Davidson's Philosophy and Chinese Philosophy: Constructive Engagement (Philosophy of History and Culture)

DAVIDSON’S PHILOSOPHY AND CHINESE PHILOSOPHY

PHILOSOPHY OF HISTORY AND CULTURE Series Editor

Michael Krausz, Bryn Mawr College Advisory Board

Annette Baier (University of Pittsburgh), Cora Diamond (University of Virginia), William Dray (University of Ottawa), Nancy Fraser (Northwestern University), Clifford Geertz (Institute for Advanced Study, Princeton), Peter Hacker (St. John’s College, Oxford), Rom Harré (Linacre College, Oxford), Bernard Harrison (University of Utah), Martha Nussbaum (University of Chicago), Leon Pompa (University of Birmingham), Joseph Raz (Balliol College, Oxford), Amélie Oksenberg Rorty (Brandeis University), Georg Henrik Von Wright (University of Helsinki) VOLUME 23

DAVIDSON’S PHILOSOPHY AND CHINESE PHILOSOPHY Constructive Engagement EDITED BY

BO MOU

BRILL LEIDEN • BOSTON 2006

This book is printed on acid-free paper.

Library of Congress Cataloging-in-Publication Data Davidson’s philosophy and Chinese philosophy : constructive engagement / edited by Bo Mou. p. cm. — (Philosophy of history and culture ISSN 0922-6001 ; 23) Includes bibliographical references and index. ISBN 90-04-15048-X (alk. paper) 1. Philosophy, Chinese. 2. Davidson, Donald, 1917- I. Mou, Bo, 1956- II. Series. B126.D34 2006 191—dc22 2005058174

ISSN 0922–6001 ISBN-13: 978-90-04-15048-5 ISBN-10: 90-04-15048-X © Copyright 2006 by Koninklijke Brill NV, Leiden, The Netherlands Koninklijke Brill NV incorporates the imprints Brill Academic Publishers, Martinus Nijhoff Publishers and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Brill provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910 Danvers MA 01923, USA. Fees are subject to change. PRINTED IN THE NETHERLANDS

In memory of Donald Davidson (1917–2003), who inspired and participated in the project

CONTENTS

Acknowledgments ........................................................................ Note on Transcription ................................................................ Contributors ................................................................................

xi xv xvii

How Constructive Engagement of Davidson’s Philosophy and Chinese Philosophy is Possible: A Theme Introduction .......................................................... Bo Mou

1

PART ONE

CONCEPTUAL SCHEMES, RELATIVISM, AND CROSS-CULTURAL UNDERSTANDING Chapter One Relativism and Its Schemes ............................ Michael Krausz Chapter Two Davidson and Chinese Conceptual Scheme .................................................................................... Koji Tanaka Chapter Three Making Room for Comparative Philosophy: Davidson, Brandom, and Conceptual Distance .................................................................................. Stephen C. Angle

37

55

73

PART TWO

PRINCIPLE OF CHARITY AND CHINESE PHILOSOPHY Chapter Four Where Charity Begins .................................... 103 David B. Wong Chapter Five Davidson’s Charity in the Context of Chinese Philosophy ................................................................ 117 Yiu-ming Fung

viii

contents PART THREE

RATIONALITY, NORMATIVITY, AND INTER-CULTURAL DISAGREEMENT Chapter Six Davidsonian Rationality and Ethical Disagreement between Cultures ............................................ 165 Samuel C. Wheeler Chapter Seven A Davidsonian Approach to Normativity and the Limits of Cross-Cultural Interpretation .................. 189 Yujian Zheng

PART FOUR

MEANING AND INTERPRETATION Chapter Eight On Two Kinds of Meaning and Interpretation .......................................................................... 207 A.P. Martinich Chapter Nine Metaphorical Use versus Metaphorical Essence: Examples from Chinese Philosophy ...................... 229 Kim-chong Chong Chapter Ten Reading the Analects with Davidson: Mood, Force, and Communicative Practice in Early China .......... 247 Yang Xiao

PART FIVE

TRUTH CONCERN AND DAO CONCERN Chapter Eleven From Donald Davidson’s Use of “Convention T” to Meaning and Truth in Chinese Language ................................................................................ 271 Chung-ying Cheng

contents

ix

Chapter Twelve Truth Pursuit and Dao Pursuit: From Davidson’s Approach to Classical Daoist Approach in View of the Thesis of Truth as Strategic Normative Goal ........................................................................................ 309 Bo Mou Index ............................................................................................ 351

ACKNOWLEDGMENTS

My deep appreciation goes to late Professor Donald Davidson whose thought and style of doing philosophy, and whose valuable participation at the earlier stages of this anthology project, as explained in my theme introduction below, have significantly inspired and contributed to the project. I am very grateful to all the other contributing authors of this volume for their valuable contributions, all of which are previously unpublished pieces written expressly for this book, and for their patience, cooperation, and understanding throughout the process, during which I have learnt a lot from them in various aspects. Their persistent support of this project and of this editor’s efforts have become especially valuable when Donald Davidson passed away at one important stage of this project, as Davidson’s participation in this project in several ways is one of the main momentums for our reflective efforts in this constructive-engagement project. I am especially indebted to Michael Krausz not merely for his role as an active contributing participant but also for his timely and effective help in determining a decent academic publisher like Brill. To effectively fulfill the constructive-engagement purpose, this anthology project is accompanied with its conference project to provide a critical discussion and engagement platform. In this way, although the anthology project per se is an independent project instead of the conference proceedings, the latter is rather one indispensable stage for the sake of fulfilling the goal of this anthology and for the sake of effectively implementing the constructive-engagement strategy. During the whole process of preparing for the anthology project including its closely related conference project as one crucial stage of critical engagement platform, we have received a large amount of support, help and assistance in various ways from various parties. During the process of reviewing the submissions, I am grateful to Wan-Chuan Fang, Yiu-ming Fung and Linhe Han for their valuable review work and their precious time. I am grateful to Xianglong Zhang, my colleague in the 2002–2005 board of the International Society for Comparative Studies of Chinese and Western Philosophy (ISCWP), for his persistent support since the conference

