9.11 as referring to use of
imponderable evidence about fulfilling the multiple requirements of ren,
as opposed to talking about learning any less complex, bedrock practices
of ritual? I would like to suggest that passage 9.11 might be understood
in both ways. The more bedrock the practices are, the less the chance that
an adult, young adult, or disciple will describe his masters teaching in the
way Yan Hui does.66 The most basic levels of learning (bedrock practices)
would not be accompanied by the sort of phenomenological account that
Yan Hui gives. For at most bedrock levels of learning, a young child will
not have the conceptual and linguistic capacities that Yan Hui exhibits in
his account. We might, however, project such an adult sensibility onto
a child, as a way of characterizing the way the child advances beyond a

Introduction

27

stage of not grasping the concepts he is being taught. But this would be
no more than a projection.67
Despite the uncertainty of judgments based on imponderable evidence, it is possible to imagine two different approaches to these judgments.
We could argue that because the correctness of such judgments is uncertain,
we should count any one judgment to be as good or accurate as any other.
In contrast, we could argue that some judgments are more cogent than
others. It is clear that Wittgenstein opts for the latter approach. For even
though he admits there will be, in general, no agreement over whether
a persons feeling are genuine, he still allows that In general predictions
arising from judgments of those with better knowledge of people will be
more correct.68 Even so, Wittgenstein indicates that judgments based on
imponderable evidence sometimes can and even must be proved in some
other way. He gives the example of the chemical structure of a substance.
Imponderable evidence might convince us of the structure, but we would
still need to be able to demonstrate the correctness of this judgment by
appealing to consequences of the claim being true.
Whether there is such an alternative form of proof when considering
the best way to live ones life is by no means clear. It does not, however,
appear incoherent to think that some teachers might play a role in establishing a tradition through reputation for being better than others at making judgments about how to live well. Traditions could be formed around
examples of such judgment, and the traditions might flourish by transmitting examples of such judgments along with hints about how to make ones
own judgments. This possibility seems to be found in the Analects accounts
of Confucius as teacher. Though there might be disagreement about the
wisdom of judgments Confucius makes, the formation of such a tradition
will rest on substantial agreement about Confuciuss excellence as a teacher.

Peaceful Agreement and Harmony

Central to Wittgensteins account of certain practices as bedrock is his related
insight that when we agree about these practices, we agree in a distinctive
way. We might say this agreement runs deeper than the sort of agreement
involved in agreeing about this or that opinion. According to Wittgenstein,
each form of life is constituted by those judgments and practices adherence
to which makes a person a practitioner of that form of life:
Disputes do not break out (among mathematicians, say) over
the question of whether a rule has been obeyed or not. People

28

Whose Tradition? Which Dao?

dont come to blows over it, for example. That belongs to the
scaffolding from which our language operates (for example,
yields descriptions.)69
So language itself, and our forms of life interconnected with it, are
based on agreements in the framework that make the form of life and related
language possible. Training in the language and the practices that make up
the form of life will insure basic levels of agreement, each appropriate to
its distinctive language and form of life.
But this form of agreement is not only practical. It is bound up with
our views about truth and falsehood. For when we judge in accordance with
those beliefs that make up our frame of reference, we believe that those
judgments correspond to reality. And we view our judgments as part of
reality not because we know that the inferences based on our framework
beliefs assure that our judgments are certain, but because this is what it
means for our judgments to correspond to reality. Consider this point,
from On Certainty:
[Everything that we regard as evidence indicates that the earth
already existed long before my birth. The contrary hypothesis
has nothing to confirm it at all. If everything speaks for an
hypothesis and nothing against it, is it objectively certain? One
can call it that. But does it necessarily agree with the world of
facts? At the very best it shows us what agreement means. We
find it difficult to imagine it to be false, but also difficult to
make use of.] [Inside brackets crossed out in ms.] What does
this agreement consist in, if not in the fact that what is evidence
in these language-games speaks for our proposition? (Tractatus
Logico-Philosophicus)70
Wittgensteins point is that the meaning of agrees with reality is constituted by the frame of reference we operate with. So in mathematics, to
agree with mathematical reality just means to operate by the rules of the
framework of mathematics.
But this view can certainly seem problematic. And Wittgenstein confronts this problem head on: his philosophical voice, tempting him toward
a problematic philosophical theory, asks:
So you are saying that human agreement decides what is true
and false?What is true or false is what human beings say; and

Introduction

29

it is in their language that human beings agree. That is agreement

not in opinions but rather in form of life.71
His point here is that agreement is crucial, as he indicates, in both judgments and definitions if people are to be able to communicate, that is, if
they are to use clear, shared concepts. He offers the example of measuring:
we need a shared definition, but we also need shared results of measuring
for us to have a concept of measurement with a communicable meaning.72
We can easily relate these claims to my earlier claims about learning.
For a novice to learn how to measure length is for the novice to learn how
to get the results the master measurer gets. And that means to have the
skills to get the results that almost all of us get as measurers. These results
are guaranteed by the character of the practices of measuring, the relative
constancy of objects we measure, and our ability to transmit the practices
of measuring from one generation to the next.
We ought not to expect the same level and type of agreement we find
in mathematics in other bodies of knowledge, for example, in aesthetics. But
each fields framework will be constituted by those judgments, definitions,
and bedrock practices around which its distinctive forms of agreement and
disagreement are possible.
In Remarks on the Foundation of Mathematics, Wittgenstein refers to
this same sort of agreement as peaceful:
The application of the concept following a rule presupposes a
custom. Hence it would be nonsense to say: just once in the
history of the world someone followed a rule (or a signpost;
played a game, uttered a sentence, or understood one; and so on).
...
It is of the greatest importance that a dispute hardly ever
arises between people about whether the colour of this object
is the same as the colour of that, the length of this rod the
same as the length of that, etc. This peaceful agreement is the
characteristic surrounding of the use of the word same. 73
The characteristic surrounding of the use of a word is just those
rules, practices, and judgments that constitute what is required to learn a
rule and how to carry it out in a way that all competent learners of the
rule will agree to and recognize as correct. Learning of bedrock practices
is constituted by peaceful agreement upon the bedrock. If the bedrock
practices and related beliefs were not agreed to, as basic to the framework

30

Whose Tradition? Which Dao?

of thinking and acting and as parts of a form of life, then they would not
be parts of that framework.
It is important to recognize that peaceful agreement and the related
phenomenon of disputes not breaking out does not indicate complete agreement about every opinion a person might have. Indeed, Wittgenstein makes
it clear that although in some frameworks disputes do not break out, they
do in others: for example, in psychology. In the previous section, I discussed
one example from psychology: attributions of feelings to others. There is
good reason to think of disputes in ethics as being akin to these, for questions of what to do next often involve weighing and balancing of competing
requirements with no clear system of judgment, even though we might distinguish between better and worse judgments. Each framework of thinking
and language has its own distinctive peaceful agreements and related range
of disputes, along with ways of resolving them. These peaceful agreements
are the basic practices and beliefs that masters convey to novices, thereby
making them competent to engage in assessment of opinions in dispute.
This sort of peaceful agreement constitutes what it is to be able to act
on and understand shared norms, which Wittgenstein says is basic for any
language: For only through a technique can we grasp a regularity.74 The
phenomenon of language is based on regularity, on agreement in action.75
But this means that language itself and the normative regularities that constitute it depend on group harmony. Williams makes this point as follows:
But even solitary practices are cultural practices and so have a
social dimension since the context of regularity and agreement in
judgment (in what is said) must be provided by a community,
that is, by a group of people reacting, judging, and behaving in
harmony. They owe their identity to this social background even
though they may be carried out by one individual at a time.
Without conformity to a group at bedrock level, normativity
is impossible.76
Although Confuciuss appeals to harmony (he) are not extensive,
it would not be hard to find some common ground with Wittgensteins
insights about the relations among agreement, normativity, and learning.
Consider the following passage, in which Youzi distinguishes between illicit
and licit harmony:
. . .. . .
.. . .

Introduction

31

Youzi said, among the uses of ritual action, harmony is most

important. The Way of the first kings because of this was most
graceful. In matters small and large, they followed this. As for
those things we shouldnt practice, achieving harmony by realizing harmony for its own sake, but not using ritual proprieties
to achieve it, this is indeed not acceptable.77
This key passage distinguishes between two approaches to harmony,
one that seeks it for its own sake and one that seeks it as interconnected
with ritual propriety. I take practices of ritual propriety to be bedrock practices, exhibiting dao, the norms of interpersonal engagement. The harmony
Confucian adherents want to embrace as fundamental to their conduct is
harmony that arises from shared agreements in a ritualized form of life,
carried out through a set of practices that embody norms of interpersonal
engagement. The sort of harmony they wish to avoid is harmony or agreement that is inconsistent with those practices.78
Although not the same, this distinction is very close to Wittgensteins
distinction between agreement in opinions and agreement in forms of life.
The latter is a deep sort of agreement, as we have seen, based on learning
of bedrock practices and beliefs that constitute the framework of norms
governing conduct and thinking and those opinions and practices that a
majority of a community might happen to go along with, although not
embedded in basic agreements in a normative framework.
Wittgenstein draws his distinction in order to resolve conceptual confusions about understanding and rule-following, but a key point he makes
in doing this is to clarify the difference between a community agreeing that
an opinion is true and a community agreeing on its form of life, including
normative frameworks that are not made up of optional beliefs. Confucius
draws his distinction in order to avoid a possible misunderstanding of the
value and role of harmony of life lived in accordance with those fundamental
norms governing human life (dao).
Confucius offers this distinction, perhaps even more clearly, in the
following passage:
. . .
A junzi (exemplary person) is harmonious (in dealing with others), but not the same as them (when others violate dao), and a
mean man might be the same as others (even when they violate
the dao), but not harmonious (in dealing with them).79

32

Whose Tradition? Which Dao?

Casual agreement is to be avoided, but a deeper agreement, or harmony, in

normative commitments and in their manifestations are requirements for
being an exemplary person.

Nothing Is Hidden
For Wittgenstein, the project of mastery of language-games is complex,
and mastery differs from language-game to language-game. So Wittgenstein
points out that certainty, a mark of mastery, differs from language-game to
language-game.80 The type of certainty of mathematics involves complete
agreement, whereas psychological certainty does not. These two forms of
certainty are not thought by Wittgenstein to rest on some fundamental
principles found in mathematics that are lacking in psychology. Rather, they
rest on the fundamental differences between the two different languagegames. Even in mathematics a teachers ability to formulate his teaching in
clear principles and definitions will come to an end. At best, the teacher
will be able to exhibit his own mastery by showing how to calculate. So,
for example, if I wish to teach my student the meaning of the formula X
+ 1 = Y, I will be able to show what that means by solving the formula
for various numbers until the novice is able to solve them himself.
In discussing how the master communicates to a novice how to follow
a rule, Wittgenstein says:
We talk and act. That is already presupposed in everything that
I am saying.
I say to him Thats right, and this expression is the bearer
of a tone of voice, a gesture. I leave him to it. Or I say No!
and hold him back.
18. And does this mean that following a rule is indefinable?
No. I can surely define it in countless ways. Only definitions
are no use to me in these considerations.81
Wittgensteins point here is that if the teacher is to communicate his
understanding to the novice, he will do that best and for the most part by
showing him how to proceed with the formula. One might think that he
could communicate his understanding by giving the student definitions of
the parts of a formula. However, Wittgensteins point against this suggestion is that multiple definitions are possible, but what we want is not for

Introduction

33

the novice to be able to define parts of the formula, rather, to be able to

carry it out.
It can seem that just being able to carry out a formula is too little, but
Wittgenstein resists this. When teaching a formula to a novice, Wittgenstein
asks, What do I mean him to do? He answers:
Now what do I mean him to do? The best answer I can give
myself to this is to carry these orders on a bit further. Or do
you believe that an algebraic expression of this rule presupposes
less?...Here there is nothing more difficult than to avoid
pleonasms and only to say what really describes something.82
For the teacher, there is an overwhelming temptation to say something more,
when everything, meaning how to carry out the formula step-by-step, has
already been described.
The temptation is to describe some abstract formula that is the content of the rule. But Wittgensteins point is that such a formula would not
be the best way to capture what the master means the student to do. For
what he means him to do, the best way to show that is by showing him
examples, the very things that following the rule results in for a master of
the practices the rule is embedded in. Another formulation would not by
itself get a person any closer to those results. Indeed, we could ask the same
question again of this formula: in uttering it, what does the master mean
the novice to do? And the best way to show that would be by applying
and producing the correct results.
This focus on practice over abstract formulation of principles is exhibited throughout the Analects. Confuciuss project rests also on the basic
notion that walking the dao is complex, too complex for any clear formula
to encompass how to do that. The variety of formulas describing what ren
(goodness) is suggests that it can be characterized differently in different
contexts and for different people. The urgency of the interlocutors questions also suggests that they are in a situation in which they have not yet
grasped the basic concepts of the dao-constituents, such as ren. They do
not need definitions, but rather, examples and practice of the proper use
of the dao-constituent terms.
Indeed, there is a wing of the Confucian tradition that rests on this
very point. In contrast to Zhu Xis emphasis on study of texts and reflection on their meaning, Wang Yangming recommended basic daily practice.
However, this practice would be the practice established by the instances
of exemplary conduct offered in the classical texts. My point here is simply

34

Whose Tradition? Which Dao?

that we can treat a masters instructions to a particular person as an example

to be heard and understood in context as a model for how to proceed in
novel contexts, not as something to be followed in a mechanical way. Wang
Yangmings approach to daily practice seems to embody an insight here
similar to Wittgensteins insight about imponderable evidence: we can be
taught how to get better by hints and suggestions and by taking as hints
and suggestions exemplary conduct of earlier masters.83
Wittgenstein notes the need for this sort of teaching when a person
has not yet understood a concept: But if a person has not yet got the
concepts, I shall teach him to use the words by means of examples and
by exercises.84 But where a novice learns how to manage imponderable
evidence, the teachers examples are marked not by paradigmatic objects
(such as, This is a table.), but rather, by examples of how to use imponderable evidence to make a judgment. The teachers judgments in particular
circumstances themselves become the examples novices use to learn how to
follow a customary way by themselves.
Even though this might seem to some a deficient form of communication of what the teacher knows, Wittgenstein would disagree: And
when I do this I do not communicate less to him than I know myself.85
For what novices basically know when they understand a concept is how
to carry on with it, to apply it to more examples, or to apply the concept
to wider areas. Nonetheless, a novice might feel that the teacher is holding
something back. Wittgensteins point is that the successful practice of using a
word is all there is to understanding it: Since everything lies open to view,
there is nothing to explain. For whatever may be hidden is of no interest
to us.86 This aphorism makes good sense in the context of an account of
understanding concepts that depends on learning techniques and customs.
Beyond that, there is nothing left to investigate.
Given the similarity of their basic approaches, it should be no surprise
to find Confucius, like Wittgenstein, being pressed to reveal his own hidden
teaching while he denies that there is anything he is holding back:
:. .
..
Do several among you consider me to hide something from you?
Actually I dont hide anything. I have no conduct that I dont
share with you. This is who I am.87
To a novice, something deep can seem left out, but if all the Master offers
is mastery of a practice, then there is nothing hidden beyond that. To dispel

Introduction

35

the sense of the hidden, the novice only needs to master the practice for
himself. And the Master teaches all novices how to do that. The Masters
understanding consists in the practices and set of judgments in particular
circumstances that he teaches.
For this reason, Confucius points out that he would rather not speak
in the course of teaching:
.... . .
. . . .
Our Master said: I intend not to speak. Zigong said: If you
do not talk, then how should the youngsters (know how to)
say anything? Our Master said, What does Heaven say? The
four seasons operate thereby; the various (living) things engender
themselves thereby. What does Heaven say?88
Heaven influences nature and all living creatures but not by speaking, just
as a master influences novices but not always by speaking. If both Confucius
and Wittgenstein are correct, their approach to teaching is not deficient;
instead, it clarifies the way to teach complex, bedrock practices to novices.
For Wittgenstein, understanding basic concepts, and so understanding
their applications to particular cases, requires applying concepts to their
range of instances, organized through family resemblance, not through necessary and sufficient conditions.89 This approach to teaching and learning
does not undermine the intelligibility of the concept, but it does require an
understanding, primarily in the form of practical mastery, of how to apply
concepts in various ways. So, Wittgenstein makes the point in his most
famous example, the concept of games, where game applies to a variety
of more or less similar things.90 To grasp the concept of game requires
mastery of the application of game to this variety of entities, along with a
sensitivity to the differences between types of games. Because of his account
of concepts and his view that philosophical investigations aim to clarify
concepts, which are usually organized through a complex web of similarities,
Wittgenstein conceives his form of philosophical investigation as traveling
criss-cross in order to grasp the complexity of language and concepts and
to avoid the temptation of stopping prematurely by settling for an overly
simplistic account of the conceptual terrain under investigation.91
It is also useful to think of Confuciuss method as a criss-cross method
of cultivation, teaching novices how to understand and put dao into practice. Confuciuss reluctance to offer single general formulae for (ren) and
other dao-constituents and his tendency to offer different characterizations

36

Whose Tradition? Which Dao?

of dao and its constitutive ideas give evidence for this interpretation, as
does Yan Huis phenomenology of learning to follow dao. The labyrinthine
character of dao rests on the complexity of the norms that constitute dao
and its language.
We might think of both thinkers as adopting a form of what James
Klagge has called constitutivism.92 This is the view that with respect to
some concepts, what counts as instantiating them is variable. In different contexts, different behaviors or qualities count as good, for example.
Goodness consists of a complex array of elements in different contexts. As
Klagge points out, although Wittgenstein never developed a settled view
of the concept of goodness, at one point, he indicates that what counts as
good may be so complex that all one can do to clarify it is to describe a
whole environment.93 This would be a possible explanation of why Yan
Hui thought finding out how to carry on and take the next step was so
difficult. It might also go some way toward explaining why it is that Confucius answers the What about ren? questions in so many different ways.
There is no doubt that Confucius brings a different set of concerns
to dao compared with the concerns Wittgenstein brings to investigating
concepts. Confucius is also interested in how to foster self-cultivation of
his interlocutors. Teaching them an entire way of life requires him to offer
different characterizations of dao-constituents to different interlocutors.94
In contrast, Wittgenstein illustrates a method of engaging in philosophical
therapy. He displays the variable instantiations of concepts, thereby reducing
philosophical puzzlement and torment.95

Shared Insights in Different Contexts

In this chapter, I have argued that Confucius and Wittgenstein hold in
common a set of basic insights about our fundamental relationship to norms
or dao. They both hold that (a) our primary relationship to norms comes
from learning bedrock practices through the guidance of a master; (b) better
ways of managing imponderable evidence can be taught; (c) harmony or
peaceful agreement is constitutive of a proper orientation to norms or dao;
and (d) no final, systematic formulation of these norms is possible, and if
it were possible, it would not help provide us with mastery of those norms.
A fundamental critique of this argument might call into question my
account of their insights similarities, for Wittgenstein is no more Confucian in his outlook than Confucius is Wittgensteinian in his. Wittgensteins
project of philosophical investigation of the problems of philosophy and

Introduction

37

Confuciuss cultivation of the dao project are worlds apart, not just in
time, but also in goals and methods. In the last section of this chapter, I
would like to provide a general response to this objection and to indicate
the ways in which these similarities provide the basis for the arguments of
subsequent chapters.
The objection supposes that distance in time and project might be
sufficient to make holding insights in common impossible. I will ignore the
problem of distance in time as that is not fundamental by itself. Distance in
time is just a symbol of fundamental differences in concepts and projects.
The criticism raises the question: what insights do the two philosophers
hold in common or share? In this context, the argument that Confucius
and Wittgenstein hold something in common rests on two fundamental
claims: (a) that the ways they formulate their views of our relations to
norms governing living well (dao) share a family resemblance; and (b) that
they follow similar procedures in addressing misunderstanding or norms. I
will address each point in turn.
I have argued that Wittgenstein and Confucius take learning to be the
basic path to understanding norms (or dao). If my claim about this point is
wrong, that should come out in criticism of my specific interpretive arguments. But beyond that, one might argue that whereas Confucius promotes
learning as a way of understanding and conforming to dao, he does not
make the sorts of philosophical claims that Wittgenstein does. Wittgenstein
claims that the ways a person learns a concept are constitutive of what the
concept means. Nowhere does Confucius make such a claim. In fact, he
appears not to be interested in the philosophical question of the relation
of learning to conceptual content. This argument supposes that holding an
insight in common requires holding it in the same way. But this requirement would seem to be too stringent. We might, for example, argue that
Democritus and Newton shared a common insight about the basic structure
of the substances without holding that they made the same claim in the
same ways about them. Moreover, I would agree that Confucius makes no
general claims about the content of concepts, but that does not undermine
the claim that he seeks to promote understanding of dao (consciously living a good life) through mastery of basic practices. Although Confuciuss
cultivation project rests on his insight about dao, which he does not define,
develop, and defend theoretically, Wittgenstein formulates a similar insight
and proceeds to use it to resolve philosophical puzzles.
One might also argue that although Wittgenstein formulates an
insight, he does not show an interest in promoting moral self-cultivation.
Yet, neither of these true claims entails that the two thinkers do not hold

38

Whose Tradition? Which Dao?

an insight in common about the relation of learning to understanding. No

doubt Wittgensteins formulations of the insight tend to be more conceptual
and aimed at countering competing philosophical views that give rise to
conceptual confusion. And Confuciuss formulations are designed help those
engaged in cultivation of dao. However, neither of these facts undermines
the claims of similarity I am indicating.
My argument is strengthened by the ways both thinkers use this similar insight to structure their projects. Confucius encourages his disciples to
undertake those forms of learning that will lead them closer to conformity
to dao, and in so doing he tends to eschew a presentation of abstract
accounts of a life conforming to dao. Wittgenstein encourages philosophers
to investigate concrete ways that concepts are learned in order to address a
range of philosophical problems.
Perhaps at this early point these arguments seem glib. In the chapters
that follow, I address a range of problems that come up in the course of
reflecting philosophically on Confuciuss approach to dao and Wittgensteins
approach to conceptual clarification. In general, I think that these two
thinkers shared insight makes for a possible, powerful collaborationwith
Wittgenstein protecting Confuciuss project from misguided philosophical
critique and Confucius providing Wittgenstein with resources for thinking
about the ethical implications of his project. I turn to these questions in
the next chapter.

Confucius, Wittgenstein,
and the Problem of Moral Disagreement

Here we may say that we have all the materials of a tragedy; and we
could only say: Well, God help you.
Wittgenstein to Rush Rhees1
To use a word without justification does not mean to use it wrongfully.
Wittgenstein, Philosophical Investigations, section 289

Introduction
In the previous chapter, I argued that Confucius and Wittgenstein share
some basic approaches to their projects despite fundamental differences in
time, context, and their reflective goals. One insight they share is that the
contents of ideals and concepts are constituted by the ways in which the
practices that instantiate them are learned. Beyond these basic insights,
nothing about ideals and concepts is hidden. Nonetheless, even though
we can display mastery of those practices and describe what this mastery
consists of, we cannot offer general formulae for the use of these concepts
or ideals because they are learned only by learning how to engage in a
complex practice. And for this reason, explanations of these concepts and
ideals are of limited help in training children, students, or adherents and
in attempts at clarification of what constitutes the concepts and ideals. If
this basic insight is correctwe cannot offer general formulae for the concepts or ideals of either philosopherthen it would be a mistake to argue
that Confucius and Wittgenstein fail to give us the types of accounts we
need: namely, foundational theories that would provide an explanation and
39

40

Whose Tradition? Which Dao?

epistemological or metaphysical justification of the practices and concepts

embedded in them. If these two philosophers indeed share this insight, the
project of trying to define such foundational theories, to establish which of
various languages and practices are correct, would be out of order. Moral
disagreement will not be addressed by using foundational theories to refute
competing theories or moral judgments but must be understood and managed within the context of this basic insight. For Confucius, that means
that moral disagreement with his teachings will be addressed within the
context and limits of learning and reflecting on the practices and ideals
that make up dao. For Wittgenstein, moral disagreement will be viewed as
a motley, to be addressed differently, depending on context and the kind
of language-game in question.
This approach to moral disagreement is not, however, without its possible problems. Confuciuss lack of interest in foundational questions might
seem to reduce his thinking to provincial, arbitrary ethics. I will, however,
argue that this view of Confuciuss project can be neutralized by appeal to
Wittgensteins realistic spirit in ethics. Nonetheless, one can wonder how
far any non-foundational, anti-theoretical approach to ethics can go by way
of offering a way to resolve moral disagreements. Indeed, some interpreters of Wittgensteins later philosophy, such as Stanley Cavell and Stephen
Mulhall, maintain that Wittgenstein offers an account of rules that offers
no authoritative way to resolve disagreements. If Cavell and Mulhall are
right, then my project of using Wittgensteins later philosophy to clarify
and defend Confuciuss approach to ethics would fail because lacking a way
to manage moral disagreement would support the critics of anti-theory in
ethics.2 In this chapter, I will discuss the approach to moral disagreement
in Confucius and Wittgenstein and will argue that Mulhalls view of later
Wittgenstein is mistaken. I will argue that resources for resolving moral
disagreement can be found in a joint Confucius-Wittgenstein approach to
ethics. First, I will examine the way that Confucius addresses disagreement
within the context of his account of the basic insights.

Confucius and Disagreement

The Analects represents Confucius as refusing to engage as teacher or critic
with some of his interlocutors. Consider his attitude toward recluses, those
who operate outside his norms governing social interaction. These encounters offer a chance to see how Confucius responded to disagreement from

Confucius, Wittgenstein, and the Problem of Moral Disagreement

41

those who operate with what he would consider a heterodox set of norms,
which, as we can see from Analects 2.16, he considers harmful:
. . .3
If you are researching on different doctrines, then this is already
harmful.
Here we can see that Confuciuss approach to disagreement involves
holding onto his own commitments, which he uses to comment on those
who operate with another approach, but he does not discuss his views with
those who operate with radically disparate commitments:
. . . .
. . . . .
. . . . .
.. . . . .
. , . , .
. . . . .4
Zilu, a disciple of Confucius, fell behind while following Confucius on the trip. He met an old man who carried a basket on
his shoulder with a cane. Zilu asked him, saying: Have you seen
my master? The old man replied, As for one, four limbs are not
hard working and cannot distinguish five types of grain who can
serve as your master? He weeded while putting aside his cane.
Zilu stood before him, while paying respect. He kept Zilu and
had him stay the night. The old man revived Zilu by killing and
cooking a fowl, made millet, and fed him. He introduced Zilu to
his two sons. The next day, Zilu reported this to his master after
catching up with the rest of the group. The Master said, They
are recluses. He sent Zilu back to see him. When he arrived, the
old man had already left. Zilu said to his Master: Not taking
office is not righteous. The distinction between older and younger
cant be abolished. As for the righteousness between sovereign and
ministers, how could it be set aside? In keeping his conduct pure,
he neglects the major relationships. As for a junzis (exemplary
persons) taking office, it involves doing his duty, righteous actions.
The way does not prevail. We already know this.

42

Whose Tradition? Which Dao?

.. . . ?
. . . . .
. ..5
The Free Spirit of Chu, Jieyu, passed Confuciuss chariot while
singing. Jieyu said, How pitiful the Phoenix is! Why has your
virtue declined to such an extent? What is past, one is not able
to criticize. What is coming, one still can pursue. Give up! Give
up! One who today engages in government service is merely
dangerous. Confucius got down from his chariot and intended
to speak with Jieyu. Quickly Jieyu hurried off. Confucius was
unable to speak with him.
In both these cases, Confucius or his disciple, Zilu, comes into contact with
a recluse. Both attempt to engage the recluses in discourse, but they run
off. Confucius is robbed of a chance to debate them. If we assume, as I do,
that for all we know these passages are fictional,6 we can imagine that they
might have been told differently. They might have involved an encounter
with Confucius and a sustained argumentative discussion. What should we
make of this? The Analects often shows the proper approach to a situation
through Confuciuss actions rather than through spelling out and defending
some principle governing how to act. So we might take this hint and try
to understand these two passages as illustrating an exemplary approach to
disagreement over the basic norms governing conduct. The recluses hold up
the ideal of purity in the face of moral decline (as in the old mans gracious
hospitality toward a stranger and Jieyus criticism of Confucius for being
a government lackey), whereas Confucius upholds fulfilling ones duties to
others and reviving dao. Confucius never is shown to engage the recluses
in discussions of these normative disagreements. Why not?7
We might find part of an answer by looking at an encounter with
Zaiwo, a disciple, whose normative commitments depart radically from
those of Confucius. This encounter is similar to the encounter with the
recluses, though in this case Confucius and Zaiwo talk a bit.
. . .
. . . . . .8
Zaiwo asked a question by commenting: The three-year mourning period, its duration is already too long. If a junzi (exemplary

Confucius, Wittgenstein, and the Problem of Moral Disagreement

43

person) does not conduct rituals for three years, the rituals will
certainly go bad. If a junzi does not engage in music for three
years, his practice of music will certainly fall apart. New grain
has already been sent to market, and the old grain is already
gone. The drill and flint stone for making fire have changed. A
full year of mourning is already enough.
. . . . . ..
. . . . .
. . .
Our Master replied: Eating rice and wearing colored silk, to you,
would be comfortable? Yes, comfortable. If you are comfortable,
then do it. As a junzi during the mourning period, he doesnt
feel satisfied in eating fat meat, is without pleasure listening to
music, conducts daily affairs without peace, and for this reason
he does not do these things. If you feel settled, then why not
do it? Zaiwo left.
. . . .
,. .
Our Master said, Zaiwos not complying with ren! A son has
already been born three years; only after this, can he leave the
bosom of his parents. The three-year mourning period is understood everywhere. And Zaiwo, was he not loved by his mother
and father for those first three years?
In this exchange, Confucius does not engage Zaiwo in any extensive debate.
He does not try to persuade him of the incorrectness of his views. Indeed,
he makes the point that if he is comfortable with a shorter mourning
period, then he should do that. Outside of the context of some shared commitments, shared dialogue, criticism, or helpful suggestions about how to
change to conform to dao would not be productive. The basic insight that
Confucius brings to this discussion is that sensitivity and responsiveness to
norms (dao) arises from learning. But just as successful reflection depends
on prior learning, successful instruction rests on accepting the authority of
ones teacher and his normative commitments.
This general approach to disagreement is encapsulated in Analects 9.30:

44

Whose Tradition? Which Dao?

. . . . .
. .
Some [who are] able to study together have not yet been able to
reach the Way. Some [are] able together to get to the Way [but]
are not necessarily able to stand firmly on the upright way. Some
[are] able firmly to stand together on the upright Way [but] are
This passage indicates limits to complete agreement. Two people might study
together the norms governing human life, but they may not make the same
progress. Even if they make the same progress, they may not have the same
commitment to upright conduct. But even if they have the same commitment to upright conduct, they need not weigh and balance considerations
about how to act in the same manner. This passage seems to me to be both
cautionary and expressive of a kind of resignation. There are indeed limits
to agreement. Confuciuss manner of addressing disagreements accepts these
limits rather than trying to resolve them.10
For moralists, however, this approach might seem inadequate. If Confucius and his followers take up this stance toward moral disagreement,
one might argue, they miss an opportunity to correct an error, but they
also miss an opportunity to demonstrate the objective, correct approach to
these norms. But because Confucius takes learning of practices of ritual to
be basic and to be mastered prior to reflection, he is skeptical about the
possibility of resolution of disagreements between those who lack shared
bedrock practices.
I have argued that Wittgenstein and Confucius maintain similar
insights of the role of bedrock practices in understanding norms. Their
shared commitments can be characterized as follows: our primary relationship to norms arises through learning.
Wittgenstein uses this insight to address philosophical accounts of normativity that produce conceptual puzzlement. For example, he argues that
an account of norms that thinks of understanding rule-following as resting
on a persons interpretation cannot make sense of the distinction between
correct and incorrect actions. For any interpretation might itself be variously
interpreted. He concludes that there must be a form of understanding a
rule that is not an interpretation. Instead, this form of understanding is
based on custom or practice, something transmitted from master to novice.
Although Wittgenstein does not develop a moral project of self-cultivation
based on this insight about our primary relation to norms, Confucius does
just this. He focuses on learning as the means by which a child or disciple

Confucius, Wittgenstein, and the Problem of Moral Disagreement

45

becomes responsive and sensitive to dao-constituting norms. Nonetheless,

those philosophers who hold that the project of philosophy must be to
engage in some foundational investigations that would allow us to resolve
fundamental normative disagreements will find this account disappointing.
In the next section, I will discuss this problem, as it shows up in Wittgensteins later philosophy.

Wittgensteins Realistic Spirit

In this section, I will discuss Wittgensteins later project in terms of Cora
Diamonds notion that his later philosophy embodies a realistic spirit
while avoiding realism. In light of that characterization, I will argue that
Wittgensteins later philosophy, embodying as it does the realistic spirit,
rather than being normatively neutral, is in service of the realistic spirits
normative commitments. In the following section, I will argue that Confuciuss teaching embodies a spirit of its own akin to Wittgensteins realistic
spirit. I argue that both are similar in the following important ways: both,
in embodying a spirit, engage in their philosophical projects out of a set
of cares and, in so doing, out of normative commitments: the protection
of common practices, the commerce of ethical life, and strong resistance to
theoretical demands for final formulations of ethical ideals and methods of
justifying them. Finally, I will raise the dilemmas of justification, mentioned
above as related to both approaches to ethics, and will argue that in both
cases, these dilemmas need to be dissolved, not solved, and that each spirit
is in a better position to do that by drawing on resources from the other.
The Realistic Spirit Frees Us from Metaphysical Realism
In this section, I discuss Cora Diamonds account of the way in which
Wittgensteins later philosophy is done in the realistic spirit. But first,
I need to explain in what sense she uses the term realistic and present
the range of contrasts she introduces to capture it. Diamonds notion of
realism is closest to literary realism, which is marked by focus on concrete,
familiar, but often ignored details of everyday life, in opposition to abstract,
imaginative, or artificial representations of life.
In its philosophical embodiment, Wittgensteins realism of the realistic spirit stands opposed to metaphysical realism of a particular sort. The
realistic spirit addresses philosophical problems by looking at human life and
understanding concepts in terms of their commerce in life, the role they

46

Whose Tradition? Which Dao?

play in thinking, asserting, and engaging in activitiesnot in terms of some

abstract philosophical account of those concepts. Wittgenstein understands
the commerce of life and the concepts embedded in it primarily in terms
of the variety of ways these concepts are learned, taught, and otherwise
acquired. The intelligibility of these concepts is then wedded to actual and
imagined learning contexts, which are constitutive of those concepts. What
it means for something to be a real thing of some kind A, then, is related
to the ways in which the concept A is typically learned. Our primary
understanding of these bedrock concepts, then, is made up of the manner
in which we have first learned how to use them, sometimes as children,
sometimes as adults. Philosophical reflection on them is not designed to
justify usage, but to clarify it on the occasion of philosophical puzzlement.
The goal of reflection is to remove puzzlement and thereby aid our being
able to find our way about in the world. Our way of knowing or finding
our way about11 with these concepts comes from those forms of learning
that constitute their content.
Diamond characterizes Wittgensteins realistic spirit as the spirit that
frees us from the laying down of metaphysical requirements, which are
connected to thinking in philosophy that what we want is somehow an
account of how things must be, things that can either justify or correct present linguistic practices. What we really want, as she puts it, lies somewhere
else, namely, in seeing how things are or might be for us.12
Central to the method of the realistic spirit, as Wittgenstein practices
it, is to look [at] and see what we actually do and say.13 So, for example,
in ethical thinking, the antidote to a priori requirements on what ethical
thinking must be like consists of looking at ordinary examples of moral
language, using examples from literature, and so forth. The realistic spirit likens those a priori philosophical requirements on ethics, such as the demand
that there be a single rational principle that grounds all moral judgments,
to forms of fantasy ripe for diagnosis, leading, it hopes, to our liberation
from them. In Diamonds view, Wittgensteins quest is not to eliminate the
hold that norms have on us, but rather, to eliminate the hold of certain
philosophical requirements on what those norms consist of. In her discussions of her engagement in ethics in this realistic spirit, she mentions the
philosophical requirement of an ideal of unbiasedness underlying ethical
argumentation.14 But Diamond says of this requirement that it is the result
of myth about argumentation: The idea that we have not got Thought
unless we can rewrite the insight as argument in some approved form is a
result of mythology of what is accomplished by argument.15 The shift in

Confucius, Wittgenstein, and the Problem of Moral Disagreement

47

our focus should be from what moral life must be like to what it actually
is in everyday life.16
Realism in the sense in which the realistic spirit practices it is,
then, whether in or out of ethics, opposed to forms of philosophical obscurantism, which depend on fantasies of how things must be. To be realistic in
this sense is to oppose appeals to a philosophical fantasy, one that appears
to make sense, but is really nonsense and that makes no difference to us
in any case.17
The Realistic Spirit Protects the Moral Insights of Common Humanity
Diamond develops her account of the realistic spirits approach to ethics to
show how bad ethical theories, with their related philosophical mythologies, cause us to construct stupid, insensitive, or crazy moral arguments,
arguments which are capable of hiding our own genuine ethical insights
from ourselves.18 She offers, as an example of such stupid or insensitive
arguments, the philosophical view that the justification for experimenting
on animals comes from recognizing that we humans have something that
the animals lack: reason. If we hold that rational beings deserve protection, we then can argue, she says, that it would be easier to experiment on
retarded humans than on normal human beings. For retarded humans either
lack intelligence or have less of it than normal human beings, who deserve
protection. This fantasy, then, leads to utter blindness to what common
humanity recognizes.19
The Realistic Spirit Examines Established Usages or Words Placed in the
Context of Our Complicated Form of Life
In response to Onora ONeills criticism20 that Diamond holds the so-called
Wittgensteinian view that all justification is relative to locally accepted
practice,21 Diamond makes the fundamentally important, self-described
impatient response that the realistic spirit opposes constructing any prior
philosophical requirements on what counts as justification. Instead, the
realistic spirit looks to examination of how justification works within our
complicated form of life.22 Diamonds response to ONeill is illuminating:
Justification, in ethics as anywhere else, goes on within lives
we share with others, but what we make count in that life is
not laid down in advance. The force of what we are able to say

48

Whose Tradition? Which Dao?

depends on its relation to the life of the words we use, the place
of those words in our lives, and we may make the words tell by
argument, by image, by poetry, by Socratic re-description, by
aphorism, by Humean irony, by proverbs, by all sorts of old and
new things. And the judgment whether we produce illumination
or obfuscation by doing so, the judgment whether there is truth
in our words or self-deception, is not in general something on
which there will be agreement.23
Crucial to Diamonds response to ONeills criticism are her claims
that (1) what we make count in life need not be laid down in advance,
(2) the force of our words depends on the life of those words in our lives,
and (3) the effectiveness of our words depends on the variety of ways we
make those words work for us in terms of argument, imagery, and so forth.
Because of these features of our thinking about what counts as important,
there is no single way, specified in advance, to reach out to others to persuade them of our ethical standpoint.
This response to ONeill reflects a central feature of the realistic spirits
approach to philosophy; it informs and is informed by the way Diamond
interprets Wittgensteins use of language-games in his philosophical investigations. For even if examination of language-games plays a role in liberating
us from preconceptions about how things must be, his examination of how
words get used is not limited to that. For some uses of words take place
outside of any language-games, and even language-games take place in the
larger context of our complicated human form of life. Diamond captures
this aspect of the realistic spirits approach to philosophical investigation
by talking about her investigation of the commerce of our lives,24 while
Stanley Cavell captures this aspect of Wittgensteins investigation of forms
of life by referring to the whirl of organism.25
The Realistic Spirit Rejects the Philosophical Ideal of
Complete Moral Agreement
Based on the way in which the realistic spirit embraces the commerce of
our lives, it also does not presuppose any prior account of the extent to
which moral agreement is desirable or even possible, according to Diamond.
This is a topic to be investigated by examining moral agreement in the
commerce of our lives. The realistic spirit recognizes that agreement plays
a different role in morality than it does in mathematics. And Diamond

Confucius, Wittgenstein, and the Problem of Moral Disagreement

49

approaches moral dialogue with the acknowledgment that the judgment

whether we produce illumination or obfuscation by doing so [by the ways
we make our words tell]...is not in general something on which there
will be agreement.26 The extent and kinds of agreement there might be in
ethics, as in language-games, generally will not be laid down in advance,
but we need to look at and see to what extent in what ways agreements
exist and function in ethics.
The Realistic Spirit Opposes Normative Ethics
Based on Diamonds characterization of the realistic spirit in ethics, what can
we say about the initial question posed in this chapter: Are Wittgensteins
philosophy and the way in which ethics can be pursued in the realistic
spirit normative or meta-ethical? My first response is that the realistic spirit
will reject this question if the question presents a pre-determined, forced
option between two different philosophical approaches to ethics. The sense
that one must approach ethics in one of these two ways and that each way
must exclude the other is a prior requirement the realistic spirit rejects
because it imposes a prior set of requirements on what ethics must be. As
to the question of how ethics must proceed, the realistic spirit responds that
we must examine what is actually possible in language and life, given the
intelligible alternatives to present practice.
To be fair, this response to the question is perhaps too fast and easy.
Instead, I will turn to the question of how the realistic spirit might, given
its basic commitments, evaluate the enterprises of normative ethics and
meta-ethics. It is clear that Diamond rejects the enterprise of meta-ethics.
In her responses to ONeills criticisms of her approach to ethics, Diamond
provides the following characterization of meta-ethics:
One of our inheritances from the Great Age of Meta-ethics is
that we take to be morally neutral such ideas as that the body
of someones moral thought is improved when his principles are
altered so that they no longer give counterintuitive results in
some case no matter how unlikely (is it permissible to destroy
an acorn that has been treated so that it will become a rational
being in two years time?) or when they are formulated more
clearly so that it can be determined what their consequences are
for such cases, or when his underived principles no longer make
use of morally irrelevant distinctions like that between human

50

Whose Tradition? Which Dao?

beings and animals....Such an ideal of moral thought is far

from being the only one and far from being some sort of obvious
consequence of what morality is. It embodies a particular view
of the relationships between morality and human nature: a view
of which human capacities are characteristically exercised in the
development of someones moral life, and more specifically of
what it is for someone to exercise his capacities as a thinking
being in that development.27
Even though she aims here at a particular inheritance of the Great Age of
Meta-ethics, Diamond rejects the project of meta-ethics as such. One way
to capture the rejection is in terms of the Mythology of Neutrality. The
project of meta-ethics was formed around the idea that one could investigate
moral concepts and language separately from making normative claims, that
is, while remaining ethically neutral. Here we might distinguish between
the extreme view, part of the Great Age, and its more modest inheritance.
Jeffrey Sayre-McCords characterization helps to make this distinction:
The range of issues, puzzles and questions that fall within
meta-ethics purview are consistently abstract. They reflect the
fact that meta-ethics involves an attempt to step back from
particular substantive debates within morality to ask about the
views, assumptions, and commitments that are shared by those
who engage in the debate. By and large, the meta-ethical issues
that emerge as a result of this process of stepping back can be
addressed without taking a particular stand on substantive moral
issues that started the process. In fact, meta-ethics has seemed
to many to offer a crucial neutral background against which
competing moral views need to be seen if they are to be assessed
properly. Some meta-ethicists early in the twentieth century went
so far as to hold that their own work made no substantive moral
assumptions at all and had no practical implications. Whether
any view that is recognizably still a view about the nature and
status of ethics could manage this is dubious. But there is no
doubt that, whatever meta-ethicss substantive assumptions and
practical implications might be, it involves reflecting on the
presuppositions and commitments of those engaging in moral
thought, talk, and practice and so abstracting away from particular moral judgments.28

Confucius, Wittgenstein, and the Problem of Moral Disagreement

51

Even in the more moderate inheritance, the meta-ethicist understands

neutrality in terms of abstraction from particular moral judgments, resulting
in a form of reflection that the realistic spirit also opposes. Abstraction from
particular moral judgments and focus on presuppositions make it tempting to offer up abstract preconceptions as requirements of moral thought
as such and to think that this form of reflection is the only possible one.
Diamond believes this account of morality is too narrow because it excludes
the heart and imagination from a central role in moral development. As a
result, the meta-ethical account of moral neutrality is anything but neutral.
It presupposes a substantive view of human nature and teleology, which it
claims to have avoided.
For similar reasons, the realistic spirit will find itself at odds with
normative ethical theory. To the extent that normative ethical theories aim
(as they traditionally have) at articulating a single a priori principle, in
terms of which all actions must be morally assessed, the realistic spirit will
be inclined to see this normative view as out of touch with the complexities of morality.
The realistic spirits moral reflection opposes both normative and metaethics, however, not only because of their narrow views of human nature, but
also and just as importantly, because of the moral blindness they encourage.
For example, consider Diamonds critique of the forms of ethical reasoning
that stem from a view that all justification requires argument. Her central
reason for rejecting these forms of reasoning is that they lead to stupid,
insensitive, or crazy moral arguments, arguments that are capable of hiding
our own genuine ethical insights from ourselves.29 Diamonds assessment
that an argument is stupid, insensitive, or crazy arises from the vantage point
of what common humanity recognizes. The realistic spirit acknowledges
what common humanity recognizes prior to and as a basis for critiquing
systematic philosophical theory, whether normative or meta-ethical.
We might then think of the realistic spirit as countering the philosophical demand for a priori, clear requirements that moral thinking must
take place and be assessed within theoretically deductive terms. In contrast,
the realistic spirit embraces the complexity of morality as lived and as
imaginable and, in doing so, seeks to protect common human recognition of basic moral insights, attitudes, styles of writing and reflection,
and practices. Even though the realistic spirit rejects ethical theory, part
of its reason for doing so rests on caring about what common humanity
recognizes. So even if the realistic spirit is not engaged in constructing
normative ethical theories, it is engaged in philosophical investigations that

52

Whose Tradition? Which Dao?

arise out of its own normative commitments. Even though the realistic
spirits project does not rise to a full-blown project of moral self-cultivation,
it does offer a therapeutic project, focused on rooting out unhealthy forms
of understanding and promoting healthy forms of understanding that are
suited to protect the variety of things and actions that goes by the term
morality.30

Confuciuss Realistic Spirit

Although not a realistic spirit in Diamonds terms, Confucius, as he speaks
and acts in the Analects, appears to be a kindred spirit to the realistic spirit.
I know of no good name for this spirit, so I will provisionally call it the
Confucian realistic spirit. This spirit, like Wittgensteins realistic spirit,
is understandable in terms of what it opposes and what it embraces. The
Confucian realistic spirit is partly embodied in its opposition to the
(xiao ren), the small-minded person looking toward profit in his relations
and conduct toward others.31 In contrast, the person who embodies the
ideal of (ren) focuses his attention on the forms of respectfulness and
reverence embodied in ritual propriety, those ritual practices spanning
the continuum from formal government ceremonies to informal, personto-person interactions. These forms of propriety vary with context and a
persons relationships with the people he encounters. The person who is
(ren) also focuses on not doing to others what he would not want
done to himself.32
Confucian Reflection Is Open-Ended
Even though these practices form a key focus of Confuciuss considerations, the Analects shows an open-ended approach to the character of
these concerns. So although (ren), understood as the sum of all of the
other virtues, seems at some points to be the sole ideal for self-cultivation,
dao seems ultimately to be the sum of all norms for a human being: to
live well is to live according to the way, and using a literal translation
of dao, to live a life. This leaves room for (ren) to be supplemented
by other ideals. For example, in some passages of the Analects, Confucius
discusses (yi), justice or righteousness. Confucius shows no effort to
define dao; instead, he illuminates those aspects of dao relevant to concerns
of the moment.33

Confucius, Wittgenstein, and the Problem of Moral Disagreement

53

Confucian Reflection Serves Needs for Self-Cultivation

Crucial to Confuciuss concern with (ren), however, is his focus on selfcultivation. When questions about (ren) are raised, they are typically
answered in terms of how the interlocutor should change his behavior to
become closer to the ideal of (ren). The questioners focus, as that of
Confucius, is practical, not theoretical. So for the most part, clarification
of (ren) is clarification by Confucius for a particular person to help him
take the next step on the path of becoming (ren) in his daily activities.
Confucian Reflection Avoids Speculation
The Confucian realistic spirit also avoids investigations into human nature
or the Heavenly Way. Consider the following passage:
. . . .
.
Zigong said: The cultural ornamentations (speech and conduct)
of the Master: they can be heard. As for our Masters words on
nature and the heavenly way, we cannot hear them.34
Confucius concerns himself instead with dao, concrete norms governing human life, specifiable on their own, separate from any theory of
human nature or of the fundamental character of the world. We also hear
that Confucius seldom spoke of the Mandate of Heaven, and so references
to what is required of us by Heaven played little role in his project, despite
the role of this concept in later thought.
One passage in the Analects even says that Confucius did not discuss
(ren):
, , .
Our Master seldom spoke of benefit, and mandate, and ren.35
This claim, however, seems to fly in the face of the fact that (ren)
is often discussed in the Analects. I take this demurral to refer to the fact
that he did not discuss the nature of (ren) in the abstract; instead,
he discussed with various individuals how, given their circumstances and

54

Whose Tradition? Which Dao?

tendencies of character, to become (ren). These characterizations of his

mode of investigation indicate a refusal to speculate about abstruse, so-called
foundational bases of or abstract definitions of fundamental norms governing human life. The Confucian realistic spirit instead focuses primarily on
learning (xue) of concrete practices and reflection on that learning.
Confucian Study Focuses on Collected Exemplars from History and Literature
The Confucian realistic spirit opposes reflection on dao independent of
learning (xue), which, although referring primarily to the study of ritual,
is not limited to that and refers as well to a broad range of studies, including
the six arts: ritual, music, archery, chariot-riding, calligraphy, and computation. All of these practices serve the development of a cultivated sensibility
designed to allow a person to live in conformity with dao, but these were
probably no more important than the study of The Book of Songs (the
canonical poetry) and The Book of Documents, a repository of moral lessons.
Confucius was also interested in history, especially the collection of narratives of exemplary conduct. However, as indicated above, for Confucius,
these forms of learning must be wedded to reflection.
Even though Confucius often emphasizes ritual as central to selfcultivation, it is an important feature of his view of self-cultivation that it
not be limited to studying ritual. The list of other forms of study provides
us with a wider set of subject matter that makes it possible to examine and
reflect on other aspects of dao, the norms governing human life.
Consider the following passages:
. . . .
An exemplary person should broadly learn the documents and
should restrain himself in accordance with ritual actions. If so,
he indeed wont deviate from the Way.36
.. . .
To study broadly, to preserve aspirations strongly, to ask questions to the point, to think about things from whats near, the
possibility of complying with ren lies in the midst of these.37
As is often the case in Analects, the answer to our question is that
the writers do not intend to offer a complete answer to the topic at hand.

Confucius, Wittgenstein, and the Problem of Moral Disagreement

55

Thus there is no reason to think that Confucius would have held that these
forms of learning encompass the whole of dao. Given his focus on broad
learning and his reluctance to define ideals, it would make sense that he
would leave the self-cultivation syllabus open-ended.
The Confucian realistic spirit has as one of its primary activities the
collection of and reflection on exemplars of wise judgment, action, and
attitude that can serve as guidance for later generations of individuals seeking to live in conformity with dao. In the Analects, Confucius and his
interlocutors discuss the exemplary conduct of a varied cast of characters,
some of whom embody the ideal of (ren) and some of whom are good
in some respects but are not obviously fulfilling the ideal of (ren). This
collection of exemplars forms a set of moral precedents; even if they need
to be interpreted, the exemplars establish a basis upon which people can
learn how to orient themselves to others and make reasonable judgments
and decisions. In large part, the text of the Analects itself is a collection of
exemplary discussions of problems associated with moral self-cultivation.
Those discussions function for later thinkers in much the same way that
The Book of Documents and The Book of Songs functioned for Confucius:
to guide reflection and action by appeal to exemplars. In his case, however,
the exemplars are not only the historical examples he mentions but also
include the variety of ways he himself intervenes to get his interlocutors to
change their conduct and reflect better about dao.
Confucian Non-Reductionist View of Dao
Connected with this broad canon of study, the Confucian realistic spirit
seeks to avoid one-sidedness in its ideals and conduct. This ideal is best
encoded in Confuciuss critique of various forms of so-called spiritual purity, which rest on absolute approval and disapproval of some limited forms
of behavior, which, though good in some aspects, do not encompass the
whole of good behavior. As he says in the Analects of the narrow moral
purists:
. .
I am different from them. I have no absolutely certain approval
or disapproval.38
And of those who associate themselves with one party or another to a
dispute, he says:

56

Whose Tradition? Which Dao?

. . . . .
An exemplary persons attitude toward the people and events in
the whole empire is without biased affirmation or biased rejection. Doing the appropriate thing is what he follows.39
Both of these passages reflect Confuciuss awareness of the problem of onesidedness in living and reflecting on what is important. It is all too tempting
to identify an ideal and then mistakenly take it to be the whole of dao.
I have argued that Confuciuss commitment to a type of practical, embodied realism about dao and its cultivation shares similarities
with Wittgensteins realistic spirit. But it should be possible to make even
clearer what it means to think of both as embodying a spirit. I turn to
this question next.

Expressing a Realistic Spirit, Fending off a Kantian Challenge

Although I have indicated what makes Confuciuss and Wittgensteins realistic spirit realistic, I have not said much about what makes it a spirit.
Diamond says little about this, but the notion is important.40 Being realistic in this sense is opposed to metaphysical realism of the sort that puts
forth requirements in advance about how things must be, but embodying a
spirit is opposed to being identified with specific theories or sets of theses.
What is meant by the term spirit? To characterize the spirit of a person is
to characterize that persons package of basic attitudes toward life, methods
used addressing reflective questions about life, style of speaking and thinking, and general sets of value commitments. In the context of philosophy,
a philosophers spirit would be characterized by how he addresses reflective
questions. A philosophers basic orientation would not be reducible to set of
philosophical theses, defended by argument, analysis, or theory, but would
reflect the language of the attitudes and methods he uses to understand and
address the reflective problems he faces. For example, Wittgenstein is suspicious of thinking that gives rise to conceptual puzzlement. He thinks that
conceptual puzzles are to be diagnosed rather than resolved. His approach
is not best expressed in terms of some philosophical theory about puzzles;
instead, his approach is best expressed in terms of an attitude toward ordinary language and life. Ordinary language and life are to be trusted and
made intelligible in terms of their variety and complexities. For these reasons, Wittgenstein is also suspicious of abstract, a priori philosophizing with

Confucius, Wittgenstein, and the Problem of Moral Disagreement

57

its tendency to seek overly simple approaches to reflective questions in the

service of a priori philosophical requirements.41
Adopting another way to clarify what is at stake in characterizing
the spirit of these thinkers, I borrow from Harry Frankfurts discussions42
of what we care about: a persons spirit, realistic or not, is really a set of
commitments, things cared about. If a person cares about something, it
is important to him.43 A person who cares about something is guided by
what he cares about: What he cares about establishes what he is willing and
unwilling to do. Moreover, caring involves both action and self-reflection.
For the person who cares about something [identifies] with what he cares
about in the sense that he makes himself vulnerable to losses and susceptible
to benefits depending on whether what he cares about is diminished or
enhanced. Thus he concerns himself with what concerns it, giving particular
attention to such things and directing his behavior accordingly.44
For example, someone who cares about abandoned pets cares about
how such pets fare and sees his own well-being as wrapped up with the
well-being of abandoned pets. He concerns himself with what benefits them
and directs his attention to benefiting abandoned pets. But a person could
also care about the sorts of things that Wittgenstein cares about: coming
into agreement with forms of life.45
In light of Frankfurts analysis, I offer the following clarification of the
realistic spirits spirit. The realistic spirits spirit consists of its distinctive set
of cares. In particular, the realistic spirit cares about and takes ordinary life
and language and the forms of guidance they provide us to be important in
the context of philosophical thinking. Much of Wittgensteins later philosophy offers examples of philosophical puzzlement that arise from not paying
attention to the variety of ways we operate with concepts in ordinary life
and language. As such, the realistic spirit senses a loss in those philosophical contexts, where what she cares about, the forms of guidance embedded
in ordinary life and language, are diminished by being treated as less than
ideal, as based on prejudice, or as not living up the ideals of logical clarity
and rigor.46 She takes guidance from ordinary language and life in reflecting
on philosophical views, which when detached from common life diminish
what she cares about. And because this care is important to her, she cannot
adopt and endorse a range of philosophical views and styles of thinking
that, if accepted, would diminish her and others appreciation of the role
of ordinary language and life in the human form of life. But these cares are
not themselves a set of philosophical theses. Caring to understand morality
in a way that reflects the complexities of our ordinary ways of talking about
moral problems, resolving them, reaching out to others, and so on is not

58

Whose Tradition? Which Dao?

the same as presenting a theory about ethics. As Wittgenstein would say,

this way of understanding morality earmarks the form of account47 the
realistic spirit cares to develop.
The fact that this form of realism, its form as spirit, is not a set of
philosophical theses is crucial in explaining why, on the one hand, it might
be easy to think, as Diamond says ONeill does, that Wittgensteinian ethicists presuppose their own set of prior ideals about what ethics must be,
but, on the other hand, the ethicists may be wrong about this. ONeill
criticizes Diamond and other Wittgensteinian ethicists as simply imposing
a different Wittgensteinian set of requirements on what justification is
in ethics. Diamond rejects this characterization of her own views, especially
because, as she says, The aim of philosophy...is liberation from laying
down this or that requirement. She adds that ONeills interpretations of
philosophy impose on philosophical writing the requirement that it say
Regardless of what things look like, if we are to have or to do such-andsuch, there must be so-and-so.48
Even if ONeills interpretation of Diamonds view is incorrect, as it
certainly seems to be, one can find some sympathy with it. After all, philosophy done in the realistic spirit is not done in a void, even though it
does not approach philosophy with a prior set of requirements in the form
of theses. The realistic spirit brings its own set of substantive cares to its
investigations and its own view of what an account of ethics should be. And
these aims, while not identifiable with a prior set of philosophical theses,
do, as Diamond seems to acknowledge, establish a kind of requirement for
what the realistic spirits adequate philosophical account of ethics would be.
It would be an account that does not cause us to call into question, based
on prior philosophical requirements, ordinary ways of talking about ethics
and of reaching out to others in situations of moral disagreement.
In ethics, the realistic spirit, understood in the way Diamond does,
confronts a problem. For although realistic spirits challenge various philosophical ideals of truth and justification put up in advance of investigation
of the role concepts play in our life, how realistic spirits can address problems of moral disagreement is not clear. Arguably, the problem of managing
disagreement increases once realistic spirits forgo appealing to those a priori
philosophical requirements often used by philosophers to resolve ethical
disagreements. Realistic spirits attempt to resolve disagreement through discussion, but they also acknowledge that disagreement in ethical judgments
is part of life and language in a way it is not in other terrains, such as
mathematics. In light of these considerations, it is possible to generate the
following dilemma for a realistic spirit:

Confucius, Wittgenstein, and the Problem of Moral Disagreement

59

Either the realistic spiritin Confuciuss or Wittgensteins versionsadopts some a priori moral principle that shows how to
resolve moral disagreement, or it does not. If the realistic spirit
adopts such a principle, then the spirit contradicts its aim of
liberation from all such requirements. If the realistic spirit does
not adopt such a principle, it has no adequate way (according to
traditional philosophical theory) to address moral disagreement.49
From the vantage point of traditional philosophical theory, it can
appear that the correct way to diagnose the problem posed in this dilemma
is in terms of the way the realistic spirit forgoes any appeal to philosophical
ideals of justification set up in advance of encountering moral disagreement.
Once a realistic spirit forgoes appeals to such theoretical ideals, so the diagnosis goes, it lacks resources to know how to address disagreement. But this
diagnosis supposes that the only way to address disagreement is by appeal
to those very sorts of philosophical ideals that the realistic spirit eschews.
I offer an alternative: the problem for the realistic spirit of managing
disagreement can be resolved by support from the resources of a Confucianstyle tradition. For Confucius has as one of his primary activities, and so
cares about, the collection of and reflection on exemplars of wise judgment,
action, and attitude that can serve as guidance for later generations of
individuals seeking to live in conformity with dao. And this very collection
can help aid in the resolution of disagreements, not by establishing clear
and distinct a priori criteria of right and wrong, but rather by creating
a repository of different types of moral exemplars. Those who choose to
emulate and reflect on such exemplars may, in so doing, learn to produce
just those sorts of modes of judgment and habits of behavior that could
improve their chances of resolving moral disagreements.50
Of course, Wittgenstein was aware of such a possibility. In Rush
Rheess report of his discussion of ethics with Wittgenstein in 1942, Rhees
describes Wittgensteins attempt to find an example of an ethical problem.
What emerges is interesting for this discussion as they end up talking about
different types of ethical problems. One problemnot the only typeis the
problem that arises when a person who is outside of any moral tradition
considers various ways to characterize and prioritize alternative actions. A
man thinks he cannot continue to devote himself to his full-time cancer
research and remain married. What follow are various forms of reasoning he
and his friend who is counseling him may engage in. We might call these
forms of reasoning for short various considerations. One consideration
is that he took his wife out of her previous life to be married to him,

60

Whose Tradition? Which Dao?

so he cannot abandon her. Another is that he is able to and should help

humanity with his research. Another is that he loves her but would not be
a good husband to her if he were to give up his lifework. Another is that
she wouldnt take abandonment or divorce so hard after all. These considerations can be contested, but each one forms the core of a justification for
one alternative or another. However, there is no way to see which choice is
correct: Here we may say that we have all the materials of a tragedy; and
we could only say: Well, God help you.51 Although Wittgenstein does
not say what these materials of tragedy consist of in this discussion, it
is not hard to find a candidate: in such cases, there does not seem to be
a correct answer. A person who lives outside any moral tradition, as the
example states, can adopt various moral stances, and whichever stance he
adopts, the other stance remains a live option, something he also could
adopt. The ethical problem in this case is the problem of which stance to
adopt and how to maintain the stance in the face of its arbitrariness and
instability. We might say that the ethical problem here is the problem of
settling on a stance and finding ways of carrying on with it that the person
can maintain without being overwhelmed by the live alternative options and
characterizations latent in the situation.
This example seems to support the realistic spirits view that moral disagreement is a permanent possibility. For just as different options are viable
for a single agent, different agents will be to able take different approaches
to this problem, and there is no one rational way to resolve it. However,
Wittgenstein does not rest with a single example. He also discusses the
case of a Christian who faces this choice. Here, he says, it might be clear
that he must stay with his wife. The problem for him is how to make the
best of this situation.52 Other ethics, he says, like Nietzsches, would give
different answers. But, he argues, there is no sense to be made of the question of which ethical system is right. Here is why: But we do not know
what this decision would be likehow it would be determined, what sort
of criteria would be used, and so on. Compare saying that it must be possible to decide which of two standards of accuracy is the right one. We do
not even know what a person who asks this question is after.53 The point
here, developed by Wittgenstein in later discussions with Rhees, is that
without any possibility of a clear criterion of right or wrong being applied
to ethical systems from the outsideall criteria are internal to the systems
they find a home inthere is no sense to be given to this question. To
say that Christian ethics are right, instead, is simply to adopt that ethical
system, not to say something about its correctness from the outside, as it
were. Wittgenstein puts this point this way:

Confucius, Wittgenstein, and the Problem of Moral Disagreement

61

Or suppose someone says, One of the ethical systems must be

the right oneor nearer to the right one. Well, suppose I say
Christian ethics is the right one. Then I am making a judgment
of value. It amounts to adopting Christian ethics. It is not like
saying that one of these physical theories must be the right one.
The way in which some reality corresponds or conflicts with a
physical theory has no counterpart here.54
This being the case, it looks like we are left with no way to think about how
to resolve moral disagreement. If it makes no sense that to say one system
is right, then it would seem that anything goes. But despite this appearance,
Wittgenstein also rejects as meaningless the claim that all such systems are
equally right. That is, he completely excludes these deeper meta-questions
from the language. Nevertheless, Wittgenstein claims in these conversations
that there will be a common core even to different ethical systems:
There is no one system in which you can study in its purity and
its essence what ethics is. We use the term ethics for a variety
of systems, and for philosophy this variety is important. Obviously different ethical systems have points in common. There
must be grounds for saying that people who follow a particular
system are making ethical judgments: that they regard this or that
as good, and so forth. But it does not follow that what those
people say must be an expression of something more ultimate.55
Wittgensteins reluctance to look for the essential core of ethics is consistent with his thinking that there is a commonality of some sort between
ethical systems and some shared criteria we call ethical. It would appear
that even though moral agreement is not grounded in some ultimate essence,
there is a need for common characteristics in terms of which we call a system
ethical. These features might be enough to manage moral disagreement
even if they did not provide an a priori principle to resolve all disagreement.
There is a possible problem with this suggestion, however. For it seems,
sometimes at least, that the realistic spirit places a priority, as a kind of
natural historian, on observation of human life, unlike Confucius, detached
from membership in any community. If the realistic spirit were to take this
approach to moral disagreement, it would, however, have to drop its stance
as an observer of the commerce of life and, instead, would have to take more
seriously the idea that the collection of moral exemplars is useless without
emulation of them. That is, if a realistic spirit were to opt for a Confucian

62

Whose Tradition? Which Dao?

or other moral cultivation-style approach to moral disagreement to solve its

own problem of disagreement, the spirit would have to face squarely the
need to think of its project, more than it otherwise would, as involving
moral self-cultivation, not just assembling reminders about and offering
clarifications of the use of concepts in our lives with words and concepts.
But perhaps this would not be so difficult since the realistic spirit operates
out of a set of cares, and even moral commitments, that might very well
make such forms of self-cultivation both desirable and natural.

Fending off a New Wittgensteinian Challenge

A possible objection to my understanding of a later Wittgensteinian ethics would claim that it distorts a key ethical commitment that pervades
Wittgensteins later philosophy. In a section of his essay, Ethics in Light
of Wittgenstein, Stephen Mulhall argues that the Wittgensteinian ethicists acknowledgment of the inescapability of moral disagreement rests on
Wittgensteins deeper analysis of the way in which application of words to
novel contexts can be carried forward in a variety of ways and his view
that rules do not by themselves determine a single way of carrying on that
is uniquely correct:
But this attempt reveals that the criteria by reference to which
we judge that resemblance [between a past accepted exemplar
and future instance of the same kind] are multiple and complex, that the significance of any given criterion being satisfied
or not in any given case is contestable, and so that the degree
of resemblance needed to ground the projection of the concept,
of a language into this (or any) context is importantly open to
individual judgement....Our grasp of the criteria for [a] term
offers us guidance, and determines what kinds of consideration
will be pertinent to the question at stake; but the responsibility
for judging whether or not the absence of certain criteria, or
the presence of what are at best analogues or primitive versions
of them, are enough for us to withhold the word in this case
ultimately rest with the individuals invited to project those
criteria into this imagined context.56
For Mulhall, this feature of Wittgensteins approach to criteria for
applying concepts, which he calls reflective, including moral concepts, shows
the pervasive ethical dimension to Wittgensteins philosophical investigations:

Confucius, Wittgenstein, and the Problem of Moral Disagreement

63

On this reflective rather than determinant understanding of criteria and grammar, there simply is no space for the straightforward
assumption of impersonal linguistic authority....If one finds
this [reflexive] model of philosophical interlocution more true
to Wittgensteins practice than its determinant alternative, then
it will be plain that any and every Wittgensteinian philosophical
exercise will place rigorous ethical demands upon its practitioners.
For it will require them to acknowledge and respect the otherness of ones otherher equal claim to authority over how to
go on with words, her claim to be taken seriously as someone
who is attempting to say something meaningful, her right to be
the final arbiter of ones claim to represent a better way of going
on, to say yes or no to ones offer of communityin everything
one says and does.57
If Mulhalls analysis is correct, then he has provided one more way to
make sense out of the pervasiveness in philosophical investigations of Wittgensteins ethical commitments.58 This commitment would be to engage in
philosophical investigation with the full recognition that ones interlocutor
has the right to be the final arbiter of any claim to represent the better way
of going on in applying concepts and making judgments.
Not only will Mulhalls analysis provide another way, implicit in the
realistic spirits approach to philosophy, to explain the pervasive ethical
character of his philosophical investigations, it also basically challenges my
argument that the realistic spirit needs to find some way to manage irresolvable disagreement, to know when it is tolerable and when it is not.59 Mulhalls account also challenges my account of the fundamental similarities of
Confuciuss project to Wittgensteins. Mulhalls Wittgenstein is aligned with
liberal individualism in a way that Confuciuss appeal to traditional, cultural
authority is not.60 But perhaps these problems do not matter so much, for,
as I will now argue, Mulhalls representation of the moral implications of
Wittgensteins view of rule-following in other ways violates the realistic
spirits approach to ethics. Rather than spelling out the implications of the
realistic spirits approach to ethics, Mulhalls view contradicts it.
The realistic spirits primary commitment in philosophical investigations is to look and see and to ensure that it not draw premature conclusions
based on a narrow range of examples or on some prior view of how things
must be. But Mulhall generalizes from a single example in Wittgenstein,
the example of the deviant pupil who persists in applying a rule in a way
different from the way the teacher accepts as correct. In this example, the
student is asked to apply the formula X + 2. The student successfully does

64

Whose Tradition? Which Dao?

so up until he reaches 1,000 and then continues with 1004, 1008, and so
forth.61 Following Cavell, Mulhall argues that Wittgenstein uses this example
to show that the teacher has no authority over the student. If the student
proceeds differently, this shows that he rejects the authority of the teacher,
both Cavell and Mulhall maintain.62
According to Cavell and Mulhall, Wittgensteins example is meant to
show that disagreement about how to go on is, in principle, always irresolvable and that the teacher cannot ever be said to represent the correct
application. In the face of disagreement, all the teacher can do is invite the
student to accept her approach. But it is never wrong if the student refuses
to do so. He has the absolute right to make his own decision.
But what ever happened to the realistic spirits appreciation of this
complicated form of life? We can imagine a continuum of cases, including
one in which the teacher demands and expects absolute conformity to his
teaching and does so as a representative of a moral community whom he
represents as teacher. These examples are a dime a dozen. The fact that a
deviant student might find it reasonable to go on in a nonstandard way that
he can explain and insists on doing does not mean that the teacher must
accept his approach or grant him any right to determine what the correct
way is. The opposite end of the continuum is the one that Mulhall discusses.
For surely what Wittgenstein wants to show by the example of deviancy is that such an example is possible and that the logic of a formula
presented outside of the context of concrete bedrock practices does not
require going on in any particular way, not that all examples of teaching
follow this model. So in this example, perhaps, we have an example where
it makes sense to say that the student takes himself to have an absolute right
to accept or refuse the authority of the teacher. But if so, this right does
not arise, I would argue, solely from the fact that the prior applications of
a rule do not by themselves determine future applications. Instead, this right
emerges, if at all, from some prior ethical commitments. In the context of
a liberal culture that resists the authority of teachers and tradition, it may
make good sense, at least in some contexts, for the teacher and student
both to agree that the student has a right to go on as he sees fit. (I would,
however, not agree that even in a liberal culture, simple arithmetic would
be such a context.)
It is, moreover, not likely that Wittgenstein takes his example to show
what Mulhall thinks it shows. Indeed, when Wittgenstein raises the question
of the criteria for applying concepts in this sentence: The way the formula
is meant determines which steps are to be taken, he indicates that it is
the kind of way we always use it, were taught to use it.63 If the way we
are taught to use it, through the master-novice relationship (see Chapter 1

Confucius, Wittgenstein, and the Problem of Moral Disagreement

65

for my discussion of this relationship), constitutes the norms we are under,

then the student is meant to go on as a novice to practice what the master
teaches him, even if the student is inclined to go on in some other way.
If I am right to argue that there is a continuum of contexts for such
an example, and the liberal context Mulhall imagines might result in a
teacher recognizing her lack of authority in the face of a student who is
inclined to carry out the formula differently from the teacher, we would
end up with a counterintuitive result that the student can construct any
kind of basic arithmetic he would want. And even if we were to allow this
possibility, clearly Wittgenstein could not accept a view that would treat
this result as the only possible one.
But if you place this teaching situation in the context of a tradition
bound by precedents and established forms of teaching authority, it makes
perfectly good sense for the teacher to respond to the student as deviant in
his interpretation. As Wittgensteins argument about rule-following shows,
however, from a logical point of view, going on with the rule might proceed
in different ways. But this argument does not show that, as a matter of fact,
it will go on. Nor does it show that if there is agreement, it is based on
the students freedom to act as arbiter of the correct interpretation of the
rule. For in what we might call a culture of authority, such freedom is
not allowed. And the meaning of terms like teacher, right interpretation,
and deviant are determined by the role these words play in this linguistic
context. Whats more, in response to Mulhalls characterization of the students right to be the final arbiter in every context of interpretation, the
realistic spirit would be required to offer a culture-of-authority example to
undermine his dogmatic approach to the question of who in these teaching
situations must be the arbiter of disagreement.
To take Mulhalls approach to this example and use of it to defend his
general, liberal principle of the individuals right to be the final arbiter of
judgments on how to apply a rule in a novel situation would be to claim that
there is a single way to handle moral disagreements. This way would treat
novel applications of principles as arbitrary if other applications are logically
possible. But Mulhall nowhere shows that the word arbitrary functions
solely in this way, and it would be strange to think that Wittgenstein or any
other realistic spirit would think that it did. In fact, beyond this problem
of using one example to generate a universal claim, Mulhalls claim about
the student being the final arbiter of how to go on with his calculation flies
in the face of what Wittgenstein says against private language:
Let us imagine the following case. I want to keep a diary about
the recurrence of a certain sensation. To this end I associate it

66

Whose Tradition? Which Dao?

with the sign S and write this sign in a calendar for every day
on which I have the sensation.I first want to observe that a
definition of the sign cannot be formulated.But still I can give
one to myself as a kind of ostensive definition.How? Can I
point to the sensation? Not in the ordinary sense. But I speak,
or write the sign down, and at the same time I concentrate
my attention on the sensationand so, as it were, point to
it inwardly.But what is this ceremony for? For that is all it
seems to be! A definition serves to lay down the meaning of a
sign doesnt it?Well, that is done precisely by the concentrating of my attention; for in this way I commit to memory the
connection between the sign and the sensation.But I commit
to memory can only mean: this process brings it about that
I remember the connection correctly in the future. But in the
present case, I have no criterion of correctness. One would like
to say: whatever is going to seem correct to me is correct. And
that only means that here we cant talk about correct.64
But, as Wittgenstein says, in the present case I have no criterion of
correctness. One would like to say: whatever is going to seem right to me
is right. And that only means that here we cant talk about right. But
the attitude of whatever seeming right to me is right is just the attitude
of a person who takes himself as final arbiter.65 What Mulhall rules in,
Wittgenstein rules out.
Perhaps the cautionary conclusion that we must come to in light of
the problem of Mulhalls account is that we should proceed with care in
articulating Wittgensteins approach to moral disagreement. We should also
come to see that how deviancy will be handled and described very much
depends on the context in which the teaching takes place. There is no single
possible approach.
I have argued, however, that whatever pressure might be felt by the
charge that the realistic spirits approach to ethics results in irresolvable disagreement can be managed. The realistic spirit can, if it sees fit, find ways to
address disagreement within the context of the basic insights it shares with
early Confucianism. For as I argued in Chapter 1, the learning of bedrock
practices and basic concepts embedded in them does not make possible a
novices questioning the masters authority. Moreover, even when a person
weighs imponderable evidence, some people are able to do that better
than others and are in a position to teach others to replicate their success.