xii

acknowledgments

project became the first one in the ISCWP “constructive engagement” international conference series. I am thankful to the Institute of Foreign Philosophy, Peking University, China, for its assuming the conference host for the originally-scheduled August-2003 conference, which had to be postponed due to the SARS outreach in spring 2003; I am especially grateful to Linhe Han for his active role as the conference-host representative in coordinating various preparations. I am grateful to the Institute of Philosophy, Chinese Academy of Social Sciences for its assuming the conference host for the re-scheduled June-2004 conference; I am thankful to Pengcheng Li, Deputy Director of the Institute of Philosophy, CASS, and Jing Sun, Director of its Research Coordination Office, for their substantial support; I am especially grateful to He Li for his active role as the conference-host representative in coordinating various preparations. I am grateful to those speakers other than the contributors to this volume, Bo Cheng, Wan-Chuan Fang, Yi Jiang and Chuang Ye for their valuable and engaging talks at the conference. My sincere thanks also go to Lian Cheng, He Li, Jian Li, and Xiwen Luo for their helpful and effective professional service as the conference session chairs, and to Jigang Shan, Jihong Lei and Xiaojian Zhang for their effective logistics supports for the conference. I am indebted to the American Philosophical Association’s Committee on International Cooperation (CIC), under the leadership of its chair Alan M. Olson, for its valuable support and cosponsorship for the above mentioned international conference project on Davidson’s philosophy and Chinese philosophy during 2002–2004 when I served as a member of the CIC. I am grateful to Roger Ames, Editor of the journal Philosophy East and West, and He Li, Editor of the Chinese journal World Philosophy, for their valuable support and help in setting precious space in their journals for publishing the “call for papers” and/or news of some ISCWP academic activities including this project. I am also thankful to Christ Caputo at the American Philosophical Association for providing the space at the APA website to post the “call for papers” for the project. I am grateful to my school, San Jose State University, and its Department of Philosophy for their various substantial supports that are related to this anthology project. A California State University Research Grant for 2003–2004 has significantly contributed to my work on this anthology. I am thankful to Phillip Willamson, who

acknowledgments

xiii

was my graduate-student assistant in spring 2005, for his professional assistance that he completed timely. My sabbatical leave in fall 2005 has enabled me to efficiently finish the final phase of the whole project. I am grateful to our editors at Brill, Marcella Mulder at the early stage and Boris van Gool and Birgitta Poelmans at the later stage, for their variety of kindly and timely professional assistance. Bo Mou Albany, California December 15, 2005

NOTE ON TRANSCRIPTION

Because of its official status in China, its relative accuracy in transcribing actual pronunciation in Chinese common speech and consequent world-wide use, we employ the pinyin romanization system in this volume for transliterating Chinese names or terms. However, those Chinese names or terms are left in their original romanizations (typically in the Wade-Giles system) in the following cases: (i) the titles of cited publications; (ii) the names whose romanizations have become conventional (such as ‘Confucius’); and (iii) the names of the writers who have had their authored English publications under their regular non-pinyin romanized names (such as ‘Fung Yulan’). The title of a cited contemporary Chinese book and essay is given in its pinyin transcription with its translation or paraphrase given in parentheses. The following rule of thumb has been used in dealing with the order of the surname (i.e., family name) and given name in romanized Chinese names: (i) for the name of a historical figure in Chinese history, the surname appears first, and the given name second (such as ‘Zhu Xi’); and (ii) for contemporary figures, we follow their own practice in this aspect when they publish in English or other Western languages (typically, the given name appears first, and the surname second). In the pinyin versions of Chinese publication titles and those proper phrases that contain two or more than two Chinese characters, hyphens may be used to indicate separate characters.

Transcription Conversion Table Wade-Giles

Pinyin

ai ch ch’

ei zh ch q x ian

hs ien

note on transcription

xvi Table (cont.)

Wade-Giles

Pinyin

-ih j k k’ p p’ szu t t’ ts, tz ts’, tz’ tzu ung yu

-i r g k b p si d t z c zi ong you

CONTRIBUTORS

Angle, Stephen C. is Director of the Mansfield Freeman Center for East Asian Studies and Associate Professor of Philosophy at Wesleyan University, USA. He received his B.A. from Yale in East Asian Studies and his Ph.D. in Philosophy from the University of Michigan. He is the author of Human Rights and Chinese Thought: A Cross-Cultural Inquiry (Cambridge U.P., 2002) and the co-editor and co-translator of The Chinese Human Rights Reader (M.E. Sharpe, 2001). Angle studies Chinese ethical and political thought from the Song dynasty though the present, and is also interested in issues in the methodology of comparative philosophy. Cheng, Chung-ying is Professor of Philosophy at University of Hawaii at Manoa, USA. He received his Ph.D. degree from Harvard University (1964). Cheng is the founder and honorary President of the International Society for Chinese Philosophy and International Society for the Yijing; he is the Editor-in-Chief of Journal of Chinese Philosophy. He is the author of many articles and books on Chinese philosophy and comparative philosophy, including Peirce’s and Lewis’s Theories of Induction (1969), Modernization and Universalization of Chinese Culture (1988, in Chinese), New Dimensions of Confucian and New-Confucian Philosophy (1991). He is co-editor of Contemporary Chinese Philosophy (2002). Chong, Kim-chong is Professor of Humanities at the Hong Kong University of Science and Technology. He was formerly with the National University of Singapore, where he served as Head of Department for several years. His interests are in ethics, Chinese philosophy, and comparative philosophy. His publications include Moral Agoraphobia: The Challenge of Egoism (Peter Lang, 1996); The Moral Circle and the Self: Chinese and Western Approaches (co-edited, Open Court, 2003); and Early Confucian Ethics (Open Court: forthcoming). Fung, Yiu-ming is Chair Professor of the Division of Humanities at the Hong Kong University of Science and Technology. He received his Ph.D. degree in philosophy from the Chinese University of Hong Kong in 1984. Fung is the author of several books, including The Methodological Problems of Chinese Philosophy (1989); Chinese Philosophy in the Ancient Period, 4 volumes (1992); Kung-Sun Lung Tzu: A Perspective