Confucius, Wittgenstein, and the Problem of Moral Disagreement

67

But even these resources aside, the Confucian tradition offers emulation of
examples as a further resource for resolving moral disagreement without
appeal to accounts of a priori ethical principles.
The Confucian tradition collects exemplars of moral conduct, along
with proverbs, exemplary dialogues, practices of memorization of canonical
passages, exemplary modes of teaching, and handing down of interpretations within the context of established lineage. The tradition also encourages children to accept the authority of parents and teachers as one way
to deal with moral disagreement without feeling compelled to contradict
itself by accepting some other single basic principle of ethics, articulated
in advance, as the only possible way to distinguish between tolerable and
intolerable disagreement.
I have not, however, argued that the realistic spirit must become Confucian, though I would insist that the realistic spirit would benefit from
an examination of the ways exemplars are collected and interpreted within
a tradition and how they might impact a persons understanding of how
to apply a rule and transmit norms. This might be part of the larger form
of life, demonstrating how moral concepts can get their meaning and how
disagreements can be dealt with. Certainly, the realistic spirit can approach
the problems of ethics in this more or less Confucian way without pain of
self-contradiction. Indeed, it is a way for the realistic spirit to give procedural
content to its commitment to protect what common humanity recognizes
without establishing prior idealized principles of justification. And in following this path, the realistic spirit need not assume in advance how much
agreement such a tradition will generate. It need only rest assured that it
has an approach to moral disagreement consistent with its own spirit, one
that avoids a dogmatic appeal to philosophical accounts of justification
adopted in advance of investigation and without falling into a Mulhall-type
dogmatic moral individualism.
This argument offers a defense of the realistic spirit, as it shows up in
Wittgensteins later philosophy, against the charge that ethics done in this
spirit cannot manage moral disagreement. If my argument about the fundamental similarities between Confuciuss realistic spirit and Wittgensteins
holds, both projects have resources to manage moral disagreement and can
do so without appeal to foundational ethical theories. It will be my burden
in the rest of this book to argue in more detail how, by utilizing forms of
critique central to Wittgensteins realistic spirit in ethics, it is possible to
resolve fundamental outstanding interpretive and evaluative debates over
Confuciuss self-cultivation and a Wittgensteinian reflective project in ethics.

68

Whose Tradition? Which Dao?

Any mode of inquiry into Confuciuss approach to self-cultivation

presupposes, however, that we can, in principle, understand the meaning of
his utterances as reported in the Analects. In the next chapter, I address the
question of how that is possible by bringing to bear on this problem resources Wittgenstein provides for thinking about meaning and understanding.

Confucius, History, and the Problem of Meaning

Nietzsches texts therefore do not describe, but in exquisitely elaborate

detail, exemplify the perfect instance of his ideal character. And this
character is none other than the character these very texts constitute:
Nietzsche himself.
Alexander Nehamas, Nietzsche: Life as Literature

Overview
In this chapter, I take up a network of connected problems having to do
with how to read and understand the Analects in light of history. Certain
models of meaning for texts, such as the Analects that naturally suggest
themselves to interpreters, create puzzles about whether the Analects sentences (sayings) have any meaning at all and, if they do have meaning,
whether we can know them.
Two recent discussions of meaning show the problem we face when
thinking about the Analects meaning: Following the model of reader
response theories of meaning, Daniel Gardner argues that there is no meaning to the sentences of the Analects, just various attributions of meaning
made by commentators over time. He would say that there is no single
meaning for any of the sentences of the Analects.1 We can refer to this view
of meaning as semantic nihilism: The Analects has no real meaning of its own.
A somewhat less extreme approach to the meaning of the Analects can
be found in John Makehams recent work on the Analects commentarial
tradition.2 Makeham draws a distinction between historical and scriptural
meaning, which he uses in defense of his semantic skepticism: We cannot
know the meanings of the sentences of the Analects. Historical meaning

69

70

Whose Tradition? Which Dao?

is what the Analects meant to its authors and audience at the time of its
writing. Scriptural meaning is what the text has meant to subsequent
readers who, one way or another through creative interpretations, believe
that they have found their own view of the Truth, writ large, revealed in
the Analects. Given the limitation of knowledge of the language and history
of the time in which the Analects was written, Makeham despairs of capturing the historical meaning of the Analects but worries that the adoption of
scriptural meanings or the acceptance that the sentences can be interpreted
variously will lead to an unlimited number of meanings, personal to each
readers view of the Truth.3
It might seem natural to conclude from this account of two different
types of meaning that the more we focus on the history of the Analects,
the less confident we can be that there are any overall meanings or knowable meanings of the sentences of the Analects. A recent work by E. Bruce
Brooks and A. Taeko Brooks, which offers a detailed account of when the
various passages of the Analects were written, shows what it would take
to find specific historical cues that would guide our dating of the text.4
Although Brooks and Brooks believe they have made a reasonable case for
the dating that they have offered, others have found their efforts speculative
at best.5 The difficulty of dating passages and using that dating to defend
context-sensitive interpretations plays well into the hands of interpreters,
like Makeham, who have found no knowable meanings whatsoever in the
Analects. I will, however, argue that Gardners and Makehams arguments
should be rejected. Here is why: There are two necessary but distinct sources
that give rise to their conclusions. The first is the indisputable set of facts
that the historical evidence we have about the meanings of the sentences in
the Analects is limited, and this limited evidence sometimes leaves us in a
situation of not being sure how to decide between possible interpretations.
As a result, different interpreters have offered different interpretations of the
Analects. But these facts of the indeterminacy of meaning and multiplicity of interpretations by themselves do not support Makehams semantic
skepticism or Gardners semantic nihilism. Only when they combine these
facts with their accounts of meaning are they able to support their conclusions. I will argue that their accounts of meaning are problematic, thereby
making their arguments for semantic nihilism and semantic skepticism
inconclusive.
They adopt the following claims about meaning:

1. The meaning of a sentence in the Analects is the speakers

(authors) intention (but these are not accessible to us at a
temporal, cultural, and linguistic distance).

Confucius, History, and the Problem of Meaning

2. The meaning of a sentence in the Analects is the readers

response to a sentence (but these change over time).

3. There are two distinct types of meaning: historical (which

we can never know) and normative (scriptural, existential)
(which is subjective and varies from person to person).

71

My argument proceeds as follows:

In the first section, I provide two of Makehams examples of the
indeterminacy of meanings of sentences in the Analects. I defend Makehams
claim that the difficulty of our determining the meanings of these sentences
rests on our ignorance of the historical context of the speaking or writing of
these sentences. I present two possible implications of this ignorance: the
sentences of the Analects (1) have no meanings of their own, or (2) have
meanings that are unknowable.
In the second section, I examine Gardners nihilistic view that the
sentences of the Analects lack any meanings of their own. I critique the relativistic basis for his claim and clarify different versions of the crucial distinction between historical and normative meaning that his account depends on.
In the third section, I examine Makehams arguments for rejecting a
speakers meaning as a basis for coming to know the meanings of the sentences of the Analects. I argue that even though his arguments possess an
initial plausibility, they ignore a crucial possibility: that the meanings of the
sentences of the Analects, like the meanings of sentences of a novel, depend
only indirectly on known historical record. As a result, our knowledge of the
Analects overall philosophical meaning and our knowledge of the meanings
of particular sentences (and terms within them) are not undermined by our
limited knowledge of the historical context in which these conversations
and anecdotes were written.
In the fourth section, I describe the Brookses recent attempt to provide the historical context of the sentences of the Analects, an effort that, if
successful, would undermine my conclusion in section three that Confucius
is a fictional character. I also argue that their attempt fails due to a lack of
any clear chain of evidence that shows the historical genesis of the Analects.
I use the lack of a chain of evidence to support further my proposal that
we read the Analects as a multi-authored philosophical novel.
In the fifth section, I present H. G. Creels account of Confucius as
a man and argue that his efforts to show that the Analects is historically
accurate fail. He unsoundly infers from the true claim that the Analects
portrayals of Confucius are realistic that they are true. I argue that the conclusion we should draw from Creels account is that the Analects presents a

72

Whose Tradition? Which Dao?

realistic,
albeit philosophical fiction concerning a character named Confucius,
revealed through his conversations with and relations to his interlocutors.
The lack of historical evidence, however, undermines our interpretations of
the Analects sentences no more or less than the absence of direct historical
evidence undermines our interpretations of other forms of fiction.
In the sixth section, I argue against Makehams solution of the problem of semantic skepticism. This section sets the stage for my alternative
solution to the problem, developed in Chapter 4.6

Indeterminacy of Meaning
A fundamental source of indeterminacy of meaning is lack of specific knowledge of context. There are a variety of aspects of the context of an utterance
that can create indeterminacy of meaning. I examine several here:

A. Situational context: I walk past a conversation in which a

professor whom I do not recognize says, The whites are
prejudiced. I might assume he is expressing his personal
attitudes about an ethnic group, when, in fact, he is talking
about a specific family named White. I miss the context and
construe him as believing the statement, The ethnic group
whites are lazy, but what he was actually saying was, The
family living down the street, the Whites, are prejudiced.
What he means as he says this sentence depends on context.

B. Situational context: I call my wife on her cell phone and

ask where she is, and she says, At the bank. I assume she
means the savings bank, but, in fact, she means she is at the
river where we keep a boat. Here, context not only affects
what she means, it also affects the meaning of her sentence
and the words in it.

C. Grammatical context: A teacher gives a test on new

vocabulary. He gives the word color and asks, What part
of speech is this word? Some students answer, a noun; some
answer, a verb. He says it is a noun, and several students
object.

D. Cultural context: Five students and I spent a month

interviewing students at Zhongshan University in Guangzhou.
The Chinese students would often tell us that they were
modern in attitude, not traditional, and that they were

Confucius, History, and the Problem of Meaning

73

independent. But when we probed further, they would

say that the important decisions they made, such as what
subject to major in or what job to take, were decisions that
the whole family made. In every case, our translator picked
fairly obvious, straightforward English terms to represent
the students sentences, but to be not traditional and
independent had different meanings for them and for us.
Only by clarifying their contexts and the contrasts their usage
implied were we able to understand the literal meanings
of their sentences along with what they actually meant as
applied in their lives and culture.

E. Theoretical context: A gay activist claims that homosexual

preferences are, for those who have them, natural because
people are born with these preferences; they do not choose
them. A Catholic respondent argues that God-given laws
require that we order our sexual desires toward the goal of
procreation; only such orderings of sexual desire are natural.
In these two occurrences of their usage, the word natural
is defined differently. In the first case, any desires that are
already part of human life are designated as natural. In the
second, those desires that are already part of human life are
natural only if they correspond to certain interpretations
of the God-given moral law. Lacking knowledge of the
speakers context and his or her spoken words/statement,
we would not know the meaning of natural to either
person.

These are not exhaustive or mutually exclusive categories. For example,

some grammatical ambiguities may at the same time also be theoretical.
Some of the problems of indeterminacy of meaning in the Analects parallel
these sorts of examples. For example in Analects 3.17, we have a problem
of situational context. Consider this passage with two possible translations:
. . . . .

A. Zigong wanted to dispense with the raw sheep for the prayer
ceremony on the first day of each moon. The Master said:
Ci, you grudge your sheep; I grudge my rituals.

B. Tzu-kung wanted to do away with the sacrificial sheep at

the announcement of the new moon. The Master said, Ssu,

74

Whose Tradition? Which Dao?

you are loath to part with the price of the sheep, but I am
loath to see the disappearance of the rite.7
Other translators follow these two patterns. Huang translates the character (yang) in A as sheep, and Lau translates it in B as price of the
sheep. These are not just problems of translation of the sort that arise as
in example D above with the Chinese students, where the one word has no
exact equivalent in English, and so the best choice we picked was misleading. Also, for this example, our translator is familiar with the Chinese as
well as the English language. The problem demonstrates one of the ways
in which knowledge and emphasis of context can affect a persons understanding of meaning.
Huang translates the sentence about the sheep literally and does not
understand or give any weight to context. In contrast, context influences
Laus translation, which refers to the price of the sheep when no word
for price can be found in the original sentence. Instead of ignoring context
and translating (yang) as sheep, Lau understands that Zigong functions
as an administrator in charge of rituals, including the budget for them.
Knowing this context by itself will not necessarily affect an interpreters
understanding of the meaning of the sentence. The interpreter must also
think that this fact is of central importance to what Confucius means,
enough to justify a change of understanding of the meaning of the sentence
and the addition of price in the translation.
This change in translation is significant. The literal translation, without
the emphasis on the context of Zigongs budget responsibilities, causes us
to think that the issue here is the life of the sheep, not the issue of saving
money. The issue of the relation of frugality to proper observance of ritual
is, however, an important issue in the rest of the Analects. Without this
knowledge, we would miss a chance to make this important connection. If
we were to follow Huangs literal translation, we might interpret Zigongs
attitude as stemming from sympathy for the sheep instead of trying to be
frugal. Here, as usual, meaning depends on context. We also encounter
other types of ambiguity in the Analects.
Disagreements about how to translate Analects 1.2 and 4.6 are dependent on grammatical ambiguities, along with theoretical disagreement over
what the characters (ren) mean. Consider 1.2.
. . . .
. . . . .
.

Confucius, History, and the Problem of Meaning

75

Makeham discusses two different understandings of (wei ren) in

the last sentence of 1.2: If we think of as the is of identity, meaning
to be, we must understand 1.2 as follows:

A. It is (xiao ti) filial piety and deference to elder brothers

that are (wei) the foundations of humanity [ (ren)].

Understanding (wei) as to make or do (general verb of action) yields:

B. It is (xiao ti) filial piety and respect of elders that

(wei ren zhi ben) are the fundamental means of putting
humaneness into practice.8

This difference in interpretation is significant. In the first reading, we must

think of (xiao ti) as prior to (ren), whereas on the second reading, (ren) is thought of as already existing prior to (xiao ti) but as
not yet applied to particular cases.9 Makeham points out that translation A
reflects the understanding of commentators Bao Xian, Wang Bi, and Huang
Kan, whereas B reflects that of Zhu Xi, who, following the Cheng brothers,
argued that (ren) is prior to (xiao ti) and that (xiao ti) is a
means of putting (ren) into practice. This interpretation rests squarely
on Zhu Xis overall explanation of Confuciuss teachings, for it depends on
commentator Zhu Xis holding both that (ren) is a complete pattern,
or complete nature, prior to human conduct, and that the application of
patterns (natures) differs from those patterns (natures).10
A third understanding comes with Liu Baonans argument that
[Al]though nature possesses humaneness [ (ren)], it still depends on
being enacted/carried out ( wei).11 Although Makeham does not venture
an alternative translation for 1.2 dependent on Lius understanding, I offer
the following:

C. It is (xiao ti) filial piety and respect of elders that

(wei ren zhi ben) are the fundamental means of
completing humanity.

These differences also manifest themselves in translations into English:

1. Filial piety and fraternal submission! Are they not the root
of all benevolent actions ()?12 (Legge)

2. Filial piety and fraternal dutysurely they are () the roots

of humaneness ().13 (Dawson)

76

Whose Tradition? Which Dao?

3. Filiality and fraternity are () the basis of ren (), are they
not?14 (Brooks and Brooks)

4. Might we not say that filial piety and respect for elders
constitute () the root of Goodness ()? (Slingerland)15

At the most general level, Makeham is claiming that lack of knowledge of historical context makes for ambiguity, and because we do not
have sufficient historical knowledge to resolve these ambiguities, we tend to
rely on arbitrary scriptural or normative readings of the text, which do
not include needed historical information. This leads Makeham to declare
that we are not able to know the meaning[s] of the sentences of the Analects. Both Makeham and Gardner use the basic condition of ambiguity to
examine the relation of the commentarial tradition to the original Analects.
Whereas Makeham is struck by the way that ambiguity undermines our
knowledge of the meaning of the text, Gardner is impressed by the way in
which ambiguity makes for a multiplicity of interpretations. On that basis,
Gardner claims that the Analects lacks meaning of its own.16 But whether
these accounts are ultimately plausible will depend on the success of the
two philosophers arguments for these conclusions. In the next section, I
will examine in detail Gardners argument for semantic nihilism, the view
that the multiplicity of interpretations of the Analects shows that the text
lacks meaning of its own.

The Threat of Semantic Nihilism

Gardner argues that the Analects has never had a single normative meaning
but that it has been read differently by different commentators at different
times.17 He says that although it is fine for commentators to attempt to say
what the true meaning of the text is, meaning the one they think provides
justified norms for us to live by, it is not acceptable for them to claim that
this meaning is the meaning of the text. There is no such definitive meaning, he says, just different meanings at different times according to different
commentators. In a footnote, Gardner quotes Frank Kermode in support of
his view: The only works we value enough to call classic are those which,
and they demonstrate by surviving, are complex and indeterminate enough
to allow us our necessary pluralities.18
Gardners claims about meaning of the Analects echo the skeptical
claims that emerged in the early twentieth-century versions of historicism,
the view that knowledge is historically conditioned: Carried even further

Confucius, History, and the Problem of Meaning

77

the recognition of the historicity of all knowledge led to the recognition that
there is no objective historical cognition but that all historical knowledge
is relative to the standpoint of the historian.19 Developing this historicist
theme, Gardner says:
English translations of the Confucian classics have tended to
present what we might call a normative reading, the aim of
which is to present the true meaning of a text, at least as the
translators apprehend it. As a result readers are easily left with
the impression that the translation in hand is the way a classic is to be understood, the only legitimate or meaningful way
the text can be read. Let me say, of course, that this sort of
normative reading has an important place, that translators have
every right to attempt to uncover what, for them, is the true
meaning of these sacred works from the Chinese tradition. But,
at the same time, translations presenting normative readings can
be historically misleading, for they obscure the simple fact that
the Confucian classics were not static, that they held different
meaning for different people in the Confucian tradition over
the course of centuries. Historically sensitive renderings of the
classics, which underscore the changing meaning of the texts
over time, place, and person, are needed to complement the
normative readings; these translations will more fully and vividly
reflect the role of the classics in the historical development of
the Confucian tradition.20
Gardner does not present an argument for his view that Confucian
classics lack true meaning in and of themselves, but he does state it clearly.
Without some argumentation, however, and without a successful response
to an obvious criticism of his view, this approach cannot stand without a
great deal of qualification. The most straightforward argument, seemingly
implicit in his account, surely is unsound. For his argument seems to be that
because over time there have been multiple interpretations of the Analects,
there is no universal significance to the text. But we know that this argument is invalid, in much the same way that the corresponding argument
that there are no universal ethical truths is invalid. It does not follow that
because people have disagreed about some historical fact or interpretation
related to an ethical truth or a scientific proof, that there is no fact of the
matter. People have disagreed about whether the earth is the center of the
universe, but that does not mean that there are no facts of the matter.

78

Whose Tradition? Which Dao?

Similarly, people have disagreed about whether it was permissible to gas

Jews, but that does not mean that there is no right or wrong about such
actions.21 These relativistic arguments want us to take seriously the fact of
disagreement. That demand seems fair enough. But such arguments do not
support the relativists conclusion that because disagreement exists there is
no fact of the matter, that is, no basic or universal meaning of the text.
There is a better way to understand the import of these disagreements. In
those cases where the disagreement is ongoing, there is no substitute for
patient examination of arguments for and against the competing views.
Certainly, it would be a mistake to claim that all readings of the Analects
are equally good. If that were true, then there would be no particular reason
to examine anyone elses views as a way of determining the correctness of
my own. I could just construct my own interpretations, based on whatever
evidence I deemed reasonable, and accept mine as correct or as correct as
any other. However, I suspect that not even the most hardcore historicist
or relativist scholar of the Analects would accept this result, just as no
reasonable person can believe that all moral views are equally good. If we
reject the relativistic argument that disagreements about a texts meanings,
based on scholars different understandings of the texts historical contexts
or their lack of historical information about the text, render the basic text
meaningless, this criticism provides a reductio ad absurdum of Gardners view.
The fact that ambiguity exists and the fact that commentators offer
multiple interpretations of the sentences of the Analects do not by themselves
prove that the original sentences attributed to Confucius and his associates
have no meaning. Although we can admit that ambiguity is a problem, it
is not the problem of the original text having no meaning of its own. For
this reason, Makehams view of the implication of ambiguity offers a more
measured analysis that avoids the problem in Gardners account. I now turn
to his arguments for semantic skepticism.

Speakers Meaning and the Threat of Semantic Skepticism

I will begin with Makehams discussion of this topic but first need to clarify
why my discussion of his extended account of meaning is quite abbreviated. In fact, in his survey of accounts of meaning, Makeham discusses a
range of accounts: speakers meaning, authors meaning, readers response,
and hearers response, as well as Gadamers hermeneutic theory. Despite this
range, I will limit my discussion to his discussion of speakers meaning. For
it turns out that the fundamental problem Makeham finds in all of these

Confucius, History, and the Problem of Meaning

79

accounts rests on our lack of sufficient historical context to decide on the

meanings of various passages, leaving the Analects ambiguous.
Makeham argues that the speakers account of meaning is problematic
as a basis on which to understand the meaning of the Analects on several
grounds:

A. Although a speakers account of meaning makes sense

out of the idea that a sentence has one meaning and one
meaning alone, this account makes subsequent reception and
development of understandings of the text problematic:
He quotes David E. Linges comments:
The basic difficulty with this theory is that it subjectifies both meaning and understanding, thus rendering unintelligible the development of tradition that
transmits the text or art work to us and influences
our reception of it in the present. When meaning
is located exclusively in the mens auctorus,22 understanding becomes a transaction between the creative
consciousness of the author and the purely reproductive consciousness of the interpreter.23

In Makehams view, a texts meaning is separate from the intentions

of its author because the meaning of the text is changed over time by readers and commentators. The correct understanding of meaning becomes
a series of changing relationships between a creator and a community of
commentators, each with its own history. Readers understanding of a text
in the present time has been changed by the history of its various interpretations. As a result, the meaning attributed to the speakers changes each time
a different person interprets the meaning. Because the exact history of the
succeeding interpretations cannot be traced, Makeham says that there is no
way the author/speakers original meaning can be known.

B. Quoting Paul Ricoeur, Makeham himself argues, moreover,

that another problem concerns the recoverability of authorial
intention from written words after the event of their original
genesis has passed. As Ricoeur has argued, writing renders
the text autonomous with respect to the intention of the
author. What the text signifies no longer coincides with

80

Whose Tradition? Which Dao?

what the author meant; henceforth, textual meaning and

psychological meaning have different destinies.24 Although
related to point A, this point concerns the fact that the
meaning of a text cannot serve as evidence of the speakers
or authors intention. Since the meaning of a text changes
through time, it has no evidential relation to the authors
intention in writing it.

C. Makeham also makes something of the way in which the

multiple authorship of the Analects throws this account
into crisis, for we now have reason to believe that there are
multiple authors of the Analects. Attempts to locate those
authors is, he argues, at best a set of guesses:
There is no clear authorial voice; indeed, there are
many authorial voices but none that can be identified with certainty. As early as Han times, it was
accepted that the Analects was the product of many
editorial hands. Some commentators (traditional and
modern) have attempted to identify the editors, but
their proposals are little more than inspired guesses.25

From this consideration, Makeham concludes that any attempt to determine

the intentions of these authors fails.
Makehams arguments B and C depend on an unstated assumption
that an account of meaning should provide us with a method for verifying
meaning claims: If we cannot use the account of meaning to show how to
verify meaning claims, then the account fails. So, because we do not know
the identities of the authors of the Analects and because we do not have
access to them and their minds, this account of meaning fails. Also, because
there is no evidential connection between the meaning of a text and the
meaning intended by its author, this account also fails.
Despite his reasonable claims that there is no way to use a speakers
theory of meaning to establish which interpretation is correct, it is hard to see
why this is a problem for that theory. There is, I suspect, no good reason to
think that a theory of meaning must show us how we are able to determine
the meaning of sentences. To take a parallel case, an account of what it is
to be a physical object may not provide a feasible, easily applied method
for determining whether something is a physical object. An account of the
being of an object is not the same as an account of the evidence required

Confucius, History, and the Problem of Meaning

81

to for us to succeed in knowing that an object is a physical object. On the

contrary, such a theory may show us that no such knowledge is possible.
Moreover, it seems clear that most of the texts of the early Chinese
tradition, which were never single-author texts, but rather, grew and developed over time until they reached their final form, have been found to offer
a largely consistent vision, even if their genesis was complicated and only
known to us in fragments.26
Nevertheless, there is some plausibility to the claim that if (a) the
meaning of a sentence only is its speakers or authors communicative intention, and if (b) we do not know who the speaker or author is, then (c) we
cannot know the intended meaning of the authors sentences. If we accept
this argument, then it puts pressure on us not to claim to know the meanings of the sentences comprising the Analects unless we first resolve the fundamental historical question: Who were the individuals who spoke/authored
the original sentences of the Analects, and what were their intentions when
they spoke or wrote those sentences? I now turn to a recent attempt to do
just that and argue that this sort of historical argument is bound to fail.

History and the Chain of Evidence Problem

Against this sort of skeptical conclusion, we might wish to challenge the
historical claims Makeham and I are making here by examining a recent
attempt by the Brookses to determine the exact authorship of the various
books of the Analects and the historical context that fixes the meanings of
its sentences. Although there is no complete agreement on the history of
the production of the text, most scholars agree that the text was written
at different periods, by different authors, and there is some consensus on
which parts of the text were written earlier and which later. For example,
most scholars hold that Books 1 through 9 or parts of them represent the
earliest strata of the text.27
Beyond affirming this rough consensus, the Brookses attempt further
to specify the dates and authorship of specific books and passages. Despite
my real interest in their effort, I will argue that the fragmentary nature of
the text, which they adequately demonstrate, cannot justify the historical
treatment they give it. We have competing fragmentary representations of
Confuciuss teaching without the historical evidence necessary to give an
accurate account of the genesis of those fragments.
The Brookses hold that the original Analects is the bulk of Book 4
of the traditional text. They argue that these passages represent a stylisti-

82

Whose Tradition? Which Dao?

cally earlier form of Chinese than the rest of the text and that they arguably were written at the time of Confuciuss death as a memorial to him.
Specific dating of this sort tends at best to be speculative, but this sort of
argument is vitiated more importantly by chain of evidence considerations
that weaken it considerably. According to scholar, translator D. C. Lau,
although there is evidence from a quote in Liji of the pre-Han dynasty
book titled Analects,28 beyond that single reference nothing more is known
about the book from early sources. The earliest detailed information about
the text derives from the first-century AD bibliographical chapter in Han
Shu, History of the Han Dynasty by Pan Ku.29
More recent archaeological finds have offered Han Dynasty versions of
Analects, but this does not change the fact that even with these early texts,
we lack any certain account of the origin of these texts. Pushing back the
dates of the texts does not resolve the chain of evidence problem by itself.30
Although we are talking about historical evidence in this case, we
might wish to consider the strength of historical evidence using the analogy of evidence in legal cases. In legal arguments about evidence, it is
important to establish the chain of possession of evidence in a case. If that
chain is broken, it is possible to have a reasonable doubt that the evidence
is authentic. We may forever have this sort of problem when it comes to
making any claims about which parts of the Analects are original. One of
the versions of the Analects, reported in Han Shu, is the Gu Lun, reportedly
discovered in the walls of Confuciuss house.31 But how reliable is this
claim? Whats more, even if it was discovered in the walls of Confuciuss
house, what would that show about the time the text was composed and
how accurately it represents the sayings of Confucius? With no continuous
verifiable chain of evidence from the time of Confucius to the writing of
the text, we have reason to doubt its authenticity. Even if there were a clear
chain of evidence, that fact would not necessarily show that the representations of Confuciuss speech were accurate. Even if the text were written at
the time that Confucius was alive, which no one suggests, we would still
not know whether the transcribed sayings were trustworthy portrayals of
Confuciuss speech. It seems possible that the representations first were part
of an oral tradition and only later written down, like Homers epic poems.
Even if that were so, there is no reason to think that the oral tradition did
not alter the sayings through the transmission process.32
Perhaps if the state of the Analects and the various versions available
pointed to the existence of some reliable chains of evidence and some reliable
indicators that these texts were accurate accounts of Confuciuss speech, we
could draw some reasonably well-supported inferences about what Confu-

Confucius, History, and the Problem of Meaning

83

ciuss exact views were. However, the state of the historical evidence makes
that impossible. This result leaves us either crippled when it comes to making authoritative claims about Confuciuss views or casting about for some
other, nonhistorical, or at least not exclusively historical, approach to the
text. Creel attempts to address this problem by arguing that the historical
authenticity of the Analects is reflected in its realistic portrayal of Confucius. If that were true, it would vitiate the lack of a chain of evidence
problem. I turn to this question: How far does the realistic portrayal of
Confucius support the claim that he actually said or must have said much
of what was attributed to him in the Analects?
Creels account may seem dated, and some may wonder why I have
not selected more recent scholarly treatments of Confuciuss life. I decided
to focus on Creels account because many accounts seem to gloss over this
historical problem. They bring the assumption to their work that some elements of the traditional picture of Confuciuss life must be true. In contrast,
Creel feels a need to offer a defense of his approach, and I suspect that
his mode of defense and the leap of faith it is bound up with represent
a version of what any contemporary biographer presupposes. So I turn to
his account because of its clarity regarding what appear to be examples of
background thinking in others accounts.33

The Character Confucius: Real or Realistic?

In order to pursue this question, I turn to one of the best attempts to use
the Analects as the authentic account of the life and teachings of Confucius,
Creels, in order to show why Creel fails to provide convincing evidence
that the Analects is historically accurate. I argue that Creels account of the
humanity of Confucius as he is represented in the Analects ought to be
understood as demonstrating the success of the Analects as a fictional or
semi-fictional text, not as an historically accurate text.
In his Confucius and the Chinese Way, Creel attempts to solve a problem
with traditional accounts of Confuciuss life.34 The large impact of Confuciuss
thinking seems hardly intelligible, given traditional accounts of the man
himself as unoriginal, lacking personal force, and failing to live up to his
announced ideals. As Creel puts the point: This man, portrayed by tradition,
seems an inadequate cause for the effects attested by history.35 Two strategies
inform Creels development of a story about the real Confucius. First, he
places him in the tradition of Enlightenment thinkers, who, influenced by
Jesuit missionaries, saw through much of the traditional accounts of Con-

84

Whose Tradition? Which Dao?

fucius. Through the Jesuits account of Confucius, Enlightenment thinkers

found a democrat who supported the ideals of freedom and equality, the
same ideals they held dear. Second, Creel looks to evidence of who Confucius
was and what he thought from the most reliable early material about him
and his life, specifically, the Analects. In addition to appealing to this most
reliable early material, he appeals to the fact that the Analects portrays a man
in conflict, a real man, a real mensch [my term].36
One of the strengths of Creels account of Confucius is his focus on
finding the humanity in Confucius as a person and as a philosopher. His
humanity as person comes across, according to Creel, in the real tensions
in his character: he was ambitious yet frank with those in positions of
power,37 zealous yet with a sense of humor,38 principled but willing to follow social conventions, even if that involved lying,39 confident yet willing
to acknowledge his ignorance.40 These tensions are essential to Creels portrait of Confucius as a real person. Without these tensions evident in the
Analects, he would be a stylized saint, the product of fantastic hagiography.
Creel sustains his focus on constructing a complex account of Confucius as a real person in much of the rest of his explication. For example, he
develops three personas in succession, of Confucius as teacher, philosopher,
and reformer. This approach has the advantage of letting us see different
aspects of Confucius without viewing him under any single defining aspect.
It also has the advantage of allowing potential conflicts within his character
to emerge. For if we follow Creel and adopt the principle that real human
beings adopt ideals that they do not always live up to, we can expect to
find Confucius as philosopher or reformer to be in conflict with Confucius
as a real person or indeed as a teacher.
Despite the need to find conflicts among these three aspects of Confuciuss project in order to develop his argument, Creels accounts of Confucius
in these three aspects are remarkably consistent. The thread that Creel finds
in these three aspects of Confuciuss life and thought, not surprisingly, are
related to one thread that he finds in Confuciuss thought. Although he
admits that the Analects, made up of fragmentary comments and dialogues,
often makes it unclear exactly how to understand Confuciuss one thread,
Creel identifies it with learning to live well in a cooperative world.41 What
this means, however, is itself complicated.
The image of a cooperative world that Creel finds in the Analects is
one akin to Kants community of ends. Each person is treated as an end
in him- or herself, and these individuals must find a way to make their
own desires consistent with the desires of the others with whom they are

Confucius, History, and the Problem of Meaning

85

in community.42 Creels Confucius is the hero of Enlightenment thinkers,

who found in his views important elements crucial to the development of
democratic individualism. In his teaching, Confucius trains his students to
be able to make their own decisions for themselves, though they do so in
the context of dialogue, a cooperative context of inquiry. In his philosophy,
Confucius is fairly agnostic about religion but steadfastly refuses to ground
his philosophy in any metaphysics or to ground his thinking in any criteria
of truth or knowledge. According to Creel, these sorts of foundations apart
from oneself would undermine Confuciuss thread, his commitment to the
authority of the individual to make his own judgments about right or
wrong in the context of a community of inquiry. Such foundations would
establish forms of authority over individuals that would establish ways to
assess the judgments and choices of individuals. To use Kants terms, these
foundations would introduce into ethics and politics heteronomy, forms
of authority not products of ones autonomy as a rational being. And by
Creels account, as a reformer, Confucius was never willing to establish
himself as the final authority on these matters.43
This aspect of Creels account is disappointing in light of his earlier
commitment to a reasonable principle that required a persons humanity be
made manifest in conflicts between ideals and realities. It is easy to imagine
that a great reformers ideals might conflict with his philosophy and with his
teaching. For the content of the ideals may not be adequately or completely
well defended in his philosophy, and the give and take of dialogue easily
results in all sorts of statements a philosopher might not wish to include
in the best accounts of his philosophical views. But Creel is silent on these
matters. We can, however, be clear about one thing: Creel admires in the
Analects the ways in which Confuciuss humanity shines through. But the
conflicts we find in Creels account of Confucius as a person do not show
up in Creels account of Confucius as a philosopher. For that to happen we
would need to find a Confucius more inclined toward philosophical struggle
and self-doubt. If we take Creels view that humanity shows itself through
a persons inner conflicts, we have to conclude that Confuciuss humanity
as philosopher is not apparent in Creels characterization of him.
Although a final adjudication of this problem will rest on an examination of the details of the Analects, there is good reason to think that Creels
account of Confucius as a philosopher fails. The best way to see the failure
is to see the conflict that arises from his Kantian account of the thread
that runs through Confuciuss thinking. Creel attributes three important
claims to Confucius:

86

Whose Tradition? Which Dao?

1. Confucius has no interest in appealing to any metaphysical

basis for dao, though there are some passages that appeal to
tian (heaven).44

2. Confucius has no interest in the epistemological question of

the criteria of truth.45

3. Confucius adopts a view of respect among members of a

cooperative community akin to Kants.46

It is hard to see how each of these three claims can be maintained. For
Kants account of a community based on respect cannot, without distorting
that account, be divorced from his epistemology or metaphysics.
Even if Creels account of Confucius as philosopher were acceptable,
his account would run into one additional significant problem, the quest for
the historical Confucius. Although there might be some initial plausibility
of treating earlier texts in the Confucian tradition as reliable indicators of
what Confucius really said and taught, Creel does not try to show that there
is any way to know that the Analects material is accurate. Instead, as many
thinkers have, he assumes that earlier sources are more accurate than later
sources, but he never shows why this should be so. It is possible both types
of sources are unreliable. We have no reliable evidence about the genesis of
the Analects sayings, nor do we have significant, independent, corroborating evidence recorded earlier. These deficiencies are enough to refute Creels
view. His alternative tactic, however, is to argue that the accuracy of the
Analects sentences comes through in the ways they successfully portray a
person who seems real, not idealized. But real has two different meanings.
A realistic portrayal is one that rings true. I can read a realistic novel,
but that does not mean the plot reveals what actually happened or that the
characters depict real people. Literature provides a clear reason for thinking
that we cannot use realistic portrayals as evidence of truth of the portrayal.
Despite Creels attempt to offer an argument for the historical truth of the
Analects account of Confucius, his arguments are not successful.
In response to this argument, defenders of Creels general approach
might want to claim that we still have enough evidence from the Analects
to take the account of Confucius there to be partly correct. Indeed, I suspect that many scholars are influenced enough by the realistic portrayal of
Confucius in the Analects to think that it must be at least partly true. This
approach, nevertheless, is usually not offered as an argument, as it is more
like wishful thinking. Michael Nylan, in her wonderful book, The Five Confucian Classics, with great care surveys the limits of the historical evidence

Confucius, History, and the Problem of Meaning

87

for early Confucian texts, but when she characterizes the Confucius of the
Analects, she describes the portrayal of him as semifictional.47 However,
even from her careful account, it is not clear which parts are fictional and
which parts are not. Are the nonfiction parts those that claim there was a
man named Confucius, born at a certain time, who later taught students?
Do nonfiction parts include certain utterances he made? The term semifictional might be justifiable if we could pin down a way to distinguish
the nonfictional parts from the fictional parts of the sayings.
Even if one grants the various points I have been making against the
historical evidence regarding the identities of the author or authors of the
Analects and whether Confucius said any of the sentences attributed to him
in the Analects, I would argue that the lack of a sufficient historical record
makes those meanings all the more accessible. Based on what we have reason to believe about the genesis of the Analectsthat it was written over
time by several authors offering competing accounts of Confuciuss teachingit is reasonable to treat the Analects as akin to a philosophical novel:
it invokes an imaginative world with decipherable meanings within the
limits of our understanding of that world, even if our historical knowledge
of the circumstances of Confuciuss discussions with his interlocutors or the
circumstances of the authorship of the Analects are incomplete. However, in
order to make this case, we need to clarify what sort of character Confucius
reflects in the Analects and what makes him seem so real. I suggest that
this view, rather than making knowledge of the meanings of the Analects
sentences inaccessible, allows the sayings of Confucius to become even more
accessible. Although I will not pursue this argument here, these same general
points could be made about Confuciuss interlocutors, who, like Confucius,
function as characters within the various truncated stories that make up a
large portion of the Analects.48
We can think of the Analects as a philosophical novel with Confucius
as the main character. We might proceed in much the same way if we were
to become convinced that Socrates never had the sorts of conversations that
Plato attributes to him. We could, nevertheless, read the early Socratic dialogues as fictional text with philosophical points. The philosophical points
of those early dialogues are independent of any historical claims about
their accuracy as depictions of Socrates actual conversations. Whats more,
their historical influence is such that even if we were to find evidence that
some of these dialogues represented the speech of a drunken aristocrat who
definitely was not Socrates, a brilliant fellow who occasionally showed up in
the agora and challenged politicians to defend their views, it is not likely
that we would change the name of the main character of those texts. They

88

Whose Tradition? Which Dao?

have a literary and philosophical life of their own, and Socrates is the character represented in those dialogues even if he never, in fact, had some or
any of those conversations. Not only does Socrates have a life of his own
as a character, his character also represents a set of attitudes and practices
of critical reflection, which have influenced the subsequent development
of Western philosophy and, arguably, continue to provide a substantially
correct account and practice of the central aspects of critical reflection.49
Thus, I argue that although it might be disturbing to discover that Socrates
was not represented correctly in these dialogues, that knowledge would
not undermine the importance of these texts or our ability to grasp their
sentences meanings.50
Even if we distinguish between Confucius as the main character of the
Analects and the historical Confucius, we are left with additional puzzles.
For example, we typically think of novels, with main characters and plot
lines, as having some unity. The unity might be narrative: the novelist tells
a single story about a main character, such as Elizabeth Bennett in Jane
Austins Pride and Prejudice or Stephen Blackpool in Charles Dickenss Hard
Times. This model does not work perfectly if we are thinking of classical
novels with traditional narrative structures because what we have in the
Analects is a set of conversational episodes, with some of the conversations
apparently contradicting other conversations. Lacking this sort of narrative
unity, we might attribute to the text the unity of the writing of a single
author, whose authorship provides a unity for what otherwise might appear
to be disconnected fragments. My model here is T. S. Eliots great poem,
The Wasteland, which is intentionally fragmented yet, nonetheless, retains
the unity of being the writing of a single author with identifiable themes
throughout the work. In this long poem, the main character can be seen
as the author, who is revealed indirectly through the lines. He expresses his
vision about various conflicts and tensions he finds in his own beliefs or in
the beliefs of his contemporaries.
This model, however, will not work, given what we have reason to
believe about the Analects, that it was written over a period of time, probably
by various authors, some of whom might have been recording oral tradition,
some adding their own sentences, and some compiling and correcting others sentences. Recognizing multiple authors helps make sense out of some
of the conflicting claims we find in the Analects. For example, one of the
most discussed and used characters in the text is (ren), but passage 9.1
says that the Master rarely discussed ren): , , , .51
But we know Confucius often discusses (ren). How are we to make
sense of this obvious contradiction? One way would be to recognize that

Confucius, History, and the Problem of Meaning

89

there were two ways of thinking about the relation of (ren) to (li) in
Confuciuss teaching. Some of the group of authors played up a relationship
between (ren) and (li), while others attempted to present an account
of Confuciuss teaching that gave little weight to (ren). This possibility
gains further weight if we can demonstrate the phenomenon of competing
accounts of Confuciuss teaching.52
In a different text from the one already under discussion, Makeham
argues that in the context of the formation of the texts that talk about
Confucius, various understandings of his teaching were being offered and
that Ru was understood as a family resemblance concept,53 covering a variety of different teachings whose authors pressed their own views of the
significance of Confuciuss teaching by constructing and circulating their
contributions to the Confucius legend.54 If Makeham is right, what these
different authors had in common was the recognition that Confucius was
an important figure and that it mattered how he and his views were to
be understood. They may also have believed that he was a person of high
moral and intellectual achievement. Although in some cases they may have
thought their views would be taken more seriously if put in the mouth
of Confucius, in other cases they may have thought that the views they
attributed to him accurately depicted what it means to be a person of high
moral and intellectual achievement. So they may have believed that their
attributions were in some sense true, even if the specific details of the dialogue were false. They may have believed that their contributions described
important behaviors/considerations that a person such as Confucius would
have accepted as correct. What this approach suggests is that we should
understand the Analects as the beginning stages of what later became a fullblown commentarial tradition. The difference between this early stage of
proto-commentary and the later stage of full-blown commentary in which
the standard text, or texts of the Analects, were already established, is that
the commentators were anonymous, and their commentarial function was
one aspect of being a contributor to the text. If Makeham is right, and
I am inclined to think that he is, we would have to include some of the
authors of Zhuangzi as contributors to these proto-commentaries.
If this rough story about the authorship of the Analects is correct,
then it would give weight to a project of understanding the text in terms
of the versions of Confuciuss teaching that arise there, their relations to
one another, and the question of which ones represent the best versions of
Confuciuss teachings. Note that this question differs from the question of
which version best represents Confuciuss views, for on this account, we
are compelled to drop the pretense of knowing for sure from our read-

90

Whose Tradition? Which Dao?

ing of the Analects what Confuciuss real historical teachings are. In this
project, we later interpreters are continuing the very early efforts, linking
ourselves to those contributors to the Analects who lived generations apart
from Confucius, to ask, Is this the best version of his view? Here we are,
on philosophical grounds, using the fragments of the Analects as a starting
point. What threads of teaching can be identified in the text? Which threads
of the text represent the highest version of the teachings therein? Which
conflicts of view ought to be seen as central to this range of views in this
text? Which version of Confuciuss teaching should be rejected as inferior?
Which version accepted as superior?
Suppose, then, that we operate by treating Confucius as a range of
multiple characters in a single text written by various authors trying to influence how Confuciuss teachings are to be understood. Would this approach
and its related acknowledgment of indeterminacy undermine our attempts
to understand meanings in terms of speaker or authorial intention? It would
if we need to know specific information about the authors in order to grasp
their meanings and intentions. But suppose that we accept that Confucius is
a set of characters or different versions of the same character, whose sayings
were possibly produced by multiple authors? The multiplicity of authors
creates a problem only if knowing their specific identities and intentions
is required to understand the meanings of the Confucius characters utterances. We know that we can read a novel and discern the meanings of a
characters words and actions even though the character is fictional. We can
do that because the novel evokes a world and set of contexts sufficient for
us to discern speakers meanings and intentions.
I would contend that the problem Makeham poses of historical indeterminacy, does not, as he despairs, contribute to the problem of indeterminacy of meaning. Instead, the Analects historical indeterminacy makes
its sentences meanings even more accessible than they would be on the
historical model of authorial intention. Consider, for example, the appeal
Makeham makes to Roland Barthes:
It is not that the Author may not come back in the Text, in
his text, but he then does so as a guest. If he is a novelist,
he is inscribed in the novel like one of his characters, figured
in the carpet; no longer privileged, paternal, aletheological, his
inscription is ludic. He becomes, as it were, a paper-author: his
life is no longer the origin of his fictions, but a fiction contributing to his work.55

Confucius, History, and the Problem of Meaning

91

Even if Confucius, assumed to be the speaker of many of the sentences

in the Analects, appears in the Analects as a guest, I have argued, that does
not undermine our ability to determine which character, or even characters,
he is and what that character or characters may have meant when he conversed with his interlocutors. In fact, if I am right, this will make our job
of deciphering the text easier than it has been until now. For we will be
freed to focus on deciphering the imagined world of the Analects without
feeling compelled to discover the actual world of the Warring States Period.
Those who are dedicated to that work will conduct historical investigations
of language, legends, and texts, and we will use their findings to help us
deepen our understanding of the imagined world of the Analects and the
teachings of the character Confucius.
Nonetheless, one may say that even though this argument relieves us
of one historical problem, that is, attempting to discover what Confucius
actually said, it does so by reaffirming the need for knowledge of history in
order to interpret the sentences of the Analects as a fictional text. Although I
agree that we need historical knowledge, I do not believe the simplistic view
that a little historical knowledge can be sufficient for defending the correct
interpretation of a text such as the Analects. There is an important sense in
which history presents problems, but those problems cannot necessarily be
resolved only by historical research methods. I turn to this argument next.

Makehams Solution to the Problem of Limited Historical Knowledge

Makehams argument, shared by the much more radical semantic nihilism
of Gardner, is that the paucity of the historical record regarding the Analects
leaves us with no access to historical meaning, but instead, to scriptural,
normative, or existential meanings, which we mistakenly tend to think
of as historical meanings. Makeham presents a more complicated view than
Gardners of the significance of interpretive difference by clarifying the difference between normativehe calls it scripturalmeaning and historical
meaning:
Why bother with the commentaries? Why not focus on the
meaning of the text that the commentaries are attempting to
recover? Those who seek to understand the Analects face a particular conundrum: should they give priority to the historical
context of the texts genesis so as to determine its historical

92

Whose Tradition? Which Dao?

meaning or to its scriptural meaning? The historical/scriptural

distinction is not new, and it has been variously employed by
scholars. By historical meaning I refer to the meaning of a text
as composed by its original author(s) and/or its original audience.
By scriptural meaning I point to the meaning realized in the
subsequent historical trajectory of that text. In what follows, I
argue that the distinction between these two types of meaning
affords us a useful tactic for containing willful interpretation and
unlimited semiosis. This distinction is, I contend, a hermeneutical
expedient; even if the historical meaning of the Analects were
recovered, there would be no criteria by which to distinguish
it from scriptural meaning.56
The hermeneutical expedient is that just by appeal to some historial evidence, if Makeham is right, we are able to constrain the variety of possible
and actual scriptural meanings of the Analects. It is easy to see why this may
be so. In a footnote, he identifies the scriptural meaning with what Yearley
refers to as existential meaning: what the text can mean for us today given
our own existential concerns. As we saw in the previous section, Makeham
despairs of finding historical evidence to support claims about historical
meaning:
Notwithstanding these observations, [about historical evidence]
I do not believe that the project of trying to recover historical
meaning is without purpose. As a means of constraining the
proliferation of meaning, it is a necessary strategy. Without
such checks, analysts would be free to disregard the restraints
of historical context and to develop interpretations limited only
by their own imagination. The point is not that we can never
completely recover and project ourselves into the horizon of
cultural others, but that there is little to support the claim that
we can do so, even minimally, for a text such as the Analects.
We must accept that our hypotheses about historical context and
historical meaning are just that: hypotheses. Furthermore, to the
extent that such hypotheses are constructed in the absence of
adequate historical data, we need to be honest about the limits
this must impose on our interpretative claims. To give a simple
example: How many passages from the Analects can be provided
with reliable historical contextualization?57

Confucius, History, and the Problem of Meaning

93

At this point, Makehams argument lapses into incoherence. For the

very grounds he used to reject the other accounts of historical meaning
that they introduce an account of meaning, which, if we were to follow
it, would leave meaning inaccessible to usis a problem for his account.
Following Makehams argument to its logical conclusion, if I cannot know
for sure the meaning of a sentence as the author intended it, then finding
historical meaning of any sort, embedded in a form of life 2,500 years ago
with only fragments of data available, would be equally impossible. He
makes it clear in several passages that we just need to find some way to
constrain meaning, and historical meaning is the best expedient. However,
he is compelled to admit that it is an expedient that does not work. In
fact, his appeal to historical meaning, given his skeptical arguments about
it, seems nothing more than an existential leap of faith:
Unless one is keen to open the floodgates to potentially unlimited semiosis by seeing the reader as the sole determinant of
textual meaning, historical context must be addressed. There
are responsible and irresponsible readings/interpretations of the
Analects; by acknowledging the legitimate boundaries of historical context, we are better able to adjudicate between competing
interpretative claims. Other things being equal, an interpretation premised on the assumption of a historical context that
can be independently verified is preferable to an interpretation
premised on the assumption of a historical context that cannot
be independently verified.58
Here, Makeham offers a way to avoid being overwhelmed by an unlimited
number of meanings. But he does not do a very good job of this. His claim
that we should accept historically verified meanings, other things being
equal, just to limit meaning seems especially weak. We could just as well
limit meanings by a toss of a coin. What has gone wrong here?
The problem that Makeham posesthat the only way to constrain
the proliferation of scriptural meanings is by requiring that, all things being
equal, those meanings also be historical and so held by those who authored
and first read those textscannot work to solve the problem of the limits
of our historical knowledge. The fundamental problem with his view is
his holding that finding the meaning of a text is the exclusively historical problem of determining meaning. That historical problem, he thinks,
is exhausted by historical investigation. However, he mischaracterizes an

94

Whose Tradition? Which Dao?

important feature of textual interpretation: the use of the principle of charity. In fact, all interpretation has an essential scriptural aspect, which is not
just existential or subjective. It will be my task to show this by appeal to
strategies of clarification of what meaning is, derived from Wittgensteins
later philosophy and from the work of Donald Davidson. In contrast with
Makeham, I hold that history poses the problem of indeterminacy of meaning, but even the most comprehensive knowledge of history cannot solve
this problem. In the next chapter, I offer an account of understanding texts
that shows under what conditions textual understanding is possible.

Wittgenstein and the Problem of

Understanding at a Distance

Shared human behavior is the system of reference by means of which

we interpret an unknown language.
Wittgenstein, Philosophical Investigations, section 206
If a lion could talk, we couldnt understand him.
Wittgenstein, Philosophical Investigations, 235

Introduction
In this chapter, I sketch a Wittgensteinian account of meaning and argue
that this account finds no place for the distinction between historical and
normative meaning. All meaning is in some sense normative. In Chapter 3,
I concluded by arguing that whereas history poses a problem for us of how
to understand texts from a distance, history cannot solve the problem. Once
the problem of discerning meaning from a historical distance is made salient,
appeals to history alone, the source of the problem in the first place, can
provide no solution. What is involved in understanding texts created 2,500
years ago? To develop an account of understanding texts from a historical
distance, I take up arguments from Donald Davidson and Wittgenstein.
My argument proceeds as follows: In the first section of this chapter,
I use resources in Wittgensteins later philosophy to present critiques of
Makehams and Gardners views on the problems encountered when trying
to interpret the meaning of classical Chinese texts. I argue that Wittgensteins appeal to meaning, as to how words are used in a language-game,
provides an alternative to Makehams and Gardners problematic accounts. In
95

96

Whose Tradition? Which Dao?

the course of my account, I appeal to the notion of interpretive charity as

essential to understanding most, that is, relatively complex, language-games.
In the second section, with appeal to and development of Davidsons
principle of interpretive charity, I offer an account of this key principle of
interpretation, which was presupposed in my account in section one on
Wittgensteins view of meaning.
This argument about interpretive charity fits into recent discussions
of the principle of charity as follows: I see Davidsons account of charity
as refining and offering more detailed specifications of how the principle
of charity functions in understanding unfamiliar languages and conceptual
schemes. The requirement is not, however, simply of maximizing attributions of truth claims across linguistic and conceptual divides. As Richard
Grandy has correctly argued, such an approach to charity would cause us
to distort intelligible falsehood. Instead of maximizing agreement between
our beliefs and the beliefs we attribute to the unfamiliar Other, we need to
maximize intelligibility by attributing to the Other beliefs, sometimes false,
which we ourselves would most likely also have in the situations we find the
Other in. This latter view fits well with Wittgensteins view that we make
the Other intelligible by associating his or her language-game, viewed from
the outside by us, as similar to one of our own. We assume, then, a shared
form of life as a basis of translation. My view develops Wittgensteins view,
which we might just call the Principle of Shared Form of Life. Nevertheless we might map our own form of life onto the Others in various ways.1
In furtherance of this argument, in the third section, I discuss what
Davidson calls the vast middle ground of our beliefs, in which he thinks
interpretive alternatives are possible. I argue, howeverand this is the distinctive argument I offer on this topicthat the principle of charity for this
area of beliefs requires that we become practitioners, not just observers, of
the Others language-game if we are to maximize intelligibility. We might
call this view the Principle of Insider Competency.
In the concluding section, I argue that even though my Wittgensteinian account of interpreting the meanings of sentences in the Analects
accepts that we do not know whether the sentences of the Analects represent
Confuciuss actual speech or teaching, my account is congenial to contemporary thinkers interested in or committed to defending Confuciuss teachings.
Although this account of meaning requires that neither the sentences of the
Analects be historically true nor true in the substance of what they teach, it
does provide a model for specifying the meanings of those sentences. My
model is based upon a plausible account of meaning that takes seriously
the way that the practice of reading requires charitable interpretation. The

Wittgenstein and the Problem of Understanding at a Distance

97

meaning of the text is constituted in part by the practices or methods of

readingwe use for understanding the text; thus, our interpretation of the
meaning of a sentence must be understood in light of these practices. This
account saves us from thinking of the meaning as basically historical, thereby
lacking the necessary evidence to be authentic or basically scriptural and
so, merely subjective.

Wittgenstein on Understanding and Meaning

Makehams reasoning about historical meaning is fueled by what Wittgenstein refers to as a philosophical picture. In one of his famous dicta, he says,
A picture held us captive, and we could not get outside it.2 A picture, in
Wittgensteins sense, provides a philosophical fantasy of how things must
be and what we must mean when we say things. This picture provides us
with a stock of philosophical intuitions about meaning and reality, but these
pictures tend to distort the way we use language. Because these pictures
represent what we easily but incorrectly imagine to be so, when we think
outside the context of concrete linguistic practices, the pictures tend to
distort those practices and the meanings of words and concepts embedded
in them.
With respect to Makehams argument, we might suppose that his
account of meaning connects with his picture that a meaning is directly
present to an interpreter (a hearer or reader) only when the interpreter is
aware of the speakers total historical/sociocultural/linguistic context. This
picture dominates Makehams account of historical meaning. However, if
I am required to know a persons total context before I can understand
that persons meaning, then it is also an insurmountable problem for me
to attempt to understand Confuciuss meaning in the Analects because he
wrote sentences in that text 2,500 years ago in a culture far from my own.
In fact, relying on that picture, it would be correct to say that I will never
understand the meanings of my wifes words. How often can I claim to
know the total context of anyones words?
Makehams picture of meaning does not capture many of the most
pedestrian ways in which we think about meaning. Addressing a similar
set of problems, Wittgenstein suggests that we think of the meaning of a
word or sentence as the way it is used in a language-game. He does not
offer this suggestion as a final, complete theory of meaning, but rather, as
a suggested corrective, which helps us to focus on the question of how our
words are used in specific contexts of everyday life. If the meaning of a

98

Whose Tradition? Which Dao?

word is its use in a language-game, then the meaning of meaning is also

dependent on its use in a language-game. As a corrective to narrow views
of the use of meaning, Wittgenstein asks us to think of the question,
What is meaning? in terms of the related question, What is it to give
an explanation of meaning?3
To think of the meaning of the sentences of the Analects in these
terms puts us on somewhat better ground than we might otherwise be when
we think of meaning in terms of an immediate grasp of a speakers intention. Consider the following case. I have learned Wittgensteins builders
language.4 I learn to apply slab to slabs, pillar to pillars, block to
blocks, etc. The application is this: when the foreman yells slab, I bring
a slab to him, and so on. In more complicated language-games, we can
add the words here and there. Possible sentences are slab, here, pillar,
there. When the foreman yells, pillar, here, I bring a pillar to him. We
can also add time indicators: now, in five minutes, and so forth, with
corresponding behavioral requirements. But suppose that in one case, the
foreman yells, in quick succession, pillar, here, now, slab, there, now. I
cannot pick up and deliver all of these objects at once. I get upset. Then
the foreman and several other workers laugh. I can explain the foremans
meanings as follows: His sentences each mean what they normally mean
(bring me a slab, now,etc.), but I now understand that he intends to play
a joke on me. I also understand that he does not seriously intend for me
to follow his requests. My attribution of meaning makes the best sense out
of his behavior. This understanding gives me, as Wittgenstein puts it, a way
to go on playing the language-game. If I know how to carry on and take
the next step, then in doing so, I exhibit in my behavior a correct understanding. For example, if his calls (i.e., impossible orders) to me count as
a joke, then I should be able to laugh, too, but not continue to follow the
orders. If I should find that I am still expected to bring all of the objects
at once, then I will know that my joke explanation of meaning was not
apt and that the bosss sentences as a collection have not been given any
clear meaning. This account is part of Wittgensteins effort to undermine
the hold that the picture of understanding as a mental process has on us.
To understand is to know how to proceed in the language-game, not to
have in mind an image, say, of the foremans meaning, nor an image of a
rule that tells me what to do next. My meaning attribution also does not
describe his state of mind, as much as it expresses my sense of the role of
his sentences in the language-game and my understanding of the appropriate responses open to me.

Wittgenstein and the Problem of Understanding at a Distance

99

In these examples, explanations of meaning are maximizers of the

sense of a statement or utterance, prompting people to carry on in relation to those statements in a way that needs no correction by the rest of
the people in their language community. For this account of meaning and
explanation of meaning to make sense, we have to suppose the existence of
a community of speakers whose language behavior (i.e., spoken and written
communication) exhibits the norms of meaning for that community. The
norms will be embedded in a variety of practices, including teaching the
language in specific ways, accepting the exemplars that constitute correct
language use, sharing a sensibility that produces agreement in most cases,
and so on. Consider the following famous characterization of Wittgensteins
view of language learning from Stanley Cavell:
We learn and teach words in certain contexts, and then we
are expected, and expect others, to be able to project them
into further contexts. Nothing ensures that this projection will
take place (in particular, not the grasping of universals nor the
grasping of books of rules), just as nothing insures that we will
make, and understand, the same projections. That on the whole
we do is a matter of our sharing routes of interest and feeling,
modes of response, senses of humour and of significance and
of fulfillment, of what is outrageous, of what is similar to what
else, what a rebuke, what a forgiveness, of when an utterance
is an assertion, when an appeal, when an explanationall in
the whirl of organism Wittgenstein calls forms of life. Human
speech and activity, sanity and community, rest upon nothing
more, but nothing less, than this. It is a vision as simple as it
is difficult, and as difficult as it is (and because it is) terrifying.5
For our purposes, the most significant result of this view of meaning
is that meanings rest on agreement in interpretation. For a sentence to have
meaning is for it to have a role in a language-game. Moreover, when a person understands the meanings of most words making up a language-game,
that person has mastered the game and knows how to carry it forward.6
Attributing meaning to texts will not, in general, present different
sorts of strategies from the ones I have introduced in my builders example,
even though the language-games of the Analects are different and more complicated. To understand the meanings of words and sentences in texts, we
have to understand the ways in which sentences and words are embedded

100

Whose Tradition? Which Dao?

in contexts of concrete language use, and we need to have strategies of

maximizing the sense of sentences in light of explicable variations on basic
meanings. And our explications must give rise to strategies of interaction
with the text and with other readers of it that count as successfully being
able to carry on as we read and further explicate the texts meanings.
Both Makeham and Gardner appeal to a distinction between historical
and normative or scriptural meaning. Both treat the historical as objective
but completely or largely inaccessible. They treat the normative/scriptural as
subjective, relativizing it to a readers received meaning. But Wittgensteins
account of meaning draws no such distinction. We assume a community of
readers and speakers and practices of reading and communicating that are
embedded in the community. Who, then, might the community of readers
of the Analects be? A community of readers has existed from the time of the
beginning of the Han Dynasty, (ca. 206 BC), through the present moment.
This community has written commentaries, which function as primers on
how to understand the Analects. That does not, however, mean or require
that any or all of the commentators should agree in their readings. We know
from studies of the commentaries that their writers goals have varied over
time. The fact that they have offered variant readings, does not, however,
undermine the fact that they studied collections of information that have
made it possible for scholars to accumulate a body of practices of reading
over time. Information about the various meanings of terms has emerged,
as have compendia of information about historical events and persons, as
well as received imaginative reconstructions of history. The contemporary
community of readers would consist of all readers who have engaged the
records left by the historical community of readers and who have become
accomplished practitioners of reading these texts. This does not mean that
contemporary readers will agree among themselves or with the variety of
prior readers of the Analects. It does, however, mean that skilled practitioners will want to be able to appeal to shared principles and practices of
reading to defend whatever explications of text that they offer. They build
on past collections written by the community of readers while finding ways
of extending it.7
What should we make, then, of the distinction between historical
meaning, on the one hand, and what is variously called scriptural, normative, or existential meaning, on the other? Such a distinction plays no clear
role in my examples. In one sense, all texts are historical, written at a time
earlier than a contemporary reading of the text. Even speech in a language
predates the understanding of it by a hearer. A community of readers will

Wittgenstein and the Problem of Understanding at a Distance

101

need to develop basic reading skills among themselves that are specific to the
texts they take up. In the abstract, it might seem that the numerous members of the community of Analects readers would be completely hampered
by their differences in historical time, language, and culture. Yet we have
first-rate studies of the semantics of early Chinese. We also have some basic
information about the practices of writing in early China and the development of communities of scholars, which were the natural homes for writers
of early texts.8 What is crucial in this context, however, is a key principle
of interpretation and understanding, one that Makehams and Gardners
accounts do not do justice to and that is implicit in my Wittgensteinian
accountthe principle of interpretive charity.

How to Understand Interpretive Charity

It is a commonplace in textual interpretation that, other things being equal,
a charitable interpretation of a text is better than one that is uncharitable.
The principle of charity requires the practitioner, other things being equal, in
the face of equally well-supported interpretations, to adopt the one interpretation that will maximize the sense and reasonableness of the text.9 That does
not mean the practitioner should override well-defended interpretations to
make the sentences in the text true or reasonable no matter what.10 Instead,
it means the practitioner must attribute to the text the one interpretation
or view that, consistent with otherwise possibly defensible interpretations, is
the most reasonable and defensible of those legitimate alternatives. Which
views are the most reasonable will, of course, be matters of debate, but
even so, that does not indicate that the role of this principle in attribution
of meaning is subjective. For even in philosophically contentious debates,
it is possible for arguments to be demonstrably weak.
The principle of charity has come under attack recently as a basis
from which to proceed in comparative philosophy. Following Davidsons
use of this principle, it has been supposed by critics that the principle of
charity will inevitably paper over deep conceptual disagreements. For as
Davidson puts it:
[T]he criterion of a conceptual scheme different from our own
now becomes: largely true but not translatable. The question
whether this is a useful criterion is just the question how well
we understand the notion of truth, as applied to language,

102

Whose Tradition? Which Dao?

independent of the notion of translation. The answer is, I think,

that we do not understand it independently at all. We recognize
sentences like Snow is white is true if and only if snow is
white to be trivially true. Yet the totality of such English sentences uniquely determines the extension of the concept of truth
for English. Tarski generalized this observation and made it a
test of theories of truth: according to Tarskis Convention T, a
satisfactory theory of truth for a language L must entail, for every
sentence s of L, a theorem of the form s is true if and only if
p where s is replaced by a description of s and p by s itself
if L is English, and by a translation of s into English if L is not
English....Since Convention T embodies our best intuition
as to how the concept of truth is used, there does not seem to
be much hope for a test that a conceptual scheme is radically
different from ours if that test depends on the assumption that
we can divorce the notion of truth from that of translation.11
We cannot make sense out of the idea that a conceptual scheme differs
from ours without appealing to the fact that this alternative scheme must
be translatable into our language, with true statements from it identified
with true statements in our own. Davidson holds that an untranslatable
conceptual scheme is a set of sentences in a language that contains words
and cultural concepts that cannot be translated into our language; any such
conceptual scheme is, therefore, unintelligible to us. According to Benjamin
Lee Whorf, to cite an example Davidson gives of someone who holds this
view, the Hopi language has no concept of time. So a wide range of our
conceptual scheme could not be translated into Hopi.12 We cannot count
a so-called scheme as a conceptual scheme without translating it into our
own language, but once we do so, our understanding of the original cannot
be a scheme of concepts radically different from our own. Later on in his
argument, Davidson allows that there might be local differences in beliefs,
which we might or might not want to call differences in concepts or differences in beliefs, but he offers no argument that forces us to see these local
differences as differences in conceptual schemes.13 Furthermore, Davidson
clearly avoids any extreme requirement to find agreement in belief no matter
what, saying that our translations are as far as possible, subject to considerations of simplicity, hunches about the effects of social conditioning, and
of course our common-sense, or scientific, knowledge of explicable error.14
In A Coherence Theory of Truth and Knowledge, Davidson distinguishes three ways in which the principle of charity constrains interpretation:

Wittgenstein and the Problem of Understanding at a Distance

1. We have no choice but to read our own logic into that of

the thoughts of the speaker.

2. When an interpreter associates a speakers sentences with

changing time and place, he takes the conditions that he
recognizes (as expressed in his own sentences) as the truth
condition of the speakers sentences.