xviii

contributors

of Analytic Philosophy (1999); and The Myth of Transcendent Immanence: A Perspective of Analytic Philosophy on Contemporary Neo-Confucianism (2003). He has also published more than 80 research papers both in Chinese and in English. Kransz, Michael is the Milton C. Nahm Professor of the Department of Philosophy at Bryn Mawr College, USA. Krausz is the author of Limits of Rightness (2000), Rightness and Reasons: Interpretation in Cultural Practices (1993), and Varieties of Relativism (with Rom Harré) (1995). He is contributing editor of Relativism: Interpretation and Confrontation (1989), The Interpretation of Music: Philosophical Essays (1993) and Is There a Single Right Interpretation? (2002). Krausz is contributing co-editor of The Concept of Creativity in Science and Art (1981), Relativism: Cognitive and Moral (1984), Rationality, Relativism and the Human Sciences (1986) and Interpretation, Relativism and the Metaphysics of Culture (1999). In 2003 a festschrift on his work was published by Rodopi: Interpretation and Its Objects: Studies in the Philosophy of Michael Krausz. Martinich, Aloysius P. is Roy Allison Vaughan Centennial Professor of Philosophy and Professor of History and Government at the University of Texas at Austin, USA. He received his Ph.D. from the University of California at San Diego. He is the author or editor of many books and articles. His books include Philosophical Writing 3rd edition (Blackwell, 2005), The Philosophy of Language 4th edition (Oxford University Press, 2001), Hobbes: A Biography (Cambridge University Press, 1999), The Two Gods of Leviathan (Cambridge: Cambridge University Press, 1992) and Communication and Reference (Berlin and New York: Walter de Gruyter, 1984). Mou, Bo is Associate Professor of Philosophy at San Jose State University, USA. After receiving B.S. in mathematics, he received M.A. from Graduate School, Chinese Academy of Social Sciences, and Ph.D. from University of Rochester, USA. Representative publications include “A Metaphilosophical Analysis of the Core Idea of Deflationism”, Metaphilosophy (2000), “The Enumerative Character of Tarski’s Definition of Truth and Its General Character in a Tarskian System”, Synthese (2001), Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions (contributing editor, 2001), and “A Re-examination of the Structure and Content of Confucius’s Version of the Golden Rule”, Philosophy East and West (2004). Tanaka, Koji is a Postdoctoral Fellow in the Philosophy Department at Macquarie University, Australia. From 2006, he will be a Lecturer

contributors

xix

in the Philosophy Department at the University of Auckland, New Zealand. He served as Reviews Editor for Studia Logica, an international journal for symbolic logic. He has published widely in leading journals and made contributions to logic, philosophy of logic, Buddhist philosophy as well as Chinese philosophy. Wheeler III, Samuel C. is Professor of Philosophy at the University of Connecticut, USA. He received a B.A. from Carleton College and a Ph.D. in Philosophy from Princeton in 1970. He has published articles on vagueness, logical form, philosophy of language, ethics, political philosophy, Plato, metaphysics, and deconstruction. He is the editor of Public Affairs Quarterly. His book, Deconstruction as Analytic Philosophy was published by Stanford University Press in 2000. Wong, David B. is a Professor and Chair of the Department of Philosophy, Duke University, USA. He received his B.A. from Macalester College and his PhD from Princeton University. Representative publications include Moral Relativity (1984), Confucian Ethics: a Comparative Study of Self, Autonomy and Community (co-editor with Kwong-loi Shun), “Coping with Moral Conflict and Ambiguity”, Ethics (1992), “Relational and Autonomous Selves”, Journal of Chinese Philosophy (2004), “Dwelling in Humanity or Free and Easy Wandering?” in Technology and Cultural Value, ed. P. Hershock et al (2003), “Pluralistic Relativism”, Midwest Studies in Philosophy (1996). Xiao, Yang is Assistant Professor of Philosophy at Kenyon College, USA. He received his Ph.D. from the New School for Social Research in 1999, and was a Post-doctoral Fellow at UC Berkeley in 1999–2000, and a Post-doctoral Fellow at the Fairbank Center at Harvard in 2002–03. His most recent publication is “How Confucius Does Things with Words: Two Paradigms of Hermeneutic Practice in the Analects and Its Exegeses,” forthcoming in Journal of Asian Studies. Zheng, Yujian is Associate Professor in the Philosophy Department, Lingnan University, Hong Kong. With BS degree in engineering mechanics, he turned to philosophy of science at the MA level in China, and finally got his PhD in philosophy from Bowling Green State University, Ohio, USA. He has numerable paper publications in the overlapping areas of dynamic rational choice theory, philosophy of mind and action, and moral philosophy. His current research interests include evolutionary and naturalist account of normativity or emergence of intentionality.

Donald Davidson and the volume editor, Bo Mou, at Davidson’s UC Berkeley office discussing this anthology project on July 17, 2003, about one month before his passing away. (Photographed by Annie Ren)

HOW CONSTRUCTIVE ENGAGEMENT OF DAVIDSON’S PHILOSOPHY AND CHINESE PHILOSOPHY IS POSSIBLE: A THEME INTRODUCTION Bo Mou

In this theme introduction to the project titled “Davidson’s Philosophy and Chinese Philosophy: Constructive Engagement”, I intend to do three things. First, I introduce the background, nature and theme of the project and explain the significance and value of the constructive engagement between Davidson’s philosophy and Chinese philosophy. Second, I examine some relevant meta-philosophical and methodological issues that are closely related to the current project, including those concerning the distinct orientations and purposes of comparative studies and their due relations, how to look at the adequacy of a prospective project in comparative studies, etc. Third, I introduce the major points and distinguishing approaches of the contributors’ essays and the rationale by which those essays are organized in this volume.

1 It is especially philosophically interesting and challenging to investigate how a constructive engagement1 between Chinese philosophy2 and Western analytic philosophy3 is possible. For one thing, Chinese

1

By ‘constructive engagement’ I mean philosophical inquiry into how, via reflective criticism and self-criticism, distinct modes of thinking, methodological approaches or points of view in different philosophical traditions or within (the complex array of different approaches of ) the same tradition, can learn from each other and make a joint contribution to the common philosophical enterprise. 2 By ‘Chinese philosophy’ here I primarily mean various movements of philosophical thought in China from the Zhou dynasty (roughly eleventh century to 256 B.C.) through the early Qing dynasty (1644–mid 19th century) and their contemporary studies and developments. 3 By ‘Western analytic philosophy’ or ‘Western philosophy in the analytic tradition’ I mean a Western mainstream philosophy from Socrates, Plato and Aristotle via Descartes, British empiricism and Kant to the contemporary analytic movement.