3. With sentences the holding true of which are not so closely

connected to changing circumstances, the interpreter assigns
interpretations that preserve truth.15

103

The underpinning of these three aspects of the principle of charity rests

on Davidsons view that massive disagreements about the truth of sentences
between a speaker and an interpreter are not possible. Disagreements are
not possible, he says, because the interpreter will not have succeeded in
making the speaker intelligible if deviation is too great: Too great deviations from consistency and correctness leave no common ground on which
to judge either conformity or difference.16 Davidsons central point here
is that if the interpreter takes the speaker to disagree with the interpreters
own beliefs, these disagreements must, of necessity, be limited in scope. In
order for both to have incompatible beliefs (B1-BN) about the same thing
(X), they must share a wide range of interconnected beliefs about X in
order for the disagreements even to arise. To give an example, if you and
I disagree about whether the Prius is the best car for the environment, we
must already agree about the basic facts: that the Prius is a Toyota, that
Toyota is a car manufacturer, that the environment can be damaged, etc.
There may be some variations between the sets of beliefs related to the Prius
that each of us holds, but our beliefs must overlap enough for us to be
able to understand each other and even for us to disagree about whether
the Prius is the best car for the environment.
An important feature of the principle of charity, as Davidson understands it, is that it offers a necessary condition of interpretation. If a so-called
interpreter does not interpret charitably in Davidsons sense, she does not
interpret at all. He makes this point about those interpretations directly
related to the interpreters canons of logic: For the only, and therefore,
unimpeachable method available to the interpreter automatically puts the
speakers beliefs in accord with the standards of logic of the interpreter....17 But Davidson also makes the same sort of claim about those sentences of the speaker that are directly connected to changes in the speakers
immediate environment: Nor, from the interpreters point of view, is there

104

Whose Tradition? Which Dao?

any way he can discover the speaker to be largely wrong about the world.
For he [the interpreter] interprets sentences [as] held true...according
to the events and objects in the outside world that cause the sentences to
be held true.18 Davidsons view of interpretation is meant to show that
there is a necessary relationship between interpretation (understanding) and
ascribing true beliefs. Given Davidsons view of charity, it is not possible
to have succeeded in interpreting a speaker without doing so charitably,
without seeing the speaker as having largely true, logically coherent beliefs
and accurate observations about the world. We can capture the logic of
his point by saying that charitable interpretation, in his sense, constitutes
understanding. So to fail to interpret charitably is to fail to understand the
speaker and, thus, to have failed to interpret the speaker.
Nonetheless, within this fundamental constraint on interpretation,
there might still be some room for differences in interpretation. For example, consider Davidsons third category from the list given above, which I
will call the great middle ground of our sentences. Davidson says that
whereas W. V. O. Quine allows that sentences not describing immediate
goings-on in the environment can be interpreted at will, he wishes to
extend the principle of charity to these sentences as well and so to preserve truth even of these sentences. Davidson says, [I]t makes for mutual
understanding, and hence for better understanding, [my emphasis] to interpret what the speaker accepts as true when we can.19 But this extension
of the principle of charity puts Davidson on different ground in his basic
explanation and defense of charity. For whereas the first two aspects of
charity concern what makes interpretation and understanding possible at
all, this extension is justified in terms of what would make for better
mutual understanding.
If we follow Davidsons three-way distinction between parts of language in which charity is required for interpretation, we might draw the following conclusions when thinking about interpretation of classical Chinese
texts: We will need to suppose in our interpretations of classical Chinese
texts at least some basic forms of inference. So, if we decide that (shu)
means Book of History, we can also expect that it contains accounts of events
from the past, that it was written, that it may or may not be accurate, and
so on. We can also expect that in translations of sentences about perceivable,
changeable circumstances, for example, descriptions of a butcher gracefully
carving an ox, we will be able associate our sentences with the Chinese
sentences readily enough. (We will need to know something of the history
of Chinese butchering, however, to check for differences in tools, methods,
etc.) It seems to me that the greatest difficulties will come in Davidsons

Wittgenstein and the Problem of Understanding at a Distance

105

middle range, those sentences that do not directly express logical inferences
or do not directly describe the environment.
When we look again at Makehams example from Analects 1.2, we
confront ambiguity in the Chinese text concerning the character (wei),
as well as questions of how to understand (ren), translated variously
as morality, benevolence, good, virtue, kindness, humanity, authoritative
personhood, and so forth. In this part of language, Davidson says, there is
a real danger of papering over conceptual differences if we take charitable
interpretation to be required to maximize understanding.20 For example, if
we take Analects 1.2 (.. Filial piety constitutes the
basis of morality.) as a sentence we would hold true, those of us who are
liberal Westerners might have a hard time affirming this sentence as true,
and the sentences we might want to affirm in its place, for example
Freedom as long as we dont harm others is the basis of moralitywould
certainly not get the meaning right. My own understanding of the sentence
is this: Filial piety, that is, behaving in a way respectful toward ones parents
in accordance with ritual propriety, is the basis in the family for generally
knowing how to do what is morally required to maintain proper relationships. But even with this interpretation, which is close to saying that learning etiquette and practicing it in the family is the basis for learning how
conduct oneself morally in all of ones relations with others, we tend to get
a sentence many of us would consider false. It is clear that to translate this
vast middle ground of sentences, we need something different from agreement about what constitutes truth if we are to get proper interpretation.
Davidsons suggestion, that we look for explicable error, will not work
either. This suggestion, which undergoes refinement in his later discussions
of passing theories (as theories designed to explain the present utterance
of a person in context), appears to take up or at least be similar to Richard
Grandys development of the principle of humanity as an alternative to
Quines principle of charity. Grandy argues that Quines principle of charity
when strictly appliedQuine argues that we interpreters assign our own
sentences we hold true to the speakers held true sentenceswill sometimes produce bad translations. In some cases, Grandy argues, we do well
to impute beliefs and desires as similar to our own as possible, but that
will require in some cases opting for intelligible falsehoods over mysterious
truths. It is crucial, Grandy claims, to take into consideration the past history of the speaker and to understand the conditions of life in that culture
and time period that would be obvious to the speaker. Only with historical knowledge can we assume that what is obvious to us in the speakers
sentences was also obvious to the speaker. He gives the example of a person

106

Whose Tradition? Which Dao?

who has recently proved true a theorem that he and others earlier thought
was false. If the theorem prover asks another person about the theorem and
the other person says it is false, this is an explainable falsehood. Grandy
also offers cases of observation sentences, for which attributing explicable
falsehood makes more sense than attributing mysterious truth.21 Although
Grandy discusses a range of examples, including one archeological example,
all of his examples that depart from Quines principle of charity are ones
in which Grandy wants to attribute to others explicable error(s). That is,
they are cases where the interpreter takes a speakers truth as obvious, but
the interpreter uses that obvious truth to attribute explicable falsehood to
the speakers sentences.
But what of situations, such as in comparative philosophy, where
study of an alternative culture and its approach to and reflection on ethics
undermine ones settled sense of what is obvious? What about those cases
that require people to set aside their own beliefs and to imagine an alternative form of life and language-games in order to understand them? In such
cases, a principle of charity might then require the interpreter not to seek
agreement between his own beliefs and those of the alternative form of life,
as Quines principle of charity has it, nor would the interpreter need to
offer explicable error, as Grandys principle of humanity requires; instead, the
interpreter would be required to master new language-games to understand
the alternative form of lifes truths and insights.
The difficulty of settling on a correct translation of Analects 1.2 points
to the difficulty of finding sentences in English that do the same work
(convey exactly the same meanings) as the sentences in Chinese do. This
difficulty is evidence that charitable interpretation is not simply one thing
but is a family of requirements for understanding. Whereas the intelligibility
of anothers speech requires attributing some relatively logical inferences to
him of the sort we ourselves recognize as apt and recognizing the same range
of relatively obvious perceptual objects we recognize in the same contexts,
charitable understanding of the middle ground of uttered sentences leaves
aside such obvious truths and falsehoods in seeking intelligibility.
In the following section, I propose two requirements for a charitable
understanding of this middle ground of language of classical Chinese texts:

1. We require the proper correlation of others forms of life

and related language-games to our own and the correlation
of forms of reasoning and styles of explanation to our own.

For example, Confuciuss sentence about (ren) might fit better into what
we would call the language-game of pedagogy or child rearing than into

Wittgenstein and the Problem of Understanding at a Distance

107

the language of theories about ethics or human nature. Moreover, his style
of discourse may correlate better with that of psychological therapy than
psychological or philosophical theory. His recommendations about (ren)
may correlate better with our ordinary discourse about how to manage
relationships than with our philosophical discourse about whether virtue
or duty is more fundamental as a moral standard.

2. We require our best possible defense in our own terms

of our interpretations of classical Chinese texts, including
Confuciuss set of teachings.

Even if we can interpret charitably by means of this process of finding the

best fit between Confuciuss language-games and our own, we still will not
have charitably understood his sentences in one additional respect. We have
not yet figured out how to master those language-games and styles of reasoning well enough to make the best possible defense of his set of teachings
because we are using our own terms. However, we will require of ourselves
our best possible defense of our interpretations. I will next turn to a more
detailed discussion of how, in light of these two additional requirements,
charity proceeds to understand the middle ground of sentences.

Interpreting the Middle Ground

Although Wittgenstein and Davidson share some views about what makes
understanding across languages possible, Wittgenstein focuses on locating
similarities in forms of life, not similarities in sentences held true. Although
Wittgenstein requires that there be enough regularity for us to call something a language, he does not require that the regularity be the same as the
regularities of our home language.22 To understand others, says Wittgenstein,
requires that we see them as engaged in activities that exhibit activities
within our shared human form of life, but not as exhibiting it in exactly
the same ways as we do:
Suppose you came as an explorer into to an unknown country
with a language quite unknown to you. In what circumstances
would you say that the people there gave orders, understood
them, obeyed them, rebelled against them, and so on?
Shared human behavior is the system of reference by means
of which we interpret an unknown language.23

108

Whose Tradition? Which Dao?

Unlike Quines or Davidsons reliance on radical translationseeing

where the strangers we confront hold true some sentences and using those
instances as a basis for associating their sentences with those that we hold
trueWittgenstein focuses on characteristic forms of behavior. We know
from his insistence on this point that language is embedded in a wide
range of behaviors.
Consider the variety of language-games in the following examples
and in others:
Giving orders, and acting on them
Describing an object by its appearance or by its measurements
Constructing an object from a description (a drawing)
Reporting an event
Speculating about an event
Forming and testing an hypothesis
Presenting the results of an experiment in tables and diagrams
Making up a story; and reading it
Acting in a play
Singing rounds
Guessing riddles
Cracking a joke, telling one
Solving a problem in applied arithmetic
Translating from one language into another
Requesting, thanking, cursing, greeting, playing.
It is interesting to compare the diversity of the tools of language and of the way[s] they are used, the diversity of kinds of
word[s] and sentence[s], with what logicians have said about the
structure of language. (This includes the author of the Tractatus
Logico-Philosophicus.24)25
Interpreting a strange language requires, then, associating the strange
language and behaviors wrapped up in it with similar but possibly somewhat
different uses in our home language. Although there is a danger of ethnocentrism in this method, there is no requirement that ones interpretations
be ethnocentric. Wittgenstein is asking us to look for common behaviors
exhibited, for example, in obeying an order, giving an order, and rebelling
against it, not behaviors merely identical to our own.
In a situation of encountering a strange culture and translating its
language into our own, we need to fulfill twin requirements: (1) locate common behaviors and simultaneously, and (2) recognize that these behaviors

Wittgenstein and the Problem of Understanding at a Distance

109

may be no more than similar to our own. What role would this approach
have in translating classical Chinese texts into English?
This principle would require that we locate in the imagined world of
the Analects the various common forms of behavior exhibited in that world
and use that as a starting point for understanding the language of the text.
It would, for example, be important to determine in what ways Confuciuss
behaviors toward his interlocutors exhibit raising questions, making suggestions to them, making criticisms of them, forming hypotheses, giving
arguments, constructing definitions, clarifying moral ideals, and so on. We
need to locate his sentences within the common behavior of mankind while
recognizing that behaviors we categorize as criticisms, clarifying moral ideals,
etc., represent no single thing but a family of more or less similar behaviors.
Moreover, we need to find rough correlates between terms he uses and our
own terms so that we can be clear, for example, in interpreting his suggestions as to what he is asking his interlocutor to do. But in doing so, we
need to be sensitive to both similarities and differences.
Following Wittgensteins approach, we maximize our understanding of
the language and practices of unfamiliar cultures by finding corresponding
similar practices and language of our own to serve as a basis of understanding
and translation. For example, we might want to translate (li) as etiquette.
But if we do so, we miss the importance (li) has for Confucius and other
teachers of tradition (Ru). In fact, Confuciuss teachings might seem
plainly crazy if we identify li with etiquette books that were common in
the United States in the mid-twentieth century. So translated, we might end
up ascribing to him sentences such as, Etiquette is central to the whole of
morality, which seem to most of us plainly false. Here we have an anomaly.
So the challenge is to figure out, given what he says about (li), what the
concept corresponds to for us. We might want to start with forms of behavior
showing respect for people in roles of legitimate power and authority over us.
Consider the behavior we teach and expect of children toward such people,
for example, their teachers, coaches, priests, rabbis, or ministers. Other things
being equal, we expect children to obey such people, thank them for help,
act politely toward them, etc., as signs of respect for someone in such a role.
This correspondence gives us a handle on how to understand (li) in some
contexts and possibly how best to translate the term.
A second set of translation issues arises when we examine Confuciuss
discourse with others. There is a set of what look like Socratic questions
that introduce various passages in the Analects: (huo wen ren). We
might translate this as Someone asked, What is (ren)? We might think
that this is a request for a definition of the concept of (ren). But if we
understand the question in this way, we end up with various conundrums.

110

Whose Tradition? Which Dao?

There are very few, if any, answers to these sorts of questions that even
look like definitions. In this case, a fairly obvious translation, arising from
word-for-word dictionary correspondences produces a problem in how to
understand the text charitably. What else might the question mean?
We get a better understanding of the meaning of this sort of question if we understand the question to mean, Someone asked about how
he might behave to get closer to being (ren)? The argument I would
offer for this interpretation is that the answers that are typically given to
these sorts of questions appear to be answers to this question, not to the
Socratic request for a definition.
Based on these two examples alone, it is clear that we need to locate
the language of the Analects in its home language-games, its key terms,
and in terms and forms of discussion. If we follow these two suggestions,
how would this approach impact our interpretation of the Analects 1.2
(..)?

Basic Meaning

The virtues of respectful
filial and fraternal
love and respect
politeness within the

family (: filial piety,

devotion to ones living

and dead forbears)
(: fraternal love,

devotion to ones eldest
brother)
As for

As for N as such

Presumably, I suppose;
perhaps; apparently

constitute

count as N, be an N

practicing

Do; commit; be engaged

in a contextually
determinate matter
or undertaking

Humane, complete moral
goodness, relationally
responsible

Humaneness, kindliness,
kind-heartedness;
benevolence, Goodness
as a moral value

What is basic and

crucial; the basic and
crucial feature;
basic features

Sentences final particle

Wittgenstein and the Problem of Understanding at a Distance

111

From this chart, it is easy to generate a range of possible interpretations of this sentence, using what Davidson calls our prior theory of
the meanings of these characters. The difficulty is to determine the most
reasonable passing theory, which attempts to settle on the most reasonable
interpretation of this sentence in context and what its author might have
meant by it. Beyond the ambiguity of the character (wei), the difficulty
of settling on the meanings of (ren) and (ben) also presents special
problems. For something to be basic or crucial is for it to be basic or crucial
in some respect relative to some specifiable goal. So being a male is crucial
for defining bachelor, and being male is a necessary condition of being
a bachelor. And exposing children to art at a young age is crucial for their
developing a refined artistic sensibility. How can we tell in this specific
context what (ben) means?
It will be important in resolving this question to clarify the context
of this statement as well as the enterprise/practice in which it is embedded. In short, using Wittgensteins term, we need to be able to determine
the language-game in which this statement plays a role. Is this statement
being used to analyze a concept, guide moral cultivation, present a theory
of moral development, or what?
Out of context, this statement could be used to do any of these
things. But there is little evidence that Confucius had much interest in
defining concepts. Moreover, although he does offer insights about stages of
moral development and clarifies distinctions between exemplary and nonexemplary persons and conduct, these do not take on the form of theories
of moral development. For example, the clearest account of stages of moral
development can be found in his autobiographical comments about his
own development. And the clarity he offers about ren) often takes
the form of making specific comments about individual conduct, not the
development of general, systematic theories. Without offering any account
to back it up, Confucius appears to be a functioning moral and pedagogical particularist.
Reflection on dao, the way to conduct ones life, plays an important
role in Confuciuss project but only, as he himself indicates, in the context
of study and work to improve oneself in terms of dao-constituting norms
(Analects 2.15). If we take this key claim as a clue about the overall genre
of the text of the Analects, we might classify it as a combination of recommendations about how to improve oneself morally and reflections about
topics central to that enterprise.
In light of this characterization, how should we understand Analects
1.2? If we think of 1.2 as a reflection on or out of study designed to improve
oneself in terms of dao-constituting norms, then I suggest the following

112

Whose Tradition? Which Dao?

understanding of (ben): It means crucial for improving oneself in terms

of dao-constituting norms and is a result of or instance of reflection on
the process of such improvement. Although the statement seems general,
we might also understand it as arising out of reflection on Confuciuss
process of self-cultivation and, in light of that, his reflection on the lives
of others. So we could contextualize the statement as follows: Based on my
own reflection on my and others process of self-cultivation, it strikes me
that filial and fraternal devotion and respect are crucial for (getting closer
to practicing, cultivating) humanity in the form of relational responsibility,
understood as complete moral goodness.
What I have done in this interpretation is try to specify the meaning
of this statement by drawing a key distinction between theory and practice
to get a more adequate account of what this sentence means. In terms of
forms of life and language-games, we could characterize this specification
as follows: Analects 1.2 articulates a norm based on Confuciuss reflection
on what is crucial for child rearing and/or self-cultivation. His reference to
filial and fraternal love and respect not only refers to practices of family
ritual, (li), practiced during the Warring States Period of China,26 but
(li) also regulates and offers proper ways to express respect, love, sadness, etc., as follows:
Being reverential without complying with ritual actions tends to
be reverential in vain. Being cautious and not complying with
ritual actions tends to be cowardly. Being brave and not complying with ritual actions tends to be rebellious. Being upright
and not complying with ritual actions tends to be too hurried.
(Analects 8.2)
Moreover, Confucius understands the ideal of (ren) in relation to
those practices of ritual, though (ren) also transcends the rituals in the
sense that he treats (ren) as partially embodied in the conduct of some
folks and in the canon of ritual propriety, but (ren) is not found only
in these actions. For example, he allows that ritual might be reformed, so
(ren) cannot be fully expressed in contemporary ritual forms. He also
holds that contemporary practices embody a falling away from (dao). So
although we cannot understand (ren) separately from (li), no ones
behavior or no concrete, contemporary (for Confucius) set of (li) need
fully embody the ideal of (ren).
Even though Confucius treats (ren) both as largely embodied in
(li) but in its ideal form independent from (li) and from historically real
persons and their conduct, this latter aspect of his approach to (ren) is

Wittgenstein and the Problem of Understanding at a Distance

113

also embodied in Confuciuss methods of inquiry and self-examination: his

practice of daily self-examination, his willingness to treat others as teachers, his commitment to showing appreciation for the goodness aspects
of a persons conduct without treating those aspects as fully good, and his
cognitive and moral humility, expressed in his steadfast refusal to consider
himself to be a sage.
We might then ask, what sort account of the relation of (ren) and
(li) does Confucius advocate? His form of reflection avoids metaphysical speculation; nonetheless, his view of the independence of (ren) from
full embodiment suggests that he is some sort of realist. In light of my
characterization of Confuciuss basic project of self-cultivation with reflection, we should, however, avoid attributing to him any theoretical views of
metaphysical reality that are not part of his reflective self-cultivation project.
Instead, his so-called realism shows up in his willingness to criticize present practices and to engage in daily self-critique, as well as in his refusal to
consider the ideal of (ren) to be completely embodied in any persons
describable conduct.
We might, following my argument in Chapter 2, refer to Confuciuss
attitudes as embodying a realistic spirit without metaphysical realism.
This phrase, which I use following Cora Diamond, to characterize Wittgensteins later approach to ethics, captures the way in which he rejects
the metaphysical realist view that distrusts non-metaphysical methods of
distinguishing between the real and the unreal and stipulate[s] that, if our
concept reality is in fact legitimate, there must be more to it than such
methods reveal.27 In contrast, Wittgensteins realistic spirit exhibits itself
in accepting that our non-metaphysical methods for distinguishing between
real and unreal are adequate for giving content to that distinction. Whereas
Wittgensteins realistic spirit is a form of what Paul Ricoeur calls a second naivet, cultivated in the face of a loss of innocence, Confuciuss is
a first naivet.28
Thus, what gives Analects 1.2 its meaning is precisely its embeddedness
in a set of ritual/self-cultivation practices and its independence from those
practices, which arises from Confuciuss methods of reflection and selfexamination. Outside of Zhou (or Han) Dynasty practices and Confuciuss
reflective practice, that sentence loses its specific meaning.29 In addition to
Analects 1.2s meaning being dependent on the practices embeddedness, its
intelligibility depends on the interpreters familiarity with those practices,
either through actual engagement in them or through an imaginative identification with such an engagement. In Confuciuss world, the relationship
between parents and child is sacrosanct; it is meant to be a harmonious
relationship between those in superior and dependent social positions, in

114

Whose Tradition? Which Dao?

which both ought to express care and respect. Children grow up learning how to navigate in this sort of social world and develop a sense of
self and responsibility that fits this set of cultural arrangements. The ideals
and concepts that undergird this set of arrangementstradition, authority, respect, reciprocity, and responsibilityare all ideals that contemporary
North Americans, for example, recognize. So we can readily understand this
rather different set of norms in general terms familiar to us, and we can
even affirm these Zhou (or Han) ideals in their most general forms, but
the specific cultural practices and related normative statements articulating
them, such as Analects 1.2, would be hard for us to hold true.
Davidson might challenge my argument by claiming that even I would
have to admit that Zhou Dynasty thinkers and I would have to agree about a
great deal for me to be able to locate our normative disagreements. Although
I dont disagree with this claimin fact, it is encoded in the Wittgensteinian view about the possibility of translation of the language of a foreign,
strange culture into our ownI would argue that the intelligibility of the
details of the language of that culture depends on our familiarity, imagined
or real, with the forms of life in which these sentences are embedded. Even
if that familiarity is connected to being able to associate Analects 1.2 with
some very general sentences that we North Americans hold true (respect for
parents is good, responsibility is good, mutual benefit is good, etc.), what
is at stake in our understanding Analects 1.2 is not just these general claims
on which we easily agree.
So the problem of understanding Analects 1.2 is not to associate Confuciuss statements with statements we personally hold true in the twenty-first
century or to explain his errors but to clarify the Confucian form of life. We
understand Confucian life by understanding how Zhou ritual practice is connected to filial love and respect, how the rituals and feelings together serve as
the basis for and model of good conduct in general, and how people would
act and understand themselves and their relations to others in this form of
life. That understanding requires that we hold as true some sentences that
Confucius holds as true, butand it is here that Davidson goes wrongit
also requires making Confuciuss language and related practices an imaginable
option for us, mastered in imagination if not in reality. That is, we need to
be able to make sense of how someone, engaged in relevant Confucian, but
foreign (to us) practices, would use these sentences and how he would hold
them to be true. Otherwise, we cannot understand them.
Supposing that my accounts of the meaning and intelligibility of Analects 1.2 and other such Analects sentences are correct, then we would have to
conclude thatpace Davidsoninterpreting these sentences does not simply

Wittgenstein and the Problem of Understanding at a Distance

115

require associating our held true sentences with Confuciuss held true sentences
or explaining his errors. Instead, by locating the place of those sentences in
a set of practices different from our own, made understandable in terms of
both their similarities and differences from our own linguistic practices, we
need to learn how to use and affirm as viable in that culture some sentences
that we are personally and in our own culture inclined to hold as false.30
We do not need to require holding these foreign language-game
sentences to be true to our own values in order to understand them. We
simply need to be able to imagine how we and others engaged in that
language-game could hold them to be true in a language-game similar to,
but different from, our own.
Following this analysis of what it takes to understand Analects 1.2, we
must infer that the principle of charity requires that for this and the rest
of Davidsons so-called middle ground of sentences, we locate those claims
in the language-games where those sentences get their meaning, understand
the general claims we and practitioners of those language-games both hold
true, master the foreign language-games, and imagine how we could hold
true those middle ground sentences in light of that mastery, despite our
inclination to hold them false.
Even if we find a charitable interpretation of our middle ground
sentences, like Analects 1.2, when we use this language-game approach, we
will, however, not have resolved all of the potential ambiguities facing us.
For example, locating the language-games, in terms of which , , and
get their meaning, does not resolve the question of the meaning of the
character (wei). I will next focus on the relation of the requirements
of interpretive charity to this character (wei), which means either to
be or to do.
In my example of the builders language-game, I argued that a person
should follow and act on the interpretation that, in that language-game,
makes best sense of his foremans behavior. I argued that, in this context, this
is simply to interpret charitably. But in addition, this interpretive principle
is a commonplace in textual interpretation, that a charitable interpretation
of a text is better than one that is not charitable. The principle of charity requires that, other things being equal, in the face of well-supported
interpretations, the interpreter should adopt the one that will maximize the
sense and reasonableness of the text. That does not mean that the interpreter
should override well-defended interpretations to make the sentences in the
text true or reasonable no matter what. Instead, it means attributing those
views to the text that, consistent with possible defensible interpretations,
make those views the most reasonable of the possible alternatives. Here I

116

Whose Tradition? Which Dao?

assume that words in the text (with their cultural associations) provide the
most basic evidence for a correct interpretation. That is, the definitions
of a texts words, even when they include the correct anthropological and
historical associations, do not provide sufficient evidence to prove that we
know for sure what the speakers or authors intended meanings were. Yet,
it is precisely because of this under-determination of interpretation of text
that it is both possible and necessary to invoke the principle of charity.
That is, only when two competing interpretations are supported by the text
(and other relevant evidence) can the principle of charity be invoked. For
example, consider once again three possible interpretations of Analects 1.2:
. .

A. It is (xiao ti) filial piety and deference to elder brothers

that are (wei) the foundation of humanity [ (ren)].

B. It is (xiao ti) filial piety and respect for elders that

(wei ren zhi ben) are the fundamental means of
putting humaneness into practice.

C. It is (xiao ti) filial piety and respect for elders that

(wei ren zhi ben) are the fundamental means of
completing humanity.

Even if we lack historical evidence to determine conclusively who

meant what by this sentence, we can address this indeterminacy by tackling
the question of which of these interpretations is the most charitable. By
answering that question, we will know what Analects 1.2 means. Of course,
that will require a philosophical debate and a winner. But life is, after all,
tough. We cannot even begin to ask what the meaning of a text is without looking at the history of its genesis, the language-games in which it is
embedded, and the interpretation that attributes to it the most reasonable
set of views.
To interpret this sort of text charitably is, then, to determine the
most reasonable version of the claims the text makes on us in light of historical evidence and in light of practices of interpretation and translation.
By limiting our interpretations to those backed by historical evidence, we
avoid the danger of reaffirming our own prejudices by projecting onto the
text interpretations that offer no substantial challenge to our own settled
cultural and personal perspectives. When we impose on ourselves a more
rigorous standard, as we study several possible interpretations and select the
most reasonable ones, we offer the best chance of understanding the texts

Wittgenstein and the Problem of Understanding at a Distance

117

claims on us, i.e., insights that we might not yet understand about how
life can be lived better.
Charitable interpretation primarily plays a role in texts for which an
authors or interpreters statements cannot be immediately understood, but
the author or interpreter is not available or willing to explain the texts
intended meanings. When the speaker can explain his meanings, interpretation is not needed. According to Ian Hackings view of interpretation,
which is meant to capture ordinary language usage, interpretation of a
sentence is required only when the hearer lacks immediate understanding,
and interpretation is not required if the speakers explanation is forthcoming and successful.31 When a speaker offers a successful explanation, there
is no need for interpretation. We typically accept the speakers explanation
unless the explanation itself is not understandable or is in conflict with other
things the person has said, either in general or concerning the statements
in need of explanation.
When the speaker of a statement is not available and the statement is not immediately understood, we must interpret it to understand
it. Although the speakers explanation is presumed to be correct, someone
elses interpretation of it does not have the same presumption. This gives
rise to the need to evaluate alternative interpretations. Of those interpretations that are consistent with (1) the text and (2) other relevant historical
evidence, including (3) the relevant language-games, the interpreter and his
audience must select an interpretation that can, as far as possible, take the
place of the speakers explanation. Typically, this will be the most charitable
interpretation, in the sense of being the interpretation that a defender of the
statement would offer as the best way to defend the position from actual
or possible criticisms.
Because we often treat a charitable interpretation as standing in for the
absent speakers explanation of meaning, we often speak of the charitable
interpretation as the speakers meaning. We presume that if the speaker were
available, the person would, as far as possible, offer the most defensible, that
is, most charitable explanation of the original meanings of sentences. So
we can treat the interpreters charitable interpretation as standing in for the
speakers explanation. In fact, the interpreters ability to present the position
charitably gives him the authority to speak for the author, with the right to
claim that his interpretation, while merely possible given the linguistic and
historical evidence, nevertheless succeeds in capturing the authors meaning.
What interpretive charity aims at is making sense of texts speech, as
written by authors and spoken by speakers. For example, we want to make
sense of the Analects, its teachings, and of Confucius, its main character. If

118

Whose Tradition? Which Dao?

I am right, the notion of making sense is a complex ideal not reducible to

Davidsonian maximization of agreement or Grandyan explication of error.
It requires more. In addition to anthropological knowledge that gives rise
to insightful correlations of language-games, it requires mastery of the target
texts styles of reasoning, including mastery of how best to speak for the
text and its authors meaning in light of actual and possible criticisms of it.
Interpretation of texts such as the Analects can be assessed in terms of these
different aspects of interpretive charity. And though a person might be said
to understand the text in merely one of the ways I have been discussing,
in the context of textual interpretation, all of them are required. That is,
all of the ways represent aspects of textual interpretation, without which a
person can be said not to have understood the text or its authors meaning.
To be schooled in understanding a text such as the Analects, then, is
to be schooled in a set of reading practices involving historical investigation
and relying on the best historical work that has come before. In the case
of competing interpretations, this practice requires providing accounts of
what the text means and willingness to submit those judgments to critical
examination. This process may result in changing of minds or, in some
cases where no decisive evidence can be offered, no difference at all, but
there is no need to perceive, as Makeham and Gardner do, an essential
gap between scriptural and historical meaning. The text of the Analects
means what it means. Interpreting by appeal to (historical) evidence and
charitable interpretation are two essential moments of determining a texts
meaning, just as they are two essential moments of determining an utterances meaning in the builders language-game. Rather than crippling the
interpretive process, the paucity of historical evidence is enabling, for it
opens the door to the principle of charity. And the principle of charity,
rather than undermining the objectivity of interpretation, is a necessary part
of what it means to determine the meaning of a text. The claim that there
can be historical meanings of the sentences of the Analects independent
of normative meanings is an illicit dualism that we should drop. When
we do so, we can get back to the work at hand, determining the meaning
of the sentences of the Analects.

Conclusion
Even though I have argued that the historical record of Confuciuss teachings
is too weak to support claims that passages in the Analects represent his actual
speech and teaching, there are two key features of this approach that should be

Wittgenstein and the Problem of Understanding at a Distance

119

congenial to contemporary philosophers interested in Confuciuss views. If my

argument works, it saves us from historicist nihilism and relativism and gives
us a way of thinking about how to discuss the meaning of Confuciuss teachings and the meaning of the Analects sentences. Moreover, my view requires
that we reject the claim that historical investigation of texts like the Analects
can proceed without philosophy. For the necessary under-determination of
meaning of these texts by evidence requires that we introduce a Principle of
Charity to use in our interpretation of the texts. As we already do this for
texts of all sorts, not to use this principle in this context would be to treat
the Analects as if it fell outside the requirement of charitable interpretation.
It would seem to be, at an intellectual level, a form of slight, of disrespect.
This interpretive project weds historical investigation with philosophical argument. For if we think that one possible interpretation is better than
others, this will rest on a philosophical argument that, rather than being an
extrinsic activity, is essential to discovering the meaning of this text. Indeed,
we should understand the Confucian commentarial tradition itself as engaged
in this very sort of activity. A key difference between contemporary Confucian researchers and traditional commentators, however, is that we do not
need to labor under the restrictive burden of seeing the Analects Confucius
as a real person, let alone as a sage who uttered only the truth. We can see
him as a character who represents a way of thinking and learning about the
proper ways to realize our humanity. Our task, as interpreters, is to take this
character seriously, in much the same way we feel required to do that with
one another. I need to know what you say, ponder its meaning, and discriminate among those understandings that would be more or less charitable.
If your form of life differs too much from my own for me to get a handle
on it, I need to receive instructions from the ground up. And with a text
such as the Analects, we do this under the real possibility that this text and
its main character may make claims on us, on the ways we think about and
live our own lives. If we take up my approach to reading the Analects, we
will have to face the following weighty philosophical question:
Of the three different interpretations of Analects 1.2, which one presents the best account of (ren)?

A. It is (xiao ti) filial piety and deference to elder brothers

that (wei) ARE the foundation of humanity [ (ren)].

B. It is (xiao ti) filial piety and respect for elders that

(wei ren zhi ben) are the fundamental means of
putting humaneness into practice (237).

120

Whose Tradition? Which Dao?

C. It is (xiao ti) filial piety and respect for elders that

(wei ren zhi ben) are the fundamental means of
completing humanity.

Unless we give, beyond the sort of historical evidence that leaves the
meaning underdetermined, a successful philosophical defense of our selection, we will have fallen into the mere subjectively selected readers responses
that Gardner embraces and Makeham is worried about. And unless our
arguments are successful, our insistence on our preferred interpretation will
be arbitrary, and our Confucius would be a mere figure we use for our own
projections. But those responses would mean we were not taking Confucius
or the Analects seriously; instead, we would be engaging in a kind of intellectual self-indulgence. No serious follower of Confucius should welcome
this arbitrary approach, not only because of its failure to take Confucius
and the Analects seriously, but also because of the sort of small-mindedness
such an approach encourages by allowing interpreters to believe that beyond
their and others personal Confucius, there exists no real Confucius to teach
and challenge them. Even those interpreters who do not consider themselves
followers of Confucius must be open to the possibility that this character
Confucius and the Analects may make claims on them. If they admit this
much, they should welcome the best accounts of what this character has to
offer. Even if we reject Confuciuss teachings about how to lead our lives, we
need to have understood what his views are and what his sentences mean.
Understanding requires agreeing about what is obvious (logical inferences
and truths about perceptual beliefs), finding ones feet with a text or authors
language-games, putting those language-games into practice, and putting the
text in its best light in the face of possible criticisms. In this context, such
an understanding constitutes making the texts meaning clear.

How to Be a Confucian Pragmatist

without Losing the Truth

After we came out of the church, we stood talking for some time
together of Bishop Berkeleys ingenious sophistry to prove the nonexistence of matter, and that every thing in the universe is merely ideal.
I observed, that though we are satisfied his doctrine is not true, it is
impossible to refute it. I never shall forget the alacrity with which
Johnson answered, striking his foot with mighty force against a large
stone, till he rebounded from itI refute it thus.
Boswell, Life of Samuel Johnson, 248

Introduction
In Chapter 1, I have argued that Confuciuss thinking and practice as represented in the Analects offers a fundamental commitment to intervention
in the conduct and reflection of his interlocutors but little interest in theorizing about dao and its constitutive ideals and their epistemological and
metaphysical bases. He does, however, invoke these ideals in the course of
intervening in the behavior and reflection of his interlocutors. Recent commentators who have noticed this feature of Confuciuss practice have also,
problematically in my mind, tended to see him as not interested in truth
and have mistakenly associated this general view with later Wittgenstein.
This view, attributed to any thinkerthat by focusing on practice, we can
or should or have to jettison appeal to truthsuffers serious difficulties.
Although not every sentence we utter is a candidate for being true or false,
very many of them are. And even if some simple language-games do not
have devices and second-order semantic concepts for attributing truth to
their sentences, there are ways for truth to be affirmed. For example, even
in Wittgensteins simple builder language-game (discussed in Chapter 4),
121

122

Whose Tradition? Which Dao?

although there is no sentence that says, It is true that he brought a slab,

in that game some things are slabs, and some things are not. And even
if Bring me a slab! is an imperative, the builder either successfully follows the imperative or he does not. That is, it is either true that he brings
the slab or it is not. A conscientious builder is interested in truth, if only
because he is interested in making sure that his actions fulfill the criteria
of the terms constitutive of his practice.
Whats more, Confuciuss linguistic practice is substantially more complicated than Wittgensteins builders. He is engaged in the question of
whether or not specific people are (ren). He never says that the answer
to this question does not matter or has no meaning. Instead, he is interested
to show that some types of action do not rise to the level of being (ren).
That is, he is interested in showing that some people are not characterized
as being (ren). Or, given the ways in which Confucius is characterizing
them, he is interested in pointing out that we do not yet know that they
are (ren). It is, I would contend, impossible to understand his concerns
in these passages apart from thinking of him as interested in the question
of whether it is true that certain types of conduct are sufficient for (ren).
And that makes him concerned with, making judgments about, truth. There
is no other way to make sense of these (ren) passages.
The basic motivation behind denying that Confucius is concerned
with truth seems to rest on the insight that Confucius has none of the
philosophical interests of Western metaphysicians, and thus, these philosophers conclude, he is not interested in the question of the nature of truth as
a subject of metaphysical and epistemological questions. But it is a mistake
to argue that because a thinker is not interested in the metaphysical or
epistemological questions of the nature of truth, he is thereby not interested
in truth. Of course, one might try to argue that only someone interested in
such questions is really interested in truth, but such arguments are questionbegging and tendentious. Most residents of Taipei and I can perfectly well
sustain full interest in the question of whether the Taipei 101 building is at
the Taipei City Hall Station, and so we care about whether this statement is
true, while we may have no interest in related metaphysical questions about
the skyscrapers essence or reality. There is no tension between this ordinary,
everyday interest in truth and a lack of interest in metaphysical questions.
In fact, because of his focus on cultivation of dao-sensitivities, Confuciuss thought is what I call theory indifferent. That is, he is not interested
in the questions we typically associate with metaphysics and epistemology.
But this approach leaves his work vulnerable to misinterpretation through
the lenses of different types of philosophical theories. In Chapter 6, I will
argue against Zhu Xis metaphysical interpretation of Confuciuss teachings

How to Be a Confucian Pragmatist without Losing the Truth

123

in the Analects. In this chapter, I argue against what is a contrasting, antimetaphysical interpretation. Both suffer from the same problem of distorting the text in the name of offering a version of the Analects that rests on
a philosophical theory that I view as problematic. In Chapter 4, I argued
that interpretation of the Analects depends on the principle of charity; this
view ultimately takes on the philosophical problem: Which of several possible, competing interpretations is more reasonable? Just as I find Zhu Xis
metaphysical interpretations of the Analects implausible, I find anti-realist
accounts of truth attributed to Confucius equally unreasonable. And because
I think that the Analects-type accounts of ethical cultivation do not require
either of these types of theories to make sense and are significantly better
off without them, I argue that neither of these accounts represents Confuciuss views.
But before I proceed to develop that argument in relation to these
anti-truth accounts of the Analects, I would like to develop a partial explanation of why this sort of anti-truth view can seem plausible. One line of
reasoning that can make this view seem plausible comes from the following
premises:

1. Confucius (or any other early Chinese thinker) shows interest

primarily in modifying conduct.

2. Confucius shows an interest in impacting conduct but does

not show interest in debating or pondering metaphysical
questions about what constitutes truth; therefore, he has no
interest in the ideas that constitute Western metaphysics,
which require acceptance of a substantive, theoretical view
of the timeless essence of truth.

3. If one rejects that there is a timeless essence to truth, then

one must reject the claim that there is truth simpliciter.

4. The best possible explanation of a thinkers lack of interest

in Western metaphysical accounts of the truth is that
the individual has adopted or operates with a worldview
inconsistent with Western metaphysics.

5. So, the best way to interpret Confucius (or any other early
Chinese thinker) is in terms of this anti-Western-metaphysics
worldview.

6. This means that we must understand Confucius (or all early

Chinese thinkers) as rejecting the timeless essence of truth,
as rejecting truth simpliciter.

124

Whose Tradition? Which Dao?

In what follows, I will argue against premises 2 and 3. Stated boldly,

premises 2 and 3 are part of a powerful philosophical picture that requires
thinkers in general, no matter what the context, either to affirm or deny
any philosophical claim. As philosophers, we can, for example, be tempted
to think that everyone either affirms or denies ethical realism. Thus, if we
can show that someone does not affirm ethical realism, his non-affirmation
proves that he must reject it. However, this set of possibilities leaves out
another possibility that can seem hard to grasp: a person does not need to
take, implicitly or explicitly, a stance on either of these theories.
One way to make sense of this is to consider the fact that we do not
formulate or take an interest in every possible statement that we could form
with the stock of concepts we operate with. We limit ourselves to considering questions that are salient and possible answers to salient questions. But
if, for a particular thinker, the question of whether some proposition p is
true is not salientthen we cannot claim that he or she thinks either p
or not-p. The question may not have come up for this person, and what
the person explicitly believes may not require an answer to this question.
So we might not be able to infer an answer from the beliefs we know the
person does entertain.1
What makes a question salient? I doubt that it is possible to give a
complete analysis of the question of salience, but it should be possible to
indicate some conditions that cause questions to lack salience. For example,
if a person A cannot understand a question, that question will not be salient
for A. Here we might distinguish two types of cases: (1) one in which the
question has not been given a clear meaning, and (2) one in which it has
been given a clear meaning in the immediate context but was not one
grasped by A in this context. We can also imagine two additional categories:
(3) one in which the question makes sense in some context, c2, but not in
the immediate context, c1, yet a person in c1 takes the question to be salient
for him in c1, but it isnt really meaningful to him in this context. (That
is, he is just confused.) Another type of case (4) would be a question that
is meaningful to a person only in a particular context but not important
or not very important to a person.
Wittgenstein offers numerous examples of types 1 and 3. Indeed, a
central feature of his method is to expose philosophical questions and related
claims as nonsense. In the Tractatus, he claims that, in philosophy, we can
fall into the illusion of thinking that statements, such as, The world is all
that is the case or In the world, there is no value have meaning when
they do not.2 In his later writings, we also get numerous examples of type

How to Be a Confucian Pragmatist without Losing the Truth

125

3. For example, in On Certainty, he gives the example of a person raising

the question of whether he really has hands:
The idealists question would be something like: What right
have I not to doubt the existence of my hands? (And to that
the answer cant be: I know that they exist.) But someone who
asks such a question is overlooking the fact that a doubt about
existence only works in a language-game. Hence, that we should
first have to ask: what would such a doubt be like?, and dont
understand this straight off.3
In most contexts, this question makes no sense. That is, we dont know what
sense to make of it. This question does not arise in ordinary contexts, and
so while we can make some sense of the question for other contexts, it makes
no sense in this context. So for example, in a situation where someone has
returned from war and is lying in a hospital bed and his arms are under
a sheet, on waking, he might wonder whether he still has hands. The fact
that a question makes sense in one context does not guarantee that it make
sense in other contexts.
Case type 2, of there being a clear meaning to a question, but not
one the person grasps, is one that occurs often. In any situation in which a
novice has not mastered a set of concepts that his teacher and others have
mastered, questions will come up that have a clear meaning, but the novice
does not understand them.
Even in those contexts in which a sentence makes sense, it might lack
salience for a person because of its lack of importance to that person.4 So,
for example, using Dretskes example, we might claim that we know that
a zebra at the zoo is a zebra despite it being possible that it is a cleverly
disguised mule. Although it is possible that the animal is a mule, this
question does not come up in ordinary contexts. And even if the question
does come up, nothing important may rest on the answer we give to it.
So, in ordinary contexts, we could claim that we know that it is a zebra.
The question of whether it is a cleverly disguised mule, while having meaning, would not be salient. Absent some special reason for thinking that
the possibility is a live one and the resolution of the question mattersa
sweepstakes might give me a $1 million if I get the answer rightthe
question is not salient for us.
When I am claiming that the question of the nature of truth or the
minds relation to the world was not salient for Confucius, I mean to sug-

126

Whose Tradition? Which Dao?

gest that it was not salient at least in the type 4 way. But it may also be
that it is not salient in the type 1 way. For example, consider Analects 5.13.
. . . .
.
Zigong said: The cultural ornamentations (speech and conduct)
of the Master: they can be heard. As for our Masters words on
nature and the heavenly way, we cannot hear them.5
This passage, like others akin to it, such as Analects 9.1, 6.22, and 7.20,
causes interpretive problems. For example, we find a passage on the nature
of people (xing) at 17.2 and one on the Mandates of Heaven
(tianming) at 2.4. We are compelled to draw some plausible distinction to
try to make sense of 5.13. I am inclined to understand this passage as an
indicator, for the writer, of which topics were centrally important for Confuciuss project. The writer may or may not have been aware of the conflicting
passages, or he might have been aware of them but wanted to indicate their
lack of central importance. One way to do that was to indicate what sorts
of things Confucius communicated to those around him. Even though this
might seem descriptive, it can also be normative. Sentences like, Wise men
use words cautiously, even though descriptive in form indicate a norm.
Similarly, given Confuciuss role in early Chinese thought, especially for his
disciples, descriptive statements about what Confucius talked about would
likely be used to indicate a norm. But what would be behind it? We might
speculate that what is behind it is that discussion of these questions would
not further the central goals of his project, cultivation of a life lived out of
dao. And we get some confirmation of this approach from Analects 11.12:
. . . . . ..
. .
Zilu asked about spirits and gods. Our Master said, You havent
yet been able to appropriately serve human beings. How could
you be qualified to serve the ancestors spirits? Zilu further
asked, Dare I ask about the problem of death? Our Master
replied, You havent understood the problem of life, so how
can you understand death?
This passage might be read variously: as a critique of Zilu or as a more
general recommendation about what comes first and is, therefore, most

How to Be a Confucian Pragmatist without Losing the Truth

127

important on the path toward cultivation of dao. I argued, in Chapter 4,

that these ambiguities need to be resolved through appeal to the principle
of charity. In what follows, I will assume that Confuciuss response to Zilu
does not just represent a particular critique of Zilu but gives a general
view about what is central to Confuciuss cultivation-of-dao project. This
approach appears partly confirmed by the limits he places on reflection in
2.15. For present purposes, my point is that if we grant this view, we have
a basis for thinking that certain questions about human nature, such as the
Ways of Heaven, death, and spirits, were not salient for Confucius given his
focus on ethical intervention, which required first of all changes in behavior.
We do not have clear evidence, however, that the questions listed in 11.12
were, then, to be discussed later, reserved for advanced students, as it were.
We might speculate that his view was that by taking care of these basic
questions of how to live life well, these abstruse questions would drop out.
Once a disciple has a fairly well-developed knack for how to serve human
beings, he also has a knack for how to serve spirits.6
If these questions are not salient for Confucius, then what follows? The
only thing that follows is that he remains indifferent to the answer to these
questions and is counseling us to remain equally indifferent. But that would
mean any of a range of claimssuch as, there are ghosts, there are no ghosts,
human nature is good, human nature is baddrop out of or are excluded
from Confuciuss cultivation project as not basic to that project. And we
can extend this account as follows: For those interpreters who wish to find
in Confuciuss teaching a presupposed accountrealist or anti-realistof
the essence of truth, mind, evaluation, or language, we have some reason to
think he would not have found these questions salient. His silence or virtual
silence on these topics does not, then, indicate that he adopted or presupposed an alternative to standard Western accounts of these topics. His lack
of interest in these questions indicates that these questions were not salient
for him. As I argued above, descriptions of Confuciuss approach to problems
are meant by the writers of the Analects to be simply examples or examples
of the best approach to questions posed in specific contexts. So we should
understand this claim about his silence as indicative of which questions are
normative for the sort of cultivation-cum-reflection project he is engaged in.
If I am right about the nature of truth topic not being salient for
Confucius, then the interpreters of early Chinese philosophy, who recognize
that certain questions are not raised by early Chinese thinkers (for example,
whether or not there is truth) and conclude from this omission that the
Chinese thinkers hold that there is no truth, are mistaken.
In section one of this chapter, I discuss Donald Munros views of
early Chinese thinkers views of truth and offer a critique of his account

128

Whose Tradition? Which Dao?

on philosophical and interpretive grounds. In sections two and three, I

discuss Chad Hansens views on Chinese philosophy and truth and argue
that this account in various ways is problematic. In section four, I discuss
the ways in which David Hall and Rogers Ames, on the one hand, and
Chad Hansen, on the other, who accept Munros account, mistakenly appeal
to Wittgenstein as a recent Western example of the anti-truth view they
attribute to early Chinese philosophy. I argue that what all three of these
thinkers wish to highlight in early Chinese thinking, its focus on changing
conduct and its lack of interest in methodical inquiry into epistemology
and metaphysics, can be more coherently captured without attributing to
the Chinese philosophers anti-truth views. In the fifth section, I end with
a discussion of a more promising approach to the pragmatic aspects of
Confuciuss thought, illustrated well in a recent essay by Yang Xiao, which
seeks to correct the problems in Hansens account. I supplement his account
with a discussion of Wittgensteins On Certainty.

Munro on Plato, Chinese Philosophy, and Truth

Munro, whose important work on early Chinese philosophy has sensitized
those of us interested in the subject to its pragmatic aspects, makes the
following argument: Because Chinese philosophers did not adopt Platos
distinction between forms and particulars, they were not interested in the
truth but only in practice:
In China, truth and falsity in the Greek sense have rarely been
important considerations in a philosophers acceptance of a given
belief or proposition; these are Western concerns. The consideration important to the Chinese is the behavioral implications
of a belief or proposition in question.7
The significance of the relativity of opposite qualities is
epistemological for Plato. Since a thing never is one thing rather
than another, the phenomenal world can never be completely
known. We should seek instead to know the basic Forms...as
opposed to particular instances of them. The Forms alone always
are what they are and nothing else.8
It is important to note how extreme Munros perspective appears to be:
In China, truth and falsity in the Greek sense have rarely been important
considerations in a philosophers acceptance of a given belief or proposi-

How to Be a Confucian Pragmatist without Losing the Truth

129

tion. Munro is claiming that rarely have Chinese philosophers been interested in whether or not a proposition is true in the Greek sense. His
claim seems, then, related to another claim, which I discussed above and
rejected, that an interest in truth is conceptually connected to adopting a
particular metaphysical picture of the world. Munro seems to be claiming
that whoever does not accept that there is a timeless realm of perfect forms
necessarily ends up rejecting truth. His argument comes from the analogy
he sees between Zhuangzis view of the indeterminacy of truth and Platos
arguments that particulars are as much one way as their opposites. He does
not offer arguments for this claim but makes broad claims about differences
between Eastern and Western philosophies. He treats some of the icons of
Western philosophy, often Plato, as (a) representing the whole tradition
and (b) offering the sole possible philosophical orientation that takes truth
seriously. The fact that Chinese philosophers reject or do not appear to be
interested in Western metaphysics, specifically Platos, is, then, invoked as an
explanation for why they adopt the anti-truth, pragmatic views that they do.
Munro rightly offers no explicit appeal to the idea that early Chinese philosophers entertained Platos position and rejected it. But Munros
interpretive approach seems reasonable only if something like that were the
case. One can opt for the opposite approach to a given approach only if
the given approach is in ones awareness. So, these interpretive arguments
make the most sense if we treat Chinese philosophers as casting about for a
metaphysical view of the world: They see two possible views but reject one
and adopt the other because they dont see truth as an important goal. Or,
perhaps the relation goes the other way around: They dont see truth as an
important consideration because they have already adopted a metaphysics
that rejects truth or that finds no place for truth in it.
But I have already argued above that this did not happen. Confucius,
for example, did not anticipate the possibility of a Platonic or quasi-Platonic
metaphysics and reject it. So we are left to cast about for a more subtle
account. One possible approach is to frame this sort of account in terms
of an argument to find the best possible explanation. If the best explanation of why Confucius lacks a quasi-Platonic metaphysics is that he holds
an anti-Platonic view of metaphysics, then, on this basis, we must attribute
this belief to him, even if we have no direct evidence of him entertaining
and then rejecting this metaphysics.
The key assumption to this sort of reasoning, however, is that the
best way to explain central features of a Chinese philosophical view is to
attribute metaphysical principles that entail or require that view. What I
mean by this can be explained in terms of a modern correlate. One can

130

Whose Tradition? Which Dao?

see that mid-twentieth-century Anglo-American philosophers tended not

to be interested in normative ethics and could wonder why. If we assume
that central elements of a philosophers thinking can be best explained by
his or her metaphysical commitments, then we could explain this silence
by claiming that these philosophers have a metaphysical view that excludes
moral truth as part of the world. We might find partial confirmation for
this explanation by seeing whether any mid-twentieth-century philosophers
hold this metaphysical view and whether any of them use their metaphysical
theories to explain and justify their approach toward ethics. We might also
want to argue that even for those philosophers who adopt no such metaphysical views or who do not use their metaphysical views to justify their
silence on questions of ethics, their silence can be equally well explained in
this way. We might attribute to them a background position, a pervasive
zeitgeist, clear in some thinkers but only implicit in the rest, that provides
the hidden background of the thought of other philosophers.
But this assumption breaks down in the case of philosophers who are
silent on questions of metaphysics and for whom metaphysical questions are
not salient. That is, they would be examples of thinkers who do not operate according to the prevailing background worldview. So we should seek
for explanations of their silence on ethics elsewhere, not in terms of their
adoption or rejection of the prevailing worldview. For by being abnormal,
their thinking stands outside of the prevailing worldview. But if we explain
their views in terms other than their relation to the prevailing worldview,
then in terms of what? The best place to look will be in terms of what they
themselves say and do as philosophers or thinkers of the distinct sort that
they are. We are left with reading their texts and attempting explanations
internal to the statements and reflective practices exhibited therein.
If we turn back to Munros argument, how can we use this modern
parable to explain what goes wrong with his analysis? His argument takes
a Western philosophical view, Platos, as definitive of a normal philosophers ability to think about truth seriously, for he takes it to be normal
that philosophers will hold views about what is important that arise out
of and are defended by their metaphysical views. After all, that is what
many philosophers do, and it would be strange if Chinese philosophers did
not do that. But normal Western philosophers adopt some form of Platos
metaphysics and are, as a result, centrally focused on the pursuit of truth.
But, so this reasoning goes, abnormal philosophers must reject truth as a
value, and they are engaged in thinking that is against this value because
of their metaphysical commitments.

How to Be a Confucian Pragmatist without Losing the Truth

131

If we suppose that classical Chinese philosophers are abnormal relative

to their Western counterparts, then if we accept my argument above that
their abnormality puts them outside but not necessarily against the prevailing
worldview, we are left with the requirement that we explain their central
commitments in terms of what they themselves say and how they proceed
in the course of their reflections. How will we be able to tell whether they
are abnormal? This will be tricky to determine in each case. But the less
they seem to embrace metaphysical inquiry in their claims and practices of
reflection, the more we can feel confident in this assumption.
The question of how to interpret classical Chinese philosophers statements and procedures, moreover, rests squarely on the question of how best
to understand the role of the principle of charity in the interpretation of
philosophical texts. And, once again, this will be tricky. For what an interpreter takes to be the most charitable reading cannot be entirely separated
from her own philosophical commitments. Normal philosophers, seeking
to find normal philosophy in the texts they are reading, will look for normal patterns of statements and procedures when the text allows for them.
Abnormal philosophers will look for weird, abnormal patterns. Nonetheless,
the more an interpreter finds anomalous statements that do not fit into a
reading, the more likely that reading is wrong. The more a reading seems
forced, the less likely will it be correct. The more an interpreter attributes
to the text optional, avoidable interpretations that are subject to inescapable
criticisms, the more she is required to develop alternative interpretations.
In constructing accounts of Confuciuss philosophy, it might make
more sense to start with the expectation that early Confucianism represents
a family of views, similar to each other in a variety of ways, and most notably sharing no metaphysical views or even a commitment to metaphysics
except where these thinkers announce such views. In what follows, I will
examine Munros central claims about the metaphysics and epistemology of
early Chinese philosophy.
A central argument in Munros account of the early Chinese view of
man is that there was a tendency, pronounced in Confucianism, to view
the mind as an evaluator. Such a mind is distinguished from a Platonic
view of mind, according to Munro, by virtue of the formers concern with
behavior and the behavioral impact of beliefs and the latters concern with
the truth of beliefs. Related to this distinction is the evaluating minds emulation of models of virtuous behavior. Whereas the truth-seeking, Platonic
mind seeks truth and understanding, the evaluating Chinese mind seeks to
emulate concrete models of virtuous behavior and accept beliefs that result

132

Whose Tradition? Which Dao?

in virtuous behavior. Munro characterizes the Platonic-Confucian divide on

model emulation as follows:
Plato was not especially concerned with the Confucian idea
of model emulation. Supposedly, a person who had developed
the habit of the disinterested pursuit of truth would have the
unbiased mind necessary for being a political official; but the
political application seems secondary to the joy of knowledge for
its own sake. Works of the early Greek poets like Hesiod were
essential parts of the Athenian educational system. They contained
descriptions of gods and men, some of whose heroic exploits
were treated as exemplary by the Athenians, just as the Confucians treated similar episodes in the Book of Odes (Shih ching).9
Plato, Munro says, in contrast, explicitly rejected such works from his educational program:
The difference between the early Platonists and Confucians can
be stated as follows: The Platonists were more concerned with
knowing in order to understand, while the Confucians were
more concerned with knowing in order to behave properly
toward other men.10
Here, Munro draws a clear distinction between the evaluating minds goal
of changing behavior and the Platonic minds goal of seeking truth. This
distinction is, however, overdrawn. Although it would be wrong to admit
no differences between Confucius and Plato on these important topics,
Munros claims are partly false. Plato never rejects model emulation in The
Republic. In fact, the bulk of the opening books of The Republic consist of
critiques of the models presented in Homers poetry. Plato does not, however, argue for getting rid of models altogether. Instead, he acknowledges
that they are important, if not unavoidable. Because of this, he argues for
modifications of the role models in Homers poetry so as to bring them in
line with virtue. And at that point in his account, he has not yet offered
any definitions of the virtues. Instead, he relies on commonsense notions
of the virtues to show why Homers heroes lack the virtue required for
real virtue. In fact, because Plato realizes that a person cannot later learn to
reflect well on the virtues without early childhood exposure to exemplars of
virtue, he thoroughly critiques the role models presented by Homers poetry.

How to Be a Confucian Pragmatist without Losing the Truth

133

Munro also overstates another difference when he states that for Plato,
a persons true bliss comes from contemplation of the Forms. And although
Plato does say that true pleasure derives from contemplation of the Forms,
he does so in the context of an argument designed to show that the overall
happiness of the soul requires that each of three parts, the rational, the
spirited, and the appetitive, must find its own pleasures. For Plato, there is
no such thing as a soul of a living human being that finds its bliss exclusively in contemplation.
Munros distinction between the Greek and early Chinese view of
mind and truth clearly rests on a caricature of Platos view. What about
Confuciuss view? I am reluctant to attribute a parallel philosophical view
of mind, belief, and truth, to Confucius. When Munro is attributing philosophical views to Confucius, he is often appealing to early Confucians in
general. And it is easier to find philosophical views about mind, human
nature, forms, and so on in post-Confucius Confucians. My point here is
that we need to exercise caution in attributing sweeping philosophical views
to Confucius based on the evidence that later Confucians held such views.
Once we assume that early Confucians, and maybe even early Chinese, all
have the same set of philosophical views or the same background worldview,
with implicit philosophical commitments, it is easy to find these philosophical views in Confuciuss teachings by reading them into the text, that is, by
question-begging, interpretive procedures.
But these general worries aside, it is clear that Confucius is not exclusively focused on changing behavior. Like Plato, he has some room for
reflection (si) in his project. Although it is not clear what is included
under this term, it seems reasonable to examine instances of Confuciuss
own reflections in the Analects in order to understand what they include.
His critique of various characterizations of (ren) that his interlocutors
offer would be one example. These almost always have the same form.
The interlocutor asks if some person is (ren). Confucius answers that
the person has some important, usually positive, quality, but then he asks
what this has to do with (ren). These instances of reflective exchange
rest on Confuciuss grasp that the ideal of (ren) requires proper conduct
and attitude over a lifetime without any failures. But Confucius presents
no detailed theory about (ren). Instead, he presses this sort of point
over and over. He seeks no final definition, however, and in that way his
project differs from Platos. He does, however, seek to help his interlocutor
overcome ethical confusion and change his behavior so that he will be able
to come closer to being (ren) in his everyday practice.

134

Whose Tradition? Which Dao?

If my argument works up to this point, it should show that Munros

characterization of the difference between Confucius and Plato on their
views of mind and truth as a goal rests on weak arguments. It is not clear
that Confucius has any such philosophical views. Munros characterization
of Plato exaggerates his lack of interest in practice, and his characterization
of Confucius exaggerates his lack of interest in reflection.
Although it is not entirely clear what Munro thinks the Confucian
evaluating mind does, he says that sometimes it seems to pick out good
features of a situation, thereby indicating that something is to be done, and
other times the evaluating mind picks out bad features, thereby indicating
that something is to be avoided. However, Munro says that the Confucian
mind makes these distinctions between good and bad without a concern
for truth. This characterization is, once again, sweeping. I wonder how this
view can be defended. It would seem to require a lot of evidence. The most
reasonable part of that evidence might be what Munro indicates when he
says that in China truth and falsity in the Greek sense have rarely been
important considerations. But this point is also not clear. It could mean
that in China, there is no concern with whether or not a belief is true.
Or it may mean that Chinese philosophers, like the Confucians, were not
interested in a theory of truth, understood in terms of the development and
defense of a metaphysics of forms, which would need to be used to explain
what truth is in the Platonic sense. This last claim seems true enough.
For example, we find no such interest in Confucius, but the more radical
claim seems far from clear. Is Confucius, for example, only interested in the
behavioral implications of the beliefs he holds? What would this mean?
How plausible is this sort of view?
I have already argued in favor of distinguishing between being interested in the truth of propositions and being interested in metaphysical
theories of truth. I have also argued that even in simple language-games,
in which language functions to bring about behavior in the contexts of
practices, there is still a basic concern for truth. Indeed, being concerned
with truth is not just one thing, but rather, a motley. Only by offering a
narrow, theoretical view of concern for truth, is Munro able to make the
radical claims that he makes. As a result, his arguments are weak. If we
adopt them, it is not because they have been successfully defended but
because they appear to offer us a glimpse of a different form of life, one
not only without the Western philosophical concern for truth, but also
one that has nipped that concern in the bud at the basic level of throwing
out truth as a desideratum in accepting propositions. But by attributing to
early Chinese philosophy in general, and to Confucius in particular, a set of

How to Be a Confucian Pragmatist without Losing the Truth

135

beliefs that make no sense, this approach nips the bud in the wrong place.
For if Munro is right, Confucius accepts the beliefs he does only because of
their influence on behavior. But surely Confuciuss interest in conduct and
changes in conduct centers on his fundamental belief that some conduct is
closer to exhibiting (ren) than other conduct. And, I would argue, deciding which conduct is closer to ren and which is not must be understood
in terms of propositions that are held to be true.
The Analects contains such propositions such as 13.27: .
, Our Master said, Firm, resolute, honest, but dull are close to
(ren). If we take this claim and try to understand its meaning just in
terms of its consequences, those would be to get their audience to work to
develop these character traits and perhaps to get them to call these traits
nearly (ren). But this interpretation would not permit the audience to
believe that these character traits are closer to (ren) than their opposites.
Nevertheless, this statement means that these traits are close to the trait of
being (ren), and the statement practically encourages us for this reason to
develop these traits. If we reduce the meaning to encouraging us to develop
these traits without any further reason-supplying content for doing this, then
we end up with a behavioral account of the meaning of these utterances,
which fails to capture the way in which Confucius, even if in a very simple
way, is pointing to a normative relationship between the specified traits and
(ren). Although he neither defines (ren) nor presents a metaphysical
or epistemological account of our knowledge of it, we cannot assume that
Confucius is making a claim whose meaning can only be captured in terms
of an intended behavioral output. Munros view confuses a plain concern
with truth with a concern to develop a metaphysical or epistemological
account of truth or of our knowledge of truth.
A second, related feature of this mistake in reasoning is that it explains
a salient feature of Confuciuss method and focus, the concern with practice, in terms of essentialist, metaphysical commitments about the nature
of mind. It might make sense for a normal, Western philosophers views
to be explained in this way because such a philosophers focus and practice
might reasonably be expected to rest on some metaphysical, epistemological
commitments, which would serve to explain and justify them. Or at the
very least, we might want to press such a philosopher for the account that
his focus and practice presuppose and require to be defensible. And such
a philosopher would have to rise to that challenge. But an abnormal philosophers focus and practice need neither be explained nor even justified in
these ways. Munros account of early Confucians, like Confucius, treats them
commitments
as having abnormal commitments, while explaining those

136

Whose Tradition? Which Dao?

as if they arose from a normal philosophical theory of the nature of the

mind. I am not saying that Munro contradicts himself by asserting and
denying the same proposition, but I am saying that his project embodies a
systematic incoherence. His insight about Confuciuss basic focus on practice
gets explained in terms of a theoretical apparatus that is inconsistent with
that focus.
The question this problem raises is how to make Confuciuss focus
on practice intelligible. The way to avoid this systematic inconsistency is to
find a way to make the practice intelligible without appealing to something
theoretical or metaphysical. That would require presenting an account of the
basic character of those practices without appealing to some foundational
account of mind and its relation to the world.
To take a similar example from Wittgenstein, if we want to characterize and explain something like Wittgensteins builders language-game in
terms of some theory of the world or language that the builders adopt and
reject, this would count as misplaced theory and misplaced abstraction. We
understand the builders language-game once we have learned its rules, not
once we have learned of some theory that, if the builders accepted it, could
be used to explain or justify their linguistic practices. I suggest the same
for Confucius. The recognition that Confucius is not offering us a Platonic,
westernized metaphysics and epistemology should not be an issue in our
attributing to him an inverted metaphysics and epistemology, but rather
in finding a way of understanding his language outside of the context of
such theoretical distinctions. That would be to see his utterances as part of
a language-game that is embedded in a form of life with a complex range
of linguistic moves, including making assertions. For us philosophers to
see what is radical in Confuciuss views, we have to look elsewhere than in
Munros inverted Platonism, with its evaluating mind that is not interested
in the truth of propositions.
We might do well to look to Chad Hansens version of this argument
under an analytic, linguistic turn to see if there are resources from his linguistic turn that will show how to defend Munros radical conclusions. My
view of the problems Munro faces rests on a diagnosis that derives in large
part from Wittgensteins later critiques of philosophy. Hansens linguistic
turn in many ways also derives from his understanding of the basic commitments embedded in Wittgensteins later philosophy. And although we
might hope for a resolution of these difficulties in Hansens development of
Munros basic position, I will argue that despite the way in which Munros
views are in some ways successfully clarified by Hansen, these basic problems
remain. I will argue that the reason that Hansens appeal to Wittgenstein is

How to Be a Confucian Pragmatist without Losing the Truth

137

unsuccessful is that he ignores some key features of Wittgensteins later focus

on practicethe concern to make language intelligible without appeal to
foundational theoriesthat cause him to inherit the problems of Munros
account of early Confucian philosophy. Beyond that, Hansen gets later
Wittgenstein wrong, which I will explain.

Hansen on Chinese Philosophy on Truth

In his A Daoist Theory of Chinese Thought, Chad Hansen begins by announcing his intention to
attribute a theory of language and mind to Chinese thinkers
that differs fundamentally from the popular western view. This
theory of language makes sense of the philosophical disputes
between the ancient philosophers. It is a very different theory.
We can explain those differences either as prima facie plausible
or as a tenable theory of language for this philosophical traditiongiven the Chinese language and their other philosophical
presuppositions.11
Like Munro, Hansen sees early Chinese philosophers as adopting a
single theory of language, an alternative to the popular Western view:12
I will attribute the following perspective on language to all Chinese philosophers of the period: Language is a social practice. Its
basic function is guiding action. The smallest units of guiding
discourse are ming names. We string ming names together in
progressively larger units. The salient compositional linguistic
structure is a dao guiding discourse. The Chinese counterpart of
interpretation is not an account of the truth conditions. Rather,
to interpret a dao is to perform it. The interpretation of a dao
guiding discourse starts from an interpretation of the ming names
that compose it. In learning a conventional name, you learn a
socially shared way of making discriminations in guiding your
action according to a dao way.13
Although this account of language might appear at first glance to be
plausibly attributed to early Chinese philosophers, including Confucius, it
is a mischaracterization of Confuciuss reflection and practice to claim that

138

Whose Tradition? Which Dao?

he held this view of language, and, if thought to derive from Wittgenstein,

it seems to get Wittgenstein wrong as well. But even more, despite the
linguistic turn, Hansens account suffers from the same sorts of problems I
attributed to Munros account.
The text of the Analects shows Confucius invoking ideals that are
constitutive in parts of dao, most often (ren), and intervening to get his
interlocutors closer to dao through changes in conduct and reflection. In a
key passage on rectification of names, he makes the point that a key failure
of government is to let nameshis primary examples in the Analects are
function names, like father, ruler, etc.become detached from the norms
governing being a father or a ruler, etc. Although Hansen holds that this
requirement reflects the intersection in Confuciuss philosophy of educational theory, theory of language, and political theory,14 it is hard to see
why his interpretation of this passage is reasonable. He gives no argument
for this claim but just characterizes this passage as containing the intersection
of these three theories. Consider the passage Analects 13.3, which Hansen
uses to represent this intersection of theories:
. . . . .
... .. .
. . . . . .
. . . . .
15. , . .
, .
Zilu asked, Suppose the ruler of Wei invited you to serve in
government, then what will you do first? Our Master answered,
Certainly to reconfirm the interpersonal relationship (of social
positions) in light of their names. Zilu said, This will be necessary?! This is your being impractical! Why must the relationships
be reconfirmed in light of their names? Our Master said, You
are so vulgar!
If the names or descriptions of interpersonal relationships
are not clarified correctly, then words from junzi, rulers, will
not be obeyed by people below them. If their orders are not
obeyed, then affairs will not get accomplished. If affairs do not
get accomplished, then ritual and music wont arise. If ritual
music cant arise, then sentences and punishments will not be
to the point, if penalties do not hit the mark, then the people
will have nowhere to put their hands and feet. Hence, if a junzi

How to Be a Confucian Pragmatist without Losing the Truth

139

has his name (title of junzi) himself, he can certainly give orders.
When he issues orders himself, they will certainly be fulfillable.
Toward his orders issued, a junzi has nowhere to be casual.
This passage consists of a sorites argument, a multi-premise syllogism. This
sort of argument is no doubt from a later period in the composition of
the content of the text. The known earlier passages lack this sort of sorites
argument. Even though we must see this passage as containing a sorites
argument of some complexity, does it contain or imply a theory of language,
and so, does this support the theory that Hansen attributes to all Chinese
philosophers, including Confucius? It is easy to see why this sorites argument is theory neutral and, as such, neither implies nor invokes Hansens
candidate theories.
Take, for example the claim that if names are not rectified, language
will not flow smoothly. Does this claim entail or presuppose Hansens early Chinese candidate theory of language? Does this statement entail, for
example, that interpretation is performance, not the assignment of truth
conditions to sentences? This statement is neutral about this theoretical
claim. It entails nothing about the nature or interpretation of language in
general. Moreover, someone who holds Hansens Western account could
easily agree with Confucius on this point. How is that possible? The most
natural reading of Confuciuss claim is the commonsense view that by making our normative terms clear, we can make clear our expectations and
requirements expressed in language. This will cause use of language to lead
to fewer conflicts of interpretation. A Western theorist of language would
not need, because of his theory that interpretation assigns truth values to
sentences, to reject this practical advice, and the theorist would, moreover,
not need to adopt Hansens theory that language is important only in guiding behavior to accept this practical advice as reasonable.
It is also crucial to see the way in which Hansens account of language, while similar in some ways to claims about language in Philosophical
Investigations, is crucially different. Wittgensteins views about language are
nuanced in a way that Hansens are not. The nuances make all the difference. For example, consider Wittgensteins famous statement: For a large
class of cases of the employment of the word meaningthough not for
allthis word can be explained this way: the meaning of a word is its use
in the language.16 The qualifications herefor a large class of cases, it can
be explained thusindicates that he is not reporting to us the essence of
language as such, and he is not saying how it must be defined, only how
it can be. These qualifications express his interest in providing a flexible

140

Whose Tradition? Which Dao?

view of language that can be used to resolve conceptual puzzles but not a
view that describes, whatever this might mean, how language is itself. This
interpretation gets additional support from what Wittgenstein says later of
his method, which is connected to his view of meaning: There is not a
single philosophical method, though there are indeed methods, different
therapies, as it were.17
Although Hansen is free to ascribe any theory of language he would
like to early Chinese philosophers, he may not, as he seems to be doing,
claim to find parallels for his view in Wittgensteins theory of language.
Wittgenstein has no such theory. It is easy to see the problem of associating
his view with Wittgensteins when it comes to his claims about what language communicates, based on his and the early Chinese view of language.
Hansen explains his view that language communicates a dao in terms of
Wittgensteins language-games:
[T]he focus at the early Confucian baseline is not on metaphysics, but on guidance. Dao does not communicate scientific
truths, but ways to perform. Dao is initially and basically a
prescriptive, not a descriptive concept. The role of language is
not representing a reality that is external to our inner psyche.
Its role is communicating and transmitting guidance to society
from social leaders through history.18
So Hansen claims that language-games communicate a guidance, not a set
of truths. Along this line, he continues:
Moreover, explanation of a persons behavior will have to focus
on the dao and the persons interpretation of it: If interpretation
of a social practice replaces belief-desire and practical reasoning
models in explaining behavior, this blending of correct (performance) and aesthetically pleasing (interpretation) is natural.
Interpretation and dao together explain action.19
In these two passages, Hansen claims that language communicates a dao and
associates that with the claim that language-games, of course a term taken
from Wittgensteins middle and later writings, communicate a guidance.
But it is important to see that Wittgensteins use of language-games would
never allow for this formulation. The opening passages of the Philosophical
Investigations exemplify his method of using language-games to clarify and
call into question just this very sort of abstract, essentialist account of lan-

How to Be a Confucian Pragmatist without Losing the Truth

141

guage. Take, for example, Wittgensteins approach to the abstract, essentialist

claim that every word has meaning by referring to something.
As Wittgenstein indicates through his method of critiquing these
claims,20 the first problem with this sort of philosophical claim is that it
seems to mean something, but we are not clear what. So we need to provide
an example of the sort of language that would consist only of words that
refer to objects. But then we immediately realize that this use of language
is limited and does not capture the whole of meaningful language. Or we
see that the word refer is used without any clear meaning. Once we clarify
what we mean by refer and the range of words that get meaning from
reference, we see that this account and language-games modeled on it do
not capture the whole of what we call language or meaning.
Even though Wittgenstein is concerned with critiquing a particular
picture of language and language learning at the outset of Philosophical
Investigations, using the picture of words referring to objects, which are their
meanings, and an overly intellectualized view of the child being engaged
in interpretation to make sense of the language of adults, his method is
general. The method Wittgenstein offers is designed to challenge any overly
simplistic, essentialist account of meaning. I suggest that we can use it to
challenge Hansens account as well. Hansen might want to argue that his
account is captured by Wittgensteins grocers language-game.21 Participants
learn to count, identify colors on a chart, associate those colors with colors
on a bin, and associate names of fruit with particular types of fruit. A person
says to the grocer: Five, red, apples. The grocer goes to the apple bin,
looks up red on a color chart, pulls out the apple bin with the color red
on it, and then counts from one to five while pulling a red apple out for
each number he recites. Shall we say that in this language-game the goal of
language is to communicate a guidance, not a set of truths?
This language-game makes possible a particular transaction between
the buyer and the grocer. But Wittgenstein is at pains to point out that
this game is not the whole of language.22 We can imagine a more complex
grocer language-game in which at the end of each day the grocer is required
to give an account of how many and which kind of fruit he sold on that
day. We can imagine that he keeps a chart during the day. When he sells
five red apples, he writes down five next to red apple, etc. At the end of
business, he adds up all of the entries and creates a total for the day. When
the owner arrives, the grocer gives him the chart and says such things as,
Today, eight red apples and four green pears. If he had added incorrectly,
upon checking the owner might say, That is false. Today there were only
seven red apples. Although it would be wrong to claim that each sentence

142

Whose Tradition? Which Dao?

of this language aims to say something true, it would be equally false to

claim that none of them do this.
Perhaps it is not difficult to understand why Hansen might take this
view of Wittgenstein. Interpreters have understood Wittgenstein as rejecting his early Tractarian picture theory of meaning, which holds that sentences are made up of names that refer to objects. As his later view rejects
this theory, we must think that Wittgenstein is defending a non-referential
account of meaning: that is, meaning is use. Recent commentators have
argued that in neither case is he offering a theory of meaning, but rather,
offering a way of thinking about meaning that can serve to help us therapeutically expunge the impulse to offer such theories. As Wittgenstein says,
Philosophy simply puts everything before us, nor deduces anything.Since
everything lies open to view there is nothing to explain.23 But even if one
does not accept this account, an interpreter needs to come to grips with
the way Wittgenstein wishes to clarify reference and truth in Philosophical
Investigations.
Wittgenstein admits that we can, if we wish, say that every word
signifies, which I assume he means refers to, something. His is not saying
that this claim is false but that it does not fully clarify differences among
different kinds of words. He counsels us, then, to examine how we operate
with different kinds of words in different contexts. This counsel suggests that
he would reject a referential view of language not because it is falsewe
can always determine some object: and refers to conjunction, etc.but
because it causes us to ignore important differences between words. In
particular, it causes us to ignore the ways in which words, which signify
objects, are embedded in language-games and forms of life and are oriented
toward the world in a variety ways, all embedded in forms of life.
By emphasizing that words get their meaning and thus their relation
to objects by being embedded in forms of life, Wittgenstein makes a claim,
which, once taken to its consequences, would undermine the claim that
language does not communicate truth. For, as he says, in reference to the
charge that he is a behaviorist, he is not trying to get rid of the mental
process because that would be to deny remembering anything:
But you surely cannot deny that. For example, in remembering,
an inner process takes place.What gives the impression that
we want to deny anything?...The impression that we wanted
to deny something arises from our setting our faces against the
picture of the inner process. What we deny is that the picture
of the inner process gives us the correct idea of the use of the

How to Be a Confucian Pragmatist without Losing the Truth

143

word remember. Indeed, were saying that this picture with

its ramifications stands in the way of our seeing the use of the
word as it is.
...To deny the mental process would be to deny the remembering; to deny that anyone ever remembers anything.24
As Wittgensteins attitude toward behaviorism holds generally, it would
arguably hold against any other reductionist or eliminationist account of
other phenomena. Against the charge that Wittgenstein is an anti-realist
when it comes to truth, we should say the same. He does not want to deny
that there is truth; instead, he wishes to deny the picture of truth that gives
us a false use of the word true.25
Another key point that Wittgenstein makes, one that tells against
Hansens account of early Chinese views of language, rests on his critique of
essentialist views of general concepts, like the concept of game. He uses this
account against any attempt to provide a one-dimensional general account
of general concepts. He says of language:
Instead of pointing out something common to all that we call
language, Im saying that these phenomena have no one thing in
common in virtue of which we use the same word for all,but
there are many different kinds of affinity between them. And
on account of this affinity, or these affinities, that we call them
all languages.26
Up to now, I have argued that Hansens account of the early Chinese
theory of language, which he claims all early Chinese philosophers adopted,
finds no support in the Analects. His account is contradicted by that text,
anddespite his references to Wittgenstein in elaborating his views of that
account of languagedoes not capture, in fact, contradicts Wittgensteins
therapeutic approach to the sort of theory that Hansens account of Wittgenstein upholds. I will next turn to the question of whether Hansens
account of the early Chinese theory of language attributes a reasonable view
of language to Confucius and other early Chinese thinkers. In Chapter 4,
I have argued that interpretive debates often face interpretive ambiguities
that get resolved, not by historical investigation but by philosophical arguments about which of the possible interpretations are the most reasonable.
Although I have argued against Hansens interpretation of Confuciuss teaching, I think it is still important to pursue the question: Even if Hansens

144

Whose Tradition? Which Dao?

account has some basis in the text, would it be the strongest possible account
of Confuciuss teaching? In the next section, I offer two significant criticisms
of his account: the first criticism is that the account rests on a category
mistake: Hansen mistakenly identifies the concept of truth with a philosophical theory of truth. The second criticism is that only by making this
first, categorical mistake, can Hansen argue that early Chinese thinkers
lacked a concept of truth.27 Moreover, despite occasional indications that
it would be problematic to foist onto the early Chinese context a concern
with the distinction between prescription and representation, his account
relies on that problematic interpretation just by claiming that the early
Chinese theory of language treats it as communicating action-guidance, not
truths. Finally, I will argue that a superior approach to the relation between
dao and truth would see the two concepts as interdependent. For no form
of action-guidance makes sense without appealing to facts on the ground,
and claiming how to represent facts on the ground makes no sense without
appealing to those norms that guide speech. My account, although not
required by the details of the Analects, is, as I maintain in the last section
of this chapter, arguably consistent with the Analects and does not rest on
the problematic Western philosophical views that Hansens account of early
Chinese philosophy rests on.