2

a theme introduction

philosophy and Western philosophy in the analytic tradition are two major philosophical traditions that have made many distinct and significant contributions. For another thing, the two philosophical traditions have been considered by many to be remote, alien or even opposed to each other; some in each tradition have taken philosophical practice in the other tradition to have merely marginal value or regard the two traditions as being essentially alien to each other. Indeed, some mistaken or at least seriously misleading stereotypes have resulted from one party’s ignorance, or lack of in-depth investigation, of the other party’s philosophy, while others have resulted from failure to recognize the genuine nature of even one’s own tradition or from some theoretical conflations. Today more and more philosophers in both traditions have realized that Chinese philosophy (or the philosophical dimension of Chinese thought) and Western philosophy (including its analytic tradition) are not essentially alien to one another: they have common concerns with a series of fundamental issues and have taken their characteristic approaches to them; thus they can learn from each other and jointly contribute to the common philosophical enterprise through constructive dialogue and engagement. Some systematic meta-philosophical discussions, especially in view of constructive engagement of Chinese philosophy and Western philosophy in the analytic tradition, have been carried out concerning the nature of philosophy in different philosophical traditions and the issue of comparative philosophical methodology.4 These have made constructive preparation at the level of meta-philosophical theory and at the level of reflective practice for carrying out further in-depth investigations, like the current project, of exactly how the Chinese philosophy and the Western philosophy in the analytic tradition can jointly contribute to the common philosophical enterprise. Note that, besides indicating a historical connection between Western philosophy in such a tradition and some methodological approaches taken in this tradition, such phrases as ‘Western analytic philosophy’ and ‘Western philosophy in the analytic tradition’ used by this writer are not intended to imply that those methodological approaches are, intrinsically or conceptually, exclusively connected with Western philosophy. 4 One recent result from such efforts is the anthology volume, Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions, Open Court, 2001, edited by Bo Mou and given a foreword by Donald Davidson. It is noted that the efforts in this connection have been made in the past dozens of years in some ways; for the sake of the purpose of this “Introduction” I do not plan to review them here.

a theme introduction

3

In the aforementioned background, and in view of the need and significance of constructive dialogue and engagement of Chinese and Western philosophy, one effective way to carry out such studies is to focus on one philosophically significant figure or one significant movement of philosophical thought in either Chinese or Western tradition in constructive comparison with various relevant thoughts and strands in the other tradition. It has been rendered especially philosophically interesting, rewarding, and significant to carry out a case investigation of constructive engagement between the philosophy of Donald Davidson in the Western analytic tradition and Chinese philosophy for a number of theoretical considerations. First, Donald Davidson (1917–2003) is known as one of the most important and influential philosophers in the twentieth century whose works involve a series of fundamental issues in philosophy. “There is no more creative or systematic philosopher at work in America today than Donald Davidson . . . Davidson has already constructed one of the most remarkable pillars of sustained philosophical reasoning to be found in any era” (Ian Hacking).5 Both those philosophers who endorse Davidson’s views and those who oppose (some of ) his views have unanimously agreed that Davidson’s philosophy is one of the most creative sources to stimulate their philosophical reflections in depth. Davidson’s works involve a series of fundamental issues and concerns in philosophy many of which various thinkers in the Chinese philosophical tradition have also explicitly or implicitly addressed and somehow made their distinct contributions to; those issues include (but are not limited to): (i) the relations between language, thought, and reality; (ii) philosophical issue of truth; (iii) meaning and reference; (iv) understanding and interpretation; (v) knowledge and objectivity; (vi) actions and events; (vii) philosophy of mind; (viii) problem of human rationality; (ix) irrationality and practical reasoning; and (x) the issue of metaphors. Their constructive engagement on those issues would jointly contribute to our understandings and approaches to them. Second, one distinct portion of Davidson’s philosophy concerning the conceptual scheme, rationality, etc. has its significant implication to the relation and engagement among distinct modes of thinking,

5 The citation is from the back cover of Donald Davidson (2001), Subjective, Intersubjective, Objective, Oxford: Clarendon Press.

a theme introduction

4

methodological approaches or points of view in different philosophical traditions including Chinese and Western philosophies. On the other hand, there are some significant meta-philosophical thoughts in Chinese philosophy in regard to how to look at different approaches and points of view. It would be especially philosophically interesting and significant to investigate how some significant meta-philosophical ideas in Davidson’s philosophy and in Chinese philosophy can contribute to (our understanding of ) constructive engagement among distinct modes of thinking, methodological approaches or points of view in different philosophical traditions as well as within (the complex array of different approaches of ) the same tradition. Third, because Davidson is a well-respected and influential philosopher in contemporary Western philosophy, a positive research result from this test case study would play a positive or even strong exemplary role for further constructive engagements of this kind both in view of methodological approach and in regard to substantial treatment of some fundamental issues and concerns in philosophy. Fourth, this project would provide an effective and unique way to look at how the philosophy of one of the major figures in the Western analytic tradition has crossed cultural and national boundaries to contribute to the common philosophical enterprise. There is another contributing consideration. Though it seems to be less theoretical and more related to Davidson’s style of doing philosophy, this consideration has indeed contributed to the fashion in which and means by which the foregoing constructive-engagement agenda has been implemented in regard to Davidson’s philosophy. It is Davidson’s very positive and encouraging attitude towards reflective criticism and challenge, insofar as my personal contact with him can tell. Let me start with some small things that I have personally experienced. About eighteen years ago, around 1987, when I still worked at the Institute of Philosophy, Chinese Academy of Social Sciences, I selected, and translated into Chinese, thirteen representative essays of Davidson’s writings, mainly from his Inquiries into Truth and Interpretation (first edition, 1984), to give a more or less systematic introduction of Davidson’s thought into the Chinese philosophical circle as well as for the sake of my personal research plan to have a more precise understanding of his thought.6 When trans-