More on Hansen
Hansens argument about truth in early Chinese philosophy intentionally
says something radical: Chinese philosophy has no concept of truth.28
Although radical, Hansens view gains some plausibility by being highly
qualified. He says,
Chinese philosophy has no concept of truth is a theoretical interpretive claim about the general character of pre-Han
philosophical activity. The argument is for the conclusion that
a pragmatic (non-truth-based) interpretation explains the general
character of the corpus better than does one that attributes to
Chinese thinkers the philosophical concerns characteristic of
the traditional (truth-based) philosophy. In part, the theory will
state how classical Chinese language explains the adoption of a
pragmatic rather than a semantic interest in language.29
These qualifications are important, but they also betray a problem with
Hansens account. They are important because they take out of the equation

How to Be a Confucian Pragmatist without Losing the Truth

145

the even more radical claims that Hansen sometimes seems to be making:
that early Chinese thinkers lack any concept of truth. For his claim is here
reduced to the claim that Chinese thinkers [lack] the philosophical concerns characteristic of the traditional (truth-based) philosophy. That means
that early Chinese philosophers lack any articulated concept of truth that
plays a role in the philosophical theories that Hansen claims they develop.
Hansen says, According to [my] theory doctrines with a truth role are
absent; hence there is no such role, no concept of truth. He characterizes
this conclusion as a claim about the fundamentally contrasting nature of
Chinese intellectual activity.30
Hansen makes it clear that his claim does not require the even stronger
claim that there is no character pair in Chinese that could be translated
as true-false. (shi-fei) can, he claims, be so translated, but such a
translation might be misleading if one were to think that shi and fei are
used as part of the same intellectual activity in Chinese thinking as they
are in Western philosophical activity.31
We might then formulate his claim as follows: Early Chinese language
allows for a true-false distinction, but when early Chinese thinkers engage in
philosophical reflection, they employ a theory of language that does not have
an articulated concept of truth, and so their philosophical reflections, inevitably influenced by their theories of language, lack any concern about truth.32
This theory is then meant to explain why early Chinese philosophers lack
interest in Western-style metaphysics and epistemology: they had no concept
of truth in terms of which the questions of metaphysics and epistemology
could be articulated. Instead, the early Chinese philosophers had a view of
language that centered on learning how to use names and the sort of social
control that arises from enforcing speech that conforms to those distinctions.
One way to test the success of Hansens account, which is based on an
argument to the best explanation, is whether there are alternative explanations that he does not adequately discuss. It is easy to see how to construct
such alternatives. Because of my interests, I will couch my points in terms of
Confucius. Hansens explanatory account depends on the crucial assumption
that Confucius is offering a philosophical theory. So one fact he attempts
to explain is why Confuciuss philosophical theory lacks any appeal to a
concept of truth that would play some role in Western metaphysics and
epistemology. His explanation is that Confuciuss theory of language lacked
any such concept. But suppose that we call into question Hansens reigning
assumption: Confucius is offering a philosophical theory. Suppose instead
that Confucius were offering a self-cultivation process designed to help his
students to become more sensitive and responsive to dao, primarily through
the cultivation of and reflection on ritual practices, among other things.33

146

Whose Tradition? Which Dao?

Suppose, furthermore, as I argued above, that questions of metaphysics and

epistemology were not salient for Confucius. We can motivate this claim
by virtue of emphasizing what commentators often emphasize, that Confuciuss project is aimed at self-cultivation.34 Where would this approach leave
Hansens inference to the best possible explanation?
Before I answer that question, let me reiterate that Confucius really has
no theory of language. The passage about rectification of names is neutral
when it comes to theories of language. To say that Confuciuss rectification
of names is not neutral would be much the same as suggesting that former
President George W. Bushs use of propaganda during his administration
shows that he had a cynical, postmodern view of language as merely expressing power, not truth. If Confucius, like the later Mohists, developed an
account of language, that would be different, just as if Bush had developed
or adopted an account of language that caused him to develop his propaganda machine, that would be different. I venture that there is no reason
to think that Confucius did any such thing.
If I am right, then Confucius both operated with an ordinary concept
of truth and was focused on promoting sensitivity and responsiveness to
dao. The best explanation for his lack of a philosophy of truth employed in
the context of a theory of language and related theory of metaphysics and
epistemology is just that he had no such theoretical interests. His focus was
practical as well as pedagogical. He offers examples, guidance, and critical
reflections to his students and, in doing so, makes claims, just as all of us
do, and even some truth claims. Here is one more example, in which it is
difficult to understand the passage purely in terms of influencing conduct.
Here, Confucius uses language to support and sustain changes in behavior,
but he does so through challenges to his interlocutors claims. These involve
reasoning, and so seem to invoke principles of reasoning. So, for example,
in the following passage, he established the proper conceptual relationship
between (ren) and (yong):
.. . .
.
The Master said, One who has the power of virtue, necessarily
has language. One who has language, does not necessarily have
the power of virtue. One who is (ren) is brave. One who is
brave is not necessarily (ren).35
There is no way to make sense of this passage as simply trying to change
performance. It is designed to clarify a conceptual relationship. Any impact

How to Be a Confucian Pragmatist without Losing the Truth

147

it has on conduct would be indirect. Here is an example of Confuciuss

use of reflection to impact conduct. Consider the following interchange:
. , .. . , .
. .
Someone commented: As for (Ran) Yong, he is of (ren) but
not elegant in speech. Our Master said: why should he use
eloquence? In dealing with others with eloquent speech he will
frequently be disliked by others. I do not know if he is achieving being of (ren), but why should he practice eloquence?36
Clearly, here, as in other passages that start with an interlocutor asking
about (ren), Confucius answers by presenting a criticism of their thinking. In this case, he points out that being eloquent is not always a good
thing and so ought not be offered up as an unqualified good. Also, he
makes it clear that no one knows whether Ran Yong is a person of
(ren). In doing this, he establishes a principle he often invokes of being
skeptical of attributions of (ren) to individuals. This mode of criticism
shows that he holds that the ideal is not realizable by ordinary people, for
it is too complex to realize in the whole of a persons life, and realizing it
requires realizing it perfectly. The important point here is that we have use
of language that is not directly designed to impact conduct but, instead,
is designed to correct mistakes in reflection. Any impact on behavior is
indirect. So, like Munro, Hansen distorts the way in which Confuciuss
use of language is designed to impact behavior. Confucius is making truth
claims throughout, and he is doing this even if he is not developing or
appealing to a semantic theory of truth.
Perhaps I have overstated the case in saying that in this passage Confucius is making truth claims, but the way we resolve this issue will depend
very much on what it means to make a truth claim. A. C. Graham argues
that there is no single term in Chinese that means what truth means in
English. But he makes the point that there are various terms in classical
Chinese that correspond roughly to some aspects of the English usage of
true. Despite acknowledging differences, Graham maintains, However
much or little the words used to assent (Chinese jan yu there is,
xin trustworthy and so forth) resemble or differ from English true,
one assents to the announcement, Dinner is on the table. if and only if
dinner is on the table.37
In his review of Hall and Amess Thinking through Confucius, in which
the authors develop an approach to truth similar to Hansens, P. J. Ivanhoe

148

Whose Tradition? Which Dao?

also points out that there are various terms in classical Chinese that express
an interest in how things really are:
The early Chinese were very interested in the way things really
are; they just had a different way of talking about such things.
A state of affairs could be jan (so) or pu jan (not so), shih
(the case) or fei (not the case), and so forth. There are significant differences between these two ways of describing the
world, but the ancient Chinese clearly had an interest in and a
way of talking about the way things really are. The same thing
applies in regard to Confuciuss use of moral paragons. It is a
distinct this-worldly way of talking about human conduct and
very different from following abstract rules. But this does not
entail doing away with beliefs that function in the same way as
our notions of true and false.38
When Ivanhoe makes this claim, he is not thinking of the concept of truth
in the same way as Hansen (and Hall and Ames) do. For Hansen and
Hall and Ames are thinking of the concept of truth as part of a Western
metaphysical theory about truth, which they correctly find lacking in Confuciuss teaching. But understood the way that Hansen and Hall and Ames
understand it, Ivanhoes and Grahams claims about a notion of truth do
not contradict their own. For the claim that my mother has a notion of
truth is not contradicted by the claim that she lacks a theoretical concept
of truth. These are clearly both possibly true. Two questions thus arise in
this context: What is at stake in this dispute about truth, and is there any
way to resolve it while protecting the best insights of both sides?

Hall and Ames and the Pragmatic Reading of Chinese Philosophy

Like Hansen, who they refer to in the development of their view, Hall and
Ames see later Wittgenstein as adopting a pragmatic view of language that
eschews semantics, and they attribute to Confucius a pragmatic view of
language. But their argument for this account, like Hansens, rests on the
dubious claim that to have a concept of truth requires that a person have
a theory of truth. So since Confucius lacks a theory of truth, he lacks the
concept. Besides, they point out, his sole concern is with use of language,
not with truth. I have argued that this sort of approach supposes that these
sorts of philosophical questions are salient for Confucius, but they are not

How to Be a Confucian Pragmatist without Losing the Truth

149

salient for him. Hall and Ames stress that Confucius is interested in the
use of language more than the truth of his claims, and as long as they limit
this to the claim that he is not interested in philosophical questions about the
nature of truth, they are right. However, it does not follow that Confucius
is not at some level operating with a concept of truth. And this would be
the case even if there is no exact correspondent term for truth in classical
Chinese. To show that Confucius has such a concept, all we need to do is
to show sufficient similarities between his uses and responses to sentences
and our own. And here I appeal to the later Wittgensteinian identification
of possession of concepts with practices and coming to understand concepts
by learning how to operate with a related set of words and sentences. Our
capacity to translate Confuciuss sentences into English requires such correlations, and once we have those correlations, it is easy to show that by
asserting, rejecting, and questioning a range of sentences parallel to our
own where we respond in the same way, Confucius operates with a concept
of truth because operating with the concept of truth means nothing more
than being able to engage correctly in commerce with a range of words and
sentences that we identify with having a concept of truth.
Perhaps what makes this conclusion disturbing is that it would seem
to require us to focus on the way Confucius uses language and the way his
usage is embedded in practices, without getting hung up on the theoretical
question of what truth is and how we can know what it is. If Wittgenstein
is right, we can have our practices and truth together within the context of
language-games, for designating which sentences count as true is not something that, for him, stands outside of these practices; instead, this type of
designation is constituted by these practices. If that is so, we can have our
pragmatism and semantics, too. Although I do not think that Confucius
was especially interested in the question of the relation between truth and
practice, we get a better account of what he does care aboutand a necessary one if we are to translate his utterances into Englishif we attribute
to him an ordinary concept of truth used for talking about dao. I will
defend the view that dao-seekingof the sort we find in Confuciusis
not separable from having a notion of truth and that this truth, in fact,
makes dao-seeking intelligible. But before I turn to that argument, I would
like to say more about what appears to motivate Hall and Ames in their
claim that Confucius and early Chinese philosophy lack a concept of truth.
Hall and Ames offer an argument about the absence of the concept
of truth that is, for the most part, identical with Hansens. In essence, they
identify the conception of truth with a theoretical account of truth. This
identification is stated clearly throughout their argument: [A] concept of

150

Whose Tradition? Which Dao?

truth has its locus within a particular theory.39 This claim is connected with
their further claim that what is at stake in their claim has to do with the
basic philosophical orientation of the Chinese: What makes the observation that the Chinese have no concept of truth distinctly nontrivial is that
this absence signals an alternative to the basic philosophical assumptions
implicated in the belief in a single-ordered world.40 Despite these strong
claims, they also admit that nothing in what they say implies that they are
denying the obvious:
Concerning the issue of propositional truth, Graham, of course,
agrees that Chinese philosophizing centres on the Way rather
than on the Truth. And we certainly agree with Graham that
this claim has nothing to do with everyday questions of fact.
Whether or not dinner is ready is a significant question in both
Chinese and Western cultures.41
This passage acknowledges that in this account of the differences between
Chinese and Western philosophies, there is no disagreement that in both
approaches everyday facts, and so truths, are not in question. Both sides
embrace everyday truths. The fundamental difference, then, appears to be
at the level of theory. Chinese philosophy develops a theory of the Way,
but Western philosophers of the Truth.
I find this approach unsettling. It suggests too strict a distinction
between Chinese and Western approaches and may itself rest on the Western distinctions between Fact and Value, on the one hand, or Theory and
Practice, on the other. But they make two more claims about the status of
their account that are both important, and, even if ambiguous, indicate a
possible successful resolution of these difficulties.
The first additional claim is their acknowledgment that everyday facts
are not themselves philosophically significant: The interesting and important issues generated by the search for truth are consequences of the strong
motivation of Western thinkers to move beyond the obvious to construct
theories.42 This claim can have two possible meanings: the first meaning is
that the philosophical issues about truth are nothing but issues of the construction of theories of truth that support foundational claims about truth
and knowledge of truth. On this view, truth would not be philosophically
important were it not for the epistemological and metaphysical project of
justifying claims of truth. The second meaning is that there is no important
question of truth outside the nexus of the Western philosophical problems.
There is certainly room here for alternative approaches to the one Hall and

How to Be a Confucian Pragmatist without Losing the Truth

151

Ames take. For one can hold that in the everyday sense, prior to taking
up questions of theory, truths matter in a way that philosophers ought to
acknowledge even if they are suspicious of theories of truth and interrelated
epistemological and metaphysical theories. In fact, it would seem a clear
responsibility of philosophers to clarify ordinary senses of truth and their
importance.
Their second additional claim is that the Jamesian pragmatic account
that they embrace as an alternative to the standard Western account of
truth is not, they suggest, really a theory of truth at all. Pragmatism, they
claim, is committed to the rejection of ontology and the representational
understanding of knowledge, along with a focus on language as the central
metaphor in which pragmatists engage in philosophical discussion.43 But
then they pose the following question:
The functional equivalent of reflections concerning truth in the
Chinese tradition is closely related to what we term pragmatism.
But, if we say that the Chinese have a pragmatic understanding
of truth, are we not denying our claim that there is effectively
no concept of theory of truth in China? We avoid this inference
by noting that the pragmatic theory of truth...is itself less
a theory of Truth than a vision of the Way....As James says,
pragmatism is a method only. As a methodos, pragmatism is
merely a way, a set of means of instruments that permits the
accomplishments of certain practical actions involved in getting
on with it.44
The key point here is that, according to Hall and Ames, James does not
offer a theory of truth, but rather, a way to get on with things.
The points that they are making can seem odd, but with suitable
qualification and emphasis, can be defended. I start first with the oddities.
It seems odd to me that they do not admit that the ordinary importance
attributed to saying how things are ought to be important to philosophers,
even as philosophers. We want, I would take it, to understand the various
things we mean when we say how things are and why they should matter
to us. Second, a problem has to do with the dichotomy that Hall and Ames
operate with in their account but which they seem also to deconstruct. At
times, they distinguish between Truth seekers and Way seekers but seem
to allow no possibilities for other reflective orientations. These orientations
appear as exhaustive and mutually exclusive. But Hall and Amess acknowledgment that pragmatism offers an alternative theory of truth, that is, not

152

Whose Tradition? Which Dao?

so much a theory as a method, suggests room for at least a three-way

split: (1) Truth seekers theory of truth, (2) Way seekers theory of the Way,
and (3) getting on with things practically. Finally, the whole framework of
discussion here seems to be predicated on the very sort of metaphysical
dichotomy between Truth and Way that appears to be a stand-in for the
Western philosophical dichotomy between Fact and Value, which Confucius
and other early Chinese thinkers did not find salient. The resolution of these
difficulties seems to me to come from understanding what Hall and Ames
might mean by getting on with things.
The notion of getting on with things can be understood in various ways. The first is a more or less utilitarian way, as in Jamess Truth
is the expedient in the way of thinking. Hall and Ames give this notion
a distinctive Chinese twist when they associate it with the idea that belief
is a habit that guides action. If the belief brings the individual into productive harmony with his or her community, it functions expediently, and
it is true in so far as it so serves.45 However one understands the details,
this account appears to be a consequentialist account of the justification
of beliefs. There is no reason to think that Confucius holds such a view.
Indeed, it would seem that he might associate it with the thinking of the
petty person (xiao ren ). And even if he did hold this view, it would be
necessary for him to offer a defense of his view against the charge that he is
conflating truth with expedience. And even recognizing that the expedient
in the Chinese case is harmony would not necessarily by itself resolve this
problem without some complex theoretical account of the relation between
harmony and truth.46
There is, however, another approach to getting on with things that
solves the possible problems central to Hall and Amess position and develops some of the suggestions that they make. The first is to acknowledge and
bring into prominence that Confucius embraces everyday truths. The second
is to escape potential problems of consequentialist accounts of truth or the
rejection of any concept of truth by recognizing that harmony of beliefs
and bedrock practices make justification of beliefs possible.47 This would be
to adopt the sort of broadly pragmatist account of truth and justification
we find in Wittgensteins later writings, especially On Certainty.48 And that
would amount to recognizing dao, understood as learned norms, as making
everyday truth claims possible. This approach would understand Confucius
to be putting into practice, as we all do when making truth claims, appeals
to norms that make truth claims about how to live well possible. We might
want to think of his approach to ethical truth claims, then, as ordinary,
embedded in practices, and not resting on theories of truth.

How to Be a Confucian Pragmatist without Losing the Truth

153

If we were to take this approach, it would make sense not to look for
implicit theories in Confuciuss teaching, even implicit pragmatic accounts
of truth, but to examine his inculcation of normative practices and related
modes of reflection, which, rather than necessarily showing some implicit
theoretical commitments, show the ways in which Confucius encourages
cultivation of practices that embody norms and reflection on those norms
that deepens understanding. One way to do that without falling into inappropriate systematizing is to examine Confuciuss practice on a case-by-case
basis. This will not leave us with a theoretical elaboration of what truth is
but will provide hints and suggestions about how to understand dao (norms
about how to live well) and truth claims made possible through appealing
to learned norms governing cultivation and reflection.
Performative Language and Background Beliefs
In the first chapter, I described the relation between a novices learning of
bedrock practices under the tutelage of a master who has mastered relevant
norms governing the practice and the novices developing a grasp of the
meaning of concepts and ideals embedded in those practices. These relationships carry over to questions of what counts as true in various contexts.
In a perfectly ordinary sense, the master will be free to count as true the
novices applications of concepts to instances that reflect his or her mastery
of how to apply the concepts. These bedrock practices will be distinguished
by widespread acknowledgment of their authority and peaceful agreement
among those who have mastered the practices. In On Certainty, Wittgenstein
characterizes truth of utterances and their justification in terms of mastery of
a background system of beliefs and related practices that function as justifiers. Some propositions hold fast, and calling them into question or even
asserting them as true in normal contexts makes no clear sense. Nonetheless,
they count as the background in terms of which we distinguish between
true and false. What we do in making true claims and justifying them gets
explained in terms of these masteries.
If we take background beliefs and bedrock practices as the norms
governing belief and action, we can find room for ingredients for a clarification of dao, the range of learned norms that are dao-constitutive, and
the ordinary truths those norms make possible. Such an account will bring
about a resolution of the dao-truth dichotomy as well.
I will try to unravel these sorts of problems, bringing our discussion
down to earth by examining a particular instance of Confuciuss use of language that will illustrate why these philosophical problems need not arise. To

154

Whose Tradition? Which Dao?

this end, I turn to Yang Xiaos extremely important contribution to the discussion of interpretive practices in the Analects, How Confucius Does Things
with Words: Two Hermeneutic Paradigms in the Analects and Its Exegeses.49

Doing Things with Words

Yang Xiao argues that within the Analects itself and throughout the Analects
commentarial tradition, two interpretive paradigms have dominated interpretive practice and interpretive theory. The first is the interpretive paradigm
that focuses exclusively on the literal interpretation of the utterances of a
speaker in understanding what a speech act means. The second is the interpretive paradigm that focuses on the total speech act situation, including
especially the basic goals to be achieved by the speaker in the speech act
so that others can understand it. Xiao associates the former approach to
interpretation with Gongxi Hua, Xianqiu Meng, Han Fei, Wing-tsit Chan,
Ya Hanzhang, Wang Yousan, and Alasdair MacIntyre; he associates the latter
approach to interpreting Confucius with Sima Qian, Zheng Xuan, Mouzi,
Huang Kan, Cheng Yi, and Zhu Xi.
Although there is no reason to think that Xiao would agree with the
details of my arguments about the problems of Hansens theory of language
or the possible resolution of them in Hall and Amess account, there is no
doubt that he would agree with my general conclusion that we must reject
Hansens account. His only claim about the problem with Hansens view is
that it is too narrow to capture how language is understood in the Analects:
For example, Chad Hansen argues that classical Chinese philosophers do not have concepts of the sentence, propositional
content, or belief; he suggests that classical Chinese philosophers
see language as strings of names, the only function of which is to
produce effects on peoples behavior. (Hansen 1983, 1985, 1992)
As we will see in the next section, this pragmatic framework is not broad
enough to give an adequate account of Confuciuss communicative and
hermeneutic practice in 11.22.50
As I am, Xiao is worried that although it is right to emphasize the
pragmatic character of the language of the Analects, it is a mistake to characterize the early Chinese theory of language in a narrow way that excludes
attribution of truth to Confuciuss utterances. The interpretive practice in

How to Be a Confucian Pragmatist without Losing the Truth

155

the Analects does not exhibit the narrow view of language Hansen attributes
to early Chinese thought.
Because Xiaos approach to the question of the pragmatic import of
language use in the Analects is to begin with examples and to generate his
claims about the pragmatic character of language use in the Analects from those
examples, his account makes it possible to ask what is going on in specific
passages without presupposing some prior account of what must be happening.
Xiao begins his analysis with a translation of Analects 11.22, divided
into three sections:

(a) Zilu asked, Should one practice immediately what one has
just learned? The Master said, As ones father and elder
brothers are still alive, how could one practice immediately
what one has just learned? Ran You asked, Should one
practice immediately what one has just learned? The Master
said, One should practice immediately what one has just
learned.

(b) Gongxi Hua said [to Confucius], When Zilu asked you,
Should one practice immediately what one has just learned?,
you said, Ones father and elder brothers are still alive.
When Ran You asked you, Should one practice immediately
what one has just learned?, you said, One should practice
immediately what one has just learned. I am confused, and
would venture to question this.

(c) The Master said, Ran You has a tendency of shrinking back
easily. This is why I was pushing him forward [with those
words]. Zilu has the energy of two men. This is why I was
holding him back [with different words].51

Xiao then articulates criteria for a successful interpretation of this passage:

First, it should provide an account of what is going on in the
exchange in 11.22a.
Second, it should provide the reasons that Gongxi Hua could
have given for his puzzlement or confusion in 11.22b, if he had
had a chance to articulate them.

156

Whose Tradition? Which Dao?

Third, it should provide an interpretation of Confuciuss response

to Gongxi Hua in 11.22c that is able to relieve Gongxi Huas
confusion.52
I will examine Xiaos argument in some detail:
We...hear Confuciuss response to Ran You: (CR passage)
Wen si xing zhi . One should practice immediately
what one has just learned.
What does Confucius do when he responds to Ran You?
Obviously, Confucius is telling Ran You to practice immediately
what he has just learned. We can put it as follows: When Confucius responds to Ran You, he intends Ran You to interpret
his words as true if and only if Ran You immediately practices
what he has just learned. This description captures the content
of what is said. This is what Austin calls the locution or
locutionary act. In general, to perform a locutionary act is to
utter a sentence that has what we call propositional content
or literal meaning.
I start out by pointing out some lack of clarity in Xiaos interpretation: He says, When Confucius responds to Ran You, he intends Ran You
to interpret his words as true if and only if Ran You immediately practices
what he has just learned. There is a problem with this formulation. The
if and only if conditional governs two positions: either

A. 1. Confucius intends Ran You to interpret his words as true.

2. Ran You immediately practices what he has just learned. or

B. 1. Ran You interprets his (Confuciuss) words as true.

2. Ran You immediately practices what he has just learned.

It is hard to know exactly what Xiao intends by these claims. A1 and

A2 have no clear relationship to each other. So perhaps the propositions
connected by the bi-conditional should really be B1 and 2. But this biconditional seems false. Ran You might immediately practice what he has
just learned in order to please Confucius, not because he interprets his words
as true. And even if Ran You interprets Confuciuss words as true, he might
not put them into practice because of moral weakness. What Xiao hopes
to capture by the bi-conditional is what he calls the propositional content

How to Be a Confucian Pragmatist without Losing the Truth

157

of Confuciuss speech act. It is clear, however, that he has not done that or
that Confuciuss locutionary act is problematic. My tendency is to see the
problem in Xiaos representation.
My suggestion is that the problem arises from thinking that Confuciuss speech act has a clear content capturable in terms of a bi-conditional.
It also seems strange to think that Confucius had the sort of speech act
intention that Xiao attributes to Confucius. Let me make this point clear
by offering an alternative account of this section of the passage.
My own reading of this passage puts it in the context of the sort of
opportunity-to-teach approach that Confucius takes to dialogical encounters
with his interlocutors. He takes the opportunity to make a suggestion to his
interlocutors about how to improve their conduct in such as way as to help
get them closer to the relevant virtue under discussion. So, for example, to
his interlocutors who ask about (ren), he gives different answers. It is
clear because of his different answers that he does not offer them full-blown
accounts of (ren) but partial accounts, what I have elsewhere referred to
as interventions designed to move them closer to (ren). This approach
reflects his practical approach to dialogue.
This approach to understanding dialogues about particular topics, like
(ren), also influences how best to interpret the questions. The typical
question form, as presented in the Analects, is A (wen X) where A is
replaced by a persons name or pronoun and X is replaced by some virtue,
say (ren). Often, the best interpretation of this question is A, the person
in question, asked how to practice or achieve X. The response then says how
Confucius thinks the person should practice the relevant virtue. The interlocutor neither asks for nor gets a complete, de-contextualized account of (ren).
So, what does Confucius intend in the exchange? He intends to get
his interlocutor to change his behavior, and he intends for his interlocutor
to understand that he (Confucius) thinks that the best way for his interlocutor in the present context to get closer to the relevant virtue is to follow
his recommendation. Lets call the recommendation the interlocutor-specific
virtue-formulation, and his intention for the interlocutor to be that he
devote himself to practicing this formulation.
This context is basically pedagogical. And the exchange needs to be
understood in those terms. Keeping this background context in mind,
we can ask what Confuciuss locutionary act is. Confucius intends that
his interlocutor believe that Confucius intends to motivate him to devote
himself to the virtue-formulation offered, not because it is the final decontextualized formulation about when a person should put learning into
practice, but because a master recommends it.

158

Whose Tradition? Which Dao?

Following this line of interpretation, we would need to interpret the

exchange with Ran You as follows:
Ran You asked, Should I practice immediately what I have just
learned? The Master said, You should practice immediately
what you have just learned.53
Crucial in my reinterpretation is that I reject the following reading of
Confuciuss response: One should practice immediately what one has just
learned. In response, one might argue that clearly the Chinese sentence is
general in that it does not specify the second-person personal pronoun. But
the absence of a pronoun does not necessarily mean generality in classical
Chinese. Context determines what the intended pronoun is.
I propose to substitute my account of the CR passage, so labeled
above, because it avoids these problems in Xiaos bi-conditional and captures,
in a way that Xiaos account does not, a pervasive feature of Confuciuss
opportunistic interventions. Clearly, his locutionary act, what he is doing
with his utterance, is to instruct, or intervene in Ran Yous ethical practice.
Xiao and I are in essential agreement about this. I now turn to his account
of the exchange with Zilu:
Now let us turn to Confuciuss answer to Zilu (CZ). We can
divide CZ into two parts: Confuciuss answer to the question
and his reason for his answer.
(CZa) How could one immediately practice what one has
just learned?
(CZb) As ones father and elder brothers are still alive.
Xiao points out correctly that although CZa is in the form of rhetorical
question, it has the force of an imperative. CZa essentially means, One
should not practice immediately what one has just learned. Is this what
CZa means?
If we follow the approach I offered for interpreting CR, we will need,
once again, to qualify this interpretation. Clearly, Confucius is instructing,
urging, or attempting to intervene in Zilus conduct; in light of this, what
is the most reasonable interpretation of CZa? I propose that the most reasonable interpretation is You should not practice immediately what you
have just learned. And I would use all the same arguments from above to
support this interpretation.

How to Be a Confucian Pragmatist without Losing the Truth

159

Instead of developing a full account of the rest of the passage, I

intend next to show how, given this account of this passage, which appears
to be a paradigmatic example of Confuciuss pragmatic use of language,
how the best way to understand it and the use of language in it requires
invoking both dao (and so the range of dao-constituting norms) and truths
interconnected.
Confuciuss assessment of Ran Yous character and tendency of behavior relies, I take it, on Confuciuss mastery of dao and his mastery of how
to teach ritual and how to enact learned ritual in particular cases. We know
that Ran You is overly cautious in enacting learning. But given Confuciuss
view on the relation between learning and reflection, it is easy to see why
this creates an additional problem for Ran You. Successful reflection and
the form of understanding that emerges from it depend first on learning
and putting into practice what one has learned. Ran Yous caution inhibits
both his conduct and his understanding, for understanding follows learning.
So Ran Yous capacity to know how to go on with what he has learned,
to understand it, and to reflect on it successfully will depend on adopting
Confuciuss suggestion.
When Confucius later responds to Gongxi Huas question about why
he offers different answers to Ran You and Zilu, he says:
The Master said, Ran You has a tendency of shrinking back easily. This is why I was pushing him forward. Zilu has the energy
of two men. This is why I was holding him back.54
What makes Confuciuss assessment possible? Moreover, does he claim that
he is saying how things really are with Ran You and Zilu?
As the Master of dao, Confucius is able to notice patterns of conduct
as fulfilling or falling short of dao-constituting norms. By being able to put
himself in the situation of those he is instructing by empathizing with them,
he knows how to teach each student to go on with his studies in conformity
with dao. From this know-how, he is able to make claims about how his
disciples fare in terms of reaching dao. Because striving to reach dao requires
a lifetime of endeavor, with step-by-step progress, Confucius offers each
disciple individual advice about the next best step for him.55 Only later, will
they will they be able to understand what that advice meant. For example,
later, after changing their conduct and learning more about reaching dao,
they may see that even more nuanced answers to the questions they pose
to Confucius are possible.

160

Whose Tradition? Which Dao?

Confuciuss answer to each of these two would not be correct if he

did not grasp how things really are for Ran You and Zilu. Although he
does not investigate the question of whether his claims are true or offer an
account of their truth, he does seem to offer advice based on how things
really are for each student, including knowledge of each ones temperament
and family situation. This particular example from the passage, whose overall
goal is to change behavior, demonstrates clearly that Confucius is offering
advice based on his knowledge of how things are for each student, and his
grasp of the details of their lives is crucial for Confuciuss cultivation project.
Confuciuss ability to grasp how things really are rests on a complicated framework of background beliefs and a related form of life, which
includes Confuciuss dao. In fact, it is Confuciuss dao that makes it possible for him to claim truly that Ran You needs to put learning into action
immediately and Zilu needs to slow down. That is, it is because of his
understanding of the norms governing correct conduct that he is able to
say truly that Zilu needs to slow down and Ran You to speed up.
Let me make this clear: This account I am offering is not Confuciuss,
who was focused on a cultivation project. It is mine. I take it, however, to
be consistent with his cultivation practice and use it to make his practice
intelligible, as well as to illustrate his genuine interest in how things really
are with his students without being interested in accounts of truth. The
account of truth that I offer is Wittgensteins in On Certainty.

Truth in On Certainty
Wittgensteins On Certainty is the result of a complex, rich effort at the
end of his life to make sense of and correct G. E. Moores accounts of
certainty and related refutation of skepticism.56 Central to Wittgensteins
critique of Moore is that Moore confuses certainty with knowledge. In
his discussions, Moore lists things we all know with certainty. They would
include claims like, The world exists, Here is a hand, uttered while the
speaker raises his hand, shaking it, and I have not lived off of the surface
of the earth. Wittgensteins investigation follows the same sort of criss-cross
method of philosophizing, clarifying his thoughts, then moving on, only to
return to the same problem in a new context, clarifying it in a new way,
and so on. Although this is one of the most enigmatic of Wittgensteins
published journals, some claims find general agreement by interpreters.
Wittgenstein distinguishes between meaningful propositions, which
are either true or false, and those sentences, which appear to be meaningful,

How to Be a Confucian Pragmatist without Losing the Truth

161

true or false propositions but instead function as a part of the background

in terms of which we distinguish between true and false propositions. Wittgensteins list of such propositions is varied, and it takes as its starting
point Moores list. But both lists do not simply correspond to, say, Kants
synthetic a priori principles, such as, Every event has a cause. While
Wittgenstein would agree that this sentence expresses a background belief,
he also includes, Here is a hand, which is contingent and, we might
think, empirical.
Wittgenstein claims that these background beliefs or claims are neither true nor false, but are, in fact, a kind of nonsense. They are nonsense
because unless they are stated in a special context, we cannot make sense
of what it might mean to assert or deny them. Here is a hand asserted
in most contexts would seem crazy, just as would wondering whether or
not I (still) have a hand. Wittgenstein wants to say two things by way of
explaining these facts. The first is that even Here is a hand provides us
with a norm governing our language-game of talking about body parts.
When I teach a child the vocabulary of body parts, I hold out my hand
and say, Here is a hand. The showing of my hand is an episode in training and makes my hand an exemplar in the language-game. But the claim,
This is X for any exemplar exhibits a norm of the language-game, not
an empirical proposition. Moreover, if someone questions the claim (How
do you, after all, know it is a hand?), we wouldnt know what to say, for
there is no evidence to give for such ostensive definitions. If someone has
learned the language and authoritatively expresses the rules through this sort
of ostensive definition or learning, that is all that can and needs to be said.
But once that ostensive teaching has taken place, it becomes possible
for the learner to begin to say things true or false about hands. She can, for
example, correctly say, Her hand has a cut or She only has one hand.
Wittgenstein is also at pains to make a point that the background
beliefs are part of a form of life, a way of acting. So, for example, my
belief that the thing I am sitting on is a chair, not a horse, is also part of
background. It makes no sense to assert or deny it in ordinary contexts.
And if I assert it, it will often be by way of training someone in the norms
for governing speech about furniture. But our primary orientation to chairs
is through action, not through beliefs. I first learn to sit on chairs, move
them around, count them, and so on. So the sentence used in ostensive
teaching, This is a chair, is not just labeling an object with a word, but
has also to be embedded in training in a form of life in which we do things
to and with chairs and in which the sentence plays an instructional role.
As Wittgenstein says:

162

Whose Tradition? Which Dao?

Giving grounds, however, justifying the evidence, comes to an

end;but the end is not certain propositions striking us immediately as true, i.e., it is not a kind of seeing on our part; it is
our acting, which lies at the bottom of the language-game.57
I do not intend to defend Wittgensteins account here, though I think it
has much to recommend it. Instead, I intend to use it to clarify how to
establish a limit to theoretical inquiry into Confuciuss dao.
Before I proceed to examine a particular passage in the Analects that
exhibits the sort of Wittgensteinian account I have been clarifying, I would
first like to connect Wittgensteins account to Confuciuss basic practice.
For Wittgenstein, the background consists not just of tacit beliefs, but also
largely tacit beliefs that consist of those norms that govern our speaking and
investigating the truth of specific empirical claims. They have the form of
sentences when we say them, but they function as background norms for
inquiry and action. So if someone has not adequately mastered them, the
proper response will not be to prove them to him. Instead, it will be to reeducate him in these norms. This re-education is necessary because without a
mastery of the norms that constitute the relevant language-game and related
forms of inquiry, it will not be possible for the interlocutor to understand
or assess moves in the language-game. Consider Wittgensteins account in
On Certainty of how to manage a student who asks questions that make
no sense. In this instance, we see a schoolboy of some sophistication who,
like Confuciuss disciples, can pose questions that show some conceptual
sophistication accompanied by some confusion.58
314. Imagine that the schoolboy really did ask and is there a
table there even when I turn around, and even when no one
is there to see it? Is the teacher to reassure him and say, Of
course there is!? Perhaps the teacher will get a bit impatient,
but think that the boy will grow out of asking such questions.
315. That is to say, the teacher will feel that this is not really a
legitimate question at all. And it would be just the same if the
pupil cast doubt on the uniformity of nature, that is to say, on
the justification of inductive arguments.The teacher would feel
that this was only holding them up, that [in] this way the pupil
would only get stuck and make no progress.And he would be
right. It would be as if someone were looking for some object
in a room; he opens a drawer and doesnt see it there; then he

How to Be a Confucian Pragmatist without Losing the Truth

163

closes it again, waits, and opens it once more to see if perhaps

it isnt there now, and keeps on like that. He has not learned to
look for things. And in the same way this pupil has not learned
how to ask questions. He has not learned the game that we are
trying to teach him.59
We dont try to answer his skeptical questions with a theory, but we do
what we can to get him to understand the language-game better, what one
can say in it, and which questions one can ask and not sensibly ask.
As goes Wittgenstein, so goes Confucius. When misunderstanding
arises, Confuciuss primary mode of responding is to train his interlocutors.
Sometimes he trains them in the use of a word. Sometimes he mentions a
form of acting that would get them closer to the ideal they are seeking to
achieve. Sometimes it is in a new form of speaking about a person or an
action. But we dont find him setting out to prove his assertions by arguing
with his students or teaching them logic.
Confucian lack of constructing proofs might seem a weakness from
some philosophical points of view, but if Wittgenstein is right, this is just
exactly what is needed in those contexts of training involving a particular
language-game and practices making up related forms of life.60 But this does
not leave us, or Confucius, without a way to say how things really are. A
correct reading of Analects 11.22 helps us to see just how, in a particular
case, Confucius says what is true. In this example, his truth claims depend
crucially on background norms that make these truth claims possible. These
background norms are constituents of dao.

Invoking Dao
In contrast with the accounts of Munro and Hansen, a speech act account of
language use in the Analects provides a way to understand how Confuciuss
project relates to the truth. Xiao points out that the different answers to
Zilus and Ran Yous questions presuppose a common principle, which he
presupposes but does not announce in his answers.
It is easy to see that behind this practical thesis, there is a more general
thesis: What has just been learned should be put into practice with the right
speed, which is neither rash nor sluggish. This is Confuciuss philosophical
doctrine of the mean regarding the temporal relationship between learning
and practice. Confucius does not explicitly state such a thesis, although
we can imagine that there could have been a sentence at the end of 11.22

164

Whose Tradition? Which Dao?

in which the Master makes this general thesis explicit, just as he does on
another occasion:
Zigong asked: Who is better: Zizhang or Zixia? The Master
said: Zizhang overshoots and Zixia falls short. Zigong said:
Then Zizhang must be the better one? The Master said: To
overshoot is as bad as to fall short. (Analects 11.16)61
So even if Confuciuss specific answers to Zilus and Ran Yous questions are different, they presuppose a more general normative claim, namely,
that a persons conduct should neither be excessive nor deficient. Xiao claims
that this is Confuciuss doctrine of the mean. He never says what he means
by doctrine, but it is clear that there is very little elaboration or defense
of this and other doctrines in the Analects, so it might be better just to
characterize this as a constituent of dao, a background belief functioning
as a normative claim. The sort of Austinian framework that Xiao brings
to his pragmatic analysis of 11.22 should help us to clarify the way in
which Confucius, in his exchanges described in 11.22, stands in relation to
this normative claim. Even though Confucius does not utter this normative
claim, his answers to Zilu and Ran You presuppose it. We might then say
that in answering these two different questions, he invokes this normative
principle, where invoking itself is a speech act. In this case, his invoking
comes in the form of a tacit appeal, which, when necessary, as in the 11.16
passage, required an appeal to the explicit norm.
If this approach to Analects 11.22 is reasonable, it shows something
fundamental about Confuciuss relation to the truth. It would be hard to
understand 11.22 without understanding Confucius as appealing to these
norms about how to live well (dao). And even in 11.16, it would be hard
to understand why he would say, To overshoot is as bad as to fall short
unless this norm were a constituent of dao. Even if we say that he merely
wants to bring about a world in which people operate by the principle of
moderation, the question is, why he would want to do that? We can imagine a number of possible answers to that question, but without attributing
to him, at least, some minimal acceptance of this principle, and the true
claim that dao requires it, making sense of why he seeks to influence Zilu
and Ran You in the ways he does becomes impossible.62
Why, then, doesnt Confucius just provide the 11.16 abstract version
of the normative principle to Zilu and Ran You? If that is his bottom line,
certainly, he ought to enunciate it. But his reluctance to enunciate it arguably makes good pedagogical sense. In seeking to intervene in their conduct

How to Be a Confucian Pragmatist without Losing the Truth

165

in such a way as to help them improve their behavior and, in so doing, get
closer to (ren), it would not help to say, To overshoot is as bad as to
fall short. For this claim, while a component of Confuciuss background
normative beliefs, is practically empty: that is, it is not by itself helpful in
guiding action. It doesnt tell Zilu to slow down or Ran You to speed up.
They will understand better the deficiencies in their conduct much later,
only retrospectively having practiced differently. For, as I have been arguing,
Confucius holds that successful reflection emerges out of study, by which
he means practice. Indeed, invoking the general principle, given that it is
empty, would have offered Zilu and Ran You an important principle, but
they could have remained ignorant about their state of moral development;
instead, Confucius gave each one the guidance he needed to continue to
struggle, slower or faster, toward ren/dao.
But there is a second way in which Xaios analysis helps us understand
how Confuciuss pragmatic project stands in relation to the truth. For if we
understand the questions asked by Zilu and Ran You as questions about
how each one should put into practice what he has just learned and Confuciuss responses as indicating how each should put this into practice as a
way to get closer to (ren) and dao, then we can see that his answers are
intended not only to influence behavior, but also to be true and to influence behavior in positive ways because they are true, based on Confuciuss
knowledge of facts about students lives and on his empathetic understanding and observation of their temperaments.
Finally, by appealing to Wittgensteins notion of background beliefs,
there is a third way in which this analysis sets a limit to inquiry, something
the pragmatic accounts I have been examining seek to do. Raising certain
types of questions about background normative beliefs requires an insistence on their correctness and a renewed effort at teaching the questioner,
whose questions betray misunderstanding of the language-games in which
these questions have their meaning, not meaningful requests for answers to
meaningful questions.
Based on these points, I offer a fundamental interpretive conclusion
about Confucius and his relation to truth, corrections of the pragmatic
understandings I have been criticizing. In Confuciuss dialogues with his
students, he interprets questions posed to him by his interlocutors, based on
the specific details he knows of their own moral progress, and he provides
answers designed to say how they truly need to alter their conduct in order
to get closer to dao in their actions and understanding. In doing this, he
uses his utterances to urge change in conduct. By urging changed conduct,
he invokes dao and its constitutive principles, which, when fully articulated,

166

Whose Tradition? Which Dao?

are indisputable for someone who has mastered dao, and yet he provides specific recommendations. As someone who has a deeper understanding of dao
than his interlocutors, he is able to make true claims about which forms of
conduct will get them closer to both dao conduct and understandingeven
if the background norms he appeals to are themselves neither true nor false.
These norms about how to live well do, however, make true-false claims
possible. As such, they are constitutive of the form of life that Confucius
inhabits and teaches. Questioning them requires a pedagogical response
including truth-telling, not a philosophical theory.
One consequence of this argument is that we must limit the use of
the law of excluded middle in forcing assent to propositions. If the question
whether p is true is salient for a person, then the demand that the person
assert p or not-p is justified. If the question whether p is true is not salient
for a person, then the demand that the person assert p or not-p is not justified. I pursue this line of argument further in Chapter 7, where I argue
that the law of excluded middle cannot be used to force Confucians into
theoretical stances on metaphysical claims.

Saving Confucius from the Confucians

. . .. .
.. . . . . .
.. . . .
Yan Yuan asked how to become morally good. Our Master said, Controlling oneself and returning to ritual practice is the way to become
morally good. On a single day, if a person has controlled himself and
returned to practicing ritual, then the whole empire would categorize
him as being morally good. Becoming morally good comes from oneself; how could it come from others? Yan Yuan said, May I hear the
details? The Master said, Dont look if it does not comply with ritual
action. Dont speak if it does not comply with ritual action. Dont act
if it does not comply with ritual action. Yan Yuan said, Although I
am not intelligent enough, please let me devote myself to these words
(instructions).
Analects 12.1
But what it is in philosophy that resists such an examination of details,
we have yet to come to understand.
Wittgenstein, Philosophical Investigations, section 52

Introduction
In recent work in English on early commentaries on Confuciuss Analects,
John Makeham and Daniel Gardner make the case for the existence of an
early understanding of the Analects and its project that, unlike alternative
commentaries during the Song-Ming period, was metaphysically silent, not
inclined to see Confucius and his disciples as appealing to deep metaphysical
167

168

Whose Tradition? Which Dao?

insights.1 In contrast to Zhu Xis commentaries, for example, the early commentaries of Zheng Xuan and He Yan tend to be philosophically spare
and not at all inclined to provide a metaphysical framework to justify or
explain the Analects teachings. Whereas Zhu Xi presents a Confucius of
some philosophical interestso this characterization goesZheng Xuans
and He Yans and their fellow commentators Confucius might seem to be
a pedestrian thinker with very little to offer philosophers.
Such an assessment of Zhu Xis superiority is not, however, philosophically neutral. I will argue that by being philosophically spare, He Yans
commentary, in particular, not only captures features of the Analects style
of thinking that Zhu Xis commentaries miss, but He Yan helps students
make sense of the depth of the Analects ethical project and the corresponding depth of the early commentarial tradition without appeal to tacit
metaphysical doctrines. Moreover, I will argue that Zhu Xis metaphysically
loaded interpretation of Analects falls prey to a trilemma concerning the
criteria for applying its concepts, which He Yans and his fellow commentators spare interpretation avoids. The trilemma arises from the following
three possibilities concerning the criteria for use of the concepts in Zhu
Xis commentary: (1) his criteria for applying the concept of self-control
are not different from our ordinary criteria, or (2) they are different, or (3)
they are not specified. In each of these three cases, his commentary suffers by unnecessarily attributing a questionable account to Confucius. In
preparation of my presentation of this trilemma, I will discuss the practical
significance of details in Confuciuss project and the depth of the practical
problems he addresses, and then I will liken Confuciuss approach to these
issues with Wittgensteins.

Details and the Limits of Reflection

Although we might think of Yan Huis request in Analects 12.1 to hear the
details of Confuciuss teaching about (ren) as perhaps nothing more than
the request of a struggling student who does not clearly understand how to
ask the right question in the right way, Yan Hui was no random student.
He was Confuciuss best student. So, even though we cannot assume that his
question would be expressed in the clearest, most apt possible way, we might
expect it to be a paradigm of student inquiry. Therefore, we should take
seriously this questions form. Note also that this question is fairly specific,
in contrast with many of Confuciuss interlocutors questions about virtues.
Often the interlocutor is described as asking about some virtue: (x (wen
x) where x just is some virtue. Confucius often takes such an opportunity

Saving Confucius from the Confucians

169

to speak about/elucidate one aspect of the virtue, namely, the aspect that
the interlocutor needs to practice more often or improve upon, given the
one-sided understanding and conduct the interlocutor exhibits and given his
development toward becoming a person who desires to fulfill the requirements for practicing the virtue without limitation. Confuciuss response to
Yan Huis first question about (ren) may indicate his understanding of
the one-sidedness of Yan Huis character and conduct. Perhaps Confucius
felt that Yan Hui did not give enough attention to the practice of ritual or
that even if he did, he had not yet reflected sufficiently on the import of
practicing ritual (li).
The approach I am taking here rests on rejecting the claims that this
or other passages offer an analysis of (ren). Confuciuss project is practical.2 By rejecting the view that Confucius presents analyses of (ren),
this approach also rejects instrumentalist and definitional analyses of the
relation between (ren) and (li). My critique of these types of analyses
of the relation of (ren) to (li) borrows from Kwong-loi Shuns Jen
and Li in the Analects. 3 Shun attributes a third analysis to Confucius,
which he calls constitutive. Basically, he claims that Confuciuss constitutive view asserts that for a community, a set of practices (li) are necessary
and sufficient for being (ren), but his key example of this relationship
is problematic. Shun claims that in a particular community, sacrificing to
ancestors is necessary and sufficient for feeling indebted to those ancestors.
But he seems to hold that where there is no conventional ritualistic means,
like making a sacrifice to honor ancestors, for expressing this attitude, there
is no way to express it at all. So he attributes to Confucius the view that
without the (li) by which we sacrifice to ancestors, we have no way to
express indebtedness to them.
Shun likens this necessity to the necessity of having linguistic practices
in order to have certain concepts, such as needing particular language to
express and understand the concept of the past. But these cases are different
in important ways because a ritual of behavior to express a feeling cannot
be equated with a linguistic practice used to comprehend a concept. I can
always say, whenever the topic of respect for parents comes up, that I am
thankful for the efforts of my parents and for their parents; this is my way
of expressing my indebtedness without having a specific ritual in our culture for doing so. My own language is adequate for those purposes. Shun
claims that we can attribute this constitutive analysis to Confucius. But
even Shun acknowledges that Confucius never articulates this view himself.
Nevertheless, we ought not to attribute any view, much less a false view,
to Confucius unless he explicitly states it, implies it, or suggests it because
he need to say it to make other things he says intelligible. It would make

170

Whose Tradition? Which Dao?

more sense to claim that Confucius is interested in the relation of (ren)

to (li) based on practical considerations. Confuciuss practical view is
simply this: Rarely will one come close to complying with (ren) if he
does not engage in (li). We find these sorts of practical formulations
throughout the Analects. If I am right to emphasize the practical orientation of Analects, then even if Shuns analysis of the relation of (ren) to
(li) were otherwise correct, it would still not explain, but would indeed
obscure Confuciuss interest in this practical relationship.
What, then, is the practical relationship? That Yan Hui should request
further clarification is certainly significant, given that he is represented as
understanding Confuciuss teaching at a high level and as more astute than
others in drawing inferences from Confuciuss teaching. What does Yan
Huis question mean? I suggest three possible meanings: (A) Details matter,
(B) The complexity of complying with the dao is such that even advanced
students need guidance, and (C) Rudimentary details consist of learned,
basic practices, which guide and limit reflection.
A. Even for advanced students, the details matter and perhaps are
of utmost importance. This conclusion gains additional support from Yan
Huis response to Confuciuss list of details: .. Yan
Hui said: Although I am not clever, I beg to devote myself () to these
words. In fact, Yan Huis request for details is offset, all the more, given
that the next passage, which addresses the same questionthis time posed
by Zhonggongends with the same verbal form of devotion to its words
but lacks any request for details.
.. . .
. . . . . .
Zhonggong asked how to become morally good. The Master
said, When going out from his gate, a person should act as if
meeting an important guest. In assigning people to do unpaid
labor, a person should act as if in charge of a major sacrifice.
Moreover, what a person does not desire he should not impose
on others. (As a consequence), he will be without complaints
by others both when he serves in a state and when he serves in
a fief. Zhonggong said, Although I am not intelligent enough,
please allow me to devote myself to these teachings.4
Although we cannot be sure of the reason, the absence of any request
for details in this second passage may be because Confuciuss answer to

Saving Confucius from the Confucians

171

Zhonggong already provides details. But nothing makes this clear. The
answer Confucius gives to Zhonggong is a grab bag of important recommendations. Certainly, it is possible to imagine a request for further information about how to fulfill any of the recommendations. Nonetheless, the
fact that Confucius frames his answer around a list concerned with how to
avoid resentment suggests that this issue was one he felt Zhonggong needed
to address. And although we might suppose that Yan Hui needed to pay
more attention to ritual as a mode of moving toward practicing (ren)
without defect, we find an example on his part of exemplary behavior as
a lover of learning, which both encompasses and transcends Zhonggongs
behavior, in Yan Huis request for details. If this is right, then how could
knowing the details be so important? I address this issue by examining in
more detail Yan Huis strengths and weaknesses as a student.
B. Not only do details matter for advanced students, but such students understand this and seek them when needed. Indeed, even advanced
students like Yan Hui find themselves in situations like this, in which they
need help from Confucius to know how to make the next step on the dao.
Consider Analects 9.11:
. . . . .
. . . .
. . . .
Yan Yuan sighed admiringly, saying, The more I look upward
toward it (Confuciuss dao), the higher it seems. The more I dig
deeper into it, the more impenetrable it seems. When I see it
ahead, suddenly it is behind. Our Master is adept at guiding
others step-by-step. He broadens our knowledge by having us
practice cultural refinement. He restrains our conduct by having us practice ritual actions. Even if I intend to complete this
instruction, I cant. When I have exhausted my capabilities, it
seems like it (our Masters dao) still stands there, majestically.
Even though I desire to follow it, I have no way.
Although this passage is sometimes understood to be a Daoist teaching
injected into the text, there is no reason to think so. All we need to suppose
is that (a) practicing (ren) places the practitioners conduct in relation
to an ideal that can never be completely realized, (b) the practices of
(ren) are complex, and (c) accomplished practitioners are required to provide
guidance to apprentices, because (d) the way forward on dao is often unclear

172

Whose Tradition? Which Dao?

and subject to distortion even from the advanced apprentices incomplete

or one-sided understandings. In the interchange in 12.1, given his general
understanding of the difficulty of knowing how to comply with (ren),
Yan Hui understands that in this case he does not know how to carry on
without a set of instructions on the details.5
C. In his teaching about the relationship between learning (xue)
and reflection (si) in a passage I have already discussed, (Analects 2.15),
Confucius makes it clear that learning the practices is fundamental:
.,. ,.
If a person merely learns (how to put moral ideals into practice)
but does not reflect on them, he will learn in vain. If a person
merely thinks (about how to become morally good) but does
not learn (how to do it), he will face trouble.6
This passage indicates that reflection needs to be grounded in learning. And
if we think of learning as learning of things concrete and practical, like
rituals, then we can well understand why Yan Huis brilliance at drawing
inferences might not always get him what he needs. For without having
learned the basic practices needed to limit and guide reflection, he will be
endangered in the sense that his reflection, which will be unlimited, will
create exhaustion (dai).7 This point is reinforced by Confuciuss complaint about reflection:
.. . . . .
Our Master said, So as to (have more time to) think about
(important matters), I had gone through the whole day without
eating and the whole night without sleeping, but it was useless.
It is not as good as learning (the details of how to practice
those things).8
So Yan Huis strong point, thinking (inference, intuition), cannot bear fruit
on its own. It needs to be accompanied by and constrained by learning. And
when Yan Hui asks to hear the details, we might expect that both he and
Confucius understand that what he needs are details about learning, about
practice. My interpretation of learning is based in large part on an observation
I share with numerous other commentators and translators. When Confucius

Saving Confucius from the Confucians

173

talks about learning, he is primarily focused on self-cultivation, a central

part of which includes mastery of ritual. In ordinary speech, the concept
of learning is flexible enough to cover other forms of learning, but I take
his interest to be limited primarily to practices involved in self-cultivation.
From these considerations, it should not, then, seem surprising that, as
matter of fact, when Confucius provides him with the details, he provides
him with a list of practices. The way forward for Yan Hui is through devoting himself to the words, and through devotion to those words he will be
performing the practices themselves.9 So what, then, is Confuciuss answer
to Yan Huis question about (ren)?
Although Confucius gives two different answers in 12.1 and 12.2 (and
others elsewhere), if we follow the logic of Confuciuss views on the relation
of thinking to learning, we must conclude that he has not really given Yan
Hui a direct answer so much as a method for getting the answer for himself.
For if the question about (ren) is a question about how it would be best
for Yan Hui at the present moment to understand (ren), but if he is not
able to understand that answer without further practice, his understanding
of the answer will only come through the process of reflection that arises
out of his devotion to the words and the related practices he needs to take
up. This approach to the philosophical question about (ren) might seem
to be a way to avoid the deep metaphysical and epistemological questions
that confront us. However, I will next show how this approach locates the
depth of the problem, not in hidden essences that need to be revealed, but
in the deep change of conduct, attitude, and self-understanding that makes
answers to questions about (ren) possible.

The Depth of Confuciuss Philosophical Problems

Important questions we face when we try to understand the Analects are
how deeply and in what ways is it philosophically significant? I will argue
here that the depth and significance of the Analects can be derived from
Confuciuss ability to lead his students along the path of dao. This view
depends on two types of considerations: (A) What sorts of reflection Confucius claims to avoid, and (B) What sorts of activities are central to his
ethical interventions. I will discuss these topics in turn.
A. We are told in Analects 5.13 that the disciples did not get to hear
his teachings on human nature or the dao of Heaven. Instead, they heard
his teachings about cultural adornment:

174

Whose Tradition? Which Dao?

. . . .
.
Zigong said: The cultural ornamentations (speech and conduct)
of the Master: they can be heard. As for our Masters words on
nature and the heavenly way, we cannot hear them.10
On the interpretation I favor, which seems the most straightforward one,
this passage indicates that Confucius did not talk about metaphysical topics,
or if he talked about such topics, he did not share them with his students.
The closest we get to an ultimate teaching, I would contend, is the invocation, without a detailed analysis, of those ideals that constitute dao.
Along with such an invocation, Confucius offers ethical interventions
designed to draw his interlocutor toward these dao-constituting ideals. I
mean by intervention the attempt to get an interlocutor to understand
better the ethical norms he is living under by recommending changes in
behavior in relation to an ethical ideal or raising skeptical questions about
mistaken identifications of an ideal with some related but different form of
conduct. I borrow this term from D. Z. Phillips:
Interventions in ethics are often needed because of our deeprooted tendency to theorize in ethics. We want to give a general,
theoretical account of Morality. We search for its essence. The
intervention we need, in that event, takes the form of reminders
of possibilities, which the so-called essence cannot account for.
Our trouble is not that we have failed to locate the real essence,
or misdescribed the essence. The trouble lies in the assumption
that there is an essence of something called Morality. We are
rescued from the futile search for it by coming to pay attention
to the heterogeneity of moral practices.11
This sort of intervention requires getting the philosopher to change his
approach to philosophy by getting him to pay attention to and care about
the heterogeneity of practices. This change in the philosophers perspective
as he pays attention to the diversity of practices gives rise to a change in
his assumptions and in what seems interesting. The trouble about essences
disappears.
Confuciuss interventions, while not concerning the problem of
essence, are designed to produce progress in complying with the dao. Interventions cause deep-seated problems to disappear but, importantly, without

Saving Confucius from the Confucians

175

appeal to theory. Interventions, in this sense, are to be differentiated from

theories, and this use of the term emphasizes that Confuciuss focus is on
changing behavior and attitudes, not on presenting a philosophical account
of the ethical ideals he is committed to. This interpretation is key to understanding the otherwise baffling statement at Analects 9.1:
, , .
Our Master seldom spoke of benefit, and mandate, and (ren).
The most reasonable interpretation of this passage involves two key points,
one about when Confucius initiates talking about (ren) and another
when Confucius responds to questions about (ren). When Confucius
initiates talking about (ren), his primary focus is on the relation of
(ren) to the adoption of specific roles, like a sage () or cultivated person
(); to other qualities a person might find worth cultivating associated
with (ren); to the practice of (ren); or to whether particular people
can be said to be (ren).12 When Confucius responds to questions about
(ren), except when he refuses to speak, most of his responses are about
the practice of (ren) and whether particular persons are (ren). In
short, his concern with (ren) largely centers on questions about how
and why to cultivate it, not on how to define it or provide a metaphysical
or epistemological foundation for it.
In light of this, can we say Confucius developed an ethical philosophy? To address this question, I will turn to the question of what sort of
problems Confucius addressed and how his approach might be thought of
as a form of reflection without a metaphysical theory. In this discussion, I will
focus on just one aspect of Confuciuss interventions, bringing a persons
conduct and motivations into conformity with the ideal of (ren). Even
if Confucius offers no detailed ethical epistemology or metaphysics, when
confronting his interlocutors deep ethical problems, he does engage in a
critical, reflective activity that is at least akin to philosophy. The depth of
these problems rests in their being (a) involved in a persons self-identity
and habits of character; (b) universal for sufficiently reflective agents; and (c)
connected to a complex ideal, (ren), that is not specifiable in a simple
formula. As a result of a through c, (d) answers to (ren) questions are
individual. I will discuss each of these points in turn.

a. These problems are deep in that they are recalcitrant and

embedded in ones self-identity as a person. A person needs to

176

Whose Tradition? Which Dao?

practice (ren)-related activities, like the activities of ritual,

to be able to reflect well on ren. But this requires changes
in deep-seated habits of behavior and attitudes. Antidotes
to deficiencies in character require deep commitment and
devotion, and for this reason Analects 12.1 shows Yan Hui
devoting himself to words and their related practices, and
Confucius, throughout Analects, emphasizes the importance
of devotion to learning (haoxue).13

b. These problems are also deep in the sense that they are
the sorts of confusing problems every morally sensitive or
reflective person confronts. It is easy, even for reflective
persons, to be confused about the content and requirements
of the ethical ideals under which they live. This last point
relates to c.

c. Confucius addresses individuals problems, however, without

offering any transparent, universal formula for resolving
them. It is clear that for a person to be (ren), the person
would have to have removed all of the defeaters of his being
(ren), including actions, character problems, deficient
forms of motivation, and so forth, to become (closer to)
(ren).14 This commitment is reflected in Confuciuss
questioning specific interlocutors identification of (ren)
with specific forms of conduct that fall short of the ideal. But
this presupposition is tautological and leaves the question
of the content of the ideal of (ren) open.15 Although
Confucius answers questions about (ren), his answers are
various. When we know enough about his interlocutor, his
answers seem designed to provide that person with guidance
helpful to him, not to provide a universal account of
(ren) that would be true at all times and for all people.
(ren) itself is a complex, absolute ideal, constitutive, at least
in part, of dao, always eluding our final grasp. Moreover,
when interlocutors ask about a specific person, whether he
is (ren), no particular person is said by Confucius to
be (ren) partly because (ren) is too complicated and
too strict to be instantiated merely by a single type of good
behavior, but also because any type of good behavior we can
introduce as possible evidence that someone is (ren) never
can represent the complete ideal of (ren).16

Saving Confucius from the Confucians

177

d. The need for individual answers to the (ren) question is

related importantly to the degree to which obstacles to
(ren) are themselves deep in the first sense of being intractable.
Moreover, moral confusion in society at large makes removal
of obstacles difficult. Confuciuss all-important discussions of
the petty person (xiao ren) betray his sense of the way
in which small-mindedness has come to be acceptable in his
society and needs to be uprooted. By giving it a name and by
interventions designed to move his interlocutors into a new set
of behaviors and understanding, he shows how important it is
to rise above pettiness and focus on living a life of virtue
(ren). Beyond the cultural tendencies, the one-sidedness of a
persons character provides for a range of individual obstacles.
And a persons level of development along the way limits his
understanding of and what he must do to come closer to
(ren). Indeed, there is reason to believe that Confuciuss
understanding, even if superior to his interlocutors, is not
itself complete. He himself denies being a sage and engages
in ongoing self-correction and devotion to learning (Analects
7.34).17 And so this means that these problems have a depth
of the second and third sort, involving a failure to understand
(ren) in its complexity and perfection.

Wittgenstein, Method, and the Depth of Our Quandaries

In the previous section, I described the structure of the ethical self-cultivation
interventions that Confucius engages in: He invokes an ethical ideal of
(ren) without giving an account of it. Based on the mere tautological principle
of ideals as something instantiated only if failures to live up to the ideal are
completely absent, as well as his own sense of how to help his interlocutors to
move closer to ren, Confucius engages in a form of reflection and intervention
that addresses deep-seated ethical problems. He does not, however, develop
a foundational ethical epistemology or metaphysics. Would his project of
being a Master to novices, so understood, count as philosophical? There are
different models of philosophical reflection. Rather than being forced to think
of Confuciuss method as non-philosophical because it is not metaphysical,
we can show the way in which it is similar to an important nonstandard
approach to philosophical problems: Wittgensteins. In this section, I wish
to make only three points: (1) Like Confucius, Wittgenstein establishes a

178

Whose Tradition? Which Dao?

complex, ideal, complete conceptual clarity that requires complete elimination of faults in clarity and attaches to this ideal a piecemeal, step-by-step
method for making progress toward realizing it; (2) he uproots the tendencies
of thought that keep one from realizing that the ideal requires a complex
set of interventions; and (3) he recognizes that because realizing this ideal
is complex, individuals can easily lose their way in attempting to realize it.
So, for Wittgenstein what is required is a form of philosophical therapy to
eliminate, one by one, obstacles to the ideals full realization.
In his later philosophy, Wittgensteins goal, like Confuciuss, is to
achieve an absolute ideal, to remove lack of clarity altogether, which is
arguably impossible for any person to achieve. He says:
For the clarity that we are aiming at is indeed complete clarity.
But this simply means that the philosophical problems completely disappear.
The real discovery is the one that enables me to break off
philosophizing when I want toThe one that gives philosophy
peace, so that it is no longer tormented by questions which bring
itself into question.Instead, a method is now demonstrated by
examples, and the series of examples can be broken off. Problems
are solved (difficulties eliminated), not a single problem.18
What Wittgenstein offers, then, in the face of the failures to achieve
the ideal of complete clarity, is a method that can function whether or not
complete clarity gets achieved. The substitute method, then, is to make
limited progress on getting clear about specific examples, but it also allows
for breaking off the investigation of examples in order to create an ersatz
clarity: situations in which one can stop doing philosophy when one wants,
in which one is not obsessed to get the complete clarity, which may be in
principle impossible to achieve.
Wittgensteins method focuses on uprooting individual problems:
deeply rooted, difficult to alter tendencies of thought that give rise to problems that tend to be intractable as well as disturbing:
The problems arising through a misinterpretation of our forms of
language have the character of depth. They are as deeply rooted
in us as in the forms of our language, and their significance is
as great as the importance of our language.19
A simile that has been absorbed into the forms of our
language produces a false appearance, and this disquiets us. But
this isnt how it is!we say. Yet this is how it has to be!20

Saving Confucius from the Confucians

179

Wittgenstein sees the conceptual confusions we face as philosophers

as arising from confusions that are as deep as our language. The depth has
two aspects. On the one hand, sentences operate with, or suggest to us,
pictures that provide us with a way to understand these sentences. On the
other hand, these modes of understanding tend to be misleading when used
to understand how these sentences actually work in concrete contexts. For
example, we think that a propositions truth conditions necessarily must
be something quite clear because otherwise the proposition could neither
be true nor false. So we think that a proposition must be a unique, perfectly clear correlate of some fact. But the way in which it is a unique
correlate, we tend to think, can only be shown by some deep analysis of
the structure and meaning of the sentence and the correlative facts. But
this illusion arises from a mistaken, abstract philosophical picture of what
a true sentence is, not from an examination of how sentences actually
function in speech in concrete contexts. When we take that requirement
of complete clarity and use it to assess specific sentences, we can end
up thinking that our ordinary sentences are not really true or meaningful. But against this illusion, which it is difficult to avoid when doing
philosophy, is the fact that sentences that fail this ideal are in perfectly
good order in most ordinary contexts. We cannot escape affirming their
truth because they are ordinary for us, but we deny that they can be true
because they violate the philosophical requirement of perfect clarity and
exactness.
In the early 1930s in lectures, Wittgenstein made the pronouncement
that a method has been found. His subsequent work must be understood
as working through a specific method that he modified and clarified over
time.21 Wittgensteins cognitive approach to these sorts of problems, unlike
Confuciuss more practical approach, is to articulate the forms of thinking
that give rise to these confusions, so as to root out their deep-seated basic
illusions and errors in reasoning and to examine the details of language
use to counter these illusions.22 He examines the details, primarily, by use
of the method of language-games to develop clear, simple examples of language use, against which it becomes possible to see how our inclination to
misunderstand language arises.
We need not look to standard philosophical theories of ethics to
attribute some sort of philosophical significance to Confuciuss method
of ethical intervention. If we wish to attribute to Confucius a philosophical approach to ethics, it should be on the model of Wittgensteins later
approach to philosophical problems: the piecemeal proposal of interventions
designed to resolve specific concrete problems and move us closer to some
ideal, the full realization of which may always elude our grasp.

180

Whose Tradition? Which Dao?

Even if one accepts my basic line of argument, it would be possible to

argue that it makes a difference that whereas Confuciuss ideal is primarily
ethical, with the goal to become (ren), Wittgensteins is non-ethical: to
clarify concepts and language. But this distinction is difficult to draw in
any clear way. Confuciuss method requires reflective clarity. And various
commentators make the point that the goal of Wittgensteins philosophical
clarification is ultimately ethical.23

Details and Commentary

If my argument is successful so far, it establishes criteria for successful
commentary on Analects, for commentaries that attempt to spell out the
Analects ethical theories, including foundational metaphysical or epistemological theory, would be mistaken as characterizations of Confuciuss views
and practice. Commentaries that, instead, focus on the details of practice
as the method of getting proper insights about dao would be acceptable if
otherwise correct. In this section I will examine two different approaches
to the Analects in the commentaries of Zhu Xi and He Yan. I will argue
both (A) that the latter form of commentary, although incomplete in its
presentation of the project of intervention exhibited in the Analects, presents
a more accurate account of that text than Zhu Xis philosophical readings
and (B) that Zhu Xis account of Confuciuss ethics suffers from a trilemma
concerning the meanings he attributes to Confuciuss utterances.24
We can find in He Yans25 and Zhu Xis respective commentaries on
Analects 12.1, with which I began this chapter, a fundamental conflict on
how to understand the passage.26
. . . .
. .. . ..
. . . . . .
.
Yan Yuan asked how to become morally good. Our Master said,
Controlling oneself and returning to ritual practice is the way to
become morally good. On a single day, if a person has controlled
himself and returned to practicing ritual, then the whole empire
would categorize him as being morally good. Becoming morally
good comes from oneself; how could it come from others? Yan
Yuan said, May I hear the details? The Master said, Dont look

Saving Confucius from the Confucians

if it does not comply with ritual action. Dont speak if it does

not comply with ritual action. Dont act if it does not comply
with ritual action. Yan Yuan said, Although I am not intelligent
enough, please let me devote myself to these words (instructions).27
From He Yans Commentary
A. Ma said, (keji) is to restrain oneself (yueshen).
Kong said, (fu) is to return. If the self is able to return
to ritual, this is practicing ren (weiren).
Ma said, If within the span of one day, they might be
turned (jiangui) [away from the rituals], how much more
if for a whole lifetime.
Kong said, to practice goodness (xingshan) rests with
oneself (zaiji)), not with others (buzairen).
B. Bao said, He knows that surely there were details
(tiaomu) and consequently asked about them.
Zheng said, These four items are the details of restraining
oneself (keji) and returning to ritual propriety (fuli).
Wang said, Reverently devoting himself to these words,
he always puts them into practice (xingzhi).
From Zhu Xis Commentary
A.
(ren) is the virtue of the original mind-and-heart
(benxin) in its wholeness. (ke) is to overcome or
subdue. (ji) refers to selfish desires of the self
(shenzhisiyu). (fu) is to return. (li) is heavenly
principle (tianli) in measured display. The practice
of ren (weiren) is the means of preserving whole the
virtue of mind-and-heart (quanqixinzhide).
Now the virtue of the mind-and-heart in its wholeness
(xinzhiquande) is nothing but heavenly principle
(tianli), and this can only be harmed by human desire
(renyu). Consequently, to practice ren (weirenzhe),
one must have the wherewithal to subdue selfish desires
(shengsiyu) and thereby return to ritual (fuyuli).

B. Master Cheng said, Master Yan asks for the details of

subduing the self and returning to ritual....([Repeats]

181

182

Whose Tradition? Which Dao?

the four injunctions.) The admonition on looking says:

The mind and heart are originally unprejudiced
(xinkuabenxu), responding to things without trace. There is
an essence (yao) to holding it fast (cao), which may be
found in the example of looking. Clouded by contact with
the outside (bijiaoyuqian), what is within changes
(qizhongzeqian). Regulate it in its dealing with
the external (kongzhiyuwai) in order to still it
internally (yianqinei). Subdue the self and return
to ritual (kejifuli), and in time one will become
true to ones nature (chengyi).
Although a detailed analysis of these commentaries would take more
space than is available here, I will point out two important features. (1) He
Yans commentary approaches the passage in a common-sense fashion. (2)
To the question, Why ask for details? Baos commentary indicates that
these four details just are that, detailed specification of the four practices
needed to return oneself to ritual propriety. This comment does not much
more than paraphrase the Analects passage. The opening comments do not
spell out metaphysical accounts of the self so much as spell out in practical, nontechnical terms what Confucius is recommending: controlling ones
conduct by returning to practicing the rites.
It might appear that these two comments present competing views
of human nature or the self. One is that the self by nature has to be subdued. The other is that the self is free to return to the rites. In the first,
we see a view of the self, embattled within itself. In the second, we see
that of self not embattled but naturally able to return to the rites and so
to (ren). Kieschnick sees a contradiction in Mas and Kongs glosses:
In the text, the reader is presented with a startling contradiction: two
distinct readings of the phrase side by side. Under the phrase to keji and
return to propriety is Goodness, we read: Ma [Rong] states, keji means
to restrain the self. Kong [Anguo] states, fu means to return; if one is
able (neng) of oneself to return to propriety, then this is Goodness.28
There is only a contradiction if one explains these two claims in terms
of something like Zhu Xis metaphysics. From an ordinary point of view,
there is no contradiction here.
In contrast, Zhu Xis commentary provides us with some essentialist
metaphysical views. (ji) is selfish desires. (li) is heavenly principle.
And with these characterizations, a metaphysical battle ensues between the
original, unselfish mind, trying to return to heavenly principle, and selfish

Saving Confucius from the Confucians

183

desires, trying to prevent the return to heavenly principle. Each of the four
details, is, moreover, given a specific metaphysical significance: Looking is
said to be founded on the essence of holding the original heart-mind fast.
Although Zhu Xis commentary might seem more profound because
it interprets 12.1 in metaphysical termsthis is Gardners assessmentI
maintain that it contradicts Confuciuss project of wedding reflection with
learning practices and with his ethical intervention, which works primarily
by using ordinary, nontechnical language to urge changes in conduct.
But even if this interpretive argument, worked out in detail, is inconclusive, that is, even if both of these readings are possible given the textual
evidence, I would argue that Zhu Xis attribution of an ethical metaphysics
to Confucius also fails on theoretical grounds, for it suffers from a trilemma:
The trilemma arises out the following three possibilities concerning the criteria for use of the concepts in his commentary: (1) his criteria for applying
the concept of self-control are not different from our ordinary criteria, or
(2) they are different, or (3) they are not specified. In each of these three
cases, Zhu Xis commentary suffers by unnecessarily attributing a questionable account to Confucius.
In developing this criticism, I appeal to Wittgensteins later strategies
of clarification of ordinary language. Keep in mind one of his key dicta,
For philosophical problems arise when language goes on holiday.29 We can
see how He Yans commentary, in its fidelity to Analects, interprets 12.1s
ordinary, philosophically unproblematic language: (li) are ritual practices, not heavenly principle (tianli). (fu) means to renew or restore
a practice, after having let it lapse. (keji) means to restrain oneself,
in some ordinary sense, where the referent of (ji) is not separately
specified as referring to some metaphysical entity, as it is not so used when
we typically uses phrases of that sort, like restrain oneself, deceive oneself,
etc. This language is understandable from the vantage point of ordinary
language and practice. For in an ordinary sense, I restrain myself just by
engaging in the rites or other activities, which requires in an ordinary sense
doing what I am not already tending to do, and so, practice. We often use
the injunction control yourself just to encourage the person to proceed
with more care and not to be easily distracted by earlier habits. But Zhu
Xis account of the meaning of self-restraint (keji) places on us the
further requirement of returning to our original mind, thereby requiring
that selfish desire be expunged and original mind activated. But what are
the criteria for applying these metaphysical concepts?30 We can pose one
of three possibilities: (1) these criteria are not different from our ordinary
criteria for self-control, or (2) they are different, or (3) Zhu Xi has not

184

Whose Tradition? Which Dao?

specified them. In each of these three cases, there is a serious problem for
Zhu Xis metaphysical account.
In the first case, if the criteria for applying the metaphysical comments
are not different from ordinary ones, then the metaphysical language that
Zhu Xi employs just presents a misleading way of expressing the ordinary
notion of self-control, which has no further meaning. In that case, we can
drop his account as adding nothing to He Yans account.
In the second case, if the criteria for applying metaphysical concepts
differ from the ordinary ones, then his interpretations will put his readers
at odds with the ordinary criteria that are and even, from a practical view,
ought to be indicated by the Analects passages. Given Confuciuss goal to
intervene with recommendations of changed behavior, it is crucial that the
concept of self-restraint (keji) is understood by his interlocutors in
an ordinary way, using ordinary criteria, not using technical metaphysical
criteria neither available to them nor spelled out by him. To offer metaphysical criteria will put his interlocutors at odds with themselves. For their
practical understanding will require them to interpret (keji) in an
ordinary, nontechnical sense, but the metaphysical account Zhu Xi offers
will require them to use competing, technical concepts. Indeed, if these
criteria differ from their ordinary correlates, one can ask whether a person
who has controlled oneself in the ordinary sense has returned to his original
mind, in Zhu Xis technical, metaphysical sense. Worse yet, a person may
have satisfied the criteria for controlling himself in the ordinary sense but
not have satisfied the criteria of the technical concepts. This approach will
leave his interlocutors in conceptual confusion, which can negatively impact
practice and in some cases foster moral skepticism.31
In the third case, if these metaphysical terms lack any clear criteria
for their fulfillment, they are meaningless. Theoretical concepts that lack
criteria for their application lack meaning.32
This trilemma, which will apply equally to any other metaphysical
concepts used by Zhu Xi, is inescapable.33 So if there is a non-metaphysical
way to interpret this Analects passage, then it should not be so interpreted.
My practical interpretation of this passage, following He Yans commentary,
captures the depth of Confuciuss project while avoiding this trilemma. This
argument should cause us to take a second look at the early commentaries
on the Analects. Their apparent superficiality provides us with a way to
begin to grasp the distinctive depth of the Analects ethical project. Moreover, a Confucius understood in terms of Wittgensteins interventions in
and against metaphysics provides a more defensible account of Confuciuss
ethical interventions.

The Dilemmas of Contemporary Confucianism

The Way Forward Is the Way Back.

Herakleitos of Ephesus

Introduction
In this chapter, I address two sets of contemporary dilemmas for Confucianism. One is articulated by Jiwei Ci, the other by Alasdair MacIntyre. The
first concerns the ways in which Western philosophers have tried to renew
Confucianism through the resources of historicism, which has the effect of
removing it from its historical roots. Arguably, this removal also brings with
it the loss of any way to claim that this revised set of views is in any way
Confucian. The second concerns the need for a metaphysical grounding
for Confucian ethics that will make it possible for Confucianism to avoid
the charge of parochialism. However, according to MacIntyre, to avoid this
charge, Confucianism will need, in the end, to adopt an account of the unity
of virtues that would bring it in line with Aristotelianism. If Confucianism
were to give an account of its unity of virtues, proving how it fits Aristotelian requirements for consideration as a legitimate philosophy, Confucianism
would not even in this case offer a distinctive account of ethics.
Both of these dilemmas raise the fundamental question of whether
and how Confucianism can sustain itself as a distinct approach to ethics
in contemporary ethical debate. My burden in this chapter will be to show
how these two sets of dilemmas arise from an optional set of questions that
Analects-style Confucianism would find lacking in salience and that the
Wittgensteinian reflection I have been utilizing in offering my account of
early Confucianism would find meaningless.

185

186

Whose Tradition? Which Dao?

Jiwei Cis Dilemmas

In his essay, The Confucian Relational Concept of the Person and Its Modern Predicament, Ci rejects recent arguments by contemporary Confucians
who argue that Confucianisms relational view of self provides a positive
moral alternative in bioethics to the Western, liberal, autonomous self.1
The contemporary versions of Confucianism that Ci discusses hold that
contemporary Confucianism can transcend its past and by doing so can
justify such departures by appeal to insights grasped from our own contemporary, interpretive resources. Although Cis essay aims to draw conclusions
about bioethics, his argument is quite general. He argues that modernized
Confucian views of self fall prey to two dilemmas.
The first dilemma for contemporary Confucianism, according to Ci,
begins with contemporary Confucianisms first-order theory about the
self, a first-order theory being a theory about what the self is. He argues:
Either contemporary Confucianism is identical in its moral commitments
to historical Confucianism, or it is not. On the one hand, if it does follow
historical Confucianism, it will not provide a moral alternative to modern
liberalism attractive to contemporary thinkers because of its commitment
to hierarchy in relationships. Such hierarchies (emperor-subject, husbandwife, father-son, elder brother-younger brother, master-student), he argues,
are unattractive in a modern, Western context and, as such, do not provide
an attractive moral alternative to liberalism. On the other hand, if modern
Confucianism does not follow historical Confucianism, Ci holds, it ceases
to deserve the name Confucianism. It becomes instead a generic form of
communitarianism, undeserving of any identification with Confucianism
and unable to supply any clear moral content to its relational view of the
self. For without specifying the traditional list of relationships that make up
the Confucian relational self and specific requirements for those inhabiting
those relationships, the Confucian account, Ci argues, is not distinct from
generic communitarianism.
The second dilemma Ci presents begins with the claim that Confucianism is either committed to a second-order essentialism or an antiessentialist, historical constructionism. A second-order theory is a theory
that characterizes the status of a theorys first-order claims. So, for example,
a Confucian might argue that the justification for the Confucian claim that
the root of moral goodness ( (ren) is filial piety (xiao) rests on the
historically constructed tradition of beliefs that make up Confucianism. This
would be a second-order claim about what justifies Confucian first-order
claims. Here is the dilemma Ci presents: If Confucianism is committed

The Dilemmas of Contemporary Confucianism

187

to essentialism, it does not present an attractive second-order alternative

to traditional forms of liberalism, which are also essentialist. But if it is
committed to historical constructionism, it departs from its own historical
sources, which are not constructionist, in a way that is unfair to liberalism.
For if we allow Confucianism to be spruced up into an attractive form of
contemporary historical constructionism, contemporary liberalism ought to
be allowed to do the same. But if contemporary liberalism were similarly
spruced up to avoid essentialist commitments, contemporary Confucianism
would not be any more palatable to those of us who are skeptical of essences
than contemporary liberalism is. From the vantage point of the essentialistanti-essentialist debate, they would be equally attractive or unattractive.
As I argued in Chapter 5, for Confucius, questions about the nature
of truth are not salient. As a result, it would not make sense to attribute
to him any claims, explicit or implicit, about the nature of truth because
he does not set out to define truth as a philosophical term. That claim,
however, does not require that Confucius does not accept that some claims
are true or that their truth is important to his deliberations and to his
self-cultivation projects. I would make the same claim about Cis question
of what sort of second-order theory of justification Confucius would have
adopted. Confucius does not find this question salient; it is beside the point.
For him, a second-order question raised about the status of justification of
any claim he makes would require the same sort of intervention that his
standard answers involve. For the real meaning of the question, from a selfcultivation perspective, would be: What does this question show about the
student/interlocutor who asks it, and which answer about how to change
his practice and improve his reflection will provide him with the best next
step in his self-cultivation practice? Or, a students question about what
is true or what is not true might mean to the Master: How will getting
taken up with these time-wasting, frivolous questions cause this student to
fall away from important practices, and how can I best get him back to
the practices that will provide him with the conduct and understanding he
needs to get closer to dao?
I am not arguing that Ci or most traditional philosophers interested
in second-order questions will find this response to his dilemmas satisfactory. Nonetheless, I would argue that for a practitioner of Confucian selfcultivation, this would be a proper, serious response. Within the context of
Confucian self-cultivation, questions that are the most salient/important are
questions about how a person can improve in practice and, on that basis,
how to improve in reflection and understanding of practice. The secondorder questions either disappear or get reframed in terms of questions about

188

Whose Tradition? Which Dao?

how specific people can improve their practice and reflection and, thereby,
get closer to dao.
In reply, Ci might want to argue that without answers to those sorts
of second-order questions about justification, Confucianism cannot justify
its truth claims. And if it cannot justify its truth claims, then it cannot
justify its self-cultivation practice as one that truly brings us closer to dao.
Instead, in this case, all that Confucians could do would be to appeal to
theirs as local practices and admit to the parochial character of Confucian
ethics. I will put off a final response to this challenge until the next section, in which I examine very similar challenges from Alasdair MacIntyre.
As for Cis dilemma of how an updated, westernized Confucianism can
be justifiably called by that name, it is necessary to examine the presuppositions of the challenge. The challenge supposes that if contemporary, westernized Confucianism is not essentially the same as some specific historically
realized form of Confucianism, it cannot really be Confucianism.2 That is,
the challenge supposes that there must be some essence to Confucianism
that the term Confucian introduces. Absent those features necessary to
Confucianism, a view does not deserve that name. However, the history of
Confucianism is a record of changing views.3 We might then do better to
think of the term Confucianism as a family resemblance term. As such,
we will be able to locate a range of positions that have gone under that
name. For example, despite key differences in frameworks, Confuciuss and
Zhu Xis teachings can fall under the term Confucianism without their
positions being identical. If that is so, then the question of whether a view
that forgoes a commitment to hierarchy in relationships is Confucian will
depend on whether those who authoritatively use that term Confucian are
willing to extend it and whether the story one tells about this new usage is
a story of successful transformation of the traditionone that respects the
past while finding ways to preserve it and making changes to fit the present
context and modern lifes exigencies. The problem of which contemporary
views that go by the name Confucian truly reflect Confucianism is more
complicated than Ci allows.
For example, Fei Xiaotongs sociology of Confucianism argues that
features of the tradition are bound up with the fact that it was developed
in the context of a largely agricultural society, where social order depended
on transmission of ethical practices and preservation of traditional forms of
authority.4 Developing a Confucian ethics within an urban context would
naturally require changes of emphasis. It does, however, remain an open
question what those changes would be and how they would get practiced,
understood, and sustained. It is by no means clear that urban Confucianism

The Dilemmas of Contemporary Confucianism

189

would require giving up all aspects of hierarchy in relationships and the

adoption of a liberal ideology of individuals as strangers. What Ci offers as
mutually exclusiveaccept or reject hierarchyreally admits a variety of
possible positions. And it is by no means clear that urban dwellers, especially
in East-Asian Confucian cultures, such as contemporary China, might not
find some modified hierarchy more to their liking and possibly even more in
line with their present practice than liberal ideology. As long as one admits
that the term Confucian is a family resemblance term, the argument that
contemporary Westerners would reject early Confucian hierarchies does not
entail that they would not adopt some form of hierarchy that might arguably be Confucian.5 Such a modern hierarchy might, for example, simply
mean that a person accepts Confucius as master/teacher and other significant
persons as representative, even in a loose way, of such forms of authority.6

MacIntyres Dilemma
Alasdair MacIntyre offers a similar dilemma for contemporary Confucians,
one that sees the essentialist account of human nature as precisely what is
called for if the Confucian moral tradition is to escape the charge of presenting nothing more than an account of local practices.7
The dilemma that arises from MacIntyres account is that Confucianism can either (a) present itself as defending nothing more than a set of
local, perhaps idiosyncratic, practices, or (b) offer deeper foundations for
the practices it defends, in which case it must show itself to its rivals as
having an objectively superior account of human nature, and especially for
MacIntyres purposes, the most definitive rival, Aristotelian tradition.
It is an interesting feature of MacIntyres account of this dilemma that
he does not see it as arising from the external question begging demands
of Aristotelians for Confucians to think the way Aristotelians do. For it is
crucial to MacIntyres account of reason-based discourse in and between
traditions that each tradition must be accountable to its own fundamental
principles and that each tradition can be rationally accused of error if and
only if it fails to conform to its own principles of rational moral inquiry.
Thus, it is a key part of MacIntyres argument that he is able to demonstrate, within the history of Confucianism itself, the Confucian philosophys
demand for an objective universal account of human nature that it can
use to provide a foundation for its moral claims and to offer as a basis of
critique for alternative traditions. Nevertheless, within a context of crosstradition debate of the sort MacIntyre is engaged in, Aristotelians demand

190

Whose Tradition? Which Dao?

that Confucians either put up or shut up. I suggest that instead, Confucians
need to show Aristotelians that their account of human nature is wrong or
severely limited in terms of the principles that Aristotelians already accept.
However these larger debates proceed, MacIntyres dilemma is clear:
either the specific forms of (li) and the related account of (ren) and
(xiao) in the Analects and in the texts of the later tradition are just descriptions of local Chinese practices, or they can be detached from those merely
local, historical manifestations and be grounded in something universal.
No doubt, this dilemma presents a real difficulty for the contemporary
reception of Confucianism in North America. I will argue, however, that
the problem MacIntyre poses rests on the more general problem in any
attempt to find a philosophical home for Confucianism. Once we identify the Analects project with specific Western philosophical views, we can
generate various versions of these dilemmas. The solution that I propose to
this problem, consistent with the account I have been offering herein, is to
utilize the non-theoretical reading of Analects that I have been developing
to fend off this challenge. As I will argue, the Confucianism of the Analects,
which, in contrast with the writings of later Confucians, presents a model of
moral cultivation and clarification that is, in an important sense, distinctly
non-theoretical. This feature of the Analects is, however, not a failure but, as
I have been arguing, one of its strengths. The Analects presents examples of
moral clarification based on Confuciuss transmission of a local dao, a set
of moral practices that provide a powerful moral compass for its practitioners. My approach to this issue is not, however, unproblematic. For if the
Analects, as the founding text of Confucianism, shows the transmission of a
local dao, and that is the best way to understand Confucianism, then with
this sort of reading, it may be difficult, if not impossible, to articulate the
philosophical significance of the Analects. I will conclude my argument with
an indication of how Analects-style Confucianism can address this problem.8
My argument will develop as follows. First, I will discuss why, as
a consequence of treating the Analects charitably as a philosophical text,
these dilemmas arise. Next, I will examine MacIntyres view of the rational
development of tradition, which, he thinks, establishes necessary conditions
of rational development for the Confucian tradition. His account of the
rational development of the Confucian tradition depends on his claim that
Confucius had a relatively small place for explicit theorizing within the
moral life itself. I argue that this claim is tendentious and misleading. The
non-theoretical approaches to philosophy that we find in the later works of
Wittgenstein and in Confuciuss Analects offer a way to defuse the dilemmas
presented by Ci and MacIntyre by putting limits on the law of excluded

The Dilemmas of Contemporary Confucianism

191

middle. I conclude by arguing that the way in which Confucius relies on

local practices, pace MacIntyre, establishes a clear basis for the objectivity
of Confuciuss claims.
For my argument to work, it will be necessary for it to show why
MacIntyres important challenges to the Confucian tradition are misplaced.
In order to make my case for this position, first it will be necessary to
articulate MacIntyres view of the inevitable development of a rational, ethical tradition, as he understands this development from the vantage point
of his Aristotelianism. I will then show why it is a mistake to think that
the Confucian tradition is in agreement with these fundamental Western
principles or traditions of moral inquiry.

Rational Development of Traditions

The image of traditions of moral inquiry that emerges in MacIntyres account
of Confucianism is in some broad sense Hegelian. He indicates that he sees
Confucius as only minimally interested in theory but, nonetheless, as the
person who initiated the tradition: Confucius, it seems generally to be
agreed, had a relatively small place for explicit theorizing within the moral
life itself. And the end internal to that life, conceived in Confucian terms,
is simply to live an excellent way.9
Although his characterization already slants Confuciuss teaching illicitly in a Hegelian direction (of which more later), this story fits MacIntyres
Hegelian view of tradition well, for what Confucius says, with his relatively
small place for explicit theorizing, gives fodder for later stages of the tradition, in which the Hegelian in-itself becomes in-itself and for-itself.10 But if
MacIntyres account of tradition is correct, Confuciuss initial moves, then,
from a rational point of view, cannot stand on their own.11 They need
theoretical elaboration, which MacIntyre finds in the work of subsequent
generations of Confucians. What is the subsequent history?
Central to MacIntyres account is his view of the history of the Confucian discussion of whether rituals (li) are particular customs or are based
on some deeper universal principle:
[T]he writer (of the preface to the 1714 edition of the Pei-hsi
tzu-i) renews the criticism of Han Yus mistake, made nine hundred years earlier, of confusing the virtue of ren with universal
love, in a way that suggests that there had been more recent
renewals of that mistake. And already in the sixteenth century,

192

Whose Tradition? Which Dao?

Ho Hsin-yin (15171579), because of his attempt to formulate

a standard by which the specific social relationships in terms of
which Confucius had defined virtues and duties could themselves
be criticized morally, a standard therefore to be defined independently of those relationships, had been accused of having in
consequence discarded four out of the five relationships crucial
to the Confucian doctrine of virtues and duties, retaining only
that between friend and friend.12
No doubt MacIntyres own view of the importance of this episode rests on
his acknowledgment that the Aristotelian tradition was forced to address a
similar problem of its own when a similar problem was posed for it.
MacIntyre also indicates the steps any tradition, A, must engage in to
defend itself in the face of challenges that arise from some other tradition, B.
First, tradition B must be able to construct a history from As point of view
of As own history, including the traditions historical events and criticisms
and replies to them. Then adherents of tradition A must question whether,
by their own rational standards, the views held in the other tradition, B,
clarified through the other traditions account of tradition As history, are
rationally acceptable. If tradition As views are found not to be rationally
acceptable by tradition B, the question that arises is whether tradition A
has the resources to resolve these perceived problems. If the answer to this
question is no, then members of tradition A must determine whether the
alternative tradition B offers a better explanation of the rational failures
of tradition A than adherents of A can offer themselves. If the answer is
no, then the adherents of tradition A, now in crisis, as perceived by its
adherents, have a reason to transfer their allegiance to tradition B. But why
would members of any tradition wish their tradition to undergo this sort
of scrutiny? According to MacIntyre, only through this sort of theoretical
encounter can truth be intelligible:
Of course, anyone who makes a claim to truth for a judgment
or theory or conception or the relationship of mind to object
expressed in these does so from some one particular point of
view, from within one particular tradition of inquiry rather
than from that of its incommensurable rivals. But what is then
claimed is not that this is how things appear in the light of the
standards of that point of view (something which the adherents
of a rival and incompatible point of view need have no reason
to deny), but how they are, a claim in terms of fundamental
ontology. It follows that any claim to truth involves a claim that

The Dilemmas of Contemporary Confucianism

193

no consideration advanced from any point of view can overthrow

or subvert that claim.
Such a claim, however, can only be supported on the basis
of rational encounters between rival and incommensurable points
of view, in which one such point of view has been vindicated in
such a way as to provide support for its claims to truth vis--vis
his rival standpoints. So that for any Aristotelian, Thomistic or
otherwise, the problem of how rational encounters between rival and
incommensurable points of view can be possible is a crucial one.13
In short, according to MacIntyre, a Confucian would feel compelled to
undergo such an encounter just because he either makes or wishes to make
truth claims. For a thinker to fail to engage in this sort of encounter is to
forgo being able to provide support for the truth of the thinkers claims.
As MacIntyre says:
Without rational encounter with some other rival theory,
whether incommensurable or not, we have not tested its claims
to truth....[ W]ithout rational encounter the rival theory
becomes a subject matter concerning which we have not achieved
that truth which is adaequatio intellectus ad rem.14
Several questions arise concerning MacIntyres account. The first is
whether his characterization of Confuciuss teaching is accurate. He says
that Confucius is only minimally interested in theory. We can wonder if
this claim is accurate and also wonder what MacIntyre means by theory.
Moreover, we can also wonder whether Confucians are interested in making truth claims in MacIntyres sense. Even by his version of Confucianism,
which he derives in part from Hall and Ames, it would seem difficult to see
why a Confucian would have an interest in defending his theory or even
accept this conception of truth as a definition of terms of the debate.15
Finally, even if Confucianism counts as a tradition, we can wonder whether
it is a tradition in MacIntyres sense. Next, I will take up the first question of whether Confucius has even a small place in this view about the
necessity of defending ones theory.

Confuciuss Small Place for Theory

As I indicated earlier in this chapter, MacIntyre claims, Confucius, it seems
generally to be agreed, had a relatively small place for explicit theorizing

194

Whose Tradition? Which Dao?

within the moral life itself. And the end internal to that life, conceived in
Confucian terms, is simply to live an excellent way.16 What would it require
for this approach to theory to have been different? MacIntyres characterization of Aristotles view is revealing:
By contrast, for Aristotelianism, although practical intelligence is
something very different from theoretical, and a large measure
of practical intelligence can be had by those lacking in theory,
nonetheless, theory, by supplying a knowledge of that telos which
is the human good, a knowledge from which the first premise
of all practical deliberation Since the good and the best is such
and such derives, not only corrects the deficiencies of practice,
but also directs us toward that kind of understanding which is
the telos of every rational being. This relationship of theory to
practice, and of both to the human telos, gives expression to
the relationship of part to part and of parts to whole in a wellordered psyche. And it is in terms of the right ordering of the
psyche that the virtues and their relationship to each other are
to be understood. This is why defectiveness in any one virtue
in an individual person, being a sign of disorder in that psyche,
is a sign of defectiveness with respect also to the other virtues.17
This passage gives us an account of what MacIntyre takes to be required
for a theory of the virtues: an account of the telos, the purpose, of human
beings, which includes an account for the psyche, its parts, and the proper
ordering of the parts to each other.
This characterization is crucial for understanding MacIntyres claims
about Confuciuss limited interest in theory in so far as it shows what he
takes Confucius to lack as a theorist and why he thinks Confucius has any
interest in theory at all. Consider MacIntyres characterization of what Confucius hasa substantive disagreement with Aristotle on the unity of the
virtues, and what he lacks, a substantive account of the well-ordered psyche:
Confucianism denies this type of strong thesis about the unity
of the virtues. A courageous man does not necessarily possess
jen, although one cannot have jen without courage. But courage
can, on Confuciuss view, be put to the service of wickedness,
without thereby ceasing to be courage, and this disagreement
with Aristotelianism arises from a way of understanding the
relationship of the virtues which has no place for and no need

The Dilemmas of Contemporary Confucianism

195

for either a conception of a substantial psyche or for the kind

of telos which is eudaimonia. And this exclusion by Confucius
of Aristotelian concepts extends, as we might expect, to Aristotles conception of the type of community in which the social
relationships are such as to give socially embodied expressions to
the virtues and which provide the arena within which movement
toward the telos takes place.18
From this passage, we can see that MacIntyre has found a Confucius who
is making negative theoretical claims about the unity of the virtues, but
why accept this view of Confucius?
Crucial to MacIntyres characterization is his claim that Confucius
has committed himself to an exclusion of Aristotelian concepts. At best,
this is just a silly anachronism, but it seems to fit MacIntyres view that
Confucius was engaged in some sort of low-level theoretical enterprise: that
he anticipated, in some small way, some sort of proto-Aristotelian theory
of the soul and rejected it in favor of his own view, that he anticipated a
proto-theory of the unity of the virtues and rejected it. Presumably, then,
his characterization of Confucius as only having a limited interest in theory
rests on these claims: I am making claims about the relation of the virtues;
Confucius denies some theoretical claim that Aristotle asserted, but Confucius has not developed the theoretical resources to give a full defense of this
claim. Is this characterization correct? Lets turn to the passages MacIntyre
refers to in defense of his account: Analects 8.10; 14.4; and 17.21.
. .....
Our Master said, Being strongly inclined to be bold and disliking to be poor tends toward disorder. As a man if one has
not practiced being morally good, and if he suffers this in the
extreme, this will certainly lead to disorder.19
.. . .

Our Master said, One who is completely virtuous will certainly

have something worthwhile to say. But one who has something
worthwhile to say will not necessarily be virtuous. One who is
morally good must also be brave. One who is brave will not
necessarily be completely morally good.20

196

Whose Tradition? Which Dao?

. .. .
. .
Zilu asked: Will a junzi (exemplary person) venerate being brave?
Our Master answered: A junzi considers righteous conduct to be
the highest. A ruler being brave without being righteous tends to
be rebellious. A mean man being brave without being righteous
will tend to be a bandit.21
Given his arguments about incommensurability, MacIntyre should be
sensitive to the question of how best to translate the bravery passages he
appeals to. In order to use these passages as a starting point for his claim
of theoretical differences between Confucius and Aristotle, his translations
must be correct. The problem is that he assumes Confuciuss meaning for
(yong) as bravery is the same as Aristotles meaning for andreia, bravery. Given MacIntyres sensitivity to questions of incommensurability and
translation, he should at least wonder how these translations could be correct. The obvious problem is that he attributes to Aristotle the claim that a
courageous man will possess all of the rest of the virtues. This claim entails a
necessary principle for translation of any term from Chinese into Aristotelian
language: Any character trait that does not also require a person to have full
virtue (hereafter FV) cannot correctly be translated into Aristotelian terms as
courage or bravery. We might represent this problem by appealing to the
following device: Courage (FV) is the character trait of being appropriate
in situations of fear where being courageous with FV requires that the agent
have all of the other virtues. Courage (not FV) is the character trait of
being appropriate in situations of fear where being courageous without FV
does not require that the agent have all of the other virtues. Based on the
passages above, Confucius claims that The petty person has his Courage
(not FV), but his type of bravery does not use justice when engaged in
stealing. But this claim is not one that contradicts anything that Aristotle
says. Aristotle is discussing courage (FV), and Confucius is discussing courage (not FV), according to MacIntyre. I conclude that by his appeal to
these courage passages, MacIntyre has not made a case for his claim that
Aristotle and Confucius are involved in a theoretical disagreement about the
unity of the virtues because the claims he juxtaposes are not talking about
the same concept and, therefore, do not contradict each other.
Nevertheless, one might think that my argument fails to address the
larger issues of theoretical disagreement MacIntyre introduces between Aristotle and Confucius, as MacIntyre claims a whole host of additional disagreements for which he does not appeal to textual evidence:

The Dilemmas of Contemporary Confucianism

197

[T]heory, by supplying a knowledge of that telos which is the

human good, a knowledge from which the first premise of all
practical deliberation Since the good and the best is such and
such derives, not only corrects the deficiencies of practice,
but also directs us toward that kind of understanding which is
the telos of every rational being. This relationship of theory to
practice, and of both to the human telos, gives expression to
the relationship of part to part and of parts to whole in a wellordered psyche. And it is in terms of the right ordering of the
psyche that the virtues and their relationship to each other are
to be understood. This is why defectiveness in any one virtue
in an individual person, being a sign of disorder in that psyche,
is a sign of defectiveness with respect also to the other virtues.
Confucianism denies this type of strong thesis about the unity
of the virtues.22
According to MacIntyre, Confucius would deny this Aristotelian claim
(A) about the relation between theory and practice. Tellingly, MacIntyre does
not quote any passages from the Analects to support this claim. Why does
he not supply quotations? First, he is likely to be aware of the fact that
Confucius does not take up the question of the relationship of theory to
practice. Thus, there is no explicit, announced view of this relationship. Second, MacIntyre may think that it is clear that the sentences on courage he
quotes show what Confucius thinks about the relationship between theory
and practice, but even if the translation he appeals to for those sentences
were otherwise correct, he would still have to show that someone who
asserts those sentences must also, on pain of inconsistency, hold that these
Aristotelian claims in the previous quotation about the relation of theory
to practice are true. At minimum, MacIntyre would need to argue for this
claim, which he does not do, and at most, he needs to be able to argue
successfully that maintaining his views of courage requires that Confucius
explicitly deny, and argue against, these claims.
The problem with this view is that one can fail to assert these Aristotlelian claims without thereby denying the truth of A. If Confucius is
not interested in the theoretical questions that are the focus of Aristotles
reflections, he could be just as uninterested in either asserting or denying
what Aristotle asserts. This question may not be one Confucius ever asked
or saw the point in asking. But even if one makes this sort of claim, it is
possible for MacIntyre to assert that Confucius ought, nevertheless, to be
interested in this question, and once the question arises, Confucius and his
followers are compelled to answer in one way or another. But for MacIntyre

198

Whose Tradition? Which Dao?

to make this argument successfully, he would have to show, against his views
of incommensurability, either that there is a common stock of questions all
ethical traditions must address, including A, or that there is some dynamic
within the Confucian tradition that requires that this question be addressed.
Alternatively, he might try to show that there is some set of implicit meanings in Confuciuss statements that indicate that he meant to take a stand
on various philosophical issues, even if those meanings need to be teased
out by later thinkers. In what follows, I will demonstrate why Confucius
need not be compelled to raise the question of whether Aristotles claims
about the relation of theory to practice are true.

Limits on the Law of the Excluded Middle: Wittgenstein

In Chapter 5, I argued that the law of excluded middle does not apply to
compel a response to either p or not p for propositions that are neither
meaningful nor salient for a person. The former possibility, that the questions are not meaningful, is what I call the stronger version of the thesis.
This view is stronger in the sense that it requires a more restrictive, and,
thus, controversial definition of meaning so as to permit an individual to
fend off demands to affirm either p or not-p in every case. The claim that
the law of excluded middle cannot compel a choice between two propositions, p or not-p, if the question, p or not-p? is not salient is a weaker
thesis. It is weaker in the sense that it does not depend on a restrictive
account of meaning.
Although Confucius does not offer the account of the law of excluded
middle that I give above, he does operate in accordance with the weaker
thesis and, in doing so, does not take up questions that he does not find
salient. I have also been arguing that the salience of the questions he does
entertain is relative to the self-cultivation project that he is engaged in
practicing and teaching.
Both the weak and strong theses can be used to justify attempts to
fend off questions that demand an either-or response. Even though I am
more inclined to use the contextualist thesis to define Confuciuss account
because his utterances seem sensitive to context so often and not reflective
of an interest in exposing nonsense, nevertheless, how the contextualist strategy works mirrors Wittgensteins strategy. I will examine that strategy first.
Wittgenstein exposes the meaninglessness of all philosophical questions. He sees them as meaningless because to him they are not really
questions and do not require responses. In adopting this approach to mean-

The Dilemmas of Contemporary Confucianism

199

ing, he claims that there are no meaningful questions that these so-called
philosophical questions ask; he recognizes that there is no answer to be
given because no meaningful question was asked. I will call this avoidance
by exposure of meaninglessness.
Wittgensteins strategy is just one type of strategy for avoiding philosophical questions, suited to a particular intellectual and cultural context
and defined in part by a highly specific set of goals, for he wished in the
early to mid-twentieth century in a larger European context to combat the
requirement to raise and answer philosophical questions. In this tradition,
raising and answering philosophical questions are cultural and intellectual
requirements. In context, Wittgenstein seeks to demonstrate to Western
philosophers how to show the bankruptcy of these requirements and how
to become oriented toward human life in such a way as to forgo the philosophical demand to find foundations for common beliefs and practices.
Wittgenstein wishes to replace those demands with a form of clarification
that rests on the bedrock of an acceptance of human life as something given.
As given, the human form of life is not in need of foundations even if it is
in need of clarifications. For despite its givenness (that is, acceptance as an
unavoidable set of facts), human life is something we tend to misunderstand
in the very process of living it. Nonetheless, from this example, it is possible to imagine a range of alternative, critical approaches to philosophical
thinking. If we think of Wittgensteins project as situated in and against
an already well-formed tradition of philosophical thinking, starting with
Socrates and ending with (take your pick) Bertrand Russell, we can imagine
similar projects at various stages of the development of philosophy, including
projects that are essentially pre-philosophical.
Pre-philosophical projects of clarifying the human form of life, like
Confuciuss, do not, however, need to do battle with the Western philosophical tradition and the range of questions constitutive of it. So prephilosophical projects do not need to do battle with forms of philosophical
thinking that lead to confusion about human forms of life. Instead, prephilosophical projects battle confusions of different sorts. For example, Confucius attempts to practice a form of reflection designed to illuminate the
relationship between common ethical practices of the tradition in which he
finds himself and this traditions related moral ideals. He uses these practices
and reflections to challenge his contemporaries fundamental moral confusions. But instead of demonstrating a method of coming into agreement
with other traditions or other forms of life and struggling against other
forms of reflection that, in a fundamental way, call the human form of life
into question, Confuciuss project is to foster serious study of practices of

200

Whose Tradition? Which Dao?

self-cultivation. He does this by relating these practices to moral ideals, and

then he carefully reflects on these ideals and their related practices. He does
all of this to correct misunderstandings that arise on the path of realizing
these ideals through their constitutive practices. Similar to Wittgensteins
reflective project, Confucius takes some features of human life and morality
as given. He feels no need to justify or prove the existence of these given
features of human life and morality. Instead, Confucius acknowledges and
clarifies these ideals to help those who feel the need to live well under the
ideals requirements to understand better how to do just that. That sort of
understanding does not require a foundational theory or a clarification of
a system of ethics. Rather, Confucius makes individual ethical interventions
designed to provide moral clarifications or revisions in ethical practice at
the exact points when his students need his guidance. Confuciuss project
has its limits, and, in terms of these limits, Confucius has no need to raise
or to answer philosophical questions that are not related to his pedagogical
project.
In the next part of this chapter, I will explain in detail how Wittgenstein seeks to avoid philosophical questions by showing their meaninglessness. Then I will show how Confuciuss project leads him not to anticipate
as important a certain range of theoretical questions. I will argue that, given
the goals of their reflective practices, both thinkers can justify not taking up
certain philosophical questions. In their very different ways, they can avoid
taking a stand on those philosophical questions ruled out as meaningless
in Wittgensteins case or irrelevant to self-cultivation in Confuciuss case.
The conceptual tool that Wittgenstein uses to undermine the demand
that he engage in philosophical theorizing of a certain sort is his view of
meaning. In his early work, Tractatus Logico-Philosophicus,23 and his later
Philosophical Investigations, he develops views of meaning that establish a
limit to what can be meaningfully said. In Philosophical Investigations, he
says, For a large class of cases of the employment of the word meaning,
though not for allthis word can be explained in this way: the meaning of
a word is its use in the language.24 It is important to pay close attention
to the qualifications Wittgenstein includes because they are essential to his
project. He does not claim that his account of the meaning of a word
provides necessary conditions of the meaning of a word. His approach is
pragmatic. In the context of the sort of clarification project he is engaged
in, this view of meaning will be helpful, but it is not the only way to think
of meaning. He offers a notion of meaning that fits a large class of cases.
Moreover, he does not say that even for this set of cases meaning is use
in a language-game. Instead, he says, meaning can be defined in this way.

The Dilemmas of Contemporary Confucianism

201

And of course, can be in this sentence implies that meaning need not be
defined in this way. Given these qualifications, what would justify defining
meaning in this way? This question is of first importance, for Wittgenstein
will use this definition of meaning to expose a range of philosophical questions as meaningless. How can an optional definition warrant this rejection?
Before answering this question, I will show how this definition can be used
to expose as meaningless a range of traditional philosophical questions.
At Philosophical Investigations, section 47-8, Wittgenstein raises the
question of whether the colored squares in a rectangle are simple or complex:

(R)ed (W)hite

(G)reen

R W

This question gains its importance from the similar issue he addresses
in the Tractatus, where he claims that complex names are reducible to a
set of simple names and complex objects are reducible to a set of simple
objects. Simple names and objects represent a metaphysical, semantic bedrock, which may be analyzed further. He claims in the Tractatus that there
must be such a metaphysical bedrock if sentences are to be meaningful
because the only way a clear meaning can be assigned to a sentence is if
the sentence is composed of names that refer to simple objects. To grasp
the simple name-object correlates of a sentence is to know exactly under
what conditions the sentence would be true. For a sentence to lack such
an analysis, he says, renders it meaningless, thus, neither true nor false.
And if there were such sentences, then the law of excluded middle would
be false. However, as the law of excluded middle must be true, he thinks,
sentences, that is, genuinely meaningful sentences, must be analyzable. He
finds this requirement so strong that he thinks he can know it to be true

202

Whose Tradition? Which Dao?

even if we have never seen any such complete analyses for sentences that
produce those bedrock, metaphysical meanings.25
What this account of meaning, the so-called picture theory of meaning, introduces is the idea of a non-contextually specified set of objects that
guarantee meaning to sentences by indicating that sentences are made up
of simple names that refer to metaphysically simple objects. It is, in part,
this Tractarian project of the discovery of metaphysically simple objects that
Philosophical Investigations 47-8 is meant to challenge. The Tractatus project
is not raising the question of whether, given certain optional criteria, this or
that set of objects is simple or complex. It is asking the question, outside of
any specific context, of whether there is a set of simple objects. Philosophical
Investigations, section 43, is designed to show that this question lacks any
meaning. That is, it lacks contextually specified criteria for complexity and
simplicity, specification of which would determine its meaning, that is, its
use, in a language-game. How does this criticism work?
Wittgensteins specific examples serve to undermine the hold that certain views of meaning exert on us, especially when doing philosophy. In
order to undermine the apparent meaning of the claim that there is a set
of absolutely simple objects, he examines a parallel example of potentially
simple objects to show that what it means to say that objects are simple
requires specification. Once the specification is given, however, the question loses its philosophical character. Either a sentence is philosophical, in
which case it lacks clear meaning, or it is meaningful, in which case it lacks
philosophical significance. How does this example support these points?
What the square example shows is that without some specification of
what we mean by simple or complex, the question lacks meaning. We
know that the squares are colored. We also know that they have shape. We
could think of the squares as a shape plus a color. In such a case, we think
of the squares as complex. Or we can think of the squares, independent of
their quality as colored, as basic building blocks of the rest of the square.
By appeal to the first criterion, we will say that the squares are complex.
By appeal to the second criterion, we will say that each square is simple.
Which answer is really true? That is, which answer is true no matter how we
specify its meaning in some specific context? Wittgensteins view is that this
question only provides us with the illusion of meaning. Once the optional
criterion of simplicity is specified, we are able to say, relative to that criterion, whether the square is simple or complex. But this is not what the
philosophical question aimed at. It aimed at asking how things are outside
of any criterion-specifying context, which, Wittgenstein maintains, is not pos-

The Dilemmas of Contemporary Confucianism

203

sible. How would this view of meaning impact MacIntyres requirement that
traditions of morality must engage in theorizing about the human psyche,
human telos, and the unity of the virtues? MacIntyre claims that the goal
of a theory of virtue is to say how things are for human beings but not
how they are as determined by some context.
In defending this claim, MacIntyre uses the dialectical strategy of vanquishing opposing views by saying that any philosophy that cannot prove
its founding on Western, rational principles should abandon its views in
favor of his version of Aristotelianism. Lets suppose that he is able to get
Confucians to believe that it is important to understand that their traditions fundamental framework is inconsistent and cannot be repaired. And
suppose further that he convinces Confucians that they can best avoid
this problem and still say most of what they want to say about ethics by
becoming Aristotelians. What would the result be? On that basis, would it
be possible for such Aristotelian converts to assert and make sense out of
truth claims detached from their distinctive meaning-conferring contexts,
substituting Aristotelian-related criteria for applying terms to objects and
situations? It is true that if Aristotelians get all contenders to convert to
Aristotelianism, there will be no one left to complain that they are using
their own local criteria to make meaningful claims. MacIntyre needs more
than this, however, for he, too, needs some way to make sense of the meaning of his own truth claims that prove those claims, given the meanings
they have, to be detachable from the conceptual framework in which they
are embedded without loss of meaning. Only by doing so would MacIntyre
be licensed to say that the truths of his claims are what he says they are in
the strong metaphysical sense he advocates for others:
Of course anyone who makes a claim to truth for a judgment
or theory or conception or the relationship of mind to object
expressed in these does so from some one particular point of
view, from within one particular tradition of inquiry rather
than from that of its incommensurable rivals. But what is then
claimed is not that this is how things appear in the light of the
standards of that point of view (something which the adherents
of a rival and incompatible point of view need have no reason
to deny), but how they are, a claim in terms of fundamental
ontology. It follows that any claim to truth involves a claim that
no consideration advanced from any point of view can overthrow
or subvert that claim.26

204

Whose Tradition? Which Dao?

But all MacIntyre can say is that these truth claims, embedded in the
Aristotelian framework, mean what they mean only from the vantage point
of this framework. They are, as such, framework-dependent truth claims. Do
they tell us how things are in themselves? If Wittgenstein is right, we do
not even know what this latter question means because meaningful truth
claims must play a role in the language-games in which they have their life.
Although the grammar of most objective truth claims is that they are neither
timed nor relativized to person or point of view, it is just as true that we
can make no sense of there being truth claims without understanding them
as embedded in particular contexts that provide for meaning and criteria
of truth. So, although we must admit that the grammar of objective truth
claims is such that they are not relativized to time, person, or point of view,
we also must acknowledge that making truth claims is always embedded
within contingent contexts.
This is the Wittgensteinian path to undermining MacIntyres challenge
to force us to take a stand on the issues of moral metaphysics.27 But I have
also argued that for Wittgenstein, this view of meaning is only one possible
view of meaning. How can such a weak claim be used to refute MacIntyres
demand for philosophers to engage in moral metaphysics? I would argue
that Wittgensteins view offers us a way to set aside MacIntyres demand
by showing how his demand can be avoided. That is, MacIntyres demand,
which he presents as something Confucians must acknowledge, for Wittgenstein would be a meaningless demand they may avoid. For if we adopt
the view of meaning as use in a language-game, then we are compelled to
acknowledge that the term objective truth can only be contextually specified as playing a role in certain types of language-games. And although it
may be true that within certain traditions and developing certain types of
theories, MacIntyres characterization of the logic of the term true is apt,
it is by no means clear that all traditions must operate with the principles
of truth that are local to his version of the Aristotelian tradition. And even
if it is a feature of the grammar of the term true, said about sentences
put forth as objectively true, that these uses of true are neither timed
nor relativized, it does not follow from this that all such uses require, as
MacIntyre insists, embedding in some theoretical debate, in which only the
winners are licensed to use that predicate of statements they affirm. Indeed,
we use the word true all the time in various non-theoretical contexts:
for example, when I claim that another person promised to meet me and
should meet me. If one were free to assert this claim only as a result of
a theoretical debate between competing accounts of promise-keeping, few

The Dilemmas of Contemporary Confucianism

205

people would be licensed to assert it. But we often make this and other
similar claims. So we must, in that case, understand that the meaning of
the term true is its function in context. The Aristotelian effort to extract
some more metaphysically basic rules of usage beyond what I have already
acknowledged from the variety of language-games in which this term is used
is an effort to reduce what is complex, the myriad family of uses of the term
true, to what is simple, its sole use as a term in fundamental ontology.
But why do the Aristotelians insist that every philosophy and tradition must
look at the meaning of true in this way? And how have they managed to
hold the entire world of philosophy hostage to their judgment of what is
and what is not a valid question?
Elsewhere, I have argued that what motivates Wittgensteins view of
meaning is an ethical impulse, locatable in his early writings and present in
his later works, to come into agreement with the world (his early formulation) or forms of life (his later formulation).28 In his later formulations, we
find him affirming that there are features of human life and language that
we cannot escape despite an impulse to do so when engaged in philosophy.
The reason that philosophers often end up in conceptual confusion is that
they live in the world and speak ordinary language, but in their work they
seek pure ideals, which they take to provide a ground for our ordinary
ways of speaking or a replacement for them. But by embodying both sets
of commitments, the ordinary ones embedded in ordinary language and the
philosophical ones embedded in the theoretical quest for idealized philosophical concepts, they may live at odds with themselves. As philosophers
do not have the option of giving up ordinary language and the everyday
aspects of life in which it is embedded, they ought to find a way of engaging
in philosophy that does not leave them at odds with themselves and their
theoretical quests at odds with their philosophical selves.29 For this reason,
Wittgenstein commits himself to the following fundamentally important
approach to philosophy, based on a conception of the importance in philosophy of acknowledging the inescapable features of the human form of life:
The real discovery is the one that enables me to break off philosophizing when I want to.The one that gives philosophy
peace, so that it is no longer tormented by questions which
bring itself in question.30
To take this approach to philosophy is to take an optional stand on the
importance of ordinary language in philosophical thinking. It is to think

206

Whose Tradition? Which Dao?

of acknowledging the entire set of inescapable features of human life and

language as a fundamentally important constraint on philosophical reflection. Wittgensteins approach to this constraint is thoroughly pluralistic. As
he says, the order he finds in language is one order out of many possible
orders:
We want to establish an order in our knowledge of the use of
language: an order for a particular purpose one out of many
possible orders; not the order. For this purpose we shall again
and again emphasize distinctions which our ordinary forms of
language easily make us overlook.31
Similarly, the sort of meaning as use in a language-game, which he
finds in ordinary language, is one type of meaning out of others. Although
an option, rejecting this view of meaning comes at the cost of doing philosophy in a way that generates a specific form of philosophical conflict and
confusion. For philosophers committed to escaping this sort of confusion
and committed to recognizing their fundamental need to accept as given
features of the human form(s) of life, the demand to understand meaning as embedded in ordinary language is inescapable. When philosophers
acknowledge that given features of ordinary human life can trump demands
to engage in forms of foundational theorizing, such philosophers will no
longer be compelled to seek to privilege, on theoretical grounds, one ethical
tradition over others.32

Limits of the Law of the Excluded Middle: Confucius

As I have indicated, Wittgensteins self-conscious efforts to undermine the
force that theory can have in philosophy represents an approach that makes
sense and is possible only when philosophical theory has already been long
practiced. His therapeutic approach rests on his seeing the problem of how
to develop a form of reflection that can function to remove oneself from the
grip of philosophical theorizing. This form of therapeutic reflection requires,
moreover, an individuals engagement in uprooting deep-seated tendencies of
thought. But the problem of the relation of reflection to theory will look
quite different at a point when ethical and moral reflection arise, prior to
the construction of well-established traditions of theoretical reflection on
the foundations of ethics and morality.33 In this different context from

The Dilemmas of Contemporary Confucianism

207

Wittgensteins, how does Confucius construct a justified form of ethical

reflection that does not require MacIntyres form of ethical theory and
debate, including the construction of a fundamental ontology?
Confuciuss path in ethics is quite different both from MacIntyres and
Wittgensteins, though his path is closer to the latter. Confuciuss project
of transmitting the ethical tradition of his teacher, the Duke of Zhou, and
the sage kings that preceded him combines embeddedness in an ethical
tradition, carrying out the key practices required by that tradition, and
reflecting on the meanings of those practices and on the meanings of its
constitutive ideals.
We get enough descriptions of the sorts of discussions that Confucius
engaged in and avoided to be able to spell out a rough outline of the contours of his form of ethical reflection. Here is what we know: Confucius
rejected reflection not arising out of the practice of self-cultivation, was
not interested in speculation, and rejected discussion of theoretical topics.
I return here to passages, some of which I discussed in earlier chapters, in
order to show how Confuciuss pedagogy impacted particular questions he
took to be salient.
. . . .
.
Zigong said: The cultural ornamentations (speech and conduct)
of the Master: they can be heard. As for our aster words on
nature and the heavenly way, we cannot hear them.34
,,,.
Our Master didnt like to talk about strange happenings, violence,
social disorder, and the supernatural, including gods and ghosts.35
,,.
The Master rarely spoke of profit, fate, or ren (as an abstract
ideal).36
Although how we are to understand these claims is a matter of interpretive controversy, these passages seem to establish parameters on what count
for Confucius as useful topics of ethical reflection. Instead of focusing on

208

Whose Tradition? Which Dao?

abstract questions about heaven and nature, he was interested in encouraging

attention to details of the proper forms of everyday conduct.
. . . .
. . . .37
Zixia said, Someone who in treating the virtuous as virtuous,
considers sexual desires lightly; in serving parents, is able to
do his utmost; in serving the monarch, is able to sacrifice his
life; in interacting with and joining together with friends and
in speaking is trustworthy; although such a person says that he
has never studied, I certainly call him learned.
. ,. ,.
If people merely learn (how to put moral ideals into practice),
but do not reflect on them, they will learn in vain. If people
merely think (about how to become morally good), but do not
learn (how to do it), they will face trouble.38
.. . . . . .
. .
Zilu asked about serving ghosts and spirits. Our Master responded,
You havent yet been able to appropriately serve (living) people.
How could you appropriately serve the ancestral spirits? Zilu
further asked, May I ask about the problem of death? Our
Master responded, You havent yet understood the problem of
life; how can you understand the problem of death?39
The general thrust of these passages is that a learner of dao should stay
focused on ethical practices and then connect his reflections to those. An
aspirant to dao should avoid reflections on abstract topics removed from questions of practice. How does this approach show up in other passages of the
Analects?
Take, for example, Confuciuss discussions of (ren). He never gives a
comprehensive account of (ren).40 In fact, it seems crucial to his account
that a person who is not a sage would not be able to do that. Instead, he
makes sure that his interlocutors do not confuse characteristics others actually

The Dilemmas of Contemporary Confucianism

209

have, which, although sometimes virtues and sometimes not, are nothing but,
at best, limitations of (ren). In these passages, he understands that fully
complying with (ren) requires complying with it in all of ones actions and
motivations involved in ones ren-governed relationships. If we understand
(ren) as complete virtue, this statement will appear self-evident. Why should
we be concerned about (ren) beyond its being complete virtue? What is
the foundation for (ren)? He does not say. What is the basis for (ren)
in human nature? He does not say. In contrast with raising these foundational questions, his approach to his interlocutors is to get them to correct
one-sidedness in their behavior and to practice (li), which is in important
ways the embodiment of (ren). So following this approach, for novices to
understand (ren) better, they first need to improve their ethical practice.
Presumably, the better their practice, the better their reflection on what
(ren) is and requires. But taking each step along the way of improvement, a
novice, just as Confucius experiences in his practice, will not have exhausted
the requirements for completely becoming (ren) or for understanding
(ren) completely. So what we have here is the following approach to ethical
pedagogy: One will come closer to fulfilling the requirements of (ren) and
understanding it by practicing the variety of constitutive (ren) practices
learned from a master. In contrast with what MacIntyre requires for making
truth claimsan ontology of virtue and a related account of the soul and
its proper ordering, which the utterer must, on pain of incoherence, claim
will not be defeated in dialectical encounters with opposing accountswill
this focus on practice and reflection make it possible for Confucius to make
truth claims about (ren)?
Confucius certainly thinks so. The autobiographical account he offers
of his own development indicates that he believes self-cultivation and related
reflection over a lifetime can bring one to an understanding of the heavenly
order:
.. . .
. ..
I have set up my mind in cultivation since I was fifteen. I have
had my stance since I was thirty. I have been no longer bewildered
since I was forty. I have understood the heavenly mandate since I
was fifty. I have thoroughly understood others words since I was
sixty. I have no longer surpassed the code rules while following
my heart and minds desires since I was seventy.41

210

Whose Tradition? Which Dao?

If one thinks, as Confucius does, that his forms of practice and reflection get one closer to a correct understanding of the ideals he is committed
to, then his approach, without theory, will not prevent him from making
truth claims. One key type of truth claim that Confucius makes is the
claim that (ren) is not entailed by this or that behavior. If Confucius
does not have a theory about (ren) that would support his claims about
(ren), how can he make such claims? Will the focus he gives to learning
and reflection on learning license him to make such truth claims?
We must assume that Confucius held that the combination of learning, practice, and reflection that he advocated did justify him in asserting
the truth of various claims he makes, including those he makes or implies
about specific behaviors not adding up to a persons being (ren).
We can, however, pose the following dilemma for Confucius. He
must either present a theory of the relationship among (ren), practice,
and reflection to justify his teachings or not. If he can, then he will be offering a theory of ethics, which we will be able to put up against MacIntyres
tradition. As a result, he will be forced to enter into the dialectic that
MacIntyre forces him into. He will thereby, however, need to develop a
theory to counter MacIntyres Aristotelian theory. In that case, he will
not have been allowed the opportunity to demonstrate the possibility of
a non-theoretical approach to ethics. However, without the development
of just such a theory, he will not be able to respond to MacIntyres challenge and, thus, will not be able to justify the commitments underlying his
pedagogy.
The only way out here is for Confuciusor his followers who wish
to escape MacIntyres dialectical demandsto appeal to justificatory features
of our practical situations, which call for acknowledgment of the basic
character and force of such justifications without theory. Of course, these
sorts of justificatory features are not far off. As soon as we desire them,
they appear.
All we need here is acknowledgment of contexts in which we rightly
employ first-order (practical) justifiers that do not rest on some second-order
(theoretical) justifiers for their justification. These are a dime a dozen. We
operate with such first-order practical justifiers all the time. I take it that I
just bought a cup of Starbucks coffee, and this is reason for thinking that
the cups contents, which I cannot see because of the plastic lid on the cup,
is coffee. How do I know it is coffee? I just bought it. If someone thinks
this is not a good enough reason, I can also have a sip. If someone is not
yet convinced, I might find further reasons, but the context will determine

The Dilemmas of Contemporary Confucianism

211

a stopping point for the intelligibility of such further questions.42 And, of

course, context matters. What justifies my view that the cup contains coffee
in the Starbucks context when there has been no publicity indicating that
Starbucks coffee really is some coffee substitute, not the real thing, will differ
from what would justify it in the context of a government lab testing products for truthful accounts of their chemical makeup and potential hazards
of drinking these products. Under normal coffee-drinking contexts, I do
not need to use the government laboratory criteria to resolve the customer
context question. Similarly, when I face the practical problem of wanting
coffee, I know where to go.43
We are embedded in practical contexts in such a way that the mere
logical possibility of the falseness of a belief does not succeed in calling our
justified beliefs into question. For example, while most people would admit
that it is possible that the zebras at the zoo are really just painted mules,
this possibility will not have enough force to undermine their belief that
they are looking at zebras.44 The fact that we can ordinarily trust signs at
zoos is enough to justify our belief that what the sign says the exhibit shows
us is true. In a changed context in which there is, for example, a contest
with money to the winners to find the fake animals at the zoo, the mere
possibility of the animal being fake will undermine our normal reasons for
accepting the validity of the signs.
If we follow this model of justification, then how will Confucian
moral reflection proceed? What people accept as a justifier will depend on
their traditions, practices, and background beliefs; narratives of exemplary
conduct; and the ways in which the tradition is taught. What masters
communicate to novices in the course of their learning will provide a set
of norms the novice can later appeal to as canonical justifiers. Questions
and challenges that make no sense within the context of such learning can
be excluded as having no clear meaning or force. So, in the same way, the
denizens of Starbucks can refuse to answer further questions about whether
the substance in the cup really is coffee. Morally reflective persons, including
Confucius, aiming to improve themselves within the context of a specific
tradition, can with justification reject certain questions of theoretical justification because the questions do not make sense within the practical, moral
context of that tradition.
This approach should be all the more persuasive for situations in
which morally reflective persons believe that the practical moral problems
they confront are, in their own right, sufficiently important, complex, and
difficult to require avoiding questions of moral theory that would deflect

212

Whose Tradition? Which Dao?

novices from the fundamental problems of self-cultivation at hand or engender skepticism. In such a tradition, a practical moral agent might judge
that the best use of time and energy is to work to improve his or others
everyday behavior in light of already accepted ideals. For example, in facing
an Aristotelians complaint that he lacks a proper account of the unity of
the virtues, a Confucian may take this complaint and the reflection required
to address it as both irrelevant to his project and not likely to produce any
positive results. No matter how one resolves the issue of whether there is a
unity to the virtues, both traditions will want to make sure that dispositions
toward courageous behavior are limited by appropriate moral standards.
The Confucian will call the petty mans theft courageous but unjust, and
the Aristotelian will call it cowardly because it is unjust. Nevertheless, both
will aim to affirm the same conduct even if what they call it differs. The
Confucian moral agent may rightly judge that he will not help himself or
others to improve through the resolution of this theoretical problem. He
will, moreover, be able to justify refusing the question on the practical
grounds that learning the tradition and mastering it do not require raising
this theoretical question.45
This practical reasoning might even be further strengthened should
the Confucian have an immediate sense of a moral crisis, arising from a
failure to engage in required practices or to understand the meaning of basic
moral ideals by reflection on those practices. So, for example, Confuciuss
concern to connect reflection on (ren) with the practice of (li) and
his related efforts to help his interlocutors to understand the ways in which
they misunderstood (ren) reflect his general sense of there being a deepseated moral crisis. In so far as a masters diagnosis of the crisis rests on his
view of the novices need for proper learning and related reflection, he will
be justified in treating MacIntyres theoretical challenges as a dangerous
red herring.46
What, then, will a Confucian say to MacIntyres challenge: that unless
and until he develops a theory and subjects it to critical scrutiny by Aristotelians, he cannot be justified in making truth claims for any of his moral
judgments? The Confucian will, if I am right about the import of practical contexts on assertions to knowledge, firmly refuse to waste time on
something irrelevant. We make truth claims all the time without subjecting
our claims to critical examination by theories. What counts as a relevant
objection to ones truth claims will be fixed by the practical context. A
Confucian can certainly, then, in the practical context in which he finds
himself, deny that his license to make truth claims has been undermined by

The Dilemmas of Contemporary Confucianism

213

his unwillingness to engage in theory, just as our Starbucks coffee drinker

can, with justification, refuse this sort of challenge. However, it is crucial
that in refusing to entertain MacIntyres challenge and in insisting on the
right to make ethical truth claims, the Confucian practitioner not get drawn
into the precise theoretical debate he seeks avoid. To achieve this result, he
will need to consider insisting that refusing the debate without theoretical
justification does not mean to refuse it wrongfully.47
Against this line of practical reasoning, MacIntyre may want to argue,
as he does in his essay, that this sort of response, which I have given to
Confucius, just collapses a distinction between truth and warranted assertability, the former of which is, in contrast to the latter, perspective independent. But making ethical claims from within a moral perspective neither
more nor less alters the logic of truth claims than making truth claims in
ethical theory does. To claim that a statement is true is to claim that it is
true independent of the particular perspective from which it is justified.
And within a particular moral tradition it is possible for there to be various forms of critical thinking, through forms of self-examination, practice,
and reflection. The practical test of whether the person is operating with a
full-blown notion of truth will be the extent to which the person is willing
to detach his or her truth claims from his or her beliefs that they are true.
This means that when he is not willing to assert that p is true for some
claim is the same as asserting, I or we merely believe that p is true. No
Confucian, having gone through the process of learning, reflection, and
self-examination at the hands of an accomplished master, would be willing
to identify the truth of Confucian claims with the adherents merely believing it. In fact, a test of the adherents understanding is whether he sees the
Confucian claim as unavoidably true and not just his private view of the
matter. It is for this reason that Wittgenstein weds understanding a concept
to the learners sense that the concept must be applied the way the master
taught it: This must shews that he has adopted a concept.48

Acknowledging Local Practices

Nonetheless, one might still have a nagging feeling that this answer evades the
deeper philosophical worry, expressed in the dilemmas of Ci and MacIntyre
introduced at the outset of this chapter. Doesnt this form of ethics reduce
Confuciuss views to a parochial ethics, a set of local customs that has no
deeper basis to make any universal claims and certainly no relevance to the

214

Whose Tradition? Which Dao?

Western tradition of ethics? Arent Confucian ritual practices, so central to

Confuciuss view of morality, simply local practices with no relevance outside
their East Asian context? The charge of parochialism rests squarely on the
view, often treated as self-evident, that (1) genuine morality is constituted
by universal principles, and (2) local practices are not universal principles,
so any ethical view constituted by local practices is not a genuine morality.
It is easy to see why someone might be committed to such a view. Local
practices not brought into conformity with genuine moral principles, one
might worry, could offer practitioners potentially immoral practices masquerading as moral practices. It is easy to see why this might be so. Moral
practices not grounded in universal principles and not corrected by reflection
on those principles could be, for all we know, immoral.
This criticism, however, supposes that because Confucian practices of
(li) are local, they lack universal resources. The criticism also supposes
that the sort of grounding that would be required to solve this problem
with local practices has to be available independent of them, typically at
the level of theory. The problem with this approach is that it arguably
misunderstands the important role that ideals play in practices. Confucius
views the practices of (li) in relation to the ideal of (ren) but also
embeds the practices of (li) in practices of self-examination, so that both
contribute to a persons progress along the path of moral self-cultivation.
The ideals that animate Confuciuss moral practices are not independent of them. That is, they are not understandable outside the practices of which they are a part. To be able to understand (ren) requires
being engaged in those practices that comply with it. Over the course
of an extended period of practice and reflection with devotion to selfcultivation, a person can gain an increasingly nuanced, deep understanding
of (ren).
The complexity of the practices of ritual and an increased understanding of these practices over time give the whole complement of Confuciuss
practices depth of character. This sort of depth, which is a by-product of
mastery over a lifetime, gives further meaning to a Confucian masters claim
to have grasped dao as it really is.49
This characterization should go some way toward undermining the
claim that without an appeal to theory, including a fundamental moral
ontology, Confuciuss practices are merely parochial in interest. For if in
practice it is possible to exhibit the ways in which practice of local rituals
constitute forms of (ren), meaning complete virtue, a Confucian will
be able to distinguish between practices that exhibit (ren) and those
that do not.

The Dilemmas of Contemporary Confucianism

215

Put to the task, without falling into the sort of theoretical debate
Confucius avoided, a defender of Confucius can use resources internal to
his or her own tradition to fend off MacIntyres challenge and continue to
claim rightfully that Confuciuss claims are true. And in so doing, moreover,
the defender can reject, as lacking salience in the Confucian tradition, the
variety of theoretical, either-or claims that are necessary both for Cis and
MacIntyres dilemmas.
What would be required to undermine Confuciuss ethical project? To
undermine Confucian ethical practice and reflection, one would need to
demonstrate that this practical project either (a) does not have the resources
for making truth claims about what is or is not (ren), or (b) tacitly
introduces a mistaken or unjustified moral ontology. I have argued that the
Confucian tradition can escape both of these problems.
I have argued that Confuciuss practices of ritual (li) point beyond
themselves to an ideal of complete interpersonal virtue, realizable in practices of ritual (li) and that these practices are embedded in a range of
practices of self-examination, designed to provide a path toward complete
interpersonal virtue. Although this account arguably shows why parochialism is not the problem that Ci and MacIntyre take it to be, there is another
way to show why Confuciuss practices are not parochial. For just as he took
the Duke of Zhou as his model for emulation, Confucius is subsequently
taken as a model for emulation by those who chose to follow him on the
path of complete interpersonal virtue.
An ethics, based on the embodiment of ideals in practices and the
meanings of those practices given by their embedded ideals, may, as it
does in Confuciuss case, be expressed by adherents emulation of previous
masters who were exemplars of the ethical tradition. To put oneself up or
to be put up by others as an exemplar of a set of practices is one way to
establish the meaning of the practices and their validity. I will discuss these
claims in turn.
There are various ways to establish, in a broad sense of that term, the
meaning of a moral ideal lived out in a set of moral practices. One way of
establishing meaning is by relating the ideals to practices. This establishes
both the ideal meaning of the practices and the concrete meaning of the
ideal. The concept of meaning that I am using here is a commonplace one.
I can ask what the meaning is of the virtue of friendliness and can answer
that in part by citing a set of instances of friendliness as central to our
use of the word friendly. I can also establish the meaning of a persons
specific friendly conduct by appeal to the ideal. But there is another way to
establish the meaning of moral ideals and practices, and that is in terms of

216

Whose Tradition? Which Dao?

the role those ideals or forms of conduct play in the life of a person who
exemplifies these practices and ideals.
A person whose life exemplifies a set of practices and ideals can function as an exemplar for later generations of those on the path to realizing
the ideals the person embodies. As an exemplar, he functions in various
important ways. His life may serve to inspire others. It may serve to instruct
others. It may also serve to communicate central insights about the ideal his
life embodies, about steps along the way to the ideal, and about the meaning of the ideal. His role in all of these functions is justified because later
practitioners understand him as the person who best expresses what it means
to live on the moral path to which they are committed. Later practitioners
may reject earlier practitioners exemplars. Later practitioners may use early
exemplars like steps of a ladder, which they can move beyond once they
have mastered those steps of the practice.50 Given the practical resources
of the sort of moral practices we have attributed to Confucius, it is easy
to see how this might happen. For, given that the ideal of (ren) is the
ideal of complete virtue in interpersonal relations and given that Confucius
himself acknowledges his failure to achieve this ideal, later practitioners may
use the life of Confucius and his teachings as stepping stones of learning
and then choose to travel beyond them. This stage of treating Confucius
as an exemplar is necessary to make sense of his teachings abstract ideals
and to motivate generations of practitioners. The form that emulation takes
for later generations of those who are on Confuciuss path might undergo
unavoidable changes because of changed contexts or deeper understandings
possible over a period of generations of practice. But the changes the emulation undergoes would have to be changes that make sense and that can be
justified in terms of the previous stage of understandings and practices of
(ren). As Confucius is reported to have said in Analects 2.11, .
.: One who infers some new knowledge by reviewing
some past knowledge can be others teacher.
The upshot of my argument is that local practices can constitute an
ideal and stand as the basis upon which the ideal is understood, taught,
and characterized.51 On this basis, one can make objective claims about the
ideal and its instances. These local practices establish the meaning of the
ideal and the basis upon which concepts of the ideal can be used to make
objective claims.
In this chapter I have challenged Cis and MacIntyres Confucian
dilemmas, which have as their bases a demand for a justifying theory. I
have done so by appealing to those practical, critical resources within Con-

The Dilemmas of Contemporary Confucianism

217

fuciuss practice and the tradition generally that give those in the tradition a
right to make Confucian claims and fend off demands to engage in theory.
We acknowledge such rights all the time. It is a philosophical illusion that
ordinary truth claims require theory. Wittgenstein sought to counter this
sort of claim, and Confucians ought, as a key mode of self-understanding,
to assert this right with confidence and without any felt need to offer a
theory to justify it.