6

The result of this work is a collection, Truth, Meaning, Actions and Events: Selections

a theme introduction

5

lating one of the essays from his preceding volume, I thought there was an error in the text; but I felt hesitant about this. For the sake of translation quality, I wrote Davidson consulting him about it. Davidson gave a prompt response and agreed that it is a real error. I would have forgotten this if it were not for Davidson’s formal acknowledgement of this in his “Preface to the Second Edition” of the above volume (second edition, 2001). During 1999–2000 when I did research at UC Berkeley, I discussed with Davidson about how to look at the nature of the enumerative character of Tarski’s truth definition. Though we have different views on the issue, Davidson always patiently explained his position and listened to my explanation of my view without a dismissive attitude.7 Such things in my personal contact with Davidson have convinced me that he sincerely welcomes reflective criticism and other challenges for the sake of constructive engagement. Indeed, Davidson considers this as one central feature of the analytic method as he understands and practices it. The analytic method in philosophy, Davidson wrote, “is a method that starts with a question or a doubt and tries to find an answer or to resolve the doubt. This sets in train attempts to find reasons for or against theses that suggest themselves as answers to the questions or resolutions of the doubts. The analytic method can engage with ideas at any level and from whatever quarter or discipline or tradition. It provokes argument and when practiced with an open mind it engenders dialogue. At its best, dialogue creates mutual understanding, fresh insights, sympathy with past thinkers, and, occasionally, genuinely new ideas. But before there can be dialogue the parties must meet. . . .”8 That is where we need to start for this constructive-engagement project: we meet with an open mind, for critical challenge and engagement, and to create mutual understanding, fresh insights and genuinely new ideas. Davidson’s style of doing philosophy has inspired me to implement the foregoing constructiveengagement strategy in an active elenchus style of dialogue with the orientation of, or emphasis on, serious critical engagement instead from the Philosophical Writings of Donald Davidson, edited and translated by Bo Mou, Beijing, China: the Commercial Press, 1993 (in Chinese). 7 This disagreement is discussed in my article “The Enumerative Character of Tarski’s Definition of Truth and Its General Character in a Tarskian System”, Synthese vol. 124, Issue 1 & 2, January 2001, pp. 91–122. 8 See Davidson’s “Foreword” to the anthology Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions, edited by Bo Mou, Open Court, 2001, p. v.

6

a theme introduction

of a mere celebration, especially in Davidson’s case. The suggested project consists of, primarily, an anthology project and, secondarily, an international conference project whose main purpose is to serve as a platform for critical discussion and engaging challenge; the anthology project is an independent project, instead of the conference proceedings, primarily for the sake of quality control, while the conference project is indispensable for implementation of the constructive engagement strategy of the anthology project, though it also serves some other purpose, being sensitive to situations and need (say, to bring good international academic-exchange opportunities to local scholars of the conference host region). In late 2001 I talked with Davidson about the idea; he loved it, especially its emphasis on the critical engagement of his thought instead of a mere celebration. He indicated that he was more than happy not merely to contribute one essay to the anthology project and deliver a talk at the conference project but also to assume the role of commentator for each of the speakers’ presentation papers and provide a reply to each of the anthology contributors’ essays.9 With all these background matters in view, let me highlight the theme and objectives of this anthology project. The central theme of this anthology project is to investigate (i) how Davidson’s philosophy and some thoughts and strands in Chinese philosophy could jointly contribute to the common philosophical enterprise in some philosophically interesting ways and (ii) how some significant metaphilosophical ideas in Davidson’s philosophy and/or in Chinese philosophy can contribute to (our understanding of ) constructive engagement among distinct modes of thinking, methodological approaches or points of view in different philosophical traditions as well as within the same tradition. The volume has the following three objectives. (1) Through the 9 Later on in 2002, as the International Society for Comparative Studies of Chinese and Western Philosophy (ISCWP) was established with its emphasis of the constructive-engagement, the conference project became one of its projects, i.e., the 1st ISCWP “Constructive Engagement” international conference, co-sponsored by Institute of Foreign Philosophy, Peking University, which also assumed the conference host, and by the American Philosophical Association’s Committee on International Cooperation. The conference was originally scheduled for holding in the summer of 2003. Because of the SARS outreach in Beijing and China in the spring of 2003, the conference had to be postponed to the summer of 2004 with the Institute of Philosophy, Chinese Academy of Social Sciences, as the conference host as well as one of its co-sponsors.

a theme introduction

7

preceding theme of the project, the volume is to investigate how Davidson’s philosophy and Chinese philosophy could jointly contribute to the common philosophical enterprise and the constructive engagement among different philosophical traditions in philosophically interesting ways. (2) Through this challenging test case of constructive engagement between Davidson’s philosophy and Chinese philosophy, the volume is to show how Chinese philosophy and Western philosophy (including its analytic tradition) are not essentially alien to one another: they have common concerns with some fundamental issues and have taken their characteristic approaches to some of those issues; they can learn from each other and jointly contribute to the common philosophical enterprise in complementary ways. (3) Through (1) and (2) above, this project is to show how the constructive engagement in comparative studies is possible and how such comparative methodology of constructive engagement concerning distinct modes of thinking, methodological approaches or points of view in different philosophical traditions or within the same tradition is important or even indispensable in general philosophical inquiry. This anthology has its several distinguishing characteristics. (1) This volume is the first of its kind to investigate at an in-depth level how a major figure in the Western (contemporary) mainstream philosophy in analytic tradition and some thoughts and strands in Chinese philosophy could jointly contribute to the common philosophical enterprise in philosophically interesting ways. (2) This volume as whole (and many an individual contributed essay in the volume) is to investigate some fundamental issues and concerns in philosophy from some distinct comparative approaches that would resort to conceptual and explanatory resources from both the analytic tradition and the Chinese tradition instead of merely from one tradition. (3) Through this case study of constructive engagement, the volume has shown how Chinese philosophy and Western mainstream tradition in the analytic tradition are not essentially alien to one another: they have many common concerns with a series of fundamental issues and could jointly contribute to the common philosophical enterprise. (4) Through (2) and (3) above, this volume has shown how the constructive engagement in comparative studies is possible and how such comparative methodology of constructive engagement is important or even indispensable in general philosophical inquiry. (5) This volume, through the foregoing (1) and (2), can play a positive or even