Fingarette on Handshaking

When meeting people in China, handshaking is normal though the

handshake may be longer and may not be as firm as a Western style
handshake. Kissing and hugging is generally OUT. Back slapping and
shoulder squeezing is OK for good friends and acquaintances. Your two
hands clasped together in front of you is also a common and formal
greeting though become familiar with this action before you employ it.
Trainor Consultant Services

Introduction
In his groundbreaking book on Confuciuss contribution to philosophy,
Confucius: The Secular as Sacred, Herbert Fingarette offers a philosophical interpretation of the fundamental insights of Confuciuss teachings in
the Analects.1 This text has occasioned enormous amounts of commentary,
some positive and some critical.2 My purpose in discussing this text in this
chapter is to distinguish between the most significant part of Fingarettes
discussion, his example of ritual (li) as handshaking, and his questionable
efforts to find in Confuciuss teachings a defensible philosophical anthropology. As my arguments in earlier chapters make clear, I find no systematic
theories in the Analects. I do not even find unsystematic ones. Central to
my argument is the claim that Confucius is represented in the Analects as
offering different teachings to different people in different contexts. I have
also argued that this fact is best explained in terms of his fundamental
project of ethical intervention.3 And even though Confucius is an ethical
interventionist, that role is compatible with his making truth claims and
invoking dao-constituting ideals in his interventions and with presupposing
views about those ideals. I have also argued that it would be strange to
219

220

Whose Tradition? Which Dao?

allow interpretations of Confuciuss teachings to proceed on the principle,

adopted by many commentators, that, contrary to what he says, he had
hidden theories or implicitly held theories that he never made explicit. For
either of these interpretive principles would allow us to apply and read into
any text any theory we might wish to project onto the text. This approach
to interpretation reduces the Analects to an inkblot that simply offers some
interpreters an occasion to find their own thinking in it.4 Fingarettes interpretations of Confuciuss teachings suffer from these types of problems, as
I will show in some detail in this chapter.
Despite flaws in Fingarettes philosophical-anthropological interpretations, I find his book important for the example he offers of Confuciuss
ritual (li) when he analyzes the contemporary, primarily Western practice
of handshaking. Though his discussion of this example occupies a mere
two pages (910), these are the most memorable pages of the book. For
contemporary North American readers, these pages give a telling example
of ritual (li), one we can appreciate because handshaking is a ritual in
our culture. Fingarette helps Westerners to do something quite important
and difficult: to acknowledge ritual (li) as an important ethical feature
of our lives. A book that succeeds in this endeavor, even in two pages, is,
from Confuciuss point of view, which I share, a book well worth reading.
Would that there were more such examples. However, to be helpful in the
context of Confuciuss project, such examples would need to demonstrate
important points in ethical interventions, not be employed to illustrate a
questionable theory, as Fingarette does.
It is important to note that the handshaking example is not one that
Confucius employs. Indeed, handshaking was not a practice in early China
and was not a practice in contemporary China but has become a practice
of some people in China fairly recently. Indeed, it was not a practice in
Western cultures until the early modern period. For Fingarettes purpose,
however, it works well as an example of ritual well-known to his audience
of contemporary Westerners. For Fingarette, handshaking is a paradigm
example, even if Western, of what he thinks is Confuciuss theory of ritual,
which Fingarette, moreover, offers as a correct account of ritual in general. In
what follows, however, I argue that the general account is faulty, but offering
the example of handshaking along with other contemporary examples of
ritual can help us acknowledge ritual as a central everyday moral practice.5
First I will examine Fingarettes handshaking example. My goal in the
first section of this chapter will be to critique aspects of Fingarettes description of handshaking. While his description offers to some extent an accurate
phenomenology of handshaking, his account is distorted by his interpreta-

Fingarette on Handshaking

221

tions, which are based on his philosophical anthropology and his related
view of the nature of language. His account of an essential connection in
the Analects between ritual and mutual respect is not borne out by the text,
as there is no clear reference to mutual respect in the Analects. Second, I
will investigate the source of the philosophical distortions that weaken his
description of handshaking. Third, I will argue that these distortions arise
from his efforts to offer a philosophical interpretation of Confuciuss project
and use this example in support of Confuciuss philosophical anthropology
of ritual. Fingarettes theory of human nature wags his description and does
so by confusing an example of an ideal exemplar of ritual with its essence.

Shaking Hands
I begin this section by quoting Fingarettes handshaking example at length.
I divide his description into six sections and then proceed to analyze his
description section by section.

A. I see you in the street; I smile, walk toward you, put out my
hand to shake yours. And beholdwithout any command,
stratagem, force, special tricks or tools, without any effort on
my part to make you do so, you spontaneously turn toward
me, return my smile, raise your hand toward mine.

B. We shake handsnot by my pulling your hand up and

down or your pulling mine but by spontaneous and perfect
cooperative action. Normally, we do not notice the subtlety
and amazing complexity of this coordinated ritual act. This
subtlety and complexity become very evident, however, if
one has had to learn the ceremony only from a book of
instructions, or if one is a foreigner from a non-handshaking
world.

C. Normally we do not notice that the ritual has life in

it, that we are present to each other, at least to some
minimal extent. As Confucius said, there are always the
general and fundamental requirements of reciprocal good
faith and respect. This mutual respect is not the same as a
conscious feeling of mutual respect; when I am aware of a
respect for you, I am much more likely to be piously fatuous
or perhaps self-consciously embarrassed; and no doubt our

222

Whose Tradition? Which Dao?

little ceremony will reveal this in certain awkwardnesses. (I

put out my hand too soon and am left with it hanging in
mid-air.) No, the authenticity of the mutual respect does not
require that I consciously feel respect or focus my attention
on my respect for you; it is fully expressed in the correct
live and spontaneous performance of the act.

D. Just as an aerial acrobat must, at least for the purpose at

hand, possess (but not think about his) complete trust in his
partner if the trick is to come off, so we who shake hands,
though the stakes are less, must have (but not think about)
respect and trust. Otherwise we find ourselves fumbling
awkwardly or performing in a lifeless fashion, which easily
conveys its meaninglessness to the other....

E. This depth of human relationship expressible in ceremonial

gesture is in good part possible because of the remarkable
specificity of the ceremony. For example, if I am your former
teacher, you will spontaneously be rather obvious in your
walking toward me rather than waiting for me to walk toward
you. You will allow a certain subtle reserve in your handshake,
even though it will be warm. You will not slap me on the
back, though conceivably I might grasp you by the shoulder
with my free hand.

F. There are indescribably many subtleties in the distinctions,

nuances and minute but meaningful variations in
gesture....It is in just such ways that social activity is
coordinated in civilized society, without effort or planning,
but simply by spontaneously initiating the appropriate ritual
gesture in an appropriate setting.

A key problem with this example of handshaking is that the cultural act
of shaking hands, which is employed to support Fingarettes general theory
of ritual (li), is not neutral. His analysis of handshaking is designed
to support his view of ritual (li) as something magical, as something
that makes sense even if metaphorically, as an instance of a Holy Rite.6
Although some ancient Chinese rituals are Holy Rites, not all of them are,
even metaphorically, and defenders of Confucius need not think that all
rituals performed during that time were holy.7 I suggest that we rewrite the
example to get rid of these exaggerated characterizations. I will proceed with
a section-by-section commentary.

Fingarette on Handshaking

223

A. I see you in the street; I smile, walk toward you, put out my
hand to shake yours. And beholdwithout any command,
stratagem, force, special tricks or tools, without any effort on
my part to make you do so, you spontaneously turn toward
me, return my smile, raise your hand toward mine.

Okay. Maybe I smile, or maybe I dont. I need to look you in the

eye, at least. You turn to me, look me in the eye or smile, and raise your
hand toward mine. The second sentence sets up some contrasts that Fingarette uses throughout the rest of his description. These are tendentious.
They play a role in his view that ritual is a form of magic. The argument
he offers goes something like this: Usually if I get you to do something,
something special has to happen. I need to trick you or coerce you, and
these ploys take effort. But in handshaking, without these special efforts, I
raise my hand, and you spontaneously raise yours. This response on your part
is magical, he contends. That is, it is not the result of some non-magical
act, like a trick or coercion.
Fingarettes use of behold here introduces his magical view of
handshaking. I make no effort to get you to shake my hand. I just put
my hand forward. You just magically respond. So Fingarette is drawing a
contrast between things I get you to do through my own efforts to make you
do them and actions of yours that my actions prompt you to do, but which
arise without any efforts by me. But why call these latter actions magical?
Your action seems much less magical if we consider the way in which
handshaking is a learned practice. Learning to shake hands requires some
effort, some learning, some correction over time, and some conscious reflection on how to do it and on its meaning before a person can get good at
it. In this way, learning to shake hands for a child is a lot like learning
games and learning to speak a language. Once we have learned it from our
parents and other adults, we know how to shake hands and when we are
being prompted to do so. But there is nothing magical in this. I also know
in chess how to respond to your check of my king without you forcing
me to do so. I also know how to form new Chinese sentences once I have
become practiced in basic Chinese sentence patterns and vocabulary. But
we do not need to think of these responses as magical.
Of course, if the person seeking to shake hands with you is a complete stranger who out of context approaches you on the street, you are
not likely spontaneously to raise your hand. Instead, you might reflexively recoil. Such responses are also not usually planned. We can think of
them as spontaneous, but we do not typically think of unplanned, reflexive

224

Whose Tradition? Which Dao?

actions of this sort as magical. If I reflexively gag at the taste and smell of
stinky (fermented) tofu, is this magical? If I reflexively refuse to take some
when it is offered, is that magical? If I instinctively take the tofu when it
is offered to me because I have been taught manners, is that magical? We
could say so if we wanted, but this way of speaking is artificial and seems
designed to give these acts a special status. But the more we see ritual acts
as by-products of training, the less magical they seem. The more we see our
responses to stinky tofu as reflexive, the less magical they seem.
We can revise the first part of Fingarettes description as follows:

A*. I see you in the street; I recognize you. I make eye contact,
walk toward you, put out my hand to shake yours. In
response, you turn toward me, make eye contact, and raise
your hand toward mine. These are episodes in the practice
of handshaking.

Once we place this activity in the context of a learned practice of greeting, including the training behind it, even if we want to call it spontaneous,
that just means that it happens as a form of mastery of a practice. We can
drop referring to it as magical. Thus, Fingarette does not give adequate
weight to the ways that spontaneity in handshakes arise from mastery of
a practice, not from anything magical.
Lets turn now to passage B.

B. We shake handsnot by my pulling your hand up and

down or your pulling mine but by spontaneous and perfect
cooperative action. Normally, we do not notice the subtlety
and amazing complexity of this coordinated ritual act. This
subtlety and complexity become very evident, however, if
one has had to learn the ceremony only from a book of
instructions or if one is a foreigner from a non-handshaking
world.

Here Fingarette distinguishes spontaneous and perfectly coordinated shakes

of the hand with ones where one person controls the others actions. By
spontaneous he means not controlled by either individual. But is there
anything mysterious here? This sort of coordination happens all the time. If
you play in a musical group, you have to learn to stop and start together
and keep the same tempo. If you are a dancer, you learn to respond to
your partners movements. In both of these cases, the ability to coordinate

Fingarette on Handshaking

225

ones action with others requires practice. When musicians learn to keep the
same tempo, is their keeping the same tempo spontaneous? No, it is the
by-product of training and practice that requires concentration of a certain
sort. It cannot be done just by willing oneself to do it because a person who
has not practiced and developed the skill to coordinate will not be able to
do so. If spontaneous means that it is not a result of painstaking practice,
then we need to drop the term spontaneous here as both musicians and
handshakers do what they do from practice. If spontaneous means that
as a result of training and practice, coordinated action arises when trained
musicians set out to keep the tempo, attending to the tempo as it develops,
not as the result of some mechanical enactment of, say, a memorized list
that they mechanically carry out, then we can call it spontaneous. But that
would just mean that it is coordinated activity of well-trained practitioners
who do not do this in the same mechanical way that novices do. But we
need not think there is anything magical about this. It is not the mechanical carrying out of a process, but rather, the exhibition of a mastery of a
practice. This sort of mastery is commonplace in a wide variety of human
activities. And so it is spontaneous in this sense of exhibiting mastery of a
practice, but this sort of spontaneity is not magical.
The subtlety of handshaking becomes evident to us, according to
Fingarette, if we compare the ease of handshaking done by a skilled handshaker with (a) the difficulty it takes to learn how to shake hands based on
written instructions, or (b) the difficulty adults have learning for the first
time to shake hands in cultures where the act is performed differently from
the way it is done in their own cultures. And this is conditionally correct: if
we start with activities learned by adults from written instructions alone, we
would tend to think of handshaking learned that way as practically impossible to learn. But the apparent subtlety, arising only from this comparison,
is artificial. It is the by-product of an optional point of comparison. Why
appeal to the aberrant case of learning handshaking as an adult through
reading instructions as evidence of subtlety? Learning handshaking as a child
would be something akin to learning childhood games like Ring around the
Rosy. These are commonplace, not surprisingly subtle, and not so difficult
to learn when we compare them to other things children learn.8 Children
learn this game readily, just as they learn to coordinate behavior in this
and other games. If we asked adults who had never played it as children
to play ring-a-ring-a-roses and to learn to do so from written instructions,
learning this would not be so easy. Would the ability to coordinate singing
and actions arise just from reading? Does the fact that, for many, reading would not produce coordinated action on the first attempt show that

226

Whose Tradition? Which Dao?

this activity is quite subtle? Once we place this activity in the context of
childrens games and the ways they are usually learneda combination of
imitation with occasional announcement of rules and experience of getting
called out when one falls down too slowlywe can see that these sorts of
learned coordinations of action are not all that subtle. Handshaking does not
go much beyond childrens games in complexity or in the sorts of learning
by imitation with a few corrective rules-of-thumb from time to time. Of
course, we can mean by the term subtle and the phrase hard to learn
from written instructions, but this is a special use of the word subtle that
does not support the claim that this sort of activity is magical.
This leaves us with the following correction of passage B:

B*. We shake hands in a coordinated pattern of movement

learned through (a) imitating adults, (b) practice, (c)
occasional announcement of rules of thumb, and (d) mature
self-examination on the meaning of this practice and related
adjustments. Thus, as in example A, mastery of practice does
not make the handshaking magical.
Lets move now to passage C.

C. Normally we do not notice that the ritual has life in

it, that we are present to each other, at least to some
minimal extent. As Confucius said, there are always the
general and fundamental requirements of reciprocal good
faith and respect. This mutual respect is not the same as a
conscious feeling of mutual respect; when I am aware of a
respect for you, I am much more likely to be piously fatuous
or perhaps self-consciously embarrassed; and no doubt our
little ceremony will reveal this in certain awkwardnesses. (I
put out my hand too soon and am left with it hanging in
mid-air.) No, the authenticity of the mutual respect does not
require that I consciously feel respect or focus my attention
on my respect for you; it is fully expressed in the correct
live and spontaneous performance of the act.

What does it mean to say that the ritual has life in it and that we are
present to one and another? Fingarette places these terms in scare quotes.
Presumably, we are to understand that the ritual is not really alive and we
are not really present to one another, not in the normal senses of these

Fingarette on Handshaking

227

words at least. But this seems strange. When I shake hands with you, we
are certainly present to one another in a literal sense. We encounter one
another. I see you, you see me, and we are touching each other, engaged
in a coordinated activity. We do not get a clear explanation of these terms
from Fingarette except that at the end the suggestion seems to be that
because the act of shaking hands has by itself a meaning to it expressing
respect, we can say it has a life. So we can say that the handshake has
life, indicating that it expresses some meaning, namely, respect. Moreover,
we are present to each other, we might say, in the mode of respecting each
other. We connect. So the handshake has life in the mode of expressing
meaning, and we connect in that we both express respect for each other.
But we can drop the more paradoxical formulations here, the ones that play
into Fingarettes magical view, by saying that the act of a handshake is a
learned, conventional means for expressing respect.
The meaning of handshaking is not something all that strange. All
sorts of gestures have meaning. We do not, however, necessarily need to say
that gestures of every sort have life in them. When the traffic cop urges
me forward by moving his hand, does that urging have life in it in any
magical sense? If we say that it has life in it: that just means that it is a
motion that has a conventional meaning that we have learned by paying
attention to its role in directing traffic.
Fingarette points out that a conscious feeling of respect is not necessary for the handshake to express respect. And he is certainly correct.
Handshaking is itself, in many circumstances, though not necessarily in
all, an act expressing respect.9 The handshake, however, has to be done in
the right way to be more than an empty gesture. Sometimes it is an empty
gesture, such as when I put out my hand and quickly shake with someone
just to fulfill a social obligation. What is missing in these sorts of cases is the
attention and care I would give to greeting a person that is typically required
for the act to be completely successful. I do not need to think to myself,
I am respecting this person to express respect, but I do typically need to
attend with some care to the person to succeed in expressing respect. The
life and presence of handshakes, if I am right, are captured in terms of
the fact that handshakes done in a certain way in certain contexts succeed
in having meaning. A successful performance expresses respect. But in this
passage, Fingarette identifies the success of the handshake, its authenticity,
with its being done spontaneously, that is, without awkwardness.
There is, however, the case when someone shakes hands awkwardly,
putting his hand out too early and waiting, but the handshake does not
thereby become any the less authentic. With the right attention and care, the

228

Whose Tradition? Which Dao?

handshake, even including the awkward wait, can be fully authentic. That
is, a person can authentically express respect awkwardly. But Fingarettes
account would not allow for this case.
Fingarettes focus on spontaneity as the source of authenticity seems
to rest on his view, derived from Gilbert Ryle, that shaking hands and its
authenticity rest solely with the public, behavioral manner of the act.10 He
rejects any appeal to the inner, but he needs some substitute for the inner,
something close enough to it that is outwardly manifested, to make sense of
the question of whether or not the ritual is authentic. For him, authenticity in handshakes is gainsaid by the spontaneity of the handshake, that is,
on the way it is performed. Moreover, his emphasis on spontaneity, as he
understands it, supports his claim that such acts are magical. In contrast, I
have argued that authenticity of a ritual act does not depend on spontaneity,
but rather, on a complex array of contextual factors that make it possible
even for awkward, non-spontaneous handshakes to be authentic.
This leaves us with the following correction of passage C:

C*. The handshake has a meaning, but not just by itself. It

can be performed smoothly or not, convincingly or not,
and authentically or not. The smooth performance is
often a measure of convincingness or authenticity, but
not always. An awkward handshake can be followed by a
correction or by additional interactions that will show it to
be convincing. And even when convincing, we can suspect
that the handshake is not authentic, but additional context
can provide support for its authenticity. Even in such cases,
the question arises whether further contexts are relevant to
determining authenticity. Thus, we can say that the meaning
of handshakes is dependent on context.

I now turn to passage D, which presents an analogy between handshaking and acrobatics.

D. Just as an aerial acrobat must, at least for the purpose at

hand, possess (but not think about his) complete trust in his
partner if the trick is to come off, so we who shake hands,
though the stakes are less, must have (but not think about)
respect and trust. Otherwise we find ourselves fumbling
awkwardly or performing in a lifeless fashion, which easily
conveys its meaninglessness to the other.

Fingarette on Handshaking

229

Here, Fingarette makes two different claims. The first is that the success of
the acrobatic act depends on the acrobat not attending to the success of the
act while performing it. The second is that lack of trust in her partner will
undermine the acrobats performance. In the case of acrobats, these claims
make some sense. Lacking trust may affect a performance. The acrobat may
hesitate and disrupt her performance. And thinking during the performance
about whether she should trust her partner may disrupt her performance
even if she does trust him. But is the handshaking case similar?
If I wonder whether I really respect the person I am shaking hands
with, that thought does not necessarily undermine my handshake. I can
choose not to express my uncertainty and put my energy into making a
convincing performance. Moreover, I can set my reservations aside. If I am
practiced enough at handshaking and on setting aside my reservations, this
will not affect my performance. In fact, if having the feeling of respect were
required to express respect, theatrical performances of successful handshakes
and various forms of dissembling would not be possible. Moreover, lacking
respect for the person I am shaking hands with does not seem to undermine the handshake in the same way lacking trust undermines an acrobatic
performance. A split second hesitation for the acrobat spells disaster in a
way that a split second hesitation by the handshaker does not. This is true
in part because, as I have already argued, there are recovery processes for
the handshaker. My awkward beginning can be supplemented by added
warmth and response later in the handshake or in the follow-up to it. Or,
I can make a self-deprecating comment, making fun of my awkwardness
as a way of negating any negative meaning it might appear to have. But
this is also possible because handshaking can convey different meanings. I
might want my handshake to convey some reservations. I might wish to
convey that I am willing to engage with the person I am shaking with, even
though I express some sense of reserve toward him. An awkward beginning
may convey something appropriate in the context. Handshaking conveys a
family of meanings, not just one meaning.
I offer, then, the following revised account of section D:

D*. Although my life depends on trusting my acrobatic partner,

my life does not depend on respecting my handshake
partner. If I lack respect for my handshaking partner, I
might not wish to shake hands but may do so to be polite,
or I might only wish to convey my sense of reserve toward
the person by offering an unenthusiastic or quick shake.
Handshaking can convey different meanings. There is no

230

Whose Tradition? Which Dao?

single ideal form of handshaking. We can recover from a false

beginning or convey different attitudes by different types
of handshakes. We can conclude that handshaking involves
mutual adjustment between handshakers.
I now turn to section E and Fingarettes depiction of the character of
the depth of a handshake.

E. This depth of human relationship expressible in ceremonial

gesture is in good part possible because of the remarkable
specificity of the ceremony. For example, if I am your former
teacher, you will spontaneously be rather obvious in your
walking toward me rather than waiting for me to walk
toward you. You will allow a certain subtle reserve in your
handshake, even though it will be warm. You will not slap
me on the back, though conceivably I might grasp you by
the shoulder with my free hand.

Fingarettes analysis of the connection between the depth of human relationship and the specificity of the ritual seems right and important. The
way in which handshaking expresses the kind of bond we have with a
person depends, however, on the relationship between the handshakers.
The way the handshake proceeds will depend, in turn, on our shared sense
of what sort of relationship we have and the norms governing expressing affection and respect for such relationships. But what is missing in
Fingarettes account is the claim that the meaning of the handshake, the
so-called depth of it, is not just a function of how it is performed, but
also a function of the context of the situation in the history of the relationship. His emphasis on the manner of the handshake as the source of
depth depends on his having ignored the way in which context establishes
meaning. If I meet a former student whom I caught cheating in my
class and who subsequently complained of my unfair, preferential grading
standards and poor teaching as an excuse for his cheating, no matter how
he approaches me, his handshake with me will seem phony unless it is
accompanied by some or all of the following: a retraction of his criticisms,
an apology, some account of his change in attitude, and/or a request for
forgiveness. If the person initiating a handshake is a colleague with whom
I enjoy a close relationship but whom I have not seen in some time, the
depth of the handshake will be a by-product not just of the act itself, but
also of this shared history. We need to rewrite passage E with an eye to

Fingarette on Handshaking

231

emphasizing the importance of context for the meaning of handshakes,

also illustrated in example C.

E*. The meaning of a specific handshake is, in good part, possible

because of the context of the ceremony. For example, if
I am your former teacher and we had a reasonably good
relationship in the past, in whatever way you shake hands
with me, unless the handshake is extremely friendly or
extremely cold, it will be interpreted by both of us in terms
of our shared understanding of the amicable history of our
relationship. For example, an extremely friendly handshake
from a former student might initiate a new understanding of
our relationship, but not without some further interactions
to support it. Thus, the meaning of handshakes is dependent
on context.

I now turn to the final passage F, in which Fingarette returns to his

emphasis on spontaneity.

F. There are indescribably many subtleties in the distinctions,

nuances and minute but meaningful variations in
gesture....It is in just such ways that social activity is
coordinated in civilized society, without effort or planning,
but simply by spontaneously initiating the appropriate
ritual gesture in an appropriate setting.

One odd feature of Fingarettes description here is that it completely

eschews any specification of the agents intentions or the way in which
the ritual interactions take place over time and require mutual adjustments
on the part of handshakers. The handshakers are, in other words, improvisational performers who intentionally undertake these performances with
certain, at least general, intentions in mind and who make adjustments
in response to each others performances. In Fingarettes description, it
appears that the actors just start the handshake, and then everything else
happens automatically. Of course, this is not how handshakes proceed.
There is an ongoing process of improvisational adjustment of each person
to the other. It is true that, as improvised, much of what happens cannot
be planned in advance, but there are ongoing efforts to make sure that this
sort of close encounter does not result in embarrassment. Because handshakes are improvised, we might think of them as spontaneous. However,

232

Whose Tradition? Which Dao?

even these sorts of improvisations are possible only as they are based on
mutual backgrounds of context, history, and improvisational skills in the
practice of handshaking.
This leaves us with the following revision of this section:

F*. There are indescribably many subtleties in the distinctions,

nuances, and minute but meaningful variations in
handshaking, the meanings of which are dependent on history
and circumstance. The whole performance of handshaking
requires ongoing, improvised, unplanned adjustments to
make sure the performance is carried off without causing
the other to lose face. Thus, as in passage D, handshaking
involves mutual adjustment between handshakers.

Now we have a revised, complete description of the ritual of handshaking, stripped of Fingarettes distortions and attributions of magic:
I see you in the street; I recognize you. I make eye contact, walk
toward you, put out my hand to shake yours. In response, you
turn toward me, make eye contact, and raise your hand toward
mine. These are episodes in the practice of handshaking. We shake
hands in a coordinated pattern of movement learned through (a)
imitating adults, (b) practice, (c) occasional announcement of
rules of thumb, and (d) mature self-examination on the meaning
of this practice and related adjustment.
The handshake has a meaning but not just by itself. It
can be performed smoothly or not, convincingly or not, and
authentically or not. The smooth performance is often a measure
of convincingness or authenticity, but not always. An awkward
handshake can be followed by a correction or by additional
interactions that will show it to be convincing. And even when
it seems to be convincing, we can suspect that the handshake is
not in some sense authentic. But additional context can provide
support for its authenticity.
Although my life depends on trusting my acrobatic partner,
my life does not depend on respecting my handshaking partner.
If I lack respect for my handshaking partner, I might not wish
to shake hands, or I might only wish to convey my sense of
reserve toward him. Handshaking can convey different meanings. There is no single, ideal form of handshaking. So we can

Fingarette on Handshaking

233

recover from a false beginning or convey different attitudes by

different types of handshakes.
The meaning of a specific handshake is in good part possible because of the context of the ceremony. For example, if
I am your former teacher, and we had a reasonably good relationship in the past, however you shake hands with me, unless
the handshake is extremely friendly or extremely cold, it will be
interpreted by both of us in terms of our understanding of the
history of our relationship. An extreme handshake might initiate
a new understanding of our relationship but not without some
further interactions to support it.
There are indescribably many subtleties of distinction,
nuance, and minute but meaningful variations in handshaking,
the meanings of which are dependent on history and circumstance. The whole performance of handshaking requires ongoing,
improvised, unplanned adjustments to make sure the performance
is carried off without causing the other to lose face.

Diagnosis I
The criticisms I have offered of Fingarettes description of handshaking
indicate a serious deficiency in Fingarettes account of the meaning and
authenticity of handshakes. I have argued that (a) he does not give adequate
weight to the ways that spontaneity in handshakes arise from mastery of a
practice (passage A), (b) aspects of the handshake he labels as magical are
forms of mastery seen in a variety of practices (passage B), (c) the meaning of
handshakes is dependent on context (passages C and E), and (d) handshaking involves mutual adjustments between handshakers (passages D and F).
It is important to understand better the source of these distortions. Not
only does Fingarettes account have difficulties, it also does not capture Confuciuss teaching. I will argue that these two problems are related. Fingarette
wishes to offer a philosophical reading of the Analects that ascribes to Confucius
a philosophical anthropology, a theory about our nature as human beings. I
have argued that Confucius has no such theory and that his ambition is primarily practical, not theoretical. But I would argue that the quest for a theory
also leads Fingarette to distort the phenomena involved in handshaking.
The theory that Fingarette finds in the Analects and endorses, based
on its relation to contemporary philosophy of language and philosophical
anthropology, is this:

234

Whose Tradition? Which Dao?

A. Men become truly human as their raw impulses are shaped by

(li).11

B. All (li) is expressive of man-to-man-ness, of reciprocal

loyalty and respect.12

C. All (li) is understood by Confucius in terms of the

metaphor of sacred ceremony or Holy Rite.13

D. Performance of (li) involves some fundamental magical

power that results in a person being able to accomplish
his will effortlessly through ritual, gesture and incantation
that is, spontaneously,...of itself.14

E. For ceremony to be authentic one must do it in a

specific way: with seriousness and sincerity, with personal
presence, that is, spontaneously.15

F. [T]he spiritual is public, outer.16

G. These magical claims of the Analects are irreducible.17

H. This magic power is the essence of human virtue.18

I have argued that Confucius offers no abstract theory of human

nature to ground his vision of ethics. Instead, he is at pains to make the
point that reflection should arise out of practice. He makes no substantive
claims about human nature. And although he does talk about (ren), he
neither claims that to be (ren) is to be truly human, nor does he claim
that every case of ritual (li) is expressive of mutual loyalty and respect.
It takes an elaborate and questionable reconstruction of the text to offer
these interpretations.
As I argue below, this account distorts the text. I leave aside the question of whether the spiritual is outer, which others have questioned.19 But
I would add that the question of whether the spiritual is inner or outer
is not really a question that Confucius had any interest in. His interest in
ethical intervention leaves this sort of ontological question aside. In addition,
Confucius never claims that all ritual expresses mutual loyalty and respect.
In the Analects, a person respects someone in a position of authority or
whose achievement of (ren) is higher than the persons own. Respect is
not typically mutual. Even passages that might come close to asserting that
respect is mutual avoid this formulation:

Fingarette on Handshaking

235

. . . , .,
., .
Master Jikang asked: To cause people to be respectful, loyal,
and industrious, how could it be done? Our Master replied, If
a ruler can face them with dignity, then they will be respectful;
if the ruler can be filial and merciful, then they will be loyal
to him. If the ruler can raise up the best ones, and educate the
incapable ones, then they will be industrious.20
The peoples respect for the ruler is not the result of the rulers respect
for them. Rather, he is dignified (zhuang ), and that in itself engenders
respect. This model of one-way respect also seems embedded in the important North Star model of rulership:
.. . .
A ruler should conduct government with virtue, should be likened to the North Star, lodging in its right position, while the
myriad stars show submissiveness to it.21
The ruler rules with virtue but receives ritualized acts of submission
from those below. This does not fit a model of mutual respect. And in the
following passage, a rulers respect toward those in a lower social position
who respect the ruler does not get mentioned in a context in which it could
be mentioned if it were important:
. . . .
. .
Our Master told High Minister Zichan, There are the four
ways of being a good ruler: in his conduct toward himself, he
is reverential; in service toward those above him, he should be
respectful; in support for the people, he should be kind; and in
employing his people, he should be righteous.22
Moreover, some passages in which respect (jing) occurs are not
concerned with respect toward another person, but rather, respectful diligence in ones own action. One might say they concern respect for the task
at hand. This includes ritual tasks.

236

Whose Tradition? Which Dao?

. . . . .
. . . . ,
. .
Sima Niu worried, asking, All others have their brothers, I only
have none. Zixia answered, I have heard this saying: Life or
death are fated, rich and honor lies in heaven. If an exemplary
person (junzi) shows respectful diligence without being
neglectful, dealing with others respectfully and complying with
ritual, all of the people in the whole empire are his brothers. Why
will an exemplary person worry about being without brothers?23
Key to this translation is the grammatical point that in this passage (jing)
lacks an object. It is, then, an intransitive verb. It means being respectfully
diligent.24 The passage also mentions dealing with others respectfully
(yurengong) but does not clearly link respecting others, as Fingarette
does, with ritual. The disconnection between mutual respect and ritual is
even clearer in Analects 3.26:
. . . . .
Our Master said, focusing on high position without generosity,
conducting ritual without respectful diligence, overlooking funeral
arrangements without sorrow, how could I look onto these things?
Confucius requires that ritual be conducted with respectful diligence, but
this would not add up to the claim, at least not obviously so, that all ritual
involves mutual respect.25 The Analects never attempts to explicitly associate
(shu), sometimes translated as reciprocity, with respect (jing) or
(gong). And the idea we find later in Mengzi, in his statement of the symmetry of respect, He who loves others is always loved by them; he who
bestows his jing on others always enjoys their jing in return (Mencius 4B:
28), is not found in the Analects.26
But perhaps this argument goes too fast. We might think, as the
traditional commentators often think, that we can trust Mengzis accounts
and the accounts of key topics from the Four Books as indicators of Confuciuss views. My own approach is to look for differences, and moreover,
to be reluctant to attribute a view to the Confucius of the Analects without
sufficient textual evidence to warrant it. Moreover, because we know that
the Analects is authored by various writers and that the direct connection

Fingarette on Handshaking

237

between utterances attributed to Confucius is impossible to establish, it

makes sense not to expect a single point of view on any topic. Moreover,
Confuciuss project of cultivation of (ren) need not directly aim at unified accounts of human nature and ritual, and so on. Caution requires
that we neither assume some unity of point of view nor unity with other
texts of the tradition. Nonetheless, we might also think that making connections between passages and seemingly disparate claims might help make
the Analects intelligible.
It appears, however, that Fingarette finds connections among loyalty,
respect, and reciprocity in the Analects. This seems evident from his claim,
quoted above, that all ritual (li) is expressive of man-to-man-ness, of
reciprocal loyalty and respect. For Fingarette, li, ren, and reciprocal loyalty
and respect are essentially connected. But we can wonder how accurate this
claim is. Resolving this question requires a closer inspection of Analects
4.15, in which Confucius claims that his teaching has one thread running
through it, and his disciple Zengzi claims that the one thread is composed
of (zhong) and (shu).
Standard translations of (shu) include fairness, reciprocity, and consideration. And several passages in the Analects explicate (shu) in terms of
the so-called negative golden rule: that which I do not desire I should not
impose on others.27 Anther passage, Analects 6.30, connects this formulation
of the golden rule with (ren):
. , . . .
. ...
. .
Zigong commented, If there is one who can broadly offer
favors to the people and is able to help the masses, what do you
think of him? Our Master said, Could he be said to be
(ren)? How could he be limited by being (ren)? Wouldnt he
certainly be a sage? Even Yao and Shun still lacked in it. Those
who are ren in establishing others desire to establish themselves.
In helping others to succeed, they desire to succeed themselves.
Being able to get examples from nearby can be called going in
the direction of becoming (ren).28
One might think that this passage seems to make a nod toward an
ideal of reciprocity. People who seek to be ren seek to establish ren in others,
not just in themselves. And they do this by being an example for others.

238

Whose Tradition? Which Dao?

They also wish their own success to be harmonized with the success of their
people. But whether this passage supports any notion of mutual respect is
not clear. Nevertheless, lets suppose that it does. And if we do suppose that
it does, then by using these passages, we could construct something like the
following argument in favor of Fingarettes claim:

I.

The one thread is composed of (zhong) and (shu).

II.

The requirement of (shu) is the requirement of respect

in all encounters.

III. The requirement of (zhong) is the requirement of good

faith in all encounters.

IV. All encounters are governed by ritual.

V.

VI. So all rituals, expressive of (ren), are governed by

(zhong) and (shu).

VII. Rituals expressive of (ren) are typically (and

paradigmatically) face-to-face rituals.

VIII. So when rituals involve two co-participants engaged in

a face-to-face ritual encounter (as in handshaking), each
participant is ritually required to display to the other
(zhong) and (shu).

IX. Because in such encounters each is ritually required to

display to the other (zhong) and (shu), such rituals
involve reciprocal good faith and mutual respect.

X. So co-participant face-to-face rituals involve mutual

respect.

Ritual is the outward expression of (ren).

This argument seems to me to be an argument that Fingarette would be

happy to endorse, despite the fact that he never offers it. Crucial to the
argument is the identification of the requirement of (shu) as the requirement of reciprocal or mutual respect. I have already argued that in the
Analects passage on respect (jing), a person jings ritual performances or
superiors. So each person is not required to jing another in a face-to-face
encounter. So the question is whether it is possible to support Fingarettes
mutual respect thesis from a consideration of (shu) passages.

Fingarette on Handshaking

239

Given the range of translations offered for (shu)altruism (Chan);

reciprocity (Tu); consideration (Waley); likening to oneself (Graham); using
oneself as a measure (Lau); and deference (Hall and Ames)the challenge
of sorting out the meaning of (shu) is large. I follow Nivison, Ivanhoe,
and Van Norden in accepting the view that in the Analects, both (zhong)
(good faith or loyalty) and (shu) need to be understood as requirements
in terms of a social hierarchy. Subordinates are required to treat superiors
with (zhong). Superiors should treat subordinates with (shu).29
My question, then, is whether, with this understanding of the usage
of (zhong) and (shu), we can make sense out of a translation of
(shu) as mutual respect. It seems to me clear that we cannot, but it takes
some complicated argumentation to show this. In the course of my discussion, I will consider the character (shu) in its basic etymological meaning
of having the attitude and feeling of and indicating a set of practices of
likening oneself to another or others. I will argue that the understanding
of (shu) as reciprocity is mistaken. Finally, I will argue that the sort of
mutuality, by which I mean the sense of ourselves as a we (that is, the
concept of two people with some shared identity) in the Analects is captured
not by egalitarian, reciprocal respect, but rather, by the acknowledgment
that we are connected by ritual, by ceremonial chains. These chains will
not be the same for each ceremonial situation we find ourselves in, and the
chaining relations need not be symmetrical. So we can be part of an ongoing community and constitute a we by being connected by ceremonial
chains and related practices without those ritual chains expressing in any
clear way mutual respect.

Non-Reciprocal Bonds
David Nivisons argument that the practice of (shu) is a practice primarily and for the most part of superiors toward subordinates, comes from his
reading of the variety of passages in the Analects connected to (shu) and
to the related negative golden rule: ...what a person does not desire to
happen to himself, he should not impose on others (Analects 12.2). Rather
than discuss each passage Nivison discusses, I will offer just one such passage
to exemplify his approach. Consider Analects 12.2.
..,..
......

240

Whose Tradition? Which Dao?

Zhonggong asked how to become morally good. Our Master

said, When going out from his [own] gate, one should act as
if meeting an important guest. In assigning people tasks, one
should act as if managing a major sacrifice. Moreover, what a
person does not desire to happen to himself, he should not
impose on others. Then he will be without complaints by others
both when he serves in a state and when he serves in a fief.
Zhonggong said, Although I am not intelligent enough, please
allow me to devote myself to these teachings.
This passage is clearly discussing the relation of a superior to a subordinate.30
That being the case, in this passage (shu) as a practice does not involve
reciprocity. Reciprocity in its most basic form is the requirement of tit for
tat, the requirement that if you scratch my back, I will scratch yours. In
this context, reciprocity would be the principle in which Zhonggong likens
the people to himself and thereby adjusts the requirements he imposes
(shi) on them. But the whole idea of requirements that he imposes on the
people involves, at least in this case, him being their superior. The people
would not be in a position to impose their will on him. So in this passage
we do not find a principle of reciprocity: If I liken you to me, you must
liken me to you. No reciprocity here.31
We might think that because of the invocation of the golden rule,
understood as later texts interpret it in terms of reversibility, that we can
derive from the logic of reversibility some notion of reciprocity. Reversibility
requirements are just the requirements that, for example, a father cannot
impose on his son what he would not have accepted as having imposed on
him by his father (or some other older relative or relevant superior). The
Xunzi offers the following: To have parents one cannot repay, and yet to
have a son and to require filial piety of him is not (shu).32 This passage from Xunzi asks us to determine whether we have a consistent view
of what we find permissible to impose on a person in a particular role. If
I am willing to have imposed on myself as a subordinate what I impose
on my subordinates, then I do not contradict myself when I formulate a
general rule for such conduct. This is test of a persons behavioral consistency
toward subordinates using imagined role reversal.
We might think of this sort of reversibility as defining the practice of
likening others to oneself. But as this gives us merely one kind of likening
others to oneself, it need not be the only way. I can liken the person who
cleans my office or our house to myself by being sensitive to her situation
and her specific inclinations and desires, so as to insure that my treatment

Fingarette on Handshaking

241

of her is not offensive to her. I need not wonder what I would desire done
to me if she or someone else were to employ me as a cleaner. There is no
need to identify likening oneself to others as imagining reversing roles, even
though imagining this might be an instance of showing sensitivity.
Although I have argued that role reversal is not necessary for likening
oneself to others, I would also argue that in any case where a role-reversal
thought experiment causes one person to reject imposing a requirement
on others, it does not follow from the result of this experiment that such
reasoning supports any claim on the other persons conduct that involves
reciprocity. The role-reversal thought experiment puts me in a position to
test my reactions to see whether they are consistent.
Suppose, however, in a particular case, that I liken myself to others
by imaginatively reversing roles. How might this reasoning express reciprocity? Suppose I, a father, ask whether I would want to be treated as I
treat my son in the context in which my sons conduct has almost always
been exemplary, he has often gone beyond standard expectations of filial
duty to benefit me, and both he and I understand this. In such a case,
the father could reason that if he were the son, he would want his father
to go beyond the normal requirements of fathers to be considerate toward
their sons. This reasoning rests on an ideal of reciprocity. I decide that if
I were the son in this case, I would reasonably hope and expect that my
father would reciprocate by benefiting mein ways that would grant me
privileges or material goods that I believe I have earnedin response to
my earlier supererogatory act benefiting him.
A requirement of reciprocity between father and son would, for example, be the requirement that the father liken himself to his son, who has
importantly benefited him in ways that matter to the father, and thereby
the son feels justified in desiring that the father not impose on him and
thereby not imposing benefit him in this way. Role-reversibility thought
experiments can incorporate factors that would support reasoning about
reciprocity, but this sort of reasoning would seem to be a special case of
role-reversal thought experiments.
Another approach might argue that the father-son relationship is a
version of a benefactor-beneficiary relationship and that the negative golden
rule, not to impose on others (subordinates) what one would not want
imposed on oneself if one were a subordinate, involves reciprocity in the
following way: Superiors both benefit and make demands on subordinates,
but superiors also benefit from their subordinates. Benefit goes in both directions. Whenever one person in the relationship benefits, the other person
becomes indebted to that person. When both benefit, even if in different

242

Whose Tradition? Which Dao?

ways, both parties become indebted to the other party. But superiors might
be in a special position of being able to demand repayment and of being
able to regulate their conduct so as to achieve a proper balance between
demands and benefits. So even if the negative golden rule governs conduct
and reflection of superiors, it still is a reflection in the name of reciprocity,
a balance of benefits and requirements in the relationship. But if this were
the case, it would require specification of the meaning of the golden rule
beyond what the text offers. I conclude from these considerations that,
although the negative golden rule could get articulated in terms of some
principle of reciprocity, the Analects does not offer us such a version.
If these arguments are correct, what impact would they have on the
argument that I offered above for the claim that co-participant, face-to-face
rituals involve mutual respect? I repeat the argument and comment on it
below.

I.

The one thread is composed of (zhong) and (shu).

II. The requirement of (shu) is to show respect in all

encounters.

III. The requirement of (zhong) is to demonstrate good

faith in all encounters.

IV. All encounters are governed by ritual.

V. Ritual is the outward expression of (ren).

VI. So all rituals, expressive of (ren), are governed by

(zhong) and (shu).

VII.
Rituals expressive of (ren) are typically (and
paradigmatically) face-to-face rituals.

VIII. So when rituals involve two co-participants engaged in

a face-to-face ritual encounter (as in handshaking), each
participant is ritually required to display to the other
(zhong) and (shu).

IX. Because in such encounters each participant is ritually

required to display to the other (zhong) and (shu),
such rituals ought to involve reciprocal good faith and
mutual respect.

X. So co-participant, face-to-face rituals ought to involve

mutual respect.

Fingarette on Handshaking

243

Based on this argument, we must first of all reject premise II. Premise II
would be true only if the requirement of (shu) did not admit of the sort
of positional differentiation in a hierarchy that it seems to presuppose. Also
premise IX would have to be rejected. Even if we accept the golden rule as a
general principle that applies to everyone, not just to those in a superior position, the golden rule requirement is not a principle of reciprocity. It does not
require you to be considerate toward me because I am considerate toward you.
I conclude from this analysis that Fingarettes view that ritual involves
reciprocal and mutual respect, no matter how initially plausible that might
be for handshakingand I have earlier argued against even this claimdoes
not even apply to Confuciuss practice and reflection on ritual.
Despite these clear differences from Fingarettes understanding of ritual, Nivison, whose views have influenced my own, nevertheless, seems to
endorse Fingarettes general approach to ritual ( li). Nivison says:
Li involve countless gestures and acts that, whether required or
simply available to me, serve as signals to you and invite response
from you in such a way as to reassure both of us that you and I
are a we; and, of course it serves this function most effectively
when these acts are expected and traditional.33
But the important qualification to stress, which follows from Nivisons
account but that he does not himself stress, is that not all acts of ritual (
li) involve (shu). A superiors performance of ritual toward a subordinate
requires (shu) without the subordinate being under the same requirement.
Instead, the subordinate is required to treat his superior with (zhong)
(good faith or loyalty). And even if we accept Fingarettes translation of
(shu) as reciprocal respect, we are forced to conclude from these arguments
that not all performances of ritual (li) require reciprocal respect, for the
requirement of (shu) governs only a superiors treatment of a subordinate.
From my point of view, the Analects is even more interesting for this
result. One might think that if we cannot find egalitarian commitments in
the Analects, it will be difficult to see this texts relevance for contemporary
Western ethical reflection. Indeed, one might even think that my analysis
commits us to relegating the Analects to the ethical graveyard. But this is
not the goal of my analysis, and I would like to end by offering an alternative appraisal.
If Nivison is right about (shu) but also right in thinking that traditional rituals (li) function to assure us that we are a we, we should
look for a mutuality in the Analects that involves moral division of labor.
We become a we not through (shu) practices alone, but also through

244

Whose Tradition? Which Dao?

the performance of ritual (li) in which we are connected through a set

of ceremonial chains. But the chains do not require or entail reciprocal
respect. In fact, they seem to involve a moral division of labor. Consider
the following quotation on this point about Western ritual from sociologist
Erving Goffman:
[F]or a complete man to be expressed, individuals must hold
hands in a chain of ceremony, each giving deferentially with
proper demeanor to the person on the right what will be received
deferentially from the person on the left. While it may be true
that the individual has a unique self all his own, evidence of
this possession is thoroughly a product of joint ceremonial labor,
the part expressed through the individuals demeanor being no
more significant than the part conveyed by others through their
deferential behavior toward him.34
The following picture of an ethical and ritual division of labor emerges
here. I receive from the person on my left what I then give to the person on
my right. This differs from the picture of reciprocity, however. I do not give
back to the person on the left what I receive from him. Contrary to what
we might expect, this picture is fully integrated with Durkheims account
of ritual, fully affirmed and quoted by Goffman, in which Durkheim holds
the following strong basic principle of ritual:
The human personality is a sacred thing; one dare not violate
it nor infringe its bounds, while at the same time the greatest
good is in communion with others.35
Indeed, Goffmans fine-grained empirical studies of interaction ritual find
these chains to occur even in relationships of superior to subordinate, for
example, in relations of doctors to nurses, in which ritual requires each
to treat the other differently, as in Hello, Doctor versus Hello, Sally.36
These ritual performances allow us to reassure ourselves that we are a we,
but they do so without the requirement that each person be under the
same ritual rule. The rules are differentiated. As it goes for us, so it goes
for Confucius. If we are to follow him today, we can draw from his model
of investigation and teaching, which focuses on details of everyday ritual
and practice and eschews abstract reflection because it is not grounded in
the details of learning and practice.

Fingarette on Handshaking

245

And while the handshaking example may be in some sense exemplary

for us, we Westerners, as much as we have been influenced by the ideals of
equality, ought not to use handshaking as a model for the whole realm of
ritual. We must acknowledge contexts in which the ritual rules that make
us a we, governed by ritual propriety, do not require that we treat each
other in exactly the same ways.
I end with one final question. If we grant that mutuality does not
require (shu) and that (shu) practices do not require or entail reciprocity, how then should we translate (shu)? Can we understand it as respect
even while accepting that this form of respect is not reciprocal or mutual?
This comes down to the question of whether a non-reciprocal practice
of likening oneself to another, a subordinate, as a way of adjusting what one
imposes on that person can be understood as respect. As usual, this sort of
translation question is complicated. But if we start with the basic idea that
such likening oneself to another is akin to, or what we call, in one sense of
the word, consideration, what the Oxford English Dictionary (OED) characterizes as regard for the circumstances, feelings, comfort, etc. of another;
thoughtfulness for another; thoughtful kindness, we can find some grounds
for reluctance to embrace the translation respect. For while the OED also
assigns another meaning to consideration: estimation; regard among men,
esteem; importance, consequencethese are clearly separate meanings. I
can be considerate in this first sense toward a person to whom I give no
consideration in the second sense. This linguistic intuition, encoded in the
OED, maps onto the Analects distinction between (jing) and (shu).
To sum up, I have argued in the first section that even for us Westerners, handshaking has various meanings. It is not always meant to show
or elicit mutual respect. I have, moreover, argued in this section that Fingarettes account of the essential connection between ritual and mutual respect
is not borne out by the text. There is no clear reference to mutual respect
in the Analects. And the talk of reciprocity (shu) also does not have any
explicit relation to respect (jing). Furthermore, in the Analects ritual when
(li) and respect (jing) are spoken of together, the Analects emphasizes
engaging in ritual with respectful diligence.

A Wittgensteinian Diagnosis
I have argued that Fingarettes description of handshaking ignores crucial
distinctions we are all inclined to draw about handshaking. He runs together

246

Whose Tradition? Which Dao?

spontaneous handshakes, non-awkward handshakes, authentic handshakes,

and handshakes that express mutual respect. He also claims that all rituals
express mutual respect. He uses the handshaking example because it is a
familiar ritual and conveniently represents his analysis of ritual, which I
extrapolated in the outline above.
I have also argued that Fingarette ignores the complex views of respect
and its relation to ritual exhibited in the Analects. Although these are distinct problems, I suspect that they have the same root: the desire to offer
philosophical support for an ideal by identifying the ideal with the essential
features of instantiations of the ideal. Let me explain.
The concepts of ritual, respect, reciprocity, and spontaneity are all
family resemblance concepts.37 Instances falling under these concepts are
various but hang together through unsystematic but describable similarities.
That being the case, it would be difficult in the abstract to specify general,
essential relations between or among these concepts. But that is precisely
what Fingarette attempts to do. I have summarized those relations above.
Fingarettes example of handshaking, at least as he describes it, supports his account by providing an arresting yet familiar example, one that
can easily help us understand the essential relationship that Fingarette thinks
obtains among ritual, spontaneity, and expressing mutual respect. But as
attractive as it is, the example and account fail to clarify the variety of ways
in which rituals function in everyday life or the variety of ways Confucius
takes ritual to function.
This sort of mistake, exposed by Wittgenstein, is a Platonizing error
behind much traditional philosophy. The basic error is to identify a motley
assortment of rituals with an ideal instance of it.38 Here is one more example
of the difficulty. Take the ritual of the marriage ceremony. The goal of
that ritual is to alter the relationship between two individuals usually in a
transparent, public event, supported by a community of onlookers. When
the officiator declares the couple husband and wife, he is not expressing
respect. When the two people say in the ceremony, I do, they are not
expressing mutual respect; they are committing themselves to this altered
relationship. One might try to hold onto Fingarettes analysis by arguing that
this way of getting married expresses mutual respect because it is voluntary
(as opposed to shotgun weddings), but this point seems to me to stretch the
thesis. The marriage ceremony may be a respectful way to get married, but
that does not mean it expresses mutual respect. Here is another example.
When Catholics enter a church, they kneel and make the sign of the cross.
These are rituals. Do they express mutual respect? Do they express respect
at all? Performing these rituals expresses faithfulness to God or individuals
intentions to show faithfulness to God.

Fingarette on Handshaking

247

So the Wittgensteinian point I wish to make here is just that rituals are a motley. In some contexts, rituals express mutual respect; in other
contexts, they do not. And the latter are not deficient for not expressing
mutual respect. In some cases, even rituals that typically express mutual
respect can express other feelings or views: openness to further conversation, for example. Again, the latter are not deficient. Some authentic ritual
expressions of mutual respect are awkward. They dont just spontaneously
happen themselves, but they express respect anyway. Awkward expressions
of mutual respect might be deficient in poise but not necessarily deficient
in terms of what they express.
One might think that this diagnosis goes too far in the direction of
detaching ritual from the sort of essential ideal that Confucius and his supporters must endorse. After all, if ritual lacks a single essential ideal, how
can it play a central role in an account of ethics? I have been arguing that
it is hard to find such an ideal in the Analects. Instead, Confucius operates
by invoking dao-constitutive ideals of various sorts, understood in ways that
will help his interlocutor(s) improve their conduct and reflection. He offers
piecemeal interventions, not systematic theories.
Furthermore, it is possible to articulate an Analects-inspired worry
about Fingarettes account of ritual, one that arises out of Confuciuss project
of ethical intervention. Key to Confuciuss project is his refusal to engage
in abstract reflection that is not grounded in learning and practice. One
danger of abstract reflection not grounded in practice and learning is that it
offers a confused account of ritual that impairs practice and reflection. If we
think that all rituals express mutual respect, we are likely to feel confused
about rituals such as marriage and genuflection. If we think that authentic
rituals are spontaneous, we may feel a need to try to make the ritual of
swearing on the Bible in court spontaneous. If we think the ritual has to
happen spontaneously, we may not see the need to intervene thoughtfully to
help our ritual co-participants recover from awkward enactments of ritual.39
Confuciuss interventions are designed to help to promote a rich and
complex grasp of how to develop the virtues key to sustaining the variety of
relationships we find ourselves in. Mastery of ritual is central to his vision,
but he does not offer and would find problematic a narrow philosophical
account of ritual that has the potential to cause confusion and impede
actual practice of ritual.
There is, however, a deeper problem with Fingarettes account of Confuciuss basic insight: Human life in its entirety finally appears as one vast,
spontaneous and Holy Rite: the community of man. This, for Confucius,
was indeed an ultimate concern; it was, he said, again and again, the only
thing that mattered more than the individuals life itself (3:17, 4:5, 6:8).40

248

Whose Tradition? Which Dao?

This approach to Confuciuss teaching treats conduct in accordance with

ritual (li) as a comprehensive ideal, with (ren), which refers to the way
in which ritual (li) is performed by the agent, as conceptually connected
to ritual (li). So there is only one thing that matters: ritual (li) and
how one performs it (ren). But this interpretation depends on taking
some passages about (ren) and ritual (li), specifically 12.1, as the final
teaching while ignoring those passages that offer alternative accounts of
(ren) or ritual (li). Moreover, Fingarette never asks the question why
Confucius provides different answers to his interlocutors questions about
(ren), and he never, as a result, raises the question of why he gives the
Analects 12.1 answer to Yan Hui but not to others who ask about (ren).
That is, he lifts some statements out of the context of the total Analects
corpus and treats them as the true teaching. Contrary to his own best
insights about the performative character of language, however, Fingarette
fails to try to understand Confuciuss dialogues with his interlocutors as
themselves made up of speech acts that involve interventions, urging them
on, holding them back, getting them to reconsider their one-sided evaluations, getting them to acknowledge an ideal that they had under-appreciated,
not a straightforward articulation of a complete account of ritual, (ren),
or human nature.41
In fact, I would like to offer the following interpretation of Analects
12.1. While it is clear that we can examine the variety of Confuciuss statements about the constitutive ideal making up dao as specifying necessary
elements of dao, it would be a mistake to think that these altogether give
us a theory of dao. We do not get the sort of systematic account of the
relation of these elements or of the basic metaphysical and epistemological
character of dao and our knowledge of it from the Analects, for these sayings and reports of dialogues do not amount to a theory. Also, we never
find any arguments that would show that Confucius has spelled out all of
the elements of dao. In fact, that sort of claim would require some sort of
theory, which was never given.
But even if Confucius, arguably, had spelled out all of daos elements,
it would not follow from the facts about those elements that his discussions
offer a theory of dao. My football coach may have mentioned all of the
crucial elements of physical training, but that does not mean that in his
training sessions he constructed or spelled out a theory of the true meaning
of football. We do not even know that he had formulated such a theory,
which he may have only discussed at home with his wife or possibly with the
captain of the team. No, Analects 12.1 does not offer an element of a theory;
it offers some counsel for Yan Hui. Given his stage of development, Yan Hui

Fingarette on Handshaking

249

needed to acknowledge the importance of (li) and devote himself to it.

And what he needed to acknowledge at the time he raised the question of
(ren) with Confucius differs from what his fellow classmates needed to
acknowledge and devote themselves to when they ask their own questions.
Surely, Analects 12.1 indicates an important aspect of Confuciuss
teachings. But what is his point? Why does the writer of 12.1 think it
important to report on this dialogue? It shows that at some moment in a
persons progress on the Way, she may need to acknowledge the importance
of (li) in the development of virtue and, as a result, devote herself to
practicing it. Failure to take up ritual in the right way can cause the adherent
some trouble. If this is a correct way to understand 12.1, then Fingarettes
use of the handshake example might be better used to help those of us who
are at a stage of moral development in which our failure to acknowledge
(li) has caused a problem. That is, we should, following Confuciuss
interventionist model, use the handshake example and related examples for
this purpose, not for the purpose of persuading us to accept a questionable
theory that runs together an example of ritual with a general ideal of ritual,
and then, in a further problematic step, confuses the general ideal with its
essence. In the next chapter, I will investigate how we might develop an
approach to ritual that sees it as complex.

Acknowledging the Given

Our Complicated Form of Ritual Life

Knowledge is in the end based on acknowledgement.

Wittgenstein, On Certainty, section 378

Introduction
In Chapter 8, I argued that Fingarettes account of handshaking in the Analects offers Westerners a striking example and helpful start at thinking about
the phenomenology of handshaking, as it provides the outlines, though in
places mistaken, of ritual interactions. I offered a diagnosis of the problems
in his account of ritual and the related phenomenology of handshaking.
In this chapter, I turn to Erving Goffmans sociology of ritual interaction
as a way of broadening and correcting Fingarettes account of ritual and
providing an analytical tool for helping us acknowledge the role of ritual
and its ethical significance in everyday interactions. Using the example of
handshaking, I take up the theme of this discussion on ritual: the human
need for acknowledgment. I argue that when thinking about the problem
of ritual, a topic discussed frequently by Confucius, from a contemporary
Western, philosophical perspective, what we need first is a strategy for bringing about an acknowledgment of ritual as central to our human form of
life and to any adequate account of living well.
In the first section of this chapter, I turn to sociologist Goffmans
taxonomy of ritual as a way to produce a perspicuous overview of ritual
in everyday life. In the second section, I draw out the implications of
this taxonomy for Fingarettes account of ritual. In the third section, I use
this taxonomy to offer examples of everyday ritual designed to supplement
251

252

Whose Tradition? Which Dao?

Fingarettes handshaking example and to offer counterexamples to his narrow account. The function of this sort of overview, the demand for which
I borrow from Wittgenstein, indicates the type of account of ritual I wish
to offer: This sort of overview is designed to clarify basic features of an
often overlooked but familiar set of features of human life along with the
concepts that play a central role it. In this case, a further goal is to foster
acknowledgmentespecially by philosophersof ritual as a basic, important feature of our moral life.

Acknowledging the Familiar

A central concern of Wittgensteins later philosophy is to bring about
acknowledgment of familiar features of our lives that can easily get overlooked or misunderstood, especially when we engage in philosophy. His
focus on description of details that challenge our philosophical intuitions is
designed, he says, to bring us back to everyday usage.1 But this formulation
can make him seem more wedded to protection of ordinary language than
he is. His examination often centers on forms of life and language that are
not our own with the intention of shedding light on our lives and language
by way of comparison. Wttgensteins fundamental goal in his later investigations is to embrace our complicated human form of life and language
as given practices.2 This strategy is not designed to provide a foundation
for what we say or do; instead, it is designed to counter the tendency in
philosophy to generate accounts of concepts and forms of thinking that
evade the real limits of human life.
In On Certainty, where Wittgenstein discusses skepticism and G. E.
Moores problematic efforts to prove it false, Wittgenstein points out the
relation between knowledge and acknowledgment.3 The solution to the
problem of skepticism results, he says, not from refuting skepticism but
from acknowledging the role that knowledge claims play within the context
of human life, as illustrated by the following sentences:
376. I may claim with passion that I know that this (for example)
is my foot.
377. But this passion is after all something very rare, and there
is no trace of it when I talk of this foot in the ordinary way.
378. Knowledge is in the end based on acknowledgment.4

Acknowledging the Given

253

Here, Knowledge in the end is based on acknowledgment simply means

that our knowledge claims are rooted in forms of life that we absorb as we
grow and accept without question:
374. We teach a child that is your hand, not that is perhaps
(or probably) your hand. That is how a child learns the innumerable language-games that are concerned with his hand. An
investigation or question, whether this is really a hand never
occurs to him. Nor, on the other hand, does he learn that he
knows that this is a hand.5
Children learn these things from their background beliefs and practices,
in terms of which knowledge claims arise and can be made intelligible. To
acknowledge these rudimentary, ungrounded forms of learning and action
as the bases of knowledge is just to accept them as given, as ungrounded
forms of reasoning and justification that are central to our lives as knowers
or sentient beings. The solution to the philosophical problem of skepticism rests on this acknowledgment, not on proving skepticism false. By
acknowledging the role that knowledge claims play in our human form
of life, we can begin to use this acknowledgment as the given basis upon
which we can expose the conceptual confusions that support and sustain
philosophical skepticism.
However, the problem of ritual is different. Although it is true that
the quest for the sort of theory of ritual that Fingarette offers (and claims
to find in Analects) can be corrected by a good dose of acknowledgment of
the complexity of ritual in our lives, the deeper problem of ritual for Western philosophers concerns its widespread neglect. The Western philosophical tradition approaches ritual with indifference. Even those philosophical
approaches to ethics that might seem ripe for embracing ritual as a central
normative category, like virtue ethics, typically find no place for ritual. We
might want to think of this failure as a form of ritual blindness, by which
I mean both blindness to the varieties of ritual in everyday life and to their
meanings and significance. The problem at its root can be addressed by
examining details of everyday ritual as a step toward acknowledging that
ritual, as a part of our human form of life, constitutes a part of our moral
sensibility while conditioning other parts of life.
Fingarettes handshaking example is a start in this direction, but his
attempt to generalize his account by grafting Confuciuss approach to ritual
onto recent accounts of the performative aspect of language is not successful.6 While the performative aspect of pronouncing two individuals husband

254

Whose Tradition? Which Dao?

and wife (or a married couple) is part of a ritual, it is by no means clear

that all performatives are parts of rituals. If I say, I agree, in response to
another persons idea, I thereby agree to it. And while this is a performative
use of language, it is not at all clear that this expression of agreement is in
every case part of some ritual. (There can be ritualized forms of expressing
agreement. I could, for example, repeatedly thank someone and nod several
times if the person says something I agree with. This would be a ritual.)
Although we need to find a pattern of rituals of which handshaking is only
one instance, we would do better to proceed first by examining a variety of
examples. From these examples, we can begin to sketch out the contours
of ritual and begin to grasp the roles of various rituals in our human form
of life. How can we do this?
Fortunately, we dont have to look far for Goffmans writings on ritual action provide us with the sorts of examples we need. His reflections
occasionally involve development of suggestions that might play roles in a
theory of ritual, but often these suggestions are not fully developed into a
theory.7 Even if this were his goal, and interpreters disagree about what his
goal was, we can certainly use his work for our own purposes, even if that
requires, as it did with Fingarettes handshaking example, some removal of
theory-fueled description that does not help us in the goal of acknowledging ritual as a pervasive feature of everyday life that might have a moral
significance we need to understand.8
In what follows, I will use Goffmans taxonomy of interaction rituals to contribute to filling out the picture of the importance of ritual to
everyday life, of which Fingarettes handshaking example is an important yet
narrow beginning. A feature of Fingarettes handshaking example that might
serve as a starting point for generalization is the fact that shaking hands is
a face-to-face ritual encounter. And Goffmans fundamental contribution to
sociology, his emphasis on the need for sociology to investigate the interaction order, comes out of his insistence that face-to-face encounters have
a rich structure of their own and are not reducible to other structures of
society.9 He speaks of this order as ritualized, and his fundamental strategies for understanding how this order functionseven in a Western contextderive in part from his examination of anthropological accounts of
everyday ritual within the Chinese context.10 His accounts give central place
to the idea that participants in face-to-face encounters need to engage in
face-protecting practices.11 In so far as his accounts are successful and give
us a reason to acknowledge and take seriously the underemphasized ritual
aspects of our own everyday behavior, they give us the very thing we need

Acknowledging the Given

255

to counter the widespread view of the irrelevance of ritual for understanding

ethics. In addition, by emphasizing the dramaturgical, performance aspects
of ritual, Goffmans accounts offer an alternative to Fingarettes emphasis on
the magical character of ritual. Moreover, Goffmans accounts offer a lucid
overview of the rich variety of everyday ritual practice in Chinese culture,
not just the type represented in handshaking.
It should not be a surprise that ritual has been under-appreciated in
the West. Philosophers have tended to distinguish between morality and
etiquette to the latters denigration. This has led to an under-appreciation
of the ethical significance of ritual. North Americans also tend to be blind
to the sorts of cultural traditions that are home to ritual. North American
middle-class attitudes, reported on by Robert Bellah et al., evidence a widespread misunderstanding of or inattention to their own culture and traditions.12 It is then no wonder that North Americans tend to misunderstand
ritual, which has its basis in culture and tradition. What Bellah calls North
Americans first-language of self-understanding is a form of individualism
or utilitarianism that tends to render these aspects of our lives opaque and
distorted. Bellah argues that his interviewees show some level of acknowledgment of their own reliance on tradition but in ways that their first-languages
make difficult for them to understand. This sort of fundamental failure to
understand the rituals of our own culture can result in a failure to support
and sustain those important features of tradition that we ought to care about
because they present to us a set of norms governing our interactions with
others, which we depend on every day. In our conduct, we acknowledge
but tend to misunderstand our cultures rituals.
Goffmans distinctive contribution to reflection on the self is to
emphasize those aspects of being a self that involve its social production.
He offers an account of the social production of the self that is multifaceted,
understanding our social production in terms of drama, ritual, and games.
Although these frames might seem to be incompatible, as games emphasize
competition and manipulation and ritual emphasizes respect for the social
order and the sacred status of those whom we encounter in ritual, these
frameworks seem reconcilable. For example, game-like calculations can serve
as performances of a sort that play a role in ritual.
Against the deep-seated tendency to distinguish ritual, including etiquette, from morality, Goffman offers an account influenced by Durkheims
view of ritual, which is based on the fundamental idea that individuals are
sacred objects, to be treated with care and caution: The human personality is a sacred thing; one dare not violate it nor infringe its bounds, while

256

Whose Tradition? Which Dao?

at the same time the greatest good is in communion with others.13 This
formulation is reminiscent of Kants second formulation of the Categorical
Imperative: Act in such a way that you treat humanity, whether in your
own person or in the person of any other, always at the same time as an
end and never merely as a means to an end. Goffman bases his view of
the human personality as sacred on his phenomenology of the meaning of
everyday, ritualized, face-saving practices. At their most general, everyday
ritual practices exhibit the demand to show regard for others in face-to-face
encounters. But beyond offering this most general characterization, Goffman
analyzes the variety of norms and ideals at play in these encounters, adherence to which constitutes what is required in various contexts to show proper
regard for others. In the two opening essays of Interaction Ritual alone, his
discussion of ritual runs the gamut from self-respect and pride,14 honor,15
dignity,16 considerateness,17 shamelessness,18 malice and spitefulness,19 disgrace,20 forgiveness and gratitude,21 to discretion, sincerity, modesty, and
self-control.22 Interaction rituals are indeed embodiments of a complex of
moral norms.
In his book Behavior in Public Places, Goffman offers an account of
the norms we operate under in our behavior in public places.23 I will begin
by reproducing Goffmans sketch of the interaction order and then will turn
to some examples [of my own], their significance, and why they deserve the
attention of ethicists and non-ethicists alike.
Goffman analyzes what he refers to as one kind of social order, where
he understands a social order to be the consequence of any set of moral
norms that regulates the way a person pursues objectives.24 Each such order
comes with its set of regulations. The order that Goffman is concerned with,
what he calls elsewhere the interaction order, is the order of a persons
handling of himself and others, during, and by virtue of, his immediate
physical presence among them; what is called face-to-face or immediate
interaction.25 Goffmans interaction order applies to any gathering of two
or more persons.
In the context of co-presence, based on bodily presence, each person
immediately gives off information to those with whom he is co-present and
receives information from them.26 As a result, each person can see the other
and can perceive how he is being seen or reacted to by the other person;
these perceptions amount to undeniable evidence about each persons comportment in the situation. Gatherings of this sort are sometimes also called
social occasions or a wider social affair, undertaking, or event bounded
in regard to place and time and typically facilitated by fixed equipment.27

Acknowledging the Given

257

Examples are a spur-of-the-moment party, an open house, a store opening,

a night at the opera, a funeral, and a wedding. These social occasions vary.

Some have fixed starting and ending points (a funeral); some do

not (an open house).

Some restrict attendance (wedding); some do not (store opening).

Some strictly limit behavior (funeral, a night at the opera); some

do not (picnic).

Some are planned ahead of time (wedding); some are spontaneous

(spur-of-the-moment party).

Some serve some serious end (wedding), and

Some are recreational (picnic), organized for their own pleasure.28

This taxonomy points to the extent and variety of encounters that are
occasions for ritual interactions. It is important, for our consideration, that
social gatherings are pervasive features of our lives. Unless we live alone and
stay inside day in and day out, we cannot avoid them. Moreover, gatherings
are various in character, and so behaviors or rituals that are fitting for one,
for example, a picnic, would be wholly unfitting for a funeral.
Goffman, however, further distinguishes between the situated and
the situational. The merely situated is what might happen to take place
in certain settings but could take place elsewhere. For example, I can read
a book in a library as well as outside it. If I read it in the library, that act
is merely situated in the library. The situational, in contrast, is what can
only take place in a specific gathering. For example, borrowing a library
book is situational relative to the gathering of particular people in the
library and requires co-presence with required others in this setting.29 The
situational aspects of gatherings and their constitutive requirement of copresence are regulated by norms that are basically moral in character. And
although moral norms relating to physical safety and harm are the most
obvious ones governing such gatherings, there is also, as Goffman indicates,
another sort of norm, given less attention:
[W]hen persons are present to one another, they can function
not merely as physical instruments but also as communicative
ones. This possibility, no less than the physical one, is fateful for
everyone concerned and in every society appears to come under

258

Whose Tradition? Which Dao?

strict normative regulation, giving rise to a kind of communicative traffic order. The rules pertaining to this area of conduct I
shall call situational proprieties.30
This fatefulness for all persons and all societies is both pervasive and
inescapable, and moreover, fundamentally moral in character. And through
his phenomenology, Goffman helps us see why this inescapable aspect of
human life is worthy of our attention and analysis, despite the fact that
this realm constitutes only a part of the moral order.31
In the course of his detailed study, Goffman offers the following sketch
of key norms governing social gatherings. The general norm is not to draw
improperly on what one owes to the situation.32 Sub-rules include the
following:

Rules of access to a bounded region limit those who have access

to the gathering. Adherence to these shows respect for the gathering itself. (Those not invited should not come to a party.)

Rules against external preoccupations limit the activities and the

attention of those co-present to those relevant to the situation. This
means those co-present cannot become involved with interests
outside of the situation (text messaging at a business meeting or at
the dinner table) or to ones divisive within the gathering (hugging
and kissing during a lecture).