8

a theme introduction

strong exemplary role for further constructive engagement of this kind both in view of methodological approach and in regard to substantial treatment of some fundamental issues and concerns in philosophy. (6) All the contributed essays in this volume are previously unpublished pieces, written expressly for this volume, and unavailable anywhere else. As indicated above, Davidson himself originally planned to contribute to this anthology by offering one unpublished essay and providing his reply to each of the contributors’ essays. His participation in this anthology project actually went much beyond this formal commitment to his prospective writings in print. During the two years since I set out to work on this project in 2001, Davidson and I had quite a few discussions via emails and get-together meetings on various things involved.10 Indeed, his style of doing philosophy and taking care of some other relevant things is truly encouraging in various ways for effectively carrying out a project of this kind. In my last meeting with Davidson at his office on July 17, 2003, I brought to him copies of several contributors’ essays; he told me that he planned to use the coming month to work out at least a detailed outline of his own contributing essay to the volume, while reading and thinking of these contributors’ challenging essays, so that I can use this outline as part of the proposal package to contact a prospective publisher. Unfortunately, his sudden passing away in the next month, on August 30 of 2003, makes it impossible for us to share and learn his thought from his planned contribution writings to this volume and see how Davidson would further develop his ideas when facing criticism and challenges. It is a sad loss to this anthology project whose successful completion Davidson himself seriously looked forward to, as everyone else who has participated in the project does. However, with the firm support from all the contributors in this volume, the anthology project together with its engagement platform plan continued; the contributors have had their consensus: besides its value on its own, the successful completion of the anthology project in a critical-engagement manner which Davidson favors would be a fitting memorial to Donald Davidson. All the 12 essays in this 10 I still clearly remember the situation of one discussion (in 2003) with Davidson on some relevant things concerning this volume at a Starbucks coffee shop at Solano Avenue in Berkeley, located roughly in the middle between his home on the Berkerly hill and my home at the Albany bay shore, which he considered to have a fair driving distance to both.

a theme introduction

9

volume have undergone their authors’ many revisions that have benefited from the successful and effective critical discussion and each other’s engaging challenges at the conference in June 2004. This anthology is a collective achievement, which comes from all the authors’ valuable contributions and from the firm support from all the contributors and other participants in various ways as indicated in the “Acknowledgements” page. Having resulted from the critical engagement that Davidson most favors, and with the mutual understanding, fresh insights and new ideas which are generated from such critical engagement and to which Davidson truly looked forward, this volume is dedicated to, and in memory of, Donald Davidson, for his important contribution to the common philosophical enterprise that has crossed cultural and national boundaries, and for his inspiration and participation in the current project in his unique way.

2 To understand the nature and significance of the current project on the constructive engagement between Davidson’s philosophy and Chinese philosophy, the reader is expected to understand several major orientations with their distinct methodological approaches in comparative studies and their due relations.11 Without pretending to exhaust all working orientations, I intend to highlight three major orientations and their distinct methodological approaches in comparative studies whose due examination, in my opinion, would be most helpful for a constructive development of comparative philosophy.12 I plan to do this by discussing 11 The major part of the content of this section is a further revision of the major account in my article, “Three Orientations and Four ‘Sins’ in Comparative Studies”, the APA Newsletter, Fall 2002, Vol. 02, No. 1, pp. 42–45, and of some passages on certain involved methodological issues in the first section of my article “A ReExamination of the Structure and Content of Confucius’ Version of the Golden Rule”, Philosophy East and West, Vol. 54, No. 2 (April 2004), pp. 218–248. 12 By ‘comparative philosophy’ I mean not merely comparative studies of different philosophical traditions but any comparative investigation concerning distinct modes of thinking, methodological approaches (perspectives, guiding principles or instruments) or substantial points of view in different traditions or within (the complex array of different approaches of ) the same tradition, though what is focused on in this volume is comparative engagement between Chinese philosophy and analytic philosophical tradition via the case of Davidson’s philosophy.

10

a theme introduction

the appropriateness of four ‘sins’ that are oft-cited, explicitly or implicitly, in critically evaluating a comparative project. The reason that I take this strategy is this: the appropriateness or legitimacy of the four ‘sins’ depends on the nature, purpose, and orientation of a comparative project that would decisively determine which kind of methodological approach should be taken and what kind of expectations are appropriate; the strategy is an effective way to identify how crucial aspects and purposes of those orientations and approaches are distinct and so, in treating one’s own comparative project or critically evaluating some other’s comparative project, to be more sensitive to its distinct purpose and orientation and thus to what it is appropriate to expect. When comparative projects are critically evaluated, there seem to be four sorts of complaint. The alleged ‘sins’ are these: (1) oversimplification; (2) over-use of external resources; (3) exaggerated distinction; and (4) blurring assimilation. They, or some of them, are sometimes taken for granted in two ways: first, it is thought that any simplifying the object of study or using external resources to characterize it is doomed to be excessive and thus deserve to be charged with negative ‘over’-character; second, it is assumed that the four complaints may be made indiscriminately in evaluating any comparative project without regard to the orientation and methodological strategy of that study. A metaphilosophical examination of the four ‘sins’ will help to effectively identify the distinct character and objectives of a variety of orientations and their approaches. The first orientation under examination aims to give a historical and descriptive account. That is, the primary concern and purpose of this type of comparative study is to accurately describe relevant historical matters of facts and pursue what thinkers in comparison actually thought, what resources were actually used (by them), and what appear to be similar and different. The orientation of this type of comparative study thus might be called ‘historical orientation’, and its methodological approach aims at accurate description of historical matters of fact. The historical orientation requires its practitioners to cover a vast range of historical data to give such ‘factual’ description. It seems that this orientation and its methodological approach are typically taken in Chinese studies or Sinology as the primary approach to Chinese and comparative philosophy. There is no wonder that the aforementioned four oft-cited ‘sins’ would be assumed relevant to those comparative projects with the

a theme introduction

11

historical orientation. First, to accurately describe something, it is taken for granted that one should not simplify what is actually complicated; in other words, simplification is always oversimplification: any simplification is guilty of being negatively excessive; and simplification is thus identical with falsification. Second, as for overuse of external resources, any conceptual or explanatory resources which are used to interpret a thinker’s idea under examination but were not actually used by the thinker herself are rendered inadequate or excessive: use of external resources is always over-use of external resources.13 Third, in this approach, exaggerated distinctness often results from over-simplification of one or both parties under comparative examination in the direction of ignoring part(s) in one tradition or account that would share something common in another tradition or account; in this way, insofar as the sin of oversimplification has been already legitimately charged, the charge of exaggeration of the due distinction (if any) between the two would be appropriate. Fourth, in this approach, blurring assimilation often results from over-use of external resources to interpret one or both parties under comparative examination, especially when the external resources used to characterize one party come from the other party; to this extent, insofar as the sin of over-use of external resources has been already legitimately charged, the resulting assimilation of blurring the distinction between the two would be also adequately charged. There would be nothing wrong or inadequate with the historical orientation and its methodological approach per se, when the orientation/approach is adequately taken as one of a number of alternative orientations/approaches, instead of the exclusive one, and when one can see its limitations in serving other distinct purposes in comparative studies. In view of this, one question would be natural: Are there any orientations and approaches other than the historical orientation that would be adequate, and, more importantly, necessary in view of certain purposes in comparative studies? With a positive answer to the question actually being presupposed in the preceding discussion, the question can be phrased in another way:

13 By ‘external resources’ I mean those resources that were not actually used by the ancient thinker under discussion when the resources are identified from the historical point of view or with the historical orientation. Nevertheless, as I explain later, using the very term ‘external’ in some situations would simply miss the point in regard to the purpose of the third orientation to be discussed.

12

a theme introduction

How are other legitimate orientations and methodological approaches possible and necessary? In the following, I focus on two other orientations and their respective methodological approaches. The second orientation in comparative studies is concerned with interpretation14 through elaborating a thinker’s ideas under examination; the primary purpose of this orientation is to enhance our understanding of a thinker’s ideas via some effective conceptual and explanatory resources, whether or not those resources were actually used by the thinker herself. It is clear that a purely historical approach does not fit here: To elaborate and understand the thinker does not amount to figuring out exactly how the thinker actually thought; instead, such interpretation and understanding might include the interpreter’s elaboration of the implications of the thinker’s point, which might not have been considered by the thinker herself, or the interpreter’s representation of the thinker’s point in clearer and more coherent terms or in a more philosophically interesting way, which the thinker herself might have not actually adopted.15 In both cases, given a thinker’s ideas (in one tradition or account) under interpretation, some effective conceptual and explanatory resources well developed in another tradition or account are consciously used to enhance our understanding of, and elaborate, the thinker’s ideas; those resources used are thus tacitly and implicitly, but constructively, in comparison and contrast to those original resources by means of which the insight or vision was somehow delivered, insofar as such comparison of the two distinct sorts of resources is not expressly and directly conducted. The term ‘constructively’ here means such tacit comparative approach intrinsically involves how the interpreter of the

14 Here I use the term ‘interpretation’ in a narrow or straightforward sense as specified here (in terms of elaborating and understanding) rather than in a broad or implicit sense in which all the three orientations discussed here could be somehow identified as ‘interpretation-concerned’. 15 Then, can these implications be said to belong to the thinker’s ideas in the text (and thus fall into what the thinker truly means/meant or what the thinker’s ideas truly has/had)? In an important sense, the answer would be yes; for these implications are truly implied by the ideas delivered by the thinker, although one can surely say that these implications were not actually expressed by the thinker, and one thus might say that they are not what the thinker actually (or truly?) means/meant. (At this point, one can see that such expressions as ‘what a certain thinker truly means/meant’ or ‘what she truly has/had’ tend to be ambiguous and vague and thus deserve clarification, especially when one intends to make claims about what a thinker truly means/meant or what her ideas truly has/had.)

a theme introduction

13

thinker’s ideas could learn from another tradition or account regarding resources to enhance the interpreter’s understanding of the thinker’s ideas; therefore, some constructive philosophical engagement between distinct resources in different traditions is tacitly involved in this orientation and its corresponding methodological approach. In this way, the so-called over-use of external resources is not necessarily a sin but might really enhance our understanding of a thinker’s ideas or clarify some original unclear or confusing expression of her ideas. Consequently, the endeavor per se of using external resources in this orientation is not automatically inappropriate and thus is not doomed to be a sin, as it would be in the historical orientation. Note that, when those explanatory and conceptual resources are used, they are not intended to assign the same degree of articulated systematization and of mastering some conceptual and explanatory resources to an ancient thinker but to enhance our understanding of her ideas delivered in the text. For this explanatory purpose, it is not merely legitimate but beneficial to employ more explicit or clearer conceptual resources to elaborate some otherwise implicit and hidden thing (say, coherence and connectedness) in a thinker’s ideas that was sometimes less clearly delivered or even ill-expressed for lack of those contemporary explanatory and conceptual resources that are unavailable to the ancient but to us.16 It is also noted that, when a thinker’s line of thought and her ideas lack in articulated systematicity in their language expressions, that does not amount to saying that the thinker’s line of thought and her ideas per se go without (implicit and hidden) coherence and connectedness deep in a thinker’s ideas. Consequently, we cannot base ourselves merely on this lack of articulated systematicity in language expression to judge that the thinker’s text itself is not a philosophical work when the text was indeed intended to deliver her reflective ideas. At this point, with the previous and current methodological considerations, some further elaborations of the thinker’s line of thought and her surrounding reflective ideas via adequate conceptual and explanatory resources available to us is genuinely needed, instead of being the mere issue of preference, for the

16 It is another matter when a thinker intentionally uses some seemingly paradoxical remarks to make some points. However, such occasions imply neither that the ideas delivered by these remarks per se are actually incoherent nor that the points in question could not be delivered effectively in clearer terms without paradoxical appearance.

14

a theme introduction

sake of enhancing our understanding of the thinker’s ideas including their due implications. As indicated in discussing the historical orientation, ‘blurring’ assimilation might result from ‘over’-use of external resources when interpreting one or both parties under comparative examination, especially when the external resources used to characterize one party come from the other party. But, for the purpose of interpretation, the resulting assimilation is not necessarily a sin but might illuminate the essential connection and common points between the assimilated ideas at the fundamental level so as to enhance our understanding of those ideas. It is clear that a comparative project with the interpretation-concerned orientation, instead of the historical orientation, is free, or indeed tends, to focus on a certain aspect, layer or dimension of a thinker’s ideas based on the purpose of the project, the reflective interest of the person who carries out the project, etc. Indeed, instead of a comprehensive coverage of all aspects or dimensions of the object of study, focusing on one aspect or dimension is a kind of simplification. Now the question is this: Is any simplification per se doomed to be indiscriminately a sin of over-simplification? It should be clear that, if the purpose of a comparative project is to focus on interpreting or elaborating one aspect or dimension instead of pretending to giving a comprehensive historical description, charging the practitioner of this project with over-simplification or doing something excessive in simplifying the coverage into one aspect or dimension would be both unfair and miss the point. Let us agree that a comparative project should be guided by some comprehensive understanding. But a comparative project taking a certain methodological perspective through focusing on one aspect of the object of study is not incompatible with a comprehensive understanding. At this point, what needs to be recognized is an important distinction between a methodological perspective as the current working perspective and the methodological guiding principle that an agent presupposes when taking the methodological perspective and that would be used by the agent to guide or regulate how the current perspective would be applied and evaluated in view of some other relevant perspective(s). One’s reflective practice per se of taking a certain methodological perspective amounts to neither reflectively rejecting some other relevant methodological perspective(s) nor presupposing an inadequate methodological guiding principle