Rules requiring that one make oneself ready for involvement in the
situation. (Being attentive, interested enough to be able to respond
to what comes up, and dressed appropriately so one can take on
whatever role comes ones way.)

Keeping oneself from going too far in a situated task and so remaining ready to do what is required in the situation. (Not becoming so
absorbed in pouring drinks and helping the cook that one forgets
to greet new guests at a dinner party.)

Requirement at parties to give oneself up to mutual engagements

and the prohibition against excluding newcomers. This amounts to
a rule that anyone present has a right to obtain attention and an
obligation to give it to any other participant.33

These requirements provide a summary of rules that govern different kinds

of social gatherings. This list mostly includes those sorts of settings in which

Acknowledging the Given

259

the rules are fairly tight. Of course, at a party, a person need not talk to
everyone. And there is some flexibility within the requirements on how a
partygoer could move through the evening, avoiding some conversations
and keeping others short.
Some gatherings are much looser in structure. For example, the gathering at a picnic allows for considerable flexibility in terms of what counts
as appropriate involvement.34 But in different communities, even in one
and the same nation, the level of looseness versus tightness of requirements
can vary greatly. Consider the different conventions concerning dress and
conduct by faculty and students in a lecture hall at a small college in the
South and one in a large California university situated near the ocean.
This list of rules demarcates a rough account of the regulations we
are under in the co-presence of others. Of course, different sub-rules will
apply in different situations. Handshaking takes place in various types of
gatherings and, depending on the type of gathering, will proceed differently;
in some gatherings, handshaking would seem abnormal. Fundamental to
these rules governing face-to-face encounters at gatherings is respect for
the gathering and its participants demonstrated by a persons readiness to
participate appropriately.
Beyond spelling out such general rules, Goffman analyzes the ways
in which a persons behavior in gatherings is ritualized. Goffmans work is
most clearly relevant to Confuciuss teaching in forcing us to acknowledge
its range in everyday life and the variety of ways ritual matters to us.35
In his essays in Interaction Ritual, where he develops his Durkheimian account of the role of ritual in face-to-face encounters,36 Goffman distinguishes between substantive rules governing interpersonal relations and
ceremonial rules. Substantive rules are rules governing conduct that prescribe
or prohibit conduct for its own sake. For example, there is a substantive
rule against stealing that rules out taking others property without right, no
matter how a person ritually conducts the theft. This sort of rule lines up
with Confuciuss notion of justice or righteousness ( yi), which can be
used to specify duties. In contrast, ceremonial rules govern how a person
expresses his character or expresses his appreciation of others in the situation.37 Goffman makes the point that the ceremonial rule governs the
ceremonial component or function of an action, not the action itself.
Thus any action might have a substantive component as well as a ceremonial component.38 I might violate a substantive rule of property by robbing
a bank, but I might conduct myself in a way that showed concern and
respect toward the teller, following our cultures ceremonial rule of being
polite to strangers.

260

Whose Tradition? Which Dao?

The ritual component of actions has two basic elements: deference

and demeanor.39 Deference, for Goffman, means a symbolic means by
which appreciation is regularly conveyed to a recipient, of this recipient,
or of something of which this recipient is taken as a symbol, extension, or
agent. Although it is often thought that deference is expressed in rituals of
submission to a person in authority, Goffman makes the point that deference
is more complicated than this: It can occur in symmetrical relationships,
between equals, and by superiors to subordinates. Examples of deference
include a patient addressing his physician as doctor, a priest blessing a
parishioner, a potential customer inviting a salesman into his house, and
a person accepting the advice of a computer technician.40 These forms of
deference express various forms of regard in various social relationships.
Goffman further divides deference into two broad types: avoidance
ritual and presentational ritual. Rituals of avoidance would involve keeping proper physical distance, preserving anothers privacy, not addressing
a person by first name if the required conditions for doing so have not
been met, and avoiding mention of embarrassing facts about the person.41
These all constitute ways one person can show deferential regard for another
person in ritual form.
A second type of ritual is the presentation ritual. In contrast with
avoidance rituals, which prohibit certain types of behavior, presentational
rituals specify what a person must do.42 Examples of presentational rituals
include nodding, saying hello, smiling when making eye contact, mentioning that someones new clothes or haircut looks nice, showing an interest in
how things are going with a person, shaking hands, and so on.
Goffman points out that these two forms of deference can come into
conflict. Asking a person about his family might violate the prohibition
on avoiding invading a persons private sphere in some contexts, but not
asking can violate the requirement to express appropriate interest in others.
Of course, this conflict is usually only abstract, and in particular cases, a
person can usually judge correctly which form of deference is required. In
cases when a person judges wrongly, that person will commit a ritual gaffe.
The second ceremonial component of behavior is demeanor. Demeanor
is conveyed through deportment, dress, and bearing. According to Goffman,
a well demeaned or properly demeaned person displays such attributes
as discretion and sincerity; modesty in claims regarding self; good sportsmanship; command of speech and physical movements; self-control over
emotions, appetites, and desires; poise under pressure, and so forth.43 These
characteristics make the person a reliable actor in face-to-face encounters.
For example, the indiscreet person cannot be relied on to keep confidences,
the insincere person cannot be relied on to mean what he says or say what

Acknowledging the Given

261

he means, the immodest person may attempt to dominate the conversation

by indicating that he has more to contribute than others, and a person who
lacks control over his speech and movements will not be able to participate
gracefully.
The demands of deference and demeanor sometimes overlap in such
a way that the violation of the requirements of deference might also be
the violation of the requirements of demeanor. A person might violate
the requirement of deference to leave others alone by expressing a lack of
control over his emotions by being overly loud.44 But these forms of ritual
engagement do not always overlap. A person sitting in a restaurant whose
demeanor is otherwise exemplary could inappropriately intrude into the
conversation at the next table, not out of a lack of self-control but out of
some legitimate interest. Or a student who normally shows proper regard
for his teacher could show up at the teachers office unwashed and smelling of beer.
Goffman sees deference and demeanor as forms of ritual interaction
that are complementary and required if a person has developed a fully
manifest social image of him- or herself. For while a person can control
how he behaves and can control his deferential conduct toward others to
some extent, he cannot control others deferential conduct toward him or
others behavior. But for me to be a respected teacher, for example, requires
that I be treated deferentially by others, something not under my control
or perhaps only indirectly and partially under my control. This insight gives
rise to Goffmans fundamental idea, discussed in chapter eight, that who
we are is a function of our position within a ritual chain including others:
[F]or a complete man to be expressed, individuals must hold
hands in a chain of ceremony, each giving deferentially with
proper demeanor to the one on the right what will be received
deferentially from the one on the left. While it may be true that
the individual has a unique self all his own, evidence of this
possession is thoroughly a product of joint ceremonial labor,
the part expressed through the individuals demeanor being no
more significant than the part conveyed by others through their
deferential behavior toward him.45

Examples and Counterexamples

At this point, I have examined Goffmans classification of types of social
gatherings, along with his classification of types of ritual interactions

262

Whose Tradition? Which Dao?

required in face-to-face encounters in social gatherings. We find ourselves

in social gatherings during many of our waking hours unless we work and
live entirely on our own. We are subject to norms of ritual interaction
not of our own making, and we sometimes suffer from failing to become
successful ritual practitioners, just as we benefit by being successful in our
ritual behavior and recipients of other successful ritual behavior toward us.
Not only does this taxonomy help us appreciate the pervasive character of ritual in everyday life because ritual pervades all types of gatherings,
this taxonomy also helps us to see the limitations of Fingarettes analysis
of ritual in a new way. In light of Goffmans taxonomy, we can say that
Fingarettes account of ritual focuses on one type of ritual requirement, the
requirement of positive deference expressed in one type of context. In this
way, his account is too narrow.
Furthermore, by emphasizing his interpretation of ceremonial acts as
happening spontaneously or magically, Fingarettes account underplays the
way that participation in everyday ceremonies has the character of a performance. Goffman argues that selves, real or contrived, are socially performed.
The distinction between my real self and some contrived phony self is
not, then, the distinction between a non-performed self and a performed
self. When we distance ourselves from some social action or role we have
performed and offer an alternative as a better, corrective presentation of
who we really are, this act of distancing is itself a new social performance,
a part of a ritualized remedial interchange. The distinction between real
and contrived self can, then, be articulated in terms of a kind of syntax, a
grammatical distinction, governing the interaction order. These social performances and the syntax they depend on indicate the grammar of the
self and can help us understand how we distinguish between the real and
contrived self.46
Performances of the self, say, in handshaking, do not escape the limits of social production. All such performances reflect what Goffman calls
dramaturgical discipline.47 The performer must remain in character during her performance despite possible distractions. Her performance needs
to be convincing. But this means that her audience is required to overlook
performance flaws and provide sufficient polite hints so as to allow the
performer to adjust her performance.48
But this performance aspect of, say, handshakes, does not get adequately integrated into Fingarettes account of handshaking. Fingarette might
want to claim that what is most important about handshakes, their magical spontaneity, gets ruled out in Goffmans account, making it impossible
to distinguish between authentic and contrived handshakes. But Goffman

Acknowledging the Given

263

argues that this way of speaking about authenticity is really just a part of
the performance of a self. I use my capacity to shake hands smoothly and
the talk of smooth handshaking as a way of making my performance all
the more convincing. But Goffman adds that while this way of speaking
provides a strength of show, it provides a poor analysis of it. Crucial to
a correct analysis is the way in which performances are structured and are
convincing, the syntax of talk about the difference between an authentic
and contrived self.49
Another important feature of Goffmans account that bears on Fingarettes analysis of ritual is the way in which he characterizes the rules of
ritual as ways of expressing regard for others a person encounters. If we
distinguish between regard and respect, we might then want to say the following: ritual requirements of demeanor and deference are ways of expressing regard for others a person encounters in gatherings. But if we think of
respect as involving the acknowledgment of achievement or position in
some hierarchy, then social ritual requires that we demonstrate regard for
those we encounter, even if we do not respect them for their achievements
or position in a hierarchy. If this account is correct, and it needs further
study, then Fingarettes account confuses respect with regard and gives
too much weight to the respect-expressing role of rituals.50 This role, strictly
speaking, is limited to rituals of a certain sort. Of course, this critique
depends on clarifying the distinction between regard and respect, and I
have not tried to do that fully here. But this basic distinction between a
general requirement of showing consideration (regard) for others a person
encounters versus respecting them for their achievements and positions in
a hierarchy is one that we can also find in the Analects distinction between
(shu) (reciprocal regard) and (jing) (respect), discussed in Chapter 8.
A final issue that Goffmans taxonomy raises in general, as well as for
Fingarettes account, is the relationship between substantive rules governing
interpersonal relations and ceremonial rules. Substantive rules, such as, Do
not steal, govern interpersonal relations independent of face-to-face encounters. But Goffman never addresses the question of the relation between
substantive and ceremonial rules within the commerce of human life. We
can admit the distinction, and we can also ask this question: In which
way would our understanding of either type of rule be affected were we
the sorts of beings who operated exclusively with substantive or ceremonial
rules? I would venture that each of these two aspects of morality provides
a context for understanding the other. Ritually expressed regard for others
has to impact individuals sense of why stealing is wrong and affects their
motivation for not stealing, just as the acknowledgment of substantive rules

264

Whose Tradition? Which Dao?

of interpersonal relationships (like Dont steal! and Dont lie!) impact

our face-to-face encounters with others. We may ask ourselves, when we
anticipate possibly violating a substantive rule (like Do not cheat!), what
it would be like to confront a person face-to-face, having to acknowledge
and apologize for our conduct. This shows that we think of these distinct
types of regulation as interconnected.

Conclusion
Understanding the full range of proprieties for the wide range of gatherings
in which people find themselves requires the mastery of a wide range of rules
and the development of a fairly complicated repertoire of behaviors. Until
very recently, even in North America, it was thought that young people
needed to read etiquette manuals to help them master this knowledge.51 That
approach has tended to fall out of favor. One might speculate that this is
because of the increasing tendency for gatherings and relationships that used
to be more tightly governed to be increasingly more loosely governed. Note
the recent trend among students on college campuses to address teachers
by their first names or the tendency to dress informally for class, whereas
in the 1950s both of these behaviors were unheard of.
When we combine this North American trend toward social informality, which I am by no means arguing against, with the tendency to
understand norms governing social behavior as either individually chosen or
validated or as being constituted exclusively by what Goffman calls substantive rules, then the realm of ritual norms that govern social gatherings tends
to suffer from individual and cultural inattention and distortion, though
these ritual norms have not disappeared.
At the most abstract level, this cultural trend, supported by philosophical theories as well as the trend toward informality, can lead to the
ease with which it becomes possible to deny the existence of this realm of
interpersonal normativity or to see it as outside the bounds of morality. If
Goffmans taxonomy is basically correct and if our moral sensibility cannot
be separated from our ritualized expression of regard for others, we live at
a cultural moment in which we cannot help but acknowledge the fact and
meaning of ritual interactions, but we lack the philosophical and cultural
resources to understand them. I have argued that Fingarettes handshaking
example helps us to acknowledge the moral character of ritual. But we also
need to think seriously about other types of examples of ritual.

Acknowledging the Given

265

Consider, for example, a situation in which a person walks into an

elevator full of strangers. In contrast with Fingarettes handshaking example,
which is what Goffman calls a positive ritual, Goffman analyzes standing
in an elevator with strangers as a negative ritual, one operating under the
requirement to leave others alone. Improper performance of positive rituals
such as smiling at or addressing any of the strangers in this situation could
result in a slight but improper performance of negative rituals, possibly
interpreted as a violation of another person, in the sense of violating that
persons privacy.52 As we started with Fingarettes model of a positive ritual,
it might seem strange that rituals of leaving others alone exist. However, as
Goffman points out, negative rituals themselves have a distinct structure.
Before developing the elevator example in more detail, I will present
the reason for thinking of the requirement to leave others alone as a type
of ritual. All rituals are designed to show respect or regard for objects of
value and have a dialogical aspect. Positive rituals require some acknowledgment or expression of appreciation on the part of the person who is their
recipient. Negative rituals, ones that involve leaving a person alone, do not
typically require a dialogical response by the recipient, but they may: if the
negative ritual performance is onerous or requires particular delicacy, an
acknowledgment may be in order.53 For example, if I have no choice in the
limited space I find myself and others in but to brush against someone, I
need to offer apologies and an explanation to make the intrusion acceptable.
But if a negative ritual is improperly performed, there may be a need for
a remedial interchange, including (roughly) the offenders assurances and
the offended persons acceptance of those assurances.54 Even an action of
not leaving another person alone can be a misstep in such a complex ritual
activity. The various aspects of negative rituals typically become clearer when
they are breached than when they are enacted; that is, we notice negative
rituals more when remedial steps are taken to correct their breaches.
In returning to my elevator example, I offer Goffmans description
of the ritual of avoiding invading anothers personal space in an elevator
occupied by strangers:
Passengers have two problems: to allocate the space equably, and
to maintain a defensible position, which means in this context
orientation to the door and center with back up against the wall
if possible. The first few individuals can enter without anyone
present having to rearrange himself, but very shortly each new
entrantup to a certain numbercauses all those present to

266

Whose Tradition? Which Dao?

shift position and reorient themselves in sequence. Leave-taking

introduces a tendency to reverse the cycle, but this is tempered
by the countervailing resistance to appearing uncomfortable in
an established distance from another. Thus, as the car empties,
passengers acquire a measure of uneasiness, caught between two
opposing inclinationsto obtain maximum distance from others and to inhibit avoidance behavior that might give offense.55
In a footnote, Goffman points out an unpublished study of elevator
behavior by John Gueldner that spells out the typical positional order of
elevator riders: the first entrant positions himself near the controls or in one
back corner, the second in the corner diagonally across from the first entrant,
the third and fourth in the remaining corners, the fifth in the middle of the
rear wall, and the sixth in the middle of the car. Members of preset groups
(he calls them withs) stay together but respect the general requirement of
facing the front, despite their presence as a group.
These ritual norms we generally abide by become clearer to us in contexts where we have violated them, and then we need to find some remedy.
Recently, at a school in China, I confronted an elevator car of fellow students
going to a restaurant together. As the door opened I could see that they
were all packed in. The elevator had just enough room for one more person
to squeeze in. I decided to wait, but in a show of solidarity, they insisted
I ride with them. (Had I declined, this might have seemed disrespectful,
an unwillingness to accept their act of generosity toward a fellow student.
So I accepted.) I stepped in, but only after entering did I realize that there
was no room to turn around. So I ended up facing the back of the elevator and all of the riders who comported themselves properly by facing the
front. Upon sizing up the awkward situation I found myself in, I realized
that I was not able to avert my eyes from their eyes in the way I normally
would have on a typical elevator ride. So I decided to make a semi-joking
comment to let them know that I realized I had just unwittingly violated
an elevator rule of avoidance.
The rules of such negative rituals can also become clear when we have
to instruct children in how to comport themselves in these contexts, sometimes by holding them back, sometimes by repositioning them, sometimes
by causing them to become quiet, and so on.
These negative rituals do not, as Fingarettes do, function by requiring
co-participants to express positive respect for each other by being present
to one another in some special sense or by needing to be spontaneous to
express authenticity. In fact, awkwardness in such contexts (for example, not

Acknowledging the Given

267

quite knowing how to make the appropriate spatial adjustment when the
elevator starts to get full) might be understandable and deserving of special
appreciation by co-riders because of the way it shows consideration for them.
This example brings to the fore an important feature of some, but
not all, everyday negative rituals. They are structured to show consideration
for others. But consideration can be shown in ways that go beyond being
present, showing respect and intimacy, and so forth. People need to be
perceptive and to have developed sensitivity toward the legitimate needs of
others in the variety of contexts in which they are present to each other.
They can also show consideration by avoiding contact or interaction with
those in their presence.
Central to negative rituals, then, is what a person avoids doing as a
way of showing consideration for others. But this aspect of rituals, avoidances that show consideration for others in ones presence, also results in
requirements that ritual actors satisfy situational requirements of demeanor.
The ritual requirements of demeanor include requirements for avoiding conduct, including dress, which would undermine the persons capacity to fulfill
other ritual requirements in the situation. They are ritualized ways of being
prepared for participation in other rituals. So, even if greeting a guest is a
way of my welcoming her to my party, trying to make her feel comfortable,
and creating ritualized beginning of party-type interactions, my demeanor
alone does not guarantee that a guest will feel comfortable or welcome. The
demeanor of showing consideration for all guests, while necessary, does not
complete my ritual requirements as a host.
Other things being equal, to show proper demeanor in a situation
is to satisfy a necessary condition of successful ritual engagement in that
situation. If I am to show proper demeanor as a host of a cocktail party for
professional colleagues at my home, I must be available to greet guests, not
be wearing a swimsuit, and not be needing a bath. I have to show demeanor
appropriate to the context as a mode of expressing consideration for my
guests. Of course, we can imagine all sorts of justifying exceptions to this
requirement. If my car had broken down and I arrived home from the beach
and show up for the party just as my guests are arriving, my lateness, along
with my frazzled state of mind, swimsuit, and dirtiness could be excused.
Showing proper demeanor in a context is just a part of the ritual
activity of the context. That is to say, showing proper demeanor in a context
qualifies people to carry out their ritual responsibilities in that context. But
being properly behaved is a way of showing those with whom individuals
come into ritual contact that they are trustworthy, considerate co-participants. As guest, I would not feel comfortable being greeted by and greeting

268

Whose Tradition? Which Dao?

in return a dirty, improperly dressed host. I am made to feel comfortable

by a properly dressed host who shows him- or herself to have removed any
impediments in the context to success in greeting me and making me feel
welcomed. Showing proper demeanor is itself a ritualized way of showing
consideration for others, but not a way of being, pace Fingarette, magically
present to others or showing respect for them. Showing proper demeanor
as a host is a ritualized way of removing an obstacle to the successful rituals
required of hosts toward their guests.
Against this analysis, one might argue that showing proper demeanor
is not itself a ritual, but rather, just a requirement of rituals. And so, it
should be no surprise that the condition of showing proper demeanor, not
itself a ritual, does not satisfy the general conditions of ritual. But I would
argue that what showing proper demeanor does accomplish in a ritualized
context is itself a ritual, even if it also functions as a necessary requirement
of successfully carrying out related rituals. Priests wearing of the clerical
collar is itself a ritual of dress meant to show the wearers potential role
in the whole economy of priestly ritual activities. Priestly demeanor, even
outside of specific rituals, is a form of ritualized behavior showing the priests
orientation and preparedness for a variety of other priestly ritual activities.
But these rituals of demeanor do not themselves express what other, more
substantive priestly ritual activities, such as serving communion or presiding
at a wedding or funeral, express.
Rituals of demeanor make it possible for other rituals to express what
they are meant to express successfully. So even if, to return to Fingarettes
example, my handshake with a student expresses my respect for him, the
ritual of my dress and comportment as a professor do not express that.
Thus, in addition to the negative ritual example, here we have another
counterexample to Fingarettes account of ritual.
Finally, I want to indicate one additional type of ritual that offers a
challenge to Fingarettes account. Goffman discusses rituals of contempt.56
He offers an example of a patient in a mental ward of a hospital who shows
contempt toward a nurse by sticking out her tongue as the nurse walks away.
We might think of this as a ritualized way of expressing something negative.
But we might also want to distinguish between conventional expressions of
disrespect, that is, gestures of various sorts, and the use of those gestures in
ritual. Not every instance of showing ones middle finger is a disrespectful
ritual. So we might think that the patients gesture is not really a ritual. I
suspect intuitions might differ here, but the more we attend to the context
of Goffmans example, in that sticking out the tongue is a standard way in
this institution to express disrespect, we can think of it as a ritual shared by

Acknowledging the Given

269

the rest of the patients witnessing it, designed both to show disrespect for
authority and solidarity amongst the patientsthe more this seems a ritual
act. Furthermore, if there were specific sorts of occasions in which this ritual
were performed by patients occupying specific roles, such as patient leaders,
that fact would strengthen the case for thinking of this gesture as a ritual
act. So it seems reasonable to think that there are rituals of contempt.57
We can also think of other such rituals. I might, for example, refuse
to shake hands with someone. This can be a ritual, too. And in specific
contexts, we can easily see such a refusal as more than just a gesture. When
the leader of Israel refuses to shake hands with the leader of Egypt, we are
witnessing a ritual of expressing and sustaining a lack of respect and a desire
to maintain diplomatic distance.
One might think that this example is not only a problem for Fingarettes account of ritual, but also for Confuciuss account. However, I have
already argued that for Confucius ritual does not necessarily express respect.
Furthermore, the Analects offers an example of a show of contempt:
. . . . .
Rubei wanted an audience with Confucius. Confucius refused him
by [using] the excuse of being sick. As the messenger was going
out the door, he sang, having fetched his se [a musical instrument] to play, deliberately causing the messenger to hear this.58
Here we have an example of intentional slight. We can imagine it as a
ritual act or as the exemplar that begins a tradition. This example shows
that Confucius was not beneath expressing contempt. And we can imagine
it as becoming a ritual means of doing so even if not so here. But we also
find this example of a negative ritual, one that operates ritually by refusing
to do the recognized ritual of respect as a way of showing disrespect:
. . .
Toward friends presenting gifts, even though it is as expensive
as a carriage or a yoke of horses, but is not sacrificial meat, he
wont pay his respects.59
This example shows a refusal to engage in the accepted ritual response to a
gift. But, as in the refusal to shake hands, the refusal to pay respect for a
gift can be a ritualized way to show disrespect. Even for Confucius, there

270

Whose Tradition? Which Dao?

are acts of disrespect. And if we take Confucius as offering the rest of us

an example of how to proceed, we can imagine a ritual to emerge out of
this example. And if we imagine that he as teacher was aware of this possibility, even for him, rituals are not essentially tied to showing respect.60
In the case of rituals of insult, there might in some cases be ways to
explain and even justify the rituals in terms of respect for a norm that the
insulter believes the insulted person to have violated. For example, Confuciuss refusal to pay respect rests on his belief that gifts supportive of ritual
sacrifice are the only sort deserving of respect. The patients sticking out her
tongue may be an expression of support for norms of patient autonomy and
respect that the patient believes the nurse to have violated.
These and other norms operate every day in guiding our conduct
in relation to others. As ritualized forms of conduct, they express regard
for others, respect for others, contempt for others, and preparedness to
engage with others ritually. An examination of them offers a rich resource
of embodied insights about ethical norms governing our relations to others.
For philosophers to ignore ritual behavior of this sort and other forms of
ritual is for them to ignore a central, invariant feature of everyday morality,
without which morality would not be what it is for us.
I have argued that the most important part of Fingarettes discussion
of Confuciuss teaching is his example of handshaking. With this example
we can begin to appreciate the existence of a realm of social norms that
we are compelled to acknowledge but often misunderstand. I also argued
that the account of the importance of handshaking that Fingarette offers
is mistaken and gets Confuciuss views of respect (jing ) wrong. Because
ritual is a motley, what we need in place of Fingarettes account is a way
to understand the variety of forms of ritual norms we function under every
day. I have turned to parts of Goffmans micro-sociology of face-to-face
interaction as a starting point for offering a perspicuous presentation of the
variety of ritual requirements we are under. This characterization makes his
work in sociology an instance of clarification of our complicated form of
(ritual) life.61 This representation, which challenges Fingarettes analysis of
ritual, provides us with additional reasons beyond the compelling one his
account offers of our need to acknowledge ritual if we are to become clear
in both practice and theory about our moral form of life.
If the overall argument of this book is correct, this acknowledgment
of ritual can be embraced by either those philosophers sympathetic to Confucius or Wittgenstein as a key step along the road to moral improvement
or philosophical clarity.

Afterword

The Way Backward or Forward

Wittgenstein or Confucius?

My argumentative strategy in this book has been to emphasize ways in

which Confuciuss insistence that understanding what it means to live in
conformity to dao arise in large part out of reflection on learned practices
of ritual. This principle establishes a limit to reflection, both in terms of
the range of topics we can cogently reflect on and in terms of the necessary steps we must go through to become capable of responsible reflection.
I have explicated and defended these Confucian commitments in terms of
similar commitments in Wittgensteins later philosophy. Nevertheless, Confucius is no Wittgensteinian, and Wittgenstein is no Confucian. Perhaps
this should be no surprise. Comparative philosophy functions in the face
of both affinities and tensions.
One way to articulate the tensions is in terms of Wittgensteins disclaimer that a philosopher is a member of no community. Wittgensteins
aphorisms are notoriously opaque, this one no less than others. Yet I take
him to be indicating here that the task of the philosopher is to offer surveyable representations of any language and concepts he or she faces. Wittgensteins focus on the project of conceptual clarification and away from proving
or disproving philosophical claims arguably takes him out of the game of
endorsing the worldview and related philosophical statements upheld by any
particular community. To be a Confucian, however, is to be a member of
the particular community of followers of Confucius.
This tension does not, however, preclude engagement between Wittgensteins later philosophy and the Analects as I have interpreted it. The
argument of this book takes up a series of topics for which, despite essential differences, engagement is not only possible, but if my arguments are
successful, even fruitful. Nonetheless, it is possible to raise the question of
what it might mean for a person to be both Confucian and Wittgensteinian.
271

272

Whose Tradition? Which Dao?

I propose that it would mean to look to the texts of Confucianism for

insights about our complicated forms of life that are overlooked or distorted
by Western philosophers due to their adoption of misleading pictures of
what it is to be human being or what it is to be moral. The texts of early
Confucianism offer not only a way to rectify our one-sided diet of stock
examples, they do so as a part of an ongoing reflective tradition, committed
to making sense not only of their own stock of examples but also of what
it means to make sense of such language and concepts. What more could
we Wittgensteinians ask for?
The sort of engagement I am proposing would also provide Confucians with a mode of philosophical investigation of our complicated forms
of human life that clarifies and, in so doing, protects those forms of life
and the concepts embedded in them, which constitute the Confucian dao.
What more could we Confucians ask for?
One thing more we might ask for is some sort of epistemological
or metaphysical grounding of those Confucian forms of life. But those of
us who see those aspirations as futile and a departure from early Confucian practice will want to deploy a harmonization of the critical strategies
available to us in the Analects and in the later writings of Wittgenstein to
further the goals of both by better understanding both.
One key topic that deserves more consideration than I have given it
here is the character and role of ritual in a life lived well. This topic can
serve to illustrate the different approaches to ritual we might find in early
Confucianism and later Wittgenstein. From an early Confucian account of
ritual, we should expect a variety of hints and suggestions about the relationship between practice of a specific set of inherited rituals and their relation
to moral ideals, for example, the relation between welcoming guests in a
specific way and the ideals of respect and consideration. Attention would
be given to the way in which modifications of a persons daily practices of
ritual might be crucial for bringing about resolution of his or her quandaries
regarding the ideals in question. These interventions and reflections would
take place, we might say, from inside the Confucian tradition.
But insider discussions are not likely to satisfy everyone. Witttgensteinian investigations into ritual, of which we have a clear example in
Wittgensteins Remarks on Frazers Golden Bough, outside of any specific
commitments to Confucianism, can prompt reflection on specific rituals and
the way they function as given practices within the human form of life.
This sort of analysis supplies us with the surveyable range of things we call
ritual and with the range of their roles within the commerce of human life.

Afterword

273

We might expect to find points of tension between these two sorts

of accounts, as well as points of overlap. Any Confucian claims about what
ritual must be like would be subjected to a Wittgensteinian investigation.
Confucians might then have to face up to the ways their own tradition
takes up one of several different, possible ways to be oriented toward ritual.
And at stages of Wittgensteinian investigation, investigators outside any
moral community and thus outside any specific ethical point of view who
are practiced in the ways of Confucius can be clearer about how their own
contingent but defensible commitments provide them with an intelligible
form of life with a distinctive set of goods and norms internal to it.
The vision I sketch here is decidedly pluralistic.1 The focus I have
given in this book on the primacy of practice over reflection and reflection
limited and conditioned by learned practices leaves unsettled how far it is
possible to critique and evaluate competing moral frameworks. This pluralism leaves us facing the perennial threat of moral skepticism or relativism,
but the framework I have sketched leaves us with stronger resources for
critique. Internally inconsistent moral frameworks require revision to be
sustainable. And any framework will be testable in terms of ordinary beliefs
and practices it aims to clarify but possibly only distorts. But even if pluralists have at hand these philosophical stock-in-trade resources for evaluation
of moral frameworks, one problem will haunt these considerations: the loss,
or even lack, of crucial moral concepts. For if moral concepts are embedded
in forms of lives and traditions, and those can undergo radical disruption or
be morally stunted, then competing moral forms of life with their different
conceptual resources will confront each other in mutual incomprehension.
But as I argue in Chapter 4, even in this case, nothing rules out roving
ambassadors moving from one community to the next to allow those who
have grown up in one moral community to come to get a firsthand familiarity with an initially incomprehensible form of life. Serving as translators,
they could then work to make competing forms of life more intelligible to
their first and second moral families. They might even convert from their
first to their second moral homes and be able to make a compelling case
to those in both moral orientations for their conversions.
To some, the sort of analytical framework I have sketched here will be
unattractive. Its pluralism, indifference to foundational questions, and reliance on ambassadorships and conversions will seem to leave too many loose
ends, too many uncertainties. Yet from my point of view, a Confucianism
that faces up to these complexities will be healthier than one that does not.
To others, Confucian commitments to the authority of culture and teachers

274

Whose Tradition? Which Dao?

will be anathema. From my point of view, however, a Wittgensteinianism

that explores conceptual terrain outside of Western cultures, embedded as
they tend to be in liberal individualism, can only help its quest for surveyable representations of language, concepts, and forms of life. But, of course,
as in all things, the proof will be in ones lived, reflective form of life.

Notes

Preface
1. Note to readers: Throughout this book, I have made an effort to avoid
sexist use of pronouns. When possible, I have used the general strategy of using
third person singular neutral pronouns and neutral plural pronouns. In quoted texts,
I have used the pronouns present in those texts to guard authenticity. Despite these
efforts, some sentences remain with he, which I use as a neutral pronoun because
it is less awkward than any other alternatives I considered.

Chapter 1
1. From Philip J. Ivanhoes appendix to Ethics in the Confucian Tradition.
See note 4 below.
2. I agree with Richard Grandy that our interpretive goal should be intelligibility, including intelligible error, not correctness of a view at any cost. See his article,
Reference, Meaning, and Belief The Journal of Philosophy 70, no. 14 (1973):
439452. For an extended discussion of the principle of charity, see Chapter 4.
3. For two examples of such an approach, see Paul Johnston, The Contradictions of Modern Moral Philosophy: Ethics After Wittgenstein, Routledge Studies in Ethics and Moral Theory (London: Routledge, 1999); and Paul Johnston, Wittgenstein
and Moral Philosophy (London: Routledge, 1989).
4. For a set of representative views that offer this approach, with an overview by the editors, see Alice Crary and Rupert J. Read, eds., The New Wittgenstein
(London: Routledge, 2000). For my own account of Wittgensteins philosophical
therapy, see James F. Peterman, Philosophy as Therapy: An Interpretation and Defense
of Wittgensteins Later Philosophical Project, SUNY Series in Philosophy and Psychotherapy (Albany: State University of New York Press, 1992).
5. Ludwig Wittgenstein, Philosophical Investigations: The German Text, with
a Revised English Translation, 4th ed., trans. and ed. G. E. M. Anscombe, P. M.
S. Hacker, and Joachim Schulte (Malden, MA: Wiley-Blackwell, [1953] 2009),
section 654.
275

276

Notes to Chapter 1

6. The text on this notion, like much of the rest of the text, offers fragments of insights, not a complete account of the need to clarify and enforce correct
usage. However, some commentators have thought that the account should be seen
as applying to the whole of language. See John Makeham, The Earliest Extant
Commentary on Lunyu: Lunyu Zheng shi zhu, Toung Pao 83 (1997).
7. For an account of the relationship between Confuciuss and Youzis views,
see William A. Haines, The Purloined Philosopher: Youzi on Learning by Virtue,
Philosophy East and West 58, no. 4 (2008): 47091.
8. Ludwig Wittgenstein, Wittgensteins Lectures on the Foundations of Mathematics, Cambridge, 1939: From the Notes of R. G. Bosanquet, Norman Malcolm,
Rush Rhees, and Yorick Smythies, ed. Cora Diamond (Chicago: University of Chicago
Press, 1976), 14.
9. Wittgenstein, Philosophical Investigations, 23940.
10. In comments on an early version of my argument, Yang Xiao pointed
out the difficulty of understanding Confucius as being interested in the learning of
children and the difficulty of likening Confuciuss view of learning with Wittgensteins. However, since Wittgensteins focus on learning contexts includes learning
by adults as well as by children of various ages, there should be no difficulty in
likening these different approaches, as I do in this book.
11. Wittgenstein, Philosophical Investigations, section 217.
12. Alasdair MacIntyre, Incommensurability, Truth, and the Conversation
Between Confucians and Aristotelians About the Virtues, in Culture and Modernity:
East-West Philosophic Perspectives, ed. Eliot Deutsch (Honolulu: University of Hawaii
Press, 1991), 104122.
13. Edward Slingerland, Virtue Ethics, the Analects, and the Problem of
Commensurability, Journal of Religious Ethics 29, no. 1 (2001): 97125.
14. David L. Hall and Roger T. Ames, Thinking Through Confucius, SUNY
Series in Systematic Philosophy (Albany: State University of New York Press, 1987);
David L. Hall and Roger T. Ames, Thinking from the Han: Self, Truth, and Transcendence in Chinese and Western Culture (Albany: State University of New York
Press, 1998).
15. Herbert Fingarette, Confucius: The Secular as Sacred (Religious Traditions
of the World) (Long Grove, IL.: Waveland Press, [1972] 1998).
16. A. C. Graham, Disputers of the Tao: Philosophical Argument in Ancient
China (La Salle, IL: Open Court, 1989).
17. Chad Hansen, A Daoist Theory of Chinese Thought: A Philosophical Interpretation (Oxford: Oxford University Press, 1992).
18. Benjamin I. Schwartz, The World of Thought in Ancient China (Cambridge,
MA: Belknap Press, 1985).
19. Robert Eno, The Confucian Creation of Heaven: Philosophy and the Defense
of Ritual Mastery, SUNY Series in Chinese Philosophy and Culture (Albany: State
University of New York Press, 1989).
20. David S. Nivison, The Ways of Confucianism (Peru, IL: Open Court,
1996).

Notes to Chapter 1

277

21. Lunyu Yinde (A Concordance to the Analects of Confucius).

Harvard-Yenching Institute Sinological Index Series, Supplement 16. (Taipei, Taiwan: Distributed by Chinese Materials and Research Aids Service Center, 1966).
Hereafter Confucius, Analects. See especially 1.14, 5.2, 5.21, 14.1, 14.3, 14.19, 15.
7, 16.2, 18.6, 18.7, 19.19.
22. Confucius, Analects, 6.24 and 9.30.
23. I operate throughout with the view that dao is constituted by a range
of ideals, embedded in and giving rise to a network of norms about how to live
ones life. For this view, see Antonio S. Cua, Moral Vision and Tradition: Essays in
Chinese Ethics (Washington, DC: Catholic University of America Press, 1998); and
Cuas essay on Confucianism in Encyclopedia of Ethics, 2nd ed., eds. Lawrence C.
Becker and Charlotte B. Becker (New York: Routledge, [1992] 2001), 28795.
24. For a development of this criticism, see Chad Hansen, A Daoist Theory
of Chinese Thought.
25. By higher-level practices, I mean practices involving weighing and balancing evidence and considerations derived from what I call bedrock practices.
26. My basic point is that the self-cultivation project sets limits to what
sorts of questions and problems one poses. This principle offers a first step in
responding to the charge that Confuciuss reflections are philosophically inferior
to those of Western philosophers or, for that matter, later thinkers in the Chinese
tradition who make a more theoretical turn. Although this is a first step, it is not
sufficient to fend off this criticism. For that, I draw on argument strategies from
Wittgensteins later philosophy.
27. Wittgenstein, Philosophical Investigations, section 201.
28. Ibid., 202. In order to avoid possible confusion over the comparison I
would make here, Confucian rituals, though expressible in terms of rules, are themselves practices. So I could formulate a rule that says, When greeting strangers,
look them in the eye and give them a firm handshake; however, the ritual is not
this rule, but rather the practice in terms of which this rule makes sense.
29. Wittgenstein, On Certainty, ed. Gertrude Elizabeth Margaret Anscombe
and Georg Henrik von Wright (New York: Harper and Row, [1969] 1972), sections 9495.
30. Wittgenstein, Philosophical Investigations, section 355.
31. Ibid., section 217.
32. Throughout Philosophical Investigations and the rest of his later writings,
Wittgenstein makes the point that explanations come to an end somewhere, but
often, the endpoints are just these contexts of learning. See Philosophical Investigations, section 1, for the first instance of his use of this principle in that text.
33. Philosophical Investigations, section 190, quoted in Meredith Williams,
Wittgenstein, Mind, and Meaning: Toward a Social Conception of Mind (London:
Routledge, 1999), 179.
34. Wittgenstein, Remarks on the Foundations of Mathematics, VI.8.
35. Williams, Wittgenstein, Mind, and Meaning, 179.
36. Ibid., 180.

278

Notes to Chapter 1

37. See Wittgenstein, On Certainty, section 263, for his description of the
normal schoolboy. This account of the deviant student comes at section 310 ff.
38. Wittgenstein, On Certainty, sections 31417.
39. Williams, Wittgenstein, Mind, and Meaning, 17980.
40. Ibid., 17980.
41. See note 23 and related text from this chapter.
42. I adopt this translation for the following reason: Although translators
have offered different translations into English of (ren), they all seem somewhat
problematic to me. (ren) seems to encompass all of those virtues that govern
interpersonal relationships. I select moral goodness because it seems to me to
capture many of our most basic intuitions about what it would be for a person to
be (ren). My translations also tend to understand the general virtue of (ren),
as well as the specific virtues that make it up, like (xiao) in terms of practices.
So I translate (xiao) as practicing or being filial. This reflects my understanding
of the practical focus of early Chinese texts, especially the Analects.
43. Youzi, quoted in Confucius, Analects, 1.2.
44. Confucius, Analects, 12.1.
45. Confucius, Analects, 2.15.
46. Wittgenstein, Philosophical Investigations, sections 123 and 309.
47. I am grateful to Yang Xiaos and Meredith Williamss comments on an
earlier version of this argument, along with various audience members in the International Society of Comparative and Western Philosophy group session at the 2009
Eastern American Philosophical Association meeting, who raised this point and
indicated possible resolutions.
48. This paragraph was added as a result of P. J. Ivanhoes perceptive comments on this issue.
49. Source text: Jiang Yihua , Huang Junlang .
. (Taipei: , 2007). From Inner Pattern, Liji, section 7 and translation
from James Legge, Li Ki, vol. 28, part 4, Sacred Books of the East (Oxford, UK:
Clarendon Press, 1885).
50. Ibid., sections 7680.
51. Wittgenstein, Wittgensteins Lectures on the Foundations of Mathematics, 14.
52. Wittgenstein, Philosophical Investigations, section 217.
53. Wittgenstein, On Certainty, sections 9495.
54. Wittgenstein, Philosophical Investigations, 236.
55. See Philosophical Investigations, 23637, where Wittgenstein makes this
distinction in reference to methods of measuring length.
56. Ibid., section 227.
57. Wittgenstein, Philosophical Investigations, 237.
58. Ibid.
59. Ibid., 239.
60. Confucius, Analects, 9.11.
61. Confucius, Analects, 2.4.

Notes to Chapter 1

279

62. E. Bruce Brooks and A. Taeko Brooks, The Original Analects: Sayings of
Confucius and His Successors, 04790249 (New York: Columbia University Press,
1997), 53.
63. Edward Slingerland, Confucius Analects: With Selections from Traditional
Commentaries, translated by Edward Slingerland (Indianapolis: Hackett, 2003), 90.
For a discussion of two different ways of interpreting this passage, exhibited in the
commentarial tradition, see Daniel K. Gardner, Zhu Xis Reading of the Analects:
Canon, Commentary, and the Classical Traditin (New York: Columbia University
Press, 2003), 14954. For my account of how to address such conflicts in interpretation, see Chapter 3.
64. See Confucius, Analects 6.22 and 13.19 for the ren questions; 2.5, 2.6,
2.7, and 2.8 for the xiao questions.
65. Confucius, Analects, 7.8 for the fen and fei explanations.
66. Confucius, Analects, 5.9.
67. For Wittgensteins rejection of this sort of projection, see Wittgenstein,
Philosophical Investigations, section 1.
68. Wittgenstein, Philosophical Investigations, 239.
69. Ibid., section 240.
70. Wittgenstein, On Certainty, section 203.
71. Wittgenstein, Philosophical Investigations, section 241.
72. Ibid., section 242.
73. Wittgenstein, Remarks on the Foundations of Mathematics, VI, section
21.
74. Ibid., VI, section 2.
75. Ibid., VI, section 39.
76. Williams, Wittgenstein, Mind, and Meaning, 201.
77. Confucius, Analects, Youzi quoted in 1.12.
78. See Peterman, Philosophy as Therapy for a sustained discussion of this sort
of project in the writings of the later Wittgenstein.
79. Confucius, Analects, 13.23, on a junzi and dao.
80. Wittgenstein, Philosophical Investigations, 236.
81. Wittgenstein, Remarks on the Foundations of Mathematics, VI, sections
1718.
82. Ibid., VI, section 19.
83. I am indebted to P. J. Ivanhoe for this point. See his The Evolution of
the Chuanxilu in his Ethics in the Confucian Tradition: The Thought of Mengzi and
Wang Yangming. 2nd ed. (Indianapolis, IN: Hackett, 2002), 14353. Both Wang
and his disciple Xu Ai, the author of Chuanxilu, held the view that the examples
of Confuciuss teaching in the Analects were to be understood as aimed at each
individual and not as a general medicine for everyone. Nonetheless, despite Wangs
instruction not to do so, Xu Ai composed the Chuanxilu in order to offer examples
of Wangs instruction to individuals, as a way of giving the hints and suggestions
that would guide disciples on their own self-cultivation. This appendix offers a

280

Notes to Chapter 2

translation of the preface of Chuanxilu, which describes Wangs and Xus similar
approach to the Analects and to imponderable evidence that I offer here.
84. Wittgenstein, Philosophical Investigations, section 208.
85. Ibid., section 208.
86. Wittgenstein, Philosophical Investigations, section 126. Other aspects of
what lie open to us in such understanding are the settled, confident disposition to
carry on correctly and the related sense that one must carry on in this way.
87. Confucius, Analects, 7.24.
88. Confucius, Analects, 17.17.
89. Wittgenstein, Philosophical Investigations, section 64.
90. Ibid., section 66.
91. Ibid., ix.
92. James C. Klagge, An Unexplored Concept in Wittgenstein, History of
Philosophy Quarterly 12, no. 4 (1995): 46986.
93. Ibid., 472, quoting Ludwig Wittgenstein, Lectures and Conversations on
Aesthetics, Psychology and Religious Belief (Berkeley, CA: University of California
Press, 1972), 7.
94. Yang Xiao, How Confucius Does Things with Words: Two Hermeneutic
Paradigms in the Analects and Its Exegeses, The Journal of Asian Studies 66, no. 2
(2007): 497532.
95. Wittgenstein, Philosophical Investigations, sections 112 and 133.

Chapter 2
1. Rush Rhees, Some Developments in Wittgensteins View of Ethics, The
Philosophical Review 74, no. 1 (1965): 23.
2. For an account of the range of views of ethics that might count as antitheoretical, see Stanley G. Clarke and Evan Simpson, Anti-Theory in Ethics and
Moral Conservatism, SUNY Series in Ethical Theory (Albany: State University of
New York Press, 1989). For an account of the anti-theoretical approach to ethics
by Wittgenstein, see James C. Edwards, Ethics Without Philosophy: Wittgenstein and
the Moral Life (Tampa: University Presses of Florida, 1982).
3. Confucius, Analects, 2.16.
4. Ibid., 18.7
5. Ibid., 18.5.
6. For a development of this argument, see Chapter 3.
7. I avoided overstating my point here by a comment of P. J. Ivanhoe.
8. Confucius, Analects, 17.19.
9. I have excluded the second part of this section, which in some other
editions is relegated to a separate passage.
10. Part of Confuciuss resignationist approach to disagreement may be connected to his suspicion of clever speech (Analects 1.3), but also to his skepticism
about reflection not wedded to learning (Analects 2.15).

Notes to Chapter 2

281

11. Wittgenstein, Philosophical Investigations, section 123: A philosophical

problem has the form: I dont know my way about.
12. Cora Diamond, The Realistic Spirit: Wittgenstein, Philosophy, and the Mind
(Cambridge, MA: MIT Press, 1991), 24.
13. Wittgenstein, Philosophical Investigations, section 66.
14. Diamond, The Realistic Spirit, 8.
15. Ibid.
16. Ibid., 27.
17. Ibid., 42.
18. Ibid., 23.
19. Ibid., 24.
20. Onora ONeill, The Power of Example, Philosophy 61 (1986): 529.
21. Diamond, The Realistic Spirit, 28.
22. Ibid., 21.
23. Ibid., 28.
24. Ibid., 35.
25. Stanley Cavell, Must We Mean What We Say?: A Book of Essays (New
York: Scribner, 1969), 52.
26. Diamond, The Realistic Spirit, 28.
27. Ibid., 305.
28. Geoff Sayre-McCord, Metaethics, in The Stanford Encyclopedia of Philosophy, ed. Edward N. Zalta (Stanford, Calif.: Metaphysics Research Lab, CSLI,
Stanford University, Fall 2008), section 1, accessed December 27, 2009, <http://
plato.stanford.edu/archives/fall2008/entries/metaethics/>.
29. Ibid., 23.
30. For an account of Wittgensteins therapeutic project, see Peterman, Philosophy as Therapy.
31. See Analects, 4.11.
32. See Analects, 12.2.
33. As I argue above, both Confucius and Wittgenstein see the norms we
operate under in the forms of ideals or concepts as labyrinthine. Their clarifications
of norms are piecemeal and motivated to help others resolve puzzlement or to help
them foster cultivation of dao.
34. Confucius, Analects, 5.13. There is a long history of commentary on this
passage. Until Zhu Xi, most commentators, including He Yan, whose common sense
interpretation in of Analects 1.2 I discuss and defend in Chapter
6, believed that this passage indicated a distinction of one sort or another between
exoteric and Daoist type esoteric teachings about things that cannot be put into
words. Zhu Xi holds that Confucius seldom spoke of these topics. In this case, I
follow Zhu Xis commonsense interpretation of this passage in . For an
account of the merits of Zhu Xis interpretation and a detailed discussion of the
history of commentary on this passage, with references to these commentaries, see
Philip J. Ivanhoe, Whose Confucius? Which Analects? in Bryan Van Norden, Confucius and the Analects: New Essays (Oxford: Oxford University Press, 2002), 11933.

282

Notes to Chapter 2

35. Confucius, Analects, 9.1.

36. Ibid., 6.27.
37. Ibid., 19.6.
38. Ibid., 18.8.
39. Ibid., 4.10.
40. In Ethics, Imagination, and the Tractatus, in The New Wittgenstein,
ed. Crary and Read, Diamond does hint at what a spirit might be by drawing a
contrast between Wittgensteins conception of ethics and the standard views that
take it to be a distinct subject matter. She says, I might do that this way: just as
logic is not, for Wittgenstein, a particular subject, with its own body of truths, but
penetrates all thought, so ethics has no particular subject matter; rather, an ethical
spirit, an attitude to the world and life, can penetrate any thought or talk (153).
This passage suggests that a spirit is an attitude a person brings to the world as
a whole. In this sense, we might want to say that the realistic spirit brings to
whatever it confronts a realistic attitude. The clarification I give here, however, connects the notion of having an attitude to the world with the notion of bringing to
ones investigations a set of normative commitments. One might also connect up
the notion of spirit with that of temperament. A person with an artistic temperament would be someone whose sensibilities are formed around the sorts of things
artists care about. A person with a realistic temperament would be a person whose
sensibilities are formed around the sorts of things a realistic person, in Diamonds
sense, cares about. Indeed, Diamond uses the terms spirit and temperament
interchangeably in this essay. (See The Realistic Spirit, 171). So these terms form a
kind of trinity meant to characterize some basic orientation a philosopher ought
to bring to the world. I am grateful to James Klagge for the suggestion about temperament and to Andrew Moser for the reference to Diamonds talk of an ethical
spirit. One caveat: I am not at all sure that the notion of spirit as an attitude to
the world as a whole gets retained in later Wittgenstein. As Diamond points out,
this formulation, itself nonsense, arises out of the early Tractarian thinking about
ethics. In this essay, she argues that this attitude is retained throughout Wittgensteins thinking, but her argument about that just points to one possible guess that
competes with various others. She argues that Wittgensteins attitude toward ethics,
being unspeakable, as manifested in his 1930s Remarks on Frazers Golden Bough,
is continuous with his earlier Tractarian view. But she never argues that this view
also shows up in Wittgensteins later conversations on ethics. Indeed, she argues
that the removal of ethical talk from empirical talk is a technique of our language,
manifest, for example, in fairy tales. That being so, it would be hard to see how
talk about ethics would still, at this later point, be thought of as nonsense. Her
complaints that some interpreters of later Wittgenstein have evaded the mystical
and ethical in his thought in this case may also to apply to her own account. This
is especially so if, for the later Wittgenstein, ethics is a motley, not just the one
thing that Diamond insists on.
41. This attitude is expressed in both of the following passages from Philosophical Investigations: The real discovery is the one which enables me to stop doing

Notes to Chapter 2

283

philosophy when I want to. The one that gives philosophy peace, so that it is no
longer tormented by questions which bring itself into question (133). And What
has to be accepted, the given, isone might sayforms of life (p. 238, section
345.). These quotes are expressions of what the realistic spirit cares about and the
form of philosophical clarifications it seeks to develop.
42. Harry G. Frankfurt, The Importance of What We Care About: Philosophical
Essays (Cambridge, UK: Cambridge University Press, 1988).
43. Ibid., 92.
44. Ibid., 83.
45. See Chapter 2, note 45 and related text.
46. Wittgenstein, Philosophical Investigations, sections 98 and 105.
47. I borrow this language from Philosophical Investigations, section 122.
48. Diamond, The Realistic Spirit, 2829.
49. Perhaps the best initial response to this dilemma is to reject it as begging
the question. After all, it assumes what is in question, that resolution of moral
disagreement requires appeal to clear principles set up in advance of confronting
the disagreement. Nevertheless, without clear, prior criteria of judgment to appeal
to, it might be hard to see how the realistic spirit can do anything but arbitrarily
manage moral disagreement.
50. It might, however, seem strange to appeal to the Confucian tradition
to help the realistic spirit address the problem of moral disagreement because the
Confucian tradition can never be anything more than one actual or possible party
to moral disputes. So my resolution to the realistic spirits problem of moral disagreement seems to amount to nothing more than arbitrarily selecting one moral
tradition over others as a way to resolve disagreement.
Despite this appearance, I would defend my approach as follows. The Confucian realistic spirit is not committed to its own moral vision, but rather to
embodying dao, whatever that requires. Its method of embodying dao depends on
recognition that whatever dao is, understanding it requires study of its embodiments
and reflection on those, based on the best resources at hand. Nothing precludes the
Confucian realistic spirit from adding to its collection of exemplars by introducing
exemplars from other times and cultures. To embody the Confucian realistic spirit
is to collect exemplars of dao, whatever those might be, and to use them in practice
and reflection on dao, whatever doing that might lead to, not to limit that collection
arbitrarily. The Confucian spirits collections provide resources for the realistic spirit
to manage disagreement without having to turn to moral theory.
51. Rush Rhees, Some Developments in Wittgensteins View of Ethics, The
Philosophical Review 74, no. 1 (1965): 23.
52. Ibid.
53. Ibid.
54. Ibid., 24.
55. Ibid., 2425.
56. Stephen Mulhall, Ethics in Light of Wittgenstein, Philosophical Papers
31, no. 3 (2002): 31314.

284

Notes to Chapter 3

57. Ibid., 319.

58. For a critique of Alice Crarys version of Mulhalls account, see Nigel Pleasants, Wittgenstein, Ethics and Basic Moral Certainty, Inquiry 51, no. 3 (2008):
24167.
59. I present an alternative to the pervasive ethical character of Wittgensteins
later philosophical investigations in Peterman, Philosophy as Therapy. I argue that
Wittgensteins interest in human forms of life reflects his early ethical ideal of coming
into agreement with the world. Diamond claims that Wittgensteins later view of
the ethical character of philosophy has to do with the way in which philosophical
investigation requires courage to face the truth. See Lawrence C. Becker and Charlotte B. Becker, Encyclopedia of Ethics, 2nd ed. (New York: Routledge, 2001), 1811.
60. For an articulation of the form this liberalism takes, see Stephen Mulhall,
Stanley Cavell: Philosophys Recounting of the Ordinary (Oxford, UK: Clarendon Press
of Oxford University Press, 1994).
61. Wittgenstein, Philosophical Investigations, section 185 ff.
62. Mulhall, Ethics in Light of Wittgenstein, 31517.
63. Wittgenstein, Philosophical Investigations, section 190.
64. Ibid., section 258.
65. There are other problems with the idea that each person is the final moral
arbiter. Presumably, if that is so, the student could, as final arbiter, reject the view
that he is the final arbiter. I cannot develop that point here, discussion of which
goes back to Plato.

Chapter 3
1. Daniel K. Gardner, Confucian Commentary and Chinese Intellectual
History, The Journal of Asian Studies 57, no. 2 (1998): 397422.
2. John Makeham, Transmitters and Creators: Chinese Commentators and
Commentaries on the Analects, Harvard East Asian Monographs, 228 (Cambridge,
MA: Harvard University Asia Center, distributed by Harvard University Press, 2003).
3. Ibid.
4. E. Bruce Brooks and A. Taeko Brooks, The Original Analects: Sayings of
Confucius and His Successors, 04790249 (New York: Columbia University Press,
1997).
5. Edward Slingerland, Why Philosophy Is Not Extra in Understanding
the Analects, Philosophy East and West 50, no. 1 (2000): 13741.
6. Although I develop my own arguments here and focus specifically on
the two texts under discussion, my general approach to these questions of meaning
and interpretation is in sympathy with the approach and arguments of Bryan Van
Norden in his Introduction to Virtue Ethics and Consequentialism in Early Chinese
Philosophy. (New York: Cambridge University Press, 2007).
7. Translation A from Chichung Huang, The Analects of Confucius: A Literal
Translation (New York: Oxford University Press, 1997) and translation B from D.
C. Lau, Confucius: The Analects, London Penguin Books, 1979.

Notes to Chapter 3

285

8. John Makeham, Transmitters and Creators, 237.

9. This understanding of the relation of (ren) to application treats
(ren) as an abstract ideal or principle that then gets applied in particular cases, even
though its character as principle or ideal is not impacted by the application. It is
self-contained and whole.
10. Ibid., 239.
11. Ibid., 299.
12. James Legge, Four Books: The Great Learning, the Doctrine of the Mean,
Confucian Analects, and the Works of Mencius with English Translation and Notes
(New York: Paragon, 1966).
13. Raymond Stanley Dawson, The Analects: Translated with an Introduction
and Notes, (Oxford, UK: Oxford University Press, 1998).
14. E. Bruce Brooks and A. Taeko Brooks, The Original Analects.
15. Edward G. Slingerland, Confucius Analects: With Selections from Traditional
Commentaries (Indianapolis, IN: Hackett, 2003).
16. Daniel K. Gardner, Zhu Xis Reading of the Analects: Canon, Commentary,
and the Classical Tradition (New York: Columbia University Press, 2003); and Daniel
K. Gardner, Confucian Commentary and Chinese Intellectual History. The Journal
of Asian Studies 57, no. 2 (1998): 397422.
17. Daniel K. Gardner, Confucian Commentary.
18. Frank Kermode, The Classic: Literary Images of Permanence and Change
(London: Faber & Faber, 1975), 121.
19. George G. Iggers, Historicism, in Dictionary of the History of Ideas:
Abstraction in the Formation of Concepts to Design Argument (New York: Scribners,
1968, 1973).
20. Gardner, Confucian Commentary, 398.
21. See Ruth Benedict, A Defense of Moral Relativism, The Journal of General Psychology 10 (1934): 5982, for a defense of this relativistic view. For a successful critique of this argument along the lines I have introduced, see James Rachels,
The Elements of Moral Philosophy, 2nd ed. (New York: McGraw-Hill 1993), 1822.
22. The meaning of mens auctorus is mind of the author.
23. David E. Linge in Hans Georg Gadamer, Philosophical Hermeneutics.
xxiixxiv. Trans. David E. Linge (Berkeley: University of California Press, 1977),
quoted by John Makeham, Transmitters and Creators, 11.
24. Ibid., 22.
25. John Makeham, Transmitters and Creators, 10.
26. For a defense of this view, in relation to the Daodejing, see Philip J.
Ivanhoe, The Daodejing of Laozi (New York: Seven Bridges Press, 2001), xv.
27. See E. Bruce Brooks and A. Taeko Brooks, The Original Analects, 202.
28. D. C. Lau, The Analects, 263.
29. Ibid.
30. See Anne Cheng, Lunyu . In Early Chinese Texts: A Bibliographical Guide, ed. Michael Loewe (The Society for the Study of Early China, and The
Institute of East Asian Studies, Berkeley: University of California, 1993).
31. Ibid.

286

Notes to Chapter 3

32. H. G. Creel, whose views I discuss below, thinks that the Analects gives
us the real Confucius. He, nevertheless, admits that the Analects represents the
writings of Confuciuss disciples. See Herrlee G. Creel, Confucius: The Man and the
Myth (Westport, CT: Greenwood Press, 1972), 291. See Harold Shadick, Review
of Confucius by H. G. Creel, The Philosophical Review (1951): 11317, for a
discussion of problems with Creels account.
33. See, for example, Annping Chin, The Authentic Confucius: A Life of
Thought and Politics (New York: Scribners, 2007), who assumes without argument
that the Analects can be taken as authentic.
34. Herrlee G. Creel, Confucius and the Chinese Way (New York: Harper,
1960).
35. Ibid., xi.
36. Ibid., xi and 62.
37. Creel, Confucius and the Chinese Way, 58.
38. Ibid., 61.
39. Ibid., 6162.
40. Ibid., 60.
41. Ibid., 123.
42. Ibid., 132.
43. Ibid., 155.
44. Ibid., 115, 122, 125, 169, and 258.
45. Creel, Confucius and the Chinese Way, 133, 136.
46. Ibid., 136.
47. Michael Nylan, The Five Confucian Classics (New Haven, CT: Yale
University Press, 2001), 364.
48. I owe this point to Philip J. Ivanhoe.
49. For an account of the recent history of rejection of the need to develop
an historical account of Socratess life in order to develop an account of Socratess
views, see Debra Nails, Socrates, in The Stanford Encyclopedia of Philosophy, ed.
Edward N. Zalta (Winter 2007).
50. In thinking about this issue, I have been influenced by Alexander Nehemass account of Nietzsche in his book, Alexander Nehamas, Nietzsche, Life as
Literature (Cambridge, MA: Harvard University Press, 1985) in which he argues
that the main character of many of Nietzsches texts, himself, is not the same as
Nietzsche the writer. Nietzsche invented a literary philosophical version of himself as
the angry destroyer of what he took to be traditional idols and critic of democratic
stupidity, the whole while being a shy, introverted, scholarly type. We can read the
Nietzsche corpus, focusing on understanding the character Nietzsche internal to the
Nietzsche corpus, just as we can read the Analects with the goal of understanding
the character Confucius.
51. Some passages treat (li) and (ren) as connected: 12.1; others as
disconnected: 12.2 and 12.3. I claim that this multiple author strategy is one way
to resolve the contradiction, not the only way.

Notes to Chapter 4

287

52. The Brookses argue for this interpretation, but think they can specify
the authors of these competing strains of thinking. I have denied that, but that
does not mean we cannot identify competing views of Confuciuss teaching in the
Analects. We dont need to know the authorship to have reason to suspect that there
exist such competing accounts.
53. John Makeham, Between Chen and Cai: Zhuangzi and the Analects, in
Wandering at Ease in the Zhuangzi, trans. Roger T. Ames (Albany: State University
of New York Press, 1998), 9294. See also Jeffrey L Richeys use of this argument
in Jeffrey L. Richey, Ascetics and Aesthetics in the Analects, Numen (2000), 164.
54. For the source of this notion of family resemblance, see Ludwig Wittgenstein, Philosophical Investigations, section 64. See also Chapter 1 at note 86.
55. John Makeham, Transmitters and Creators, 20, note 47.
56. Ibid., 910.
57. Ibid., 13.
58. Ibid., 16.

Chapter 4
1. Following Frank Cioffis criticisms of Wittgensteins approach to anthropology, Brian Clack argues that Wittgensteins approach to understanding rituals of
other cultures suffers from the way in which it too easily resolves puzzlement over
foreign rituals by associating them with familiar rituals of our own. I reject this
criticism since it seems to identify some of Wittgensteins mistaken associations in
his Remarks on Frazers Golden Bough with adequate instances of his method.
See Brian R. Clack, Wittgenstein, Frazer, and Religion (Palgrave Macmillan, 1999),
Chapter 4, Perspicuous Representation.
2.Wittgenstein, Philosophical Investigations, section 115.
3.Ludwig Wittgenstein, Preliminary Studies for the Philosophical
Investigations,Generally Known as the Blue and Brown Books (Oxford: Basil Blackwell, 1958), 1.
4. Ludwig Wittgenstein, Philosophical Investigations: The German Text, With
a Revised English Translation, section 2.
5. Stanley Cavell, Must We Mean What We Say?: A Book of Essays (New
York: Scribner, 1969), 52.
6. In his Conditions Handsome and Unhandsome: The Constitution of Emersonian Perfectionism, (La Salle, IL: Open Court University of Chicago Press, 1990),
Stanley Cavell argues against Kripkes community agreement view of meaning. There
is more wrong with Cavells account than I can address here. He emphasizes that for
Wittgenstein, agreement in forms of life is just agreement in deep-seated tendencies
of response, not agreements about any specific judgment or response. The communitys agreement about a judgment, then, is not the sort of agreement that Wittgenstein has in mind. I discuss the question of the role of agreement in Wittgensteins

288

Notes to Chapter 4

later philosophy in Chapter 1. It is clear that Wittgenstein distinguishes agreement

in beliefs and agreement in forms of life. But this distinction does not map onto
the distinction between what a community requires of its normal members and what
tendencies of responses exist for any human being no matter what community. I
have argued that it maps onto the distinction between bedrock practices and beliefs
and practices and beliefs justified in terms of them.
7. This account sounds vaguely MacIntyrean. I admit some similarities,
but in contrast to MacIntyre, I do not think that such a community of readers
needs to be identical to the tradition that treats its target texts as canonical. The
community of the Analects readers overlaps the so-called Confucian tradition but
is not identical to it. The community of readers would not then all hold that its
target texts embodied the truth. For this aspect of commentarial traditions that I
am excluding form the community of the Analects readers, see John B. Henderson,
Scripture, Canon, and Commentary: A Comparison of Confucian and Western Exegesis
(Princeton, NJ: Princeton University Press, 1991). For MacIntyres view of tradition,
see Alasdair MacIntyre, After Virtue: A Study in Moral Theory (Notre Dame, IN:
University of Notre Dame Press, 1984), Chapters 1415.
8. I assume, but dont argue here, that we have a reliable sketch of the
history of those schools that formed during the Warring States period to provide
education of a class of shi, scholars, who sought training from teachers with some
skills and wisdom useful to ruling a state. See Michael LaFargue, Tao and Method:
A Reasoned Approach to the Tao Te Ching (New York: State University of New York
Press, 1994); Robert Eno, The Confucian Creation of Heaven: Philosophy and the
Defense of Ritual Mastery, SUNY Series in Chinese Philosophy and Culture (Albany:
State University of New York Press, 1989); and Cho-Yun HSU, Ancient China in
Transition: An Analysis of Social Mobility, 722222 B.C. (Stanford, CA: Stanford
University Press, 1965). I will not be able to develop this account in detail here.
For purposes of my argument, all that is needed is the presumption that such an
account is available. Such an account will be general. But used in the right way,
and given my approach to the Analects as a work of philosophical fiction, that is
all that is needed to indicate the forms of common behavior of mankind exhibited
in the text.
9. I will speak of the principle of charity as if it were one principle. Strictly
speaking, I am just discussing one version of the principle of charity or perhaps
one of the charity principles. For a discussion of the versions of the principle, see
Richard Feldman, Principle of Charity, in Routledge Encyclopedia of Philosophy
(London: Routledge, 1998).
10. On this criticism of Quines version of the principle, see Richard Grandy,
Reference, Meaning, and Belief, The Journal of Philosophy 70, No. 14 (1973):
439452. Grandy calls his version of the principle, the principle of humanity.
11. Donald Davidson, On the Very Idea of a Conceptual Scheme, in Inquiries Into Truth and Interpretation, ed. Ernest Lepore (1984), 194.
12. Davidson, On the Very Idea, 184.
13. Ibid., 192, 196.

Notes to Chapter 5

289

14. Ibid., 196.

15. Donald Davidson, A Coherence Theory of Truth and Knowledge, in
Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, ed. Ernest
Lepore (Oxford: Basil Blackwell, 1986), 316.
16. Ibid., 316.
17. Ibid., 316.
18. Ibid., 317.
19. Ibid., 316, my emphasis.
20. See my discussion of this passage in Chapter 3.
21. Ibid., 444 ff.
22. Wittgenstein, Philosophical Investigations, section 207.
23. Ibid., 206.
24. That is, Wittgenstein himself.
25. Ibid., section 23.
26. To keep things simple, I am assuming that the Analects articulates norms
embedded in Warring States period China, but it may be that the views here reflect
or also reflect later Han Dynasty norms, as articulated by its compilers and authors.
27. Alice Crary, ed. Wittgenstein and the Moral Life: Essays in Honor of Cora
Diamond (Boston: MIT Press, 2007), 9.
28. Paul Ricur, The Conflict of Interpretations, Northwestern University
Studies in Phenomenology & Existential Philosophy (Evanston, IL: Northwestern
University Press, 1974), 294ff.
29. This is not to say that it could not be given a different meaning in
another context.
30. For a critique of Putnams views of incommensurability that resolves
potential incommensurability problems by appeal to our ability to learn, as children
do, new languages, which we can (as masters of both) move between, see Paul
Feyerabend, Putnam on Incommensurability, The British Journal for the Philosophy
of Science 38, no. 1 (1987): 7592.
31. In making this point, I have been influenced by Ian Hacking, The
Parody of Conversation, in Truth and Interpretation: Perspectives on the Philosophy
of Donald Davidson, ed. E. Lepore (1986).

Chapter 5
1. In thinking about this principle of salience, I have benefited from the
important writings of Quentin Skinner. See James Tully, Meaning and Context:
Quentin Skinner and His Critics (Cambridge, UK: Polity Press, 1988).
2. The literature on Wittgensteins account of nonsense presents a range
of views on why Wittgenstein held that all of his propositions in his early work
lacked meaning. One view would hold that they lacked meaning because they do
not picture a set of possible facts. Another would hold that they lacked meaning
because certain of their signs had not been given a meaning. The former view, then,
has to cope with the question of whether Wittgenstein offers a picture theory of

290

Notes to Chapter 5

meaning, which he later abandons as nonsense, based on that theory itself. There
is a long history of thinking that the Tractatus is, for this reason, incoherent. The
latter view claims that he has no theory of meaning. He, instead, thinks, in some
ordinary, commonsense way, that in some cases we have given meaning to signs,
and in some cases we have not. However, he does not need to offer a theory of
meaning to support this view. He is just using our ordinary, non-theoretical concept
of meaning. For the former, see G. E. M. Anscombe, An Introduction to Wittgensteins
Tractatus, 4th ed. (London: Hutchinson, 1971); for the latter, see Cora Diamonds
essay Criss-cross Philosophy in Erich Ammereller and Eugen Fisher, Wittgenstein
at Work: Method in the Philosophical Investigations (London: Routledge, 2004).
3. Wittgenstein, On Certainty, section 24.
4. See Stewart Cohen, Contextualism and Skepticism, Nos, 34, Philosophical Issues 10 (2000), for a discussion of contextual aspects of salience that
includes Dretskes example. Cohen argues that standards for knowing some claim to
be true depend on context and on whether a question has been raised and whether
an answer to the question really matters.
5. See Chapter 6, note 10 and related text for a discussion of commentary
on this text.
6. The notion of certain questions dropping away is central to Wittgensteins approach to philosophy in both his early and later periods. We might even
consider distinguishing between those questions that are basic by distinguishing (a)
which questions one can fruitfully address earlier (b) by virtue of which otherwise
abstruse, unanswerable questions, get dissolved. This approach to Confucius would
bring the two projects together more clearly than my present discussion would
indicate. Another possibility, however, is that once a person learns the rituals that
pertain to spirits, he need only then reflect on these rituals meanings. I do not
try to resolve this difference here as both seem consistent with the Analects and
both seem possible strategies for addressing a Zilu-type question. Philip J. Ivanhoe
offers the first sort of interpretation of this passage in Death and Dying in the
Analects, in Amy Olberding and Philip J. Ivanhoe, eds., Mortality in Traditional
Chinese Thought (Albany, NY: SUNY Press, 2011).
7. Donald J. Munro, The Concept of Man in Early China (Stanford, CA:
Stanford University Press, 1979), 55.
8. Ibid., 56.
9. Munro, Concept of Man, 54.
10. Ibid.
11. Chad Hansen, A Daoist Theory of Chinese Thought (Oxford, UK: Oxford
University Press, 1992), 3.
12. There is no reason to think that there is a Western view of language. Is
early Wittgensteins language Western? Is Nietzsches? Is later Wittgensteins?
13. Hansen, A Daoist Theory, 34.
14. Ibid., 65.
15. Following Legge.
16. Wittgenstein, Philosophical Investigations, section 43.