a theme introduction

15

which would render irrelevant other relevant methodological perspectives (if any).17 We have discussed three ‘sins’ (i.e., ‘over-simplification,’ ‘over-use of external resources,’ and ‘blurring assimilation’) that might be charged against a comparative project with the interpretation-concerned orientation. How about the other one, the sin of ‘exaggerated distinction’? This case is more complicated than what it may appear. This ‘sin’, as discussed before, is connected with the sin of oversimplification when the comparative project assumes the historical orientation. But when a comparative project takes the interpretationconcerned orientation and does ‘simplify’ the object of study by focusing on one aspect of the object of study, is it automatically guilty of the sin of ‘exaggerated distinction’? The preceding distinction between the methodological perspective and the methodological guiding principle is helpful here again. What is at issue is whether the interpreter has assumed an adequate methodological guiding principle to guide and regulate how to look at the relation between the current methodological perspective used as a working perspective and other relevant methodological perspective(s) that would point to other aspects of the object of study. Consequently, when one evaluates a comparative project, what really matters is for one to look at what kind of methodological guiding principle is presupposed behind the working perspective; only when this is examined can the charge of ‘exaggerated distinction’ be adequately evaluated. Now let us move onto the third orientation under examination, i.e., constructive-engagement-concerned orientation aiming at joint contribution to common philosophical issues. The primary purpose of this orientation in comparative studies is to see how, through reflective criticism and self-criticism, both sides under comparative examination could jointly and constructively contribute to some commonly concerned issues of philosophy,18 rather than to focus on providing 17 For a detailed and systematic discussion of the distinction between the methodological perspective and the methodological guiding principle and its implications, see Bo Mou (2001), “An Analysis of the Structure of Philosophical Methodology— in View of Comparative Philosophy”, in Bo Mou ed. Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions, Chicago: Open Court, pp. 337–364. 18 It is arguably right that many issues that were traditionally identified as some ‘unique’ issues in different traditions have turned to be concerned primarily with different aspects, layers or dimensions of some commonly concerned, more general issues of philosophy, especially from a more broadly philosophical vantage point. This is one point that I have endeavored to make and illustrate in my several writings mentioned above.

16

a theme introduction

a historical or descriptive account of each or on interpreting some ideas historically developed in a certain tradition or account. Typically, in comparatively addressing a certain commonly concerned issue of philosophy, some substantial ideas historically developed in distinct philosophical traditions or accounts are explicitly and directly compared with the aim of showing how they could jointly and complementarily contribute to the common concern in some philosophically interesting ways. Insofar as constructive engagement in dealing with various common concerns and issues of philosophy is most philosophically interesting, this comparative orientation and its methodological strategy directly, explicitly and constructively conducts philosophical engagement and is thus considered to be most philosophically interesting. To highlight the characteristic features of a comparative project with this as its primary orientation, let us examine the appropriateness of three charges, among the aforementioned four, that have been sometime or even often brought against comparative projects with this orientation, that is, the ‘sin’ of oversimplification, the ‘sin’ of over-use of external resources, and the ‘sin’ of blurring assimilation. A typical procedure of conducting a philosophically constructive engagement in such comparative projects could be both conceptually and practically divided into three phases: (i) the pre-engagement phase in which certain ideas in different traditions or accounts that are relevant to the common concern under examination and thus to the purpose of the project are focused on and identified; (ii) the engagement phase in which those ideas internally engage with each other in view of that common concern and the purpose to be served; and (iii) the post-engagement phase in which those distinct ideas from different sources are now absorbed or assimilated into a new approach to the common concern under examination. The three ‘sins’ aforementioned may be considered to be typically associated with different phases. The ‘sin’ of over-simplification regarding a certain idea identified from a certain tradition may be typically associated with reflective efforts in the pre-engagement phase; the ‘sin’ of over-use of external resources regarding elaborating a certain idea from a certain tradition may be typically associated with reflective efforts in the engagement phase; and the ‘sin’ of blurring assimilation may be typically associated with reflective efforts in the post-engagement phase. Now let me briefly evaluate the appropriateness of the three ‘sins’ respectively in the corresponding three phases; looking at the

a theme introduction

17

‘sins’ in this way will help to highlight features of comparative projects primarily with the constructive-engagement-concerned orientation. (1) In the pre-engagement phase, it might be not only legitimate but also adequate or even necessary to have simplification and abstraction of some ideas in one tradition or account into such a perspective: this perspective per se is presented in most relevant terms to the common concern addressed, and the purpose served in a constructive-engagement-concerned comparative project, while without involving those irrelevant elements in the tradition or account from which such a perspective comes, though those irrelevant elements in that tradition might be relevant to figuring out the point of those ideas. The reasons are these. First, the primary concern of the project is not with how such an idea is related to the other elements in the source tradition or account but with how it is relevant to approaching the commonly concerned philosophical issue. Second, while one needs to understand the point of an idea in the context in which it was raised, once one understands the point (either through employing data provided by projects with the first two orientations or through one’s own background project with one of the first two orientations), there would be no present purpose served by discussing background. Third, it is clear that such an approach per se does not imply denying the social and historical integrity of the idea in the source tradition; the point is that the existence of such integrity cannot automatically guarantee an indiscriminate priority or even relevance of expressly addressing it in any comparative projects without regard to their orientations and purposes. (2) In the engagement phase, relevant perspectives from different source traditions would constructively engage each other. From each party’s point of view, the other party is something external without; but, from a more broadly philosophical vantage point and in view of the commonly concerned issue, the distinct views may be complementary within. In this context, the term ‘external’ would miss the point in regard to the purpose here: the pivotal point is not this or that distinct perspective but the issue (and its comprehensive approach) to whose various aspects those perspectives point; in view of the issue, all those perspectives are internal in the sense that they would be complementary and indispensable to a comprehensive understanding. (3) In the post-engagement phase, some sort of assimilation typically results from the preceding constructive engagement; that is,

18

a theme introduction

such assimilation would adjust, blur and absorb different perspectives into one new approach as a whole; this would be what is really expected in this sort of constructive engagement in comparative studies, instead of a sin. It should be noted that, if a comparative project which explicitly has one of the preceding orientations is considered as a projectsimplex in comparative studies, a comparative project in philosophical practice might be a complex that goes with a combination of two or more orientations. For example, a comparative project concerned with a historical figure often consists of such a combination