Notes to Chapter 5

291

17. Ibid., section 133.

18. Ibid., 86.
19. Ibid.
20. Ibid., sections 124.
21. Ibid., section 2.
22. Ibid., section 3.
23. Ibid., section 126.
24. Ibid., sections 30506.
25. For two excellent recent accounts of Wittgensteins views of truth within
languages, see Michael N. Forster, Wittgenstein on the Arbitrariness of Grammar
(Princeton, NJ: Princeton University Press, 2004); and Sara Ellenbogen, Wittgensteins Account of Truth, SUNY series in philosophy (Albany: State University of
New York Press, 2003).
26. Ibid., 65.
27. Chad Hansen, Chinese Language, Chinese Philosophy, and Truth,
The Journal of Asian Studies 44, no. 3 (1985): 491519.
28. Ibid., 492.
29. Ibid.
30. Ibid., 494.
31. Ibid., 495.
32. Hansen appeals to Wittgenstein for the insight that views of language
influence philosophical thinking generally, but he quotes a passage from Philosophical
Investigations that suggests that he either misunderstands Wittgenstein or discounts a
central tenet of Wittgensteins philosophical therapy. For he paraphrases Wittgenstein
as claiming to support his account with his dictum that philosophy is the bewitchment of the intellect by means of grammar (Hansen, Chinese Language, 493).
Of course, Wittgensteins point is that we should eschew constructing philosophical
theories about how things must be based on mistaken pictures of language.
33. See my account of Confuciuss project in Chapter 2.
34. P. J. Ivanhoe, Confucian Moral Self-Cultivation, The Rockwell Lecture
Series (New York: Peter Lang, 1993).
35. Confucius, Analects, 14.4.
36. Ibid., 5.5
37. A. C. Graham, Disputers of the Dao: Philosophical Argument in Ancient
China (La Salle, IL: Open Court, 1989), 395.
38. Philip J. Ivanhoe, Review: Thinking Through Confucius by David L.
Hall; Roger T. Ames, Philosophy East and West 41, no. 2 (1991): 252.
39. David L. Hall and Roger T. Ames, Thinking from the Han: Self, Truth,
and Transcendence in Chinese and Western Culture (Albany: State University of New
York Press, 1998), 105.
40. Ibid., 116.
41. Ibid., 114.
42. Ibid., 105.
43. Ibid., 145.

292

Notes to Chapter 5

44. Ibid.
45. Ibid., 145.
46. I am not arguing that it would be impossible to supply such a theory. Based on Analects 13.23, one might be able to generate such an account:
... Toward others, an exemplary person (junzi)
is harmonious but not always in agreement, while a petty man intends to casually
agree with others but without harmony. My main points here are simply that the
view would not be consequentialist, and Confucius does not seem interested in
developing such views.
47. See my discussion of this point in Chapter 1.
48. Wittgensteins relation to pragmatism has been recently explored by Russell Goodman. The account of Wittgensteins ambivalence toward pragmatism is
complicated, but for present purposes, it is worth mentioning that Wittgensteins
identification of meaning with use did not for him amount to an identification of
truth with usefulness of beliefs. See Russell B. Goodman, Wittgenstein and William
James (Cambridge, UK: Cambridge University Press, 2002), 158. For Wittgensteins
comment on pragmatism, see Wittgenstein, On Certainty, section 422. In Remarks
on Philosophical Psychology, Wittgenstein makes his opposition to pragmatism clear
when he says that he is not saying a proposition is true if it is useful (Part 1,
section 266).
49. Yang Xiao, How Confucius Does Things with Words: Two Hermeneutic
Paradigms in the Analects and Its Exegeses, The Journal of Asian Studies 66, no. 2
(2007): 497532.
50. Xiao, How Confucius, 502, n.7
51. Ibid., 501.
52. Ibid.
53. Either might work as a translation, but I would like to propose that
mine is better in that it captures this pervasive feature of these sorts of questions
to Confucius. Moreover, it captures a pervasive feature of Confuciuss responses. In
this particular passage, the translation is complicated by the follow-up questions by
Gongxi Hua. But there is no need to mistranslate the first part of the passage, which
reports these earlier conversations, with Gongxi Huas misunderstanding of them.
54. I am still using Xiaos translation here, but I have dropped his interpolation of with these words at the end of each sentence. He needs this interpolation to support his speech act interpretation of the passage, but the text does not
contain these phrases.
55. For Confuciuss account of his own stages of deepening his understanding
of dao, see Analects 2.4.
56. Wittgenstein, On Certainty, throughout.
57. Ibid., section 204.
58. See Chapter 1 for a discussion of the age of Wittgensteins and Confuciuss novices and the question of what counts as a background practice being
taught to a novice.
59. Wittgenstein, On Certainty.

Notes to Chapter 6

293

60. This weakness seems to show in various passages, among them Analects
17.21 and 18.67. In these passages, arguments might seem necessary to refute challenges, but arguments are lacking or not well developed. But this is as we should
expect if Confucius understands these failures of understanding to first require a
change in practice.
61. Xiao, How Confucius, 510.
62. Although I dont argue for this claim here, I suspect that any account
that does not attribute truth claims to Confucius will reduce to some form of ethical non-cognitivism. Although articulating this view in some of its forms might
be funMourn for three years. Rah! Two years. Boo!this attribution would
be uncharitable.

Chapter 6
1. I borrowed this title from D. L. Hall and R. T. Ames, Getting It Right:
On Saving Confucius from the Confucians. Philosophy East & West 34, no. 1
(1984): 323. Although the authors and I agree that Confucius needs saving, our
analysis of why differs. For my commentary on their view, see Chapter 5, Hall
and Ames and the Pragmatic Reading of Chinese Philosophy.
2. For a detailed articulation of this view, see Kim-chong Chong, The
Practice of Jen, Philosophy East and West 49, no. 3 (1999).
3. Kwong-loi Shun, Jen and Li in the Analects, Philosophy East and West
43, no. 3 (1993): 45779.
4. Analects 12.2.
5. There is no reason to understand this passage as invoking a Daoist-like
mysticism. (See Arthur Waley, The Analects of Confucius, New York: Vintage Books,
1989, 3839.) We might suppose, instead, that Confucius is at a higher stage of
mastery of the complex skills and practices involved in the dao. Yan Hui is at a lower
stage, but we dont know exactly how these stages work and what they are. What
Yan Hui offers us is an account, from the advanced novice perspective, of what it
is like to struggle at a stage lower than the most advanced stage. We should not,
however, take his account to be anything but an account from his advanced novice
stage of mastery. This view of the pedagogy required to get novices to advanced
levels of skill fits nicely with recent work on the stages of advancement on the way
toward mastery of complex, practical skills. See H. L. Dreyfus, A Five-Stage Model
of the Mental Activities Involved in Directed Skill Acquisition, Berkeley: University
of California, Operations Research Center, 1980.
6. Most translators translate (xue) as learn, except where context
demands study. Translators D. C. Lau, The Analects (Hong Kong: The Chinese
University Press, 2000); Arthur Waley, The Analects of Confucius (New York: Vintage Books, 1989); and Edward G. Slingerland, Confucius Analects: With Selections
from Traditional Commentaries (Indianapolis, IN: Hackett Pub. Co, 2003) all reject
study as being overly intellectual. But Ames and Rosemont opt for study because,

294

Notes to Chapter 6

they say, it is a process word, whereas learn is an achievement word, and, they
claim, the Analects offers a process view of the world. (See their Analects, p. 230).
But learn in its various forms is not always an achievement word. I am learning
Chinese itself can refer to a process.
7. For Philip J. Ivanhoes account of Confucian self-cultivation and its relation to learning, see his Rockwell lectures, P. J. Ivanhoe, Confucian Moral SelfCultivation, vol. 3, The Rockwell Lecture Series (Indianapolis, IN: Hackett, 2000),
especially Chapter 1. For an account of the forms of learning that Confucius and his
followers supported, see Robert Eno, The Confucian Creation of Heaven: Philosophy
and the Defense of Ritual Mastery. SUNY Series in Chinese Philosophy and Culture
(Albany: State University of New York Press, 1989), 5263. For D. C. Laus account
of the close relation between learning and practice, see Lau (2000), xliii ff. Also see
Slingerland 2003 throughout, but especially his commentary to Analects 1.1 and his
glossary entry for learning on p. 239. For a detailed analysis of various passages
that support this approach, see Kim-chong Chong, The Practice of Jen, 1999.
8. Analects 15.31.
9. That Confucius holds that there is an important relationship between
words and practices should come as no surprise, given the emphasis he gives to
(zhengming), rectifying names, which at least articulates the requirement of
applying functional words, which invokes norms of the function they introduce
only to those instances that fulfill those norms. To devote oneself to those words
requires devotion to the associated norms and, so, to the practices they require.
But in the case of ethical terms, like both (ren) and (li), whose norms of
usage are connected to practices that embody those norms, devotion to words is
connected to devotion to practices.
10. See my comments on the supporting passage 9.1, which claims that Confucius seldom spoke of benefit, fate, or ren in Chapter 2 at note 131. Of course,
passage 5.13 has been variously interpreted. My reading here follows He Yan. For a
discussion of the history of interpretations of this passage, see P. J. Ivanhoe, Review:
Thinking Through Confucius by David L. Hall and Roger T. Ames. Philosophy East
and West 41, no. 2 (1991): 24154. Ivanhoe makes the point that commentaries
on the Analects reflect the variety of metaphysical commitments that commentators bring to that text. He concludes that translation will involve the philosophical
project of teasing out metaphysical views. My only addition to this argument, based
on my arguments in Chapter 4, would be to claim that we need also to interpret
and translate these texts in light of the strongest philosophical arguments for and
against these optional metaphysical views.
11. See D. Z. Phillips, Interventions in Ethics (London: Blackwell, 1992), viii.
12. This interest in exemplars of (ren) is no doubt connected to Confuciuss claim (7.22) that if he is among others, he can find models ( shi) among
them, both in the forms of those who provide examples of conduct to follow and
examples of conduct to avoid.
13. I depart from the usual translation of (haoxue) as love of learning because in English the word love is multiply ambiguous but even in its

Notes to Chapter 6

295

strong senses expresses the idea that the person is pleased by or prefers the object
of his love to other things. However, it is clear from the discussions of Yan Huis
(haoxue) that it also requires something like a strong commitment, and that
is captured by the term devotion; thus, devotion to learning is preferable here.
14. Since Waley and Chan have made this point clear, it has become common to distinguish between (ren) understood as a specific virtue along others
and (ren) understood as the most general virtue, which includes all particular
virtues. My claims here and throughout this essay concern the latter, not the former. See Waley, The Analects of Confucius, 2729. See Wing Tsit Chan, Chinese
and Western Interpretations of Jen (Humanity), Journal of Chinese Philosophy 2
(1975): 107129.
15. This formula for (ren) would work equally well for any normative
concept. For example, a person is logical provided that his conduct and thinking
lack any of those defects that would make his behavior illogical. Although Confucius
does not himself articulate this principle, it is implicit in the way he rejects persons
or forms of conduct as exemplifying (ren). See Analects 1.3, 4.7, 5.5, 5.8, 5.19,
7.34, 12.20, 14.28, 15.9, 15.35, 15.36, and 19.15.
16. One exception to this is his claims that the brothers Boyi
Shuqi sought and achieved (ren), but given the rest of his skeptical claims about
(ren), we should think either that these men had the status of the mythological
sage kings or that they achieved (ren) merely in their specific actions of refusing
to take the throne over the other. Waley likens them to sage kings by claiming that
they are legendary figures, and so in that way not real people.
17. I cannot in detail defend my approach here but will say a couple of
things. The claim that Confucius was a sage contradicts his own self-descriptions
but is based on claims of some hidden teachings and pedagogical strategies in
which he attempts to encourage his interlocutors by claiming to be like them. All
of this is part of an orthodox commentary project, like those of other religious
traditions, designed to interpret the Analects and other classical texts as embodying a completely true and consistent account of the Truth. This approach requires
commentators to claim that Confucius did not mean what he said. Any effort to
read texts under this principle gives too much interpretive license, however, and
allows us to read into the text any meaning we happen to think represents the
Truth. Also it assumes, prior to interpretation and evaluation, that the views in
the text are true. For these reasons I reject this approach. In thinking about this
issue, I benefited from John B. Henderson, Scripture, Canon, and Commentary: A
Comparison of Confucian and Western Exegesis (Princeton, NJ: Princeton University
Press, 1991), especially 18486.
18. Wittgenstein, Philosophical Investigations, section 133.
19. Ibid., section 112.
20. Ibid., section 111.
21. See Wittgenstein et al., Wittgensteins Lectures, Cambridge, 19301932:
From the Notes of John King and Desmond Lee (Totowa, NJ: Rowman and Littlefield,
1980), 21; Wittgenstein, Philosophical Investigations, section 133.

296

Notes to Chapter 6

22. For an account of the cognitive therapeutic aspects of Wittgensteins

method and view of his therapeutic project, see Eugen Fischer, A Cognitive SelfTherapyPhilosophical Investigations, Sections 13897, in Wittgenstein at Work:
Method in the Philosophical Investigations, eds. E. Ammereller and E. Fischer (London: Routledge, 2004).
23. For this detailed argument, see Chapter 1.
24. This strategy reflects an approach in which I differ from many interpreters
of the Analects. Because philosophical texts, including the Analects, are ambiguous,
the selection of the correct or best justified interpretation depends on an appeal
to the principle of charity, which requires that we attribute the most defensible
of possible interpretations to the text and its author if we are to claim to have
understood the text. (See Chapters 3 and 4.) So we cannot just stop with offering
a possible reading of the text. We must show why, of any two possible readings,
one embodies a more defensible philosophical view than the other. In this essay, I
rule out Zhu Xis possible reading of the Analects by showing how it suffers from
a trilemma, from which, I suspect, it cannot escape. In addition, as I indicated
above, I think his approach allows for more interpretive license than could ever be
warranted for any text.
25. For the sake of simplicity, in what follows, I intend by He Yan to mean
He Yan along with his fellow commentators.
26. I use Gardners text of these two commentaries but offer my own translation. Gardners He Yan commentary comes from the Lunyujijie (Collected Commentaries on the Analects) and his Zhu Xi commentary from Lunyu Jizhu (Collected
Explanations of the Analects.)
27. Confucius, Analects, 12.1.
28. From John Kieschnick, Analects 12.1 and the Commentarial Tradition,
The Journal of the American Oriental Society 112, no. 4 (1992): 56776.
29. Wittgenstein, Philosophical Investigations, section 38.
30. As I intend it, criteria can refer to any number of types of indicators
of correct usage, from formal specifications of the conditions for applying a term
to informal precedents established by usage.
31. The confusion is, of course, the confusion of whether a person has
engaged in the practices successfully. The skepticism, which I cannot discuss here
in detail, arises from invoking abstract considerations, the correctness of which are
open to question. Confuciuss suspicion of reflection not grounded on practice is
an antidote to this sort of skepticism. For an account of the skepticism that arises
from ethical theories, see Annette Baier, Theory and Reflective Practices, and
Doing without Moral Theory, Postures of the Mind (Minneapolis: University of
Minnesota Press, 1985), 20745.
32. The general requirement that concepts must be associated with explicit
rules in order to be meaningful is mistaken. For in ordinary contexts, concepts
often get their meaning from the role they play in complex forms of behavior that
give concepts meaning, including precedents, as I indicated in note 32.Theoretical
concepts, however, typically are spelled out in terms of explicit rules because their

Notes to Chapter 7

297

meaning arises from their role in theory, not from their embeddedness in informal
practices.
33. This trilemma is also perfectly general and applies to other passages in
the Analects for which metaphysical interpretations may be offered.

Chapter 7
1. Jiwei Ci, The Confucian Relational Concept of the Person and Its Modern Predicament, Kennedy Institute of Ethics Journal 9, no. 4 (1999): 32546.
2. In addition to the argument I offer here about the changing character
of Confucianism, it is not at all clear that any view that passes as Confucian needs
to be historically realized if that means practiced. A philosopher might offer a new
version of Confucianism that, in fact, has never been practiced.
3. Although I find their semantic theories wanting (see Chapters 3 and
4), John Makehams and Daniel Gardners accounts of the history of the Analects
commentarial tradition show one aspect of those changes. Changes in the tradition
were reflected in changes of interpretation of the Analects and other sacred texts.
4. Xiaotong Fei, From the Soil: The Foundations of Chinese Society: A Translation of Fei Xiaotongs Xiangtu Zhongguo, With an Introduction and Epilogue, trans.
Gary G. Hamilton and Wang Zheng (Berkeley: University of California Press, 1992).
5. Robert Neelly Bellah, Habits of the Heart: Individualism and Commitment
in American Life (Berkeley: University of California Press, 1985) makes a case for
the view that the vocabulary of American middle-class individualism covers over
more traditional commitments in tension with our individualism.
6. For my discussion of the pervasive character of this form of authority in
true-false language-games, see Chapters 2 and 5.
7. Alasdair MacIntyre, Incommensurability, Truth, and the Conversation
between Confucians and Aristotelians about the Virtues, in Culture and Modernity:
East-West Philosophic Perspectives, ed. Eliot Deutsch (Honolulu: University of Hawaii
Press, 1991): 104122.
8. Or better yet, once again, I will address this question. Much of this book
directly or indirectly addresses this question.
9. MacIntyre, Incommensurability, Truth, (1991), 106.
10. A way to capture this sense that Confucius had only a small place for
theory can be found in May Sims Remastering Morals with Aristotle and Confucius
(Cambridge: Cambridge University Press, 2007). Sim argues that despite his lack
of interest in metaphysics, Confucius implicitly understands (59) and implicitly
uses (62) Aristotles metaphysical categories, including the category of substance.
This strikes as the only possible way to defend MacIntyres point, and Sim does
this with great care. The notion of implicit use strikes me as especially problematic,
however, since elsewhere Sim claims that Confucius lacks a metaphysic (4546).
This shows that her claim importantly relies on the notion of implicit use. But
this would be no different from claiming that my grandmother implicitly uses the

298

Notes to Chapter 7

category of substance every time she talks about tables and chairs. I would, however,
deny this. Her talk might be so translated, but that is not the same thing. On this
general issue, see Wittgensteins discussion of why, when we think about a broom,
we dont think about the brooms parts in Philosophical Investigations, section 60.
11. MacIntyre, Incommensurability, Truth, 106.
12. Ibid., 120.
13. Ibid., 113.
14. Ibid., 112.
15. For my discussion of Halls and Amess account of the Confucian view
of truth, see Chapter 5.
16. MacIntyre, Incommensurability, Truth, 106 (my emphasis).
17. Ibid., 106.
18. Ibid., 106.
19. Confucius, Analects, 8.10.
20. Ibid., 14.4.
21. Ibid., 17.21.
22. MacIntyre, Incommensurability, Truth, 106.
23. Ludwig Wittgenstein, Tractatus Logico-Philosophicus, rev. ed. (London:
Routledge & Kegan Paul, [1921] 1961).
24. Wittgenstein, Philosophical Investigations, section 43.
25. Wittgenstein, Tractatus, sections 2.0211 and 3.23.
26. MacIntyre, Incommensurability, Truth, (1991), 113.
27. For a related discussion of Wittgensteins ethics without metaphysics,
see Chapter 2.
28. James F. Peterman, Philosophy as Therapy: An Interpretation and Defense
of Wittgensteins Later Philosophical Project, SUNY Series in Philosophy and Psychotherapy (Albany: State University of New York Press, 1992).
29. For my use of this form of argument to save Confucius from Confucians, see Chapter 6.
30. Wittgenstein, Philosophical Investigations, section 133.
31. Ibid., section 132.
32. My account of Wittgensteins pluralism has been influenced by the way in
which he, like Rudolf Carnap, takes languages and their grammars to be plural. See
Rudolf Carnap, The Logical Structure of the World; and Pseudoproblems in Philosophy,
trans. Rolf A. George (Berkeley: University of California Press, 1967). As a result of
holding this view, Wittgenstein takes philosophical corrections of ordinary language
as based on recommendations to change the grammar of ordinary language, not as
discoveries of some feature of the world. I have also been influenced by Gordon
Bakers later essays on Wittgensteins later therapeutic pluralism: Gordon P. Baker,
Wittgensteins Method: Neglected Aspects (Oxford: Blackwell, 2004).
33. One might wonder what justifies my claim that Confucius is engaged
in ethical or moral reflection. MacIntyre mentions that one form of incommensurability between Confucian and Greek traditions results in our not being free to
use the term moral to characterize early Chinese thought or practice. Following

Notes to Chapter 7

299

Henry Rosemont (Henry Rosemont, Jr., Reply to Professor Fingarette. Philosophy

East and West 28, no. 4 (1978): 51519), MacIntyre claims that the early Chinese
had no single term for morality, and so, no concept of it: And Henry Rosemont
has drawn our attention to the importance of the fact that the classical Chinese
language has no terms for, and that correspondingly Confucian texts contain no
discussion of, the most familiar Western moral concepts, including that of morality itself (108). However, it doesnt follow that because a person has no single
word for something, that person has no concept of it. Confucius demonstrates a
concept of morality by reflecting on and speaking about the variety of virtues and
vices he focuses onwith the overarching purpose of living toward the normative
force of dao. In so doing, he manifests a concept of morality without having a
specific word for it.
It is also relevant that Confucius discusses both dao and de, which are later
combined to function as the translation for the Western concept of morality. His
failure to put these together in a single term does not show that he had no concept
of what they share.
34. Confucius, Analects, 5.13.
35. Ibid., 7.21.
36. Ibid., 9.1.
37. Ibid., 1.7.
38. Ibid., 2.15.
39. Ibid., 11.12.
40. For a detailed defense of this claim, see Chapter 1.
41. Confucius, Analects, 2.4.
42. See Peter Unger, Contextual Analysis in Ethics, Philosophy and Phenomenological Research 55 (1995): 126.
43. See Peter Kleins articulation of this point in Peter D. Klein, Human
Knowledge and the Infinite Regress of Reasons, Nos 33 (1999): 297325: Now,
I grant that there are occasions when it is absurd to ask for reasons for a belief.
Roughly, those are the occasions in which it is clear that the conversational presuppositions are not to be questioned. For example, when we are distinguishing
features of waking states from features of dream states, it is absurd to ask whether
we can tell the difference. But it does not follow that such questions are always
inappropriate. Indeed, when the presuppositions of the conversational context are
revealed, they can be questioned. Thus, one can grant what I think Aristotle is
suggesting, namely that demonstration can take place only within a context of
agreed upon presuppositions and that it is absurd to ask for reasons to justify
those presuppositions within that kind of a context. He is right. But, of course,
the contextual situation can change.
44. This example was made famous by Fred Dretske in Epistemic Operators,
Journal of Philosophy 67 (1970): 100723.
45. Not all Aristotelians accept MacIntyres view of the metaphysical requirements that accompany moral inquiry. See Rosiland Hursthouse, On Virtue Ethics
(Oxford: Oxford University Press, 1999).

300

Notes to Chapter 8

46. I follow Confucius in his use of dangerous. See Analects 2.15. For an
example of the sort of truth claim Confucius makes, see my discussion of Analects
13.3 in Chapter 5 at note 14.
47. I am paraphrasing part of a sentence from Philosophical Investigations,
section 289. [T]o use a word without justification does not mean to use it wrongfully. See my discussion of this dictum in Chapter 2.
48. Ludwig Wittgenstein, Remarks on the Foundations of Mathematics (rev.
ed.), trans. G. H. von Wright, Rush Rhees, and G. E. M. Anscombe (Cambridge,
MA: MIT Press, 1978), VI.8. For a fuller discussion, see Chapter 2.
49. For a characterization of this depth, see Chapter 6, and for a characterization of the realistic spirit behind these claims, see Chapter 2.
50. I take this ladder metaphor from the concluding passages of Wittgensteins Tractatus.
51. For an additional discussion of Wittgensteins constitutivism in relation
to Confuciuss teaching, see Chapter 1.

Chapter 8
1. Herbert Fingarette, Confucius: The Secular as Sacred (Religious Traditions
of the World) (Prospect Heights, IL: Waveland Press, Inc., 1998).
2. For an influential criticism, see Benjamin I. Schwartz, The World of
Thought in Ancient China (Cambridge, MA: Belknap Press of Harvard University
Press, 1985).
3. See Chapter 6 of this book.
4. See Chapter 3.
5. For a discussion of the character of moral practices implicit in Wittgensteins later philosophy, see D. Z. Phillips and H. O. Mounce, Moral Practices (New
York: Schocken Books, 1970).
6. A second example that Fingarette offers to show that ritual is magic
is the example of a teacher requesting that a student fetch a book. But, of course,
there is nothing magical here either.
7. Fingarettes (1998) appeal to the idea of Holy Rite derives from his etymological analysis of the character (li), but his analysis of this characters meaning
suffers from the problem of any such genetic analysis of meaning. It supposes that
the early root meaning is retained in later usage, either literally or metaphorically.
But if we follow Wittgenstein and think of our basic notion of meaning as use,
we will not want to use root metaphors to establish meaning except in those cases
where usage will bear that out. The meaning of (li) changes over time and
eventually comes to be identified with social norms and laws. In light of this history, we need not feel compelled to think of the etymological meaning of (li)
as governing all of its uses. See Masayuki Satos detailed history of the meaning of
(li) in The Confucian Quest for Order: The Origin and Formation of the Political

Notes to Chapter 8

301

Thought of Xun Zi, book 58, Sinica Leidensia series (Leiden, NL; Boston, MA: Brill
Academic, 2003), chapter three.
8. For Wittgensteins use of this example to illustrate what he means by
language-game, see Philosophical Investigations, section 7.
9. For a recent discussion of the way in which ritual does not always require
sincerity to be successful, see Adam B. Seligman, Robert P. Weller, Michael J. Puett,
and Bennett Simon, Ritual and Its Consequences: An Essay on the Limits of Sincerity
(Oxford: Oxford University Press, 2008).
10. He claims to find this view in Analects, but this aspect of his account
has come under attack. See Schwartz, The World of Thought in Ancient China, 1985,
chapter 3.
11. Fingarette, Confucius: The Secular as Sacred, 1998, 7.
12. Ibid.
13. Ibid., 67.
14. Ibid., 3, 8.
15. Ibid., 8.
16. Ibid., 46.
17. Ibid., 4.
18. Ibid., 1.
19. See this chapter, note 2.
20. Confucius, Analects, 2.20.
21. Ibid., 2.1.
22. Ibid., 5.16.
23. Ibid., 12.5.
24. I borrow this grammatical analysis from Christoph Harbsmeier, ed.,
Thesaurus Linguae Sericae: An Historical and Comparative Encyclopaedia of Chinese
Conceptual Schemes, accessed August 6, 2009, <http://tls.uni-hd.de/>.
25. See the translation of the Analects with accompanying grammatical analyses at Thesaurus Linguae Sericae (TLS).
26. I borrow this translation and this general point from S. Y. Chan, The
Confucian Notion of Jing (Respect), Philosophy East and West 56, no. 2 (2006):
229. Chan claims that, according to Confucius, all persons are deserving of respect
according to the Analects framework, but she bases this on the claim that everyone
is possibly a sage. But Confucius never makes this point, and, given the hierarchies
involved in his use of jing (respect), this democratizing move would have to be seen
as alien to the text. I am not, however, arguing that, given the concept of jing in
the Analects, he could not respect everyone. The evidence from the text only gives
us important examples of objects of jing, not an analysis of the limits of the concept. We do, however, know that the ethical interventions Confucius is recognized
for having engaged in did not involve him intervening to encourage everyone to
respect everyone else. If his use of the concept jing reserves it for higher-ups, then
that use would be ruled out by the concept itself. If his concept includes any object
of worth, as Chan claims, then it is not ruled out. Although both accounts are

302

Notes to Chapter 8

consistent with the text, it is clear that even if it could be argued without contradicting any passage in Analects that Confucius thought (or was motivated by the
belief ) that everyone must jing (respect) everyone, he is not presented as doing so
or as recommending it.
27. Confucius, Analects 15.24. It is noteworthy that by most accounts of the
genesis and structure of the Analects, this is a relatively late passage. Brooks and
Brooks, Original Analects, 1997, 136137 and 149, and Bryan W. Van Norden,
Unweaving the One Thread of Analects, 4:15, in Bryan W. Van Norden, Confucius
and the Analects (Oxford: Oxford University Press, 2002), argue that Analects 4.15
is a late passage. What we find in these two passages may then be an attempt to
find some unified thread not readily apparent in earlier passages.
28. Confucius, Analects 6.30.
29. Van Norden, Unweaving the One Thread. For David Nivisons analysis
of the usage of these characters in the Analects, which I follow, see his Golden
Rule Arguments in Chinese Moral Philosophy, in The Ways of Confucianism, ed.
Bryan W. Van Norden (Chicago: Open Court, 1996), 5976. For Philip. J. Ivanhoes account, see his Reweaving the One Thread of the Analects, Philosophy
East and West 40, no. 1 (1990): 1733. I have simplified Nivisons account in
one way. For reasons unclear to me, Nivison allows that a person should practice
(zhong) in relations between superiors and equals and practice shu in relations
between subordinates and equals. He neither argues that the passages in the Analects
support the inclusion of equals in each practice, nor does he discuss his reason for
mentioning them. I assume he treats equals as limiting cases of superiors and subordinates. I discuss below the cases of usage of (zhong) and (shu) in which
encounters are not between equals. For Philip J. Ivanhoes recent revision of his
earlier arguments, see his The Golden Rule in the Analects, in Confucius Now:
Contemporary Encounters with the Analects, ed. David Edward Jones (Chicago: Open
Court, 2008), 81107.
30. Nivison, Golden Rule Arguments, in Van Norden, 1996, 65. In my
discussion here, I leave out Nivisons discussion of zhong practices as practices of
subordinates to superiors. For that argument, see his discussion of Analects 5.19 in
Nivison, Golden Rule Arguments, in Van Norden, 1996, 66.
31. Without explanation, Nivison in Golden Rule Arguments holds that
the requirement of (zhong) applies to a subordinate toward a superior or equal.
The requirement of (shu) applies to subordinates and equals. But he never
explains why he thinks that these requirements are appropriate toward equals. If a
superior, who by his nature has attained a superior status, can impose his will on
subordinates, who, other things being equal, are required to do his bidding, then
we can make sense of limitations on his ability to impose his will on a subordinate
as arising from this feature of the superior-subordinate relationship. It is less clear
to me that the equal-equal relationship requires any parallel limitation. Equals are
not, other things being equal, required to do the bidding of equals.
32. Ibid., 68.
33. Nivison, Golden Rule Arguments, 76.

Notes to Chapter 9

303

34. Erving Goffman, Interaction Ritual: Essays in Face-to-Face Behavior. (New

York: Pantheon Books, 1982), 8485.
35. Quoted from mile Durkheim, The Determination of Moral Facts,
Sociology and Philosophy, trans. D. F. Pocock (New York: Free Press, 1974), 37, in
Interaction Ritual: Essays on Face-to-Face Behavior, ed. Erving Goffman (New York:
Pantheon, 1982), 73.
36. Goffman, Interaction Ritual, 6061.
37. Wittgenstein, Philosophical Investigations, section 66.
38. Ibid., section 100.
39. I develop a related criticism of the confusions caused by Zhu Xis interpretation of the Analects in Chapter 6.
40. Fingarette, Confucius: The Secular as Sacred, 1998, 17.
41. See Chapters 1 and 2 of this book.

Chapter 9
1. Wittgenstein, Philosophical Investigations, sections 109 and 116.
2. Ibid., 238.
3. Ludwig Wittgenstein, On Certainty, ed. Gertrude Elizabeth Margaret
Anscombe and Georg Henrik von Wright. (New York: Harper and Row, [1969]
1972). For a detailed account of Wittgensteins views in this text, see Avrum Stroll,
Moore and Wittgenstein on Certainty (New York: Oxford University Press, 1994).
4. Wittgenstein, On Certainty.
5. Ibid.
6. See Chapter 8 in this book.
7. In Theoretical Continuities, Randall Collins claims that Goffmans primary work was taxonomic, awaiting a later theorist to explain the phenomenon
he categorized. Collins explains the continuity of Goffmans writings in terms of
his elaboration of Durkheimian commitments concerning the importance of ritual
in everyday life (43). In his Goffman as Systematic Social Theorist, Giddens, in
contrast, argues that Goffman has been misunderstood as he developed a systematic view of the sociological enterprise and needs to be understood as engaged in
theory. Both of these essays are reprinted in Gary Alan Fine and Gregory W. H.
Smith, Erving Goffman (London: Thousand Oaks, CA: Sage, 2000). I would argue
that Goffmans taxonomic work offers a starting point for constructing a perspicuous overview of everyday ritual even if his efforts at a system are problematic. For
example, his acknowledgment that there are rituals of insult seems to contradict
his basic systematic claims about the point of ritual. For more on this, see below.
8. Some critics, like Alasdair MacIntyre in his The Self as Work of Art in
Fine and Smith, Erving Goffman, 2000, and After Virtue: A Study in Moral Theory
(South Bend, IN: Notre Dame Press, 1984) argue that Goffmans dramaturgical
view of self makes it impossible to understand ourselves as moral agents. As role
players, we can always step outside any role. And as having no substantial self, we

304

Notes to Chapter 9

have no place from which to understand fundamental moral commitments. This

criticism seems to ignore other points of view that Goffman brings to his analyses
and the interrelation between his ritual and dramaturgical analyses. On this point,
see Ann Branamans essay, Goffmans Social Theory, in The Goffman Reader, ed.
and with a preface and introduction by Charles C. Lemert and Ann Branaman
(Malden, MA: Blackwell, 1997).
9. We might distinguish two types of claims here. One is the claim that
the syntax of ritual interactions can be understood on their own terms. The other
is that the ritual interaction order does not receive or offer influence from other
parts of culture and society, and so cannot be even partly explained in terms of its
relation to other aspects of society and culture. The former claim can be true even
if the latter is false. I assume that Goffman has to mean that the syntax of ritual
interaction can be explored on its own, not that no aspect of the ritual order can
be explained in terms of its relation to other aspects of society and culture.
10. See Goffmans On Face Work, reprinted in Erving Goffman, Interaction
Ritual: Essays in Face-to-Face Behavior (New Brunswick, NJ: Aldine Transaction,
2005). Not all rituals need to be face-to-face, however, for Goffmans account of
ritual to be persuasive. We need only accept the common-sense idea that face-toface ritual is basic and any other sort arises from it. We might think that rituals
designed to send good wishes to ghosts and spirits are not face-to-face, but this
sort of ritual would develop out of face-to-face rituals.
11. Ibid., 13.
12. Robert Neelly Bellah, with 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).
13. Quoted from Durkheim, The Determination of Moral Facts, 37, in
Goffman, Interaction Ritual, 69.
14. Goffman, Interaction Ritual, 9.
15. Ibid., 10.
16. Ibid.
17. Ibid.
18. Ibid., 11.
19. Ibid., 14.
20. Ibid., 19.
21. Ibid., 22.
22. Ibid., 77.
23. Erving Goffman, Behavior in Public Places: Notes on the Social Organization of Gatherings (Westport, CT: Greenwood Press, 1980).
24. Ibid., 18.
25. Ibid.
26. Ibid., 1516.
27. Ibid., 18.
28. Ibid., 1819.

Notes to Chapter 9

305

29. Ibid., 22.

30. Ibid., 2324.
31. Ibid., 24.
32. Ibid., 194.
33. Ibid., 195.
34. Ibid., 199.
35. One might ask whether the fact of Goffmans later analysis of the rules
governing gatherings is meant to replace his earlier work on ritual. In his presidential address (1983), he continues to speak of interaction in terms of ritualization
of behavior (3), even though the primary work of that essay is, like his account of
behavior in public, an attempt at constructing a natural history of the interaction
order: A less windy effort, equally general but naturalistically based, is to try to
identify the basic substantive units, the recurrent structures and their attendant processes. What sorts of animals are to be found in the interactional zoo? What plants
in this particular garden? (Erving Goffman, The Interaction Order: American
Sociological Association, 1982 Presidential Address, American Sociological Review,
48 (no. 1, Feb. 1983), 6). The totality of Goffmans work gives the impression of
his wanting to capture, like Wittgensteins later work does, the character of that
complex order, without any ideological commitment that freezes a persons grasp of
it in just one frame of reference. It can be understood in various ways, and those
ways complement one another.
36. Goffman, Interaction Ritual.
37. Ibid., 54.
38. Ibid., 56.
39. Ibid., 56.
40. Ibid., 59.
41. Ibid., 6265.
42. Ibid., 71.
43. Ibid., 77.
44. Ibid., 82.
45. Ibid., 85.
46. This aspect of Goffmans view parallels Wittgensteins view of the function
of grammatical investigations, which function in part to correct misleading pictures
of the use of concepts. So we can be misled into thinking that the self is something
inner, hidden behind all performance. We can then be led to develop an account of
the hidden self. But Wittgenstein indicates that what is hidden is of no interest to
us. It can appear that he denies the inner self. But his grammatical investigations
are meant to clarify how we distinguish between inner self and outer performance
in our ordinary talk. Goffmans investigation of the syntax of the interaction order
has a similar motivation.
47. Erving Goffman, The Presentation of Self in Everyday Life (Harmondsworth, UK: Penguin, [1959] 1971), 216.
48. Ibid., 23134.

306

Notes to Afterword

49. Ibid., 76.

50. It would be possible to make this point as well in terms of two forms
of respect, one quite general, which we are required to express toward anyone we
encounter, and the sort that reflects our sense of the specific social position or
accomplishment of a person. The advantage I find in making the distinction the way
I have here is just that it highlights a difference in the ways these ideals function,
one that, in some respects, seems to parallel the Analects terminological difference
between shu and jing . See this discussion in Chapter 8.
51. Arthur M. Schlesinger, Learning How to Behave: A Historical Study of
American Etiquette Books (New York: Cooper Square Publishers, 1968), 95.
52. Goffman, Relations in Public: Microstudies of the Public Order (New York:
Harper and Row, 1971), xvii, 396, 6263.
53. Ibid., 64.
54. Ibid., 64.
55. Ibid., 32.
56. Goffman, Interaction Ritual, 88.
57. The need to distinguish between conventional gestures and rituals was
pointed out to me by Philip J. Ivanhoe.
58. Confucius, Analects, 17.18.
59. Ibid., 10.16.
60. Rituals of contempt also create trouble for Goffmans Durkheimian
account of ritual, which explains ritual in terms of protection of persons who are
treated as sacred. He calls rituals of contempt negative rituals of deference. But
this term just shows that there are rituals that fall outside his analysis.
61. For Frank Cioffis discussion of Goffmans examples that makes this very
point, see Information, Contemplation, and Social life in his Wittgenstein on Freud
and Frazer (Cambridge: Cambridge University Press, 1998), 1946.

Afterword
1. For another recent defense of Confucian of pluralism, see Bryan W. Van
Norden, Virtue Ethics and Consequentialism in Early Chinese Philosophy (New York:
Cambridge University Press, 2007), 315360.

Bibliography

Ames, Roger T. Wandering at Ease in the Zhuangzi. Albany: State University of

New York Press, 1998.
Ames, Roger T., and Henry Rosemont. The Analects of Confucius: A Philosophical
Translation. New York: Ballantine Publishing Group, 1998.
Ammereller, Erich, and Eugen Fisher. Wittgenstein at Work: Method in the
Philosophical Investigations. London: Routledge, 2004.
Anscombe, G. E. M. An Introduction to Wittgensteins Tractatus. 4th ed. London:
Hutchinson, (1959) 1971. Page references are to the 1971 edition.
Baier, Annette. Theory and Reflective Practices. In Postures of the Mind, 20727
Minneapolis: University of Minnesota Press, 1985.
. Doing without Moral Theory. In Postures of the Mind, 22845.
Minneapolis: University of Minnesota Press, 1985.
Baker, Gordon P. Wittgensteins Method: Neglected Aspects. Edited and introduced by
Katherine J. Morris. Oxford, UK: Blackwell, 2004.
Becker, Lawrence C., and Charlotte B. Becker, eds. Encyclopedia of Ethics. 2nd ed.
New York: Garland Publishing, 1992. New York: Routledge, 2001. Page references are to the 2001 edition.
Bellah, Robert Neelly, with 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.
Benedict, Ruth. A Defense of Moral Relativism. The Journal of General Psychology
10 (1934): 5982.
Boswell, James. The Life of Samuel Johnson. First published 1791. Edited with introduction and notes and by Christopher Hibbert, Penguin English Library,
1979. Reprint, New York: Penguin Classics, 1986. Page citations are to the
1986 edition.
Branaman, Ann. Goffmans Social Theory. In Erving Goffman, The Goffman
Reader. Edited and with a preface and introduction by Charles C. Lemert
and Ann Branaman. Malden, MA: Blackwell, 1997.
Brooks, E. Bruce, and A. Taeko Brooks. The Original Analects: Sayings of Confucius
and His Successors. New York: Columbia University Press, 1998.

307

308

Bibliography

Carnap, Rudolf. The Logical Structure of the World; Pseudoproblems in Philosophy.

Translated by Rolf A. George. Berkeley: University of California Press, 1967.
Cavell, Stanley. Conditions Handsome and Unhandsome: The Constitution of
Emersonian Perfectionism. Chicago: Open Court, University of Chicago Press,
1990. First published as The Carus Lectures in 1988. Page references are to
the 1990 edition.
. Must We Mean What We Say?: A Book of Essays. New York: Scribner, 1969.
Chan, S. Y. The Confucian Notion of Jing (Respect). Philosophy East and West
56 (2006): 22953.
Chan, Wing Tsit. Chinese and Western Interpretations of Jen (Humanity). Journal
of Chinese Philosophy 2 (1975): 10729.
Cheng, Anne. Lunyu . In Early Chinese Texts: A Bibliographical Guide, edited
by Michael Loewe, 31323. Berkeley: The Society for the Study of Early
China and the Institute of East Asian Studies, University of California, 1993.
Chin, Annping. The Authentic Confucius: A Life of Thought and Politics. New York:
Scribner, 2007.
Chong, Kim-chong. The Practice of Jen. Philosophy East and West 49, no. 3 (1999):
298316.
Ci, Jiwei. The Confucian Relational Concept of the Person and Its Modern
Predicament. Kennedy Institute of Ethics Journal 9, no. 4 (1999): 32546.
Cioffi, Frank. Information, Contemplation, and Social life. In Wittgenstein on
Freud and Frazer. Cambridge: Cambridge University Press, 1998.
Clack, Brian R. Wittgenstein, Frazer, and Religion. New York: Palgrave Macmillan,
1999.
Clarke, Stanley G., and Evan Simpson. Anti-Theory in Ethics and Moral Conservatism.
SUNY Series in Ethical Theory. Albany: State University of New York Press,
1989.
Cohen, Stewart. Contextualism and Skepticism. Philosophical Issues, 10: Skepticism
(Supplement to vol. 34 of Nos, 2000): 94107.
Confucius, and Edward Slingerland. Confucius Analects: With Selections from
Traditional Commentaries. Translated by Edward Slingerland. Indianapolis,
IN: Hackett, 2003.
Crary, Alice, ed. Wittgenstein and the Moral Life: Essays in Honor of Cora Diamond.
Boston: The MIT Press, 2007.
Crary, Alice, and Rupert J. Read, eds. The New Wittgenstein. London: Routledge,
2000.
Creel, Herrlee G. Confucius and the Chinese Way. New York: Harper, 1960.
. Confucius: The Man and the Myth. Westport, CT: Greenwood Press, 1972.
Cua, Antonio S. Confucianism. In Encyclopedia of Ethics, edited by Lawrence C.
Becker and Charlotte B. Becker. New York: Garland Publishing, 1992. 2nd
ed., New York: Routledge, 2001. Page references are to the 2001 edition.
. Moral Vision and Tradition: Essays in Chinese Ethics. Washington, DC:
Catholic University of America Press, 1998.

Bibliography

309

Davidson, Donald. A Coherence Theory of Truth and Knowledge. In Truth and

Interpretation: Perspectives on the Philosophy of Donald Davidson, edited by
Ernest Lepore. Oxford, UK: Basil Blackwell, 1986.
. On the Very Idea of a Conceptual Scheme. In Donald Davidson, Inquiries
Into Truth and Interpretation, edited by Ernest Lepore. Oxford: Clarendon, 1984.
Dawson, Raymond Stanley. The Analects: Translated with an Introduction and Notes.
Oxford: Oxford University Press, 1998.
Diamond, Cora. The Realistic Spirit: Wittgenstein, Philosophy, and the Mind.
Cambridge, MA: The MIT Press, 1991.
Dretske, Fred. Epistemic Operators. Journal of Philosophy 67 (1970): 100723.
Edwards, James C. Ethics Without Philosophy: Wittgenstein and the Moral Life. Tampa:
University Presses of Florida, 1982.
Dreyfus, Stuart E., and Hubert L. Dreyfus. A Five-Stage Model of the Mental
Activities Involved in Directed Skill Acquisition. Berkeley: University of
California, Operations Research Center, 1980.
Ellenbogen, Sara. Wittgensteins Account of Truth. SUNY Series in Philosophy. Albany:
State University of New York Press, 2003.
Eno, Robert. The Confucian Creation of Heaven: Philosophy and the Defense of Ritual
Mastery. SUNY Series in Chinese Philosophy and Culture. Albany: State
University of New York Press, 1989.
Fei Hsiao-tung. From the Soil: The Foundations of Chinese Society. A Translation of
Fei Xiaotongs Xiangtu Zhongguo. Translated and with an Introduction and
Epilogue by Gary G. Hamilton and Wang Zheng. Berkeley: University of
California Press, 1992.
Feldman, Richard. Principle of Charity. In Routledge Encyclopedia of Philosophy.
London: Routledge, 1998.
Feyerabend, Paul. Putnam on Incommensurability. The British Journal for the
Philosophy of Science 38, no. 1 (1987): 7581.
Fine, Gary Alan, and Gregory W. H. Smith, eds. Erving Goffman. Sage Masters of
Modern Social Thought. London: Sage, 2000.
Fingarette, Herbert. Confucius: The Secular as Sacred. New York: HarperSanFrancisco,
a division of Harper Collins, 1972. Reissued Prospect Heights, IL: Waveland
Press, 1998. Page references are to the 1998 edition.
Fischer, Eugen. A Cognitive Self-TherapyPhilosophical Investigations. Sections
138197. In Wittgenstein at Work: Method in the Philosophical Investigations.
Edited by Erich Ammereller and Eugen Fischer, 86126. London: Routledge,
2004.
Forster, Michael N. Wittgenstein on the Arbitrariness of Grammar. Princeton, NJ:
Princeton University Press, 2004.
Frankfurt, Harry G. The Importance of What We Care About: Philosophical Essays.
Cambridge: Cambridge University Press, 1988.
Gadamer, Hans Georg. Philosophical Hermeneutics. Translated by David E. Linge.
Berkeley: University of California Press, 1977.

310

Bibliography

Gardner, Daniel K. Confucian Commentary and Chinese Intellectual History. The

Journal of Asian Studies 57, no. 2 (1998): 397422.
. Zhu Xis Reading of the Analects: Canon, Commentary, and the Classical
Tradition. New York: Columbia University Press, 2003.
Goffman, Erving. Behavior in Public Places: Notes on the Social Organization of
Gatherings. Westport, CT: Greenwood Press, 1980.
. The Goffman Reader. Edited and with a preface and introduction by Charles
C. Lemert and Ann Branaman. Malden, MA: Blackwell, 1997.
. The Interaction Order (1982 Presidential Address). American Sociological
Review 48 (1983): 117.
. Interaction Ritual: Essays in Face-to-Face Behavior. New York: Pantheon
Books, 1982. New Brunswick, NJ: Aldine Transaction, 2005. Citations are
to the 2005 edition.
. The Presentation of Self in Everyday Life. Harmondsworth, UK: Penguin,
1971.
. Relations in Public: Microstudies of the Public Order. New York: Harper
and Row, 1971.
Goodman, Russell B. Wittgenstein and William James. Cambridge: Cambridge
University Press, 2002.
Graham, A. C. Disputers of the Tao: Philosophical Argument in Ancient China. La
Salle, IL: Open Court Publishing Company, 1989.
Grandy, Richard. Reference, Meaning, and Belief. The Journal of Philosophy 70,
no. 14 (1973): 43952.
Hacking, Ian. The Parody of Conversation. In Truth and Interpretation: Perspectives
on the Philosophy of Donald Davidson. Edited by E. Lepore. Oxford, UK:
Basil Blackwell, 1986.
Haines, William A. The Purloined Philosopher: Youzi on Learning by Virtue.
Philosophy East and West 58, no. 4 (2008): 47091.
Hall, David L., and Roger T. Ames, Getting It Right: On Saving Confucius from
the Confucians. Philosophy East and West 34, no. 1 (1984): 323.
. Thinking from the Han: Self, Truth, and Transcendence in Chinese and Western
Culture. Albany: State University of New York Press, 1998.
. Thinking Through Confucius. SUNY Series in Systematic Philosophy.
Albany: State University of New York Press, 1987.
Hansen, Chad. Chinese Language, Chinese Philosophy, and Truth. The Journal
of Asian Studies 44, no. 3 (1985): 491519.
. A Daoist Theory of Chinese Thought: A Philosophical Interpretation. Oxford:
Oxford University Press, 1992.
Henderson, John B. Scripture, Canon, and Commentary: A Comparison of Confucian
and Western Exegesis. Princeton, NJ: Princeton University Press, 1991.
He Yan Lunyu jijie (Collected Explanations of the Analects). Xing
Bing . Lunyu zhushu . Sibu beiyao edition.
Hsu, Cho-Yun. Ancient China in Transition: An Analysis of Social Mobility, 722222
B.C. Stanford, CA: Stanford University Press, 1965.

Bibliography

311

Hursthouse, Rosiland. On Virtue Ethics. Oxford: Oxford University Press, 1999.

Huang, Chichung. The Analects of Confucius: A Literal Translation. New York: Oxford
University Press, 1997.
Iggers, Georg G. Historicism. In Dictionary of the History of Ideas, edited by
Philip P. Wiener, vol. II (Depotism to Law, Common), 45664. New York:
Charles Scribners Sons, 1973.
Ivanhoe, Philip. J. Confucian Moral Self-Cultivation. The Rockwell Lecture Series,
vol. 3. Indianapolis: Hackett, 2000.
. The Daodejing of Laozi. New York: Seven Bridges Press, 2001.
. Death and Dying in the Analects. In Mortality in Traditional Chinese
Thought, edited by Amy Olberding and Philip J. Ivanhoe. Albany: State
University of New York Press, 2011.
. Ethics in the Confucian Tradition: The Thought of Mengzi and Wang Yangming.
Atlanta: Scholars Press, 1990. Indianapolis, IN: Hackett, 2002. References
are to the 2002 edition.
. The Golden Rule in the Analects. In Confucius Now: Contemporary
Encounters with the Analects, edited by David Edward Jones. Chicago: Open
Court, 2008: 81107.
. Review: Thinking Through Confucius by David L. Hall and Roger T.
Ames. Philosophy East and West 41, no. 2 (1991): 24154.
. Reweaving the One Thread of the Analects. Philosophy East and West
40, no. 1 (1990): 1733.
. Whose Confucius? Which Analects? In Confucius and the Analects: New
Essays, edited by Bryan W. Van Norden. New York: Oxford University Press,
2002: 11933.
Jiang Yihua , Huang Junlang . . Taibei: ,
2007.
Johnston, Paul. The Contradictions of Modern Moral Philosophy: Ethics after
Wittgenstein. Routledge Studies in Ethics and Moral Theory. London:
Routledge, 1999.
. Wittgenstein and Moral Philosophy. London: Routledge, 1989.
Kermode, Frank. The Classic. London: Faber & Faber, 1975.
Kieschnick, John. Analects 12.1 and the Commentarial Tradition. The Journal of
the American Oriental Society 112, no. 4 (1992): 567576.
Klagge, James C. An Unexplored Concept in Wittgenstein. History of Philosophy
Quarterly 12, no. 4 (1995): 469486.
Klein, Peter D. Human Knowledge and the Infinite Regress of Reasons. Nos 33
(1999): 297325.
LaFargue, Michael. Tao and Method: A Reasoned Approach to the Tao Te Ching.
Albany: State University of New York Press, 1994.
Lau, D. C. The Analects (new bilingual edition). Hong Kong: Chinese University
Press, 2000.
Legge, James. Four Books: The Great Learning, the Doctrine of the Mean, Confucian
Analects, and the Works of Mencius. New York: Paragon Books, 1966.

312

Bibliography

. Li Ki. Sacred Books of the East, vol. 28, part 4. Oxford, UK: Clarendon
Press, 1885.
. The She King (Book of Poetry). The Chinese Classics: With a Translation,
Critical and Exegetical Notes, Prolegomena, and Copious Indexes. Volume 4.
London: Trubner, 1871.
. The Shoo King (Book of Historical Documents). The Chinese Classics: With
a Translation, Critical and Exegetical Notes, Prolegomena, and Copious Indexes.
Volume 3. London: Trubner, 1865.
MacIntyre, Alasdair. After Virtue: A Study in Moral Theory. South Bend, IN:
University of Notre Dame Press, 1984.
. Incommensurability, Truth, and the Conversation Between Confucians
and Aristotelians about the Virtues. In Culture and Modernity: East-West
Philosophic Perspectives, 104123, edited by Eliot Deutsch. Honolulu:
University of Hawaii Press, 1991.
Makeham, John. Between Chen and Cai: Zhuangzi and the Analects. In Wandering
at Ease in the Zhuangzi, edited by Roger T. Ames, 75100. Albany: State
University of New York Press, 1998.
. The Earliest Extant Commentary on Lunyu: Lunyu Zheng Shi Zhu.
Toung Pao 83 (1977): 260299.
. Transmitters and Creators: Chinese Commentators and Commentaries on the
Analects. Harvard East Asian Monographs, 228. Cambridge, MA: Harvard
University Asia Center, 2003.
Mulhall, Stephen. Ethics in Light of Wittgenstein. Philosophical Papers 31, no. 3
(2002): 293321.
. Stanley Cavell: Philosophys Recounting of the Ordinary. Oxford: Oxford
University Press, 1994.
Munro, Donald J. The Concept of Man in Early China. Stanford, CA: Stanford
University Press, 1979.
Nails, Debra. Socrates. In The Stanford Encyclopedia of Philosophy, edited by Edward
N. Zalta, 2007. Available at <http://plato.stanford.edu/entries/socrates/>.
Nehamas, Alexander. Nietzsche, Life as Literature. Cambridge, MA: Harvard
University Press, 1985.
Nivison, David. Golden Rule Arguments in Chinese Moral Philosophy. In The
Ways of Confucianism, 5976, edited by Bryan W. Van Norden. Chicago, IL:
Open Court, 1996.
Nylan, Michael. The Five Confucian Classics. New Haven, CT: Yale University
Press, 2001.
ONeill, Onora. The Power of Example. Philosophy 61, (1986): 529.
Peterman, James F. Philosophy as Therapy: An Interpretation and Defense of Wittgensteins
Later Philosophical Project. SUNY Series in Philosophy and Psychotherapy.
Albany: State University of New York Press, 1992.
Phillips, D. Z. Interventions in Ethics. London: Blackwell, 1992.
Phillips, D. Z., and H. O. Mounce. Moral Practices. New York: Schocken Books,
1970.

Bibliography

313

Pleasants, Nigel. Wittgenstein, Ethics and Basic Moral Certainty. Inquiry 51, no.
3 (2008): 241267.
Rachels, James. The Elements of Moral Philosophy. New York: Random House, 1986.
2nd ed. New York: McGraw-Hill, 1993. Page references are to the 1993
edition.
Rhees, Rush. Some Developments in Wittgensteins View of Ethics. The
Philosophical Review, 74, no. 1 (1965): 1726.
Richey, Jeffrey L. Ascetics and Aesthetics in the Analects. Numen (2000): 16174.
Ricoeur, Paul. The Conflict of Interpretations. Northwestern University Studies in
Phenomenology and Existential Philosophy. Evanston, IL: Northwestern
University Press, 1974.
Sato, Masayuki. The Confucian Quest for Order: The Origin and Formation of
the Political Thought of Xun Zi. Sinica Leidensia, vol. 58, Leiden, NL:
Brill, 2003.
Sayre-McCord, Geoff. Metaethics. In The Stanford Encyclopedia of Philosophy,
edited by Edward N. Zalta. Stanford, CA: Metaphysics Research Lab, CSLI,
Stanford University, Fall 2008 edition. Available at <http://plato.stanford.edu/
archives/fall2008/entries/metaethics/>.
Schlesinger, Arthur M. Learning How to Behave: A Historical Study of American
Etiquette Books. New York: The Macmillan Company, 1946. New York:
Cooper Square Publishers, 1968. Page references are to the 1968 edition.
Schwartz, Benjamin I. The World of Thought in Ancient China. Cambridge, MA:
Belknap Press, 1985.
Seligman, Adam B., with Robert P. Weller, Michael J. Puett, and Bennett Simon.
Ritual and Its Consequences: An Essay on the Limits of Sincerity. Oxford: Oxford
University Press, 2008.
Shadick, Harold. Review of Confucius by H. G. Creel. The Philosophical Review
(1951): 11317.
Shun, Kwong-loi. Jen and Li in the Analects. Philosophy East and West 43, no. 3
(1993): 45779.
Slingerland, Edward G. Confucius Analects: With Selections from Traditional
Commentaries. Indianapolis, IN: Hackett, 2003.
. Virtue Ethics, the Analects, and the Problem of Commensurability.
Journal of Religious Ethics 29, no. 1 (2001): 97125.
. Why Philosophy Is Not Extra in Understanding the Analects. Philosophy
East and West 50, no. 1 (2000): 137141.
Stroll, Avrum. Moore and Wittgenstein on Certainty. New York: Oxford University
Press, 1994.
Tully, James. Meaning and Context: Quentin Skinner and His Critics. Cambridge,
UK: Polity Press, 1988.
Unger, Peter. Contextual Analysis in Ethics. Philosophy and Phenomenological
Research 55 no. 1 (1995): 126.
Van Norden, Bryan W. Confucius and the Analects: New Essays. Oxford: Oxford
University Press, 2002.

314

Bibliography

. Virtue Ethics and Consequentialism in Early Chinese Philosophy. New York:

Cambridge University Press, 2007.
Waley, Arthur. The Analects of Confucius. New York: Vintage Books, 1989.
Williams, Meredith. Wittgenstein, Mind, and Meaning: Toward a Social Conception
of Mind. London: Routledge, 1999.
. Wittgensteins Philosophical Investigations: Critical Essays on the Classics.
Lanham, MD: Rowman & Littlefield, 2007.
Wittgenstein, Ludwig. The Blue and Brown Books: Preliminary Studies for the
Philosophical Investigations. Oxford, UK: Basil Blackwell, 1958.
. On Certainty. Edited by Gertrude Elizabeth Margaret Anscombe and Georg
Henrik von Wright. Translated by Denis Paul and G. E. M. Anscombe.
London: Basil Blackwell, 1969. New York: Harper and Row, 1972. Page
citations are to the 1969 edition.
. Remarks on the Foundations of Mathematics. Edited by G. H. von Wright,
Rush Rhees, and G. E. M Anscombe. Translated by G. E. M. Anscombe.
Oxford, UK: Blackwell, 1956. Revised edition, Cambridge, MA: The MIT
Press, 1978. Page citations are to the1978 edition.
. Remarks on Frazers Golden Bough. Translated by A. C. Miles and Rush
Rhees, The Human World 3 (1971): 37. Also in Philosophical Occasions, 1912
1951. Edited by James C. Klagge and Alfred Nordmann. Indianapolis, IN:
Hackett, 1993.
. Remarks on the Philosophy of Psychology. Edited by G. E. M. Anscombe
and G. H. von Wright. Translated by G. E. M. Anscombe. Oxford: Basil
Blackwell, 1980.
. Tractatus Logico-Philosophicus. In Annalen der Naturphilosophie, 1921. First
English edition, Kegan Paul, Trench and Trbner, 1922. Revised edition,
London: Routledge & Kegan Paul, 1961. Page citations are to the 1961
edition.
. Wittgensteins Lectures, Cambridge, 19301932: From the Notes of John King
and Desmond Lee. Totowa, NJ: Rowman & Littlefield, 1980.
. Wittgensteins Lectures on the Foundations of Mathematics. Cambridge, 1939:
From the Notes of R. G. Bosanquet, Norman Malcolm, Rush Rhees, and Yorick
Smythies. Copyright 1975 and edited by Cora Diamond, 1976. Chicago:
University of Chicago Press edition, 1989. Page citations are to the 1989
edition.
. Philosophische Untersuchungen (Philosophical Investigations). Translated by
G. E. M. Anscombe, P. M. S. Hacker, and Joachim Schulte. Oxford, UK:
Basil Blackwell, Ltd., 1953. Revised 4th ed. by P. M. S. Hacker and Joachim
Schulte. Chichester, UK: Blackwell/John Wiley & Sons, 2009. Page and section references are to the 2009 edition.
Yang Xiao. How Confucius Does Things with Words: Two Hermeneutic Paradigms
in the Analects and Its Exegeses. The Journal of Asian Studies 66, no. 2
(2007): 497532.

Index

indicates sections in the Analects.

acknowledgment, of given features
of language and teaching, 9, 153,
210, 239, 263, 270; basic need for,
25155
agreement, 5, 23; in/with forms of life,
5, 29, 205; in some cases complete,
23, 24, 28, 32; limits to complete
agreement, 23, 27, 44, 4849, 99;
peaceful agreement, 28, 29, 30, 36,
153; need for in morality, 61, 65, 67,
99; relation to intelligibility, 96; relation to truth, 28. See also harmony
Ames, Roger. See Hall, David and
Ames, Roger
Analects:
1.2, 14, 74, 105, 106, 11020;
2.11, 216; 2.15, 16, 111, 172;
2.16; 2.20, 235, 2.1, 235; 2.4,
18, 25; 3.17, 73; 3.26, 236; 4.6,
74; 5.13, 126, 173; 5.6, 235;
6.30, 237; 7.22, i. 7.34, 177;
8.2, 112; 8.10, 195; 14.4, 195;
17.21, 196; 9.1, 126, 175; 9.6,
269; 9.11, 25, 26, 171; 9.30,
43; 10.16, 269; 11.12, 12627;
11.16, 164; 11.22, 155, 163; 12.1
15, 167, 168, 180, 248, 249; 12.2,
23940; 12.5, 236; 13.3, 138;
13.27, 135; 17.18, 269; 19.12,
17; MacIntyre on, 19596

Bellah, Robert Neelly, 255

bedrock practices, 5, 611; Confuciuss
use of, 24; constitutive of beliefs
and action, 6; learning of, 1122;
relation to imponderable evidence,
2223; not themselves justified, 7,
1011, 21
Cavell, Stanley, 40, 48, 64, 99
ceremonial ritual chains, 23944. See
also ritual
Ci, Jiwei, 18589
Confucius. See Analects. See also Creel,
Herrlee, G.
Creel, Herrlee G., 71, 8386
crisscross method of teaching, 35, 160,
290
Cua, Anthony, 277n23
Daimond, Cora, 8, 4552, 56, 58,
113, 282n40.
dao: 45; constituents of, 59, 26,
3336, 153, 163, 247, 277n23;
invoking, 138, 16366, 219, 248;
not defined, 55. See also norms
Davidson, Donald, 9596, 101104,
107, 11114
definition: Wittgenstein on, 11, 16,
29, 3233, 66, 161; Confuciuss
avoidance of, 54, 109, 13233

315

316

Index

depth of reflective problems: in Confucius, 9, 168, 173, 17577, 184;

in Fingarette, 222, 230, in Wittgenstein, 17780
Dewey, John, 7
Eno, Robert, 7
essences, Wittgensteins indifference
to, 61, 139; Confucian indifference
to, 12223, 127, 17374; confusion with ideals, 221, 249. See also
essentialism, foundations
essentialism, 18687. See also family
resemblance
ethical interventions: in Wang Yangming, 1; in Analects, 121, 127, 158,
164, 17477, 200, 234; Phillips, D.
Z. on, 174; philosophical, in Wittgenstein, 17980; defense of, 183,
187; and truth, 219
evidence, imponderable, 22; relation
to expert judgment, 23, role in
Confuciuss projects, 2, 4445, 178,
179, 190
examples: teaching novice with, 9, 21,
27, 3334, 171; use of, as exemplars, 55, 67; use of, in conceptual
clarification, 46, 6366, 108, 220.
See also learning
family resemblance, 35, 37, 89, 188,
246. See also essentialism; essence
Fei, Hsiao-Tung, 188
Fingarette, Herbert, 21949 passim; on
handshaking, 22122; on Confuciuss account of ritual, 234; criticisms of, 23349; See also Goffman;
handshaking
form(s) of life: acknowledgment of,
206, 25154, 27074; agreement
and harmony in, 2731; as basis of
translation, 96, 106107, 114, 119;
caring about and appreciating, 57,
64; complicated character of, 4748

foundational theories, 37, 21, 3940,

67, 137, 150; appeal to, by Confucian tradition, 180, 189, 206: of no
use to Confucius, 85, 136, 17577,
199200, 207209, 252, 273; rejection of, by realistic spirit, 4549, 54
frameworks. See normative frameworks.
Gadamer, Hans Georg, 78
Gardner, Daniel K., 6970, 7678,
91, 95, 100101, 11819, 167, 183
Goethe, Faust, 21
Goffman, Erving: account of interaction ritual order, 25659;
influenced by anthropology of
Chinese ritual, 254; on deference
and demeanor in rituals, 26061
goodness. See ren
Graham, A.C., 7
Grandy, Richard, 96, 105, 106
Hall, David and Ames, Roger, 7, 128,
14754, 193
handshaking: as a learned practice, 22326; authenticity of, not
dependent on spontaneity, 22628,
23132; meaning of, dependent on
context, 22628; meaning of, various, 22830
Hansen, Chad, 7, 128, 13649, 154,
155, 163
He Yan, 168, 18082, 184
hidden, nothing is, 3236, 39, 130,
173, 220, 295, 305
historical evidence: Brookess account
of the Analects history, 8183; chain
of evidence problem for Analects
histories, 8283, 86; not needed for
speakers meaning, 8791. See also
Creel, Herrlee G.
interpretation, required only when
speakers meaning is not clear,
11617

Index

interpretive charity, principle of, 2, 89;

Davidson on, 96, 101107; import
for Analects interpretation, 11520
interventions: defense of, 183, 187;
in Analects, 121, 127, 158, 164,
17477, 200, 234; in Wittgenstein,
17980; in Wang Yangming, 1;
Phillips, D. Z. on, 174; relation of,
to truth, 219
invoking ideals, 138, 159, 16366,
174, 219
Ivanhoe, P. J. 14749, 239, 279n83,
281n34
justification without theory, 21013.
See also foundational theories
Kermode, Frank, 76
Klagge, James, 36, 282n40
language-games: Wittgensteins appeal
to, 34, 6, 1617, 23, 28, 32,
4849, 161, 179; Hansen on, 140
42; interpretive charity in, 10620;
truth in, 12122, 134, 149, 204;
use of, to counter semantic skepticism and nihilism, 9699
Lau, D.C., 74, 82, 239
law of excluded middle, limits of, 166,
198, 201, 20613
learning. xue (), 511, 17273,
293n6, 294n7; adult versus child,
56; as constitutive of concepts, 11,
14, 23, 37, 46; as mastering bedrock
practices, 1122, 24, 153; devotion
to (haoxue, ), 5, 15, 17677,
171, 294n13; from hints, tips, etc.,
10, 15, 22, 26, 27, 34, 153, 262,
272; See also bedrock practices;
crisscross method; handshaking;
reflection
li (, ritual). See ritual
Liji () The Record of Rituals,
1820

317

MacIntyre, Aladasdair: Aristotelian

criticism of Confucianism, 185213;
on the difference between truth and
warranted assertability, 213; on truth
claims, 19299
Makeham, John, on indeterminacy of
meaning, 72, 75, 76; on historicalscriptural, semantic dualism, 6971,
118
master-novice, in Confucius, 5, 7, 9,
1617, 2426, 35, 44, 208, 211;
in Wittgenstein, 917, 2021, 29,
3235, 44, 6465, 125, 211, 194;
meaning: as use, 9799, 115, 139,
142, 201, 206; dependent on
context, 74; 202; historical, 6971,
9193, 97, 100, 118; indeterminacy of, 6976, 77, 78, 93; of
ideals, 21416; not exhausted by
behavioral output, 135; of interaction rituals, 256, 264; of shaking
hands, 22333; relation to formulae,
3233; relation to practice, 9, 16,
21, 29, 113; relation to interpretive charity, 8791, 99101, 118;
scriptural (or normative), 6970,
9192; of speakers utterance, 78; of
speakers utterance, not knowable,
7891. See also interpretive charity;
language-games; semantic nihilism;
semantic skepticism
meta-ethics, 3, 4951, 61
Mulhall, Stephen, 6267
Munro, Donald J., 12837; on
Chinese view of mind, 13135; on
truth, 188204
mutuality, ideal of, 23945; relation to
zhong () and shu (), 23943
Nehemas, Alexander, 69
Nivison, David, 7, 239, 243, 302n29
nonsense, 8, 124, 125, 184, 201203,
211; contrasted with lack of salience,
124, 125. See also salience

318

Index

norm(s), 410, 25, 26, shared, 3032,

16063; constitutive view of 36,
273; Confucian commitment to,
4145, 15253; central to teaching, 6567, expressed in descriptive language, 126; expressible in
different ways, 15858; governing
ritual, 25670; as constituting dao,
31, 3754, 111, 138, 158, 16063,
230; of language, 99, 144, 161,
162, 164; sensitivity to, 8, 35,
43, 122, 146, 267. See also dao;
learning; normative disagreements;
normative frameworks
normative disagreements, 42, 45, 114
normative frameworks, 3031, 188,
255, 273
normative ethics, 3, 49, 130
Nylan, Michael, 8687
New Wittgensteinians, 4, 6268
ONeill, Onora, 4758
Ontology, 151, 192, 204205, 207,
209, 21415
parochialism, problem of, 185, 188,
21315
Phillips, D. Z., 174
philosophical therapy, 1, 52, 59, 143,
206
Platonism, 12836
Pleasants, Nigel, 4, 284n58
puzzlement (conceptual, ethical,
philosophical): in the Analects, 38,
133, 15560; 162, 199, 296n31;
in Wittgenstein, 3, 16, 36, 4446,
5657, 17980, 205, 253, 28n33;
produced by Confucian commentaries, 18384
Quine, W. V. O., 104108
realism, 79, 45, 56, 113, 283n50. See
also realistic spirit

realistic spirit: as spirit, 5658,

282n40; in Confucius, 5256; in
Wittgenstein, 8, 9, 40, 4552;
Kantian critique of, 5262
reflection si (), 7, 133, 172; consistent with truth claims, 147, 15253,
163, 188, 204, 20815; contrasted
with philosophical theorizing, 79,
3940, 5155, 13748, 175, 244,
247, 27274; in Song Confucianism, 33; limited by learning, 916,
4346, 127, 131, 159, 164, 17073,
20612; Munro on, 137; used to
resolve disagreement, 59, 11213.
See also Analects 1.2; Analects 2.15;
puzzlement
ren (, goodness, humaneness): 8, 14,
15, 21, 26, 33, 35, 36, 106, 201; in
(weiren), 7475, 105, 11520,
18081; logic of discussions of,
5254, 88, 13338, 14647; relation to ritual (li), 89, 107, 11213;
speech acts, related to, 15465; 234.
See also Fingarette
Ricoeur, Paul, 79, 113
ritual li (): Confucius on, 1516,
25, 31, 52, 54, 11213, 138, 145,
159, 16772, 180, Goffman on,
25170; MacIntyres worry about,
191, 21315; need for acknowledgement of, 25170. See also Fingarette;
handshaking.
Rosemont, Henry, Jr., 298n33
salience, 124, 125, 185, 198, 215
Sato, Masayuki, 300n7
Schwarz, Benjamin J., 7
self-cultivation, 45; Confucius on, 79,
15, 3638, 44, 5255, 173; salience,
relation to, 198, 200, 20714; truth,
relation to, 18788; Wittgenstein on,
5, 9, 3738, 52, 11113
semantic nihilism, 6970, 7677, 91
semantic skepticism, 6970, 72, 7881

Index

Shun, Kwong-loi, 169, 237

silence, 127, 130. See also hidden,
nothing is
Sims, May, 297n10
Slingerland, Edward, 7, 76, 284n5
social hierarchy, as expressed in zhong
and shu, 239, 245
shu (, consideration), translations of,
237. See also mutuality
truth, in the Analects, 121, 21213;
concept of, distinguished from
theory of, 149, 152; ordinary interest in, 122, 128, 129, 134, 14748,
156; theory of, 102, 134, 144, 147,
148, 15152
Van Norde, Bryan, 284n6
Wang Yangming, 1
Whitehead, Alfred North, 7
Williams, Meredith, I, 1112, 1314

319

Wittgenstein, Ludwig, Blue and Brown

Books, 98; Lectures, Cambridge,
19301932, 179; On Certainty, 1,
10, 28, 125, 128, 15253, 160,
162, 251, 278, 292n48; Philosophical Investigations, 6, 10, 39, 58,
6466, 95, 9798, 107, 13942,
143, 167, 178, 183, 200202,
205, 282n41; Remarks on Fraziers Golden Bough, 272, 282n40,
287n1; Remarks on the Foundations
of Mathematics, 12, 29, 32, 213;
Rush Rhees, Some Developments,
39, 5961
Xiao, Yang, 15458, 16365
Yan Hui, 15, 24, 25, 26, 36, 168,
169, 17073, 176, 248
Youzi (Master You), 5, 9, 14, 17
Zhu Xi, 75, 154, 168, 18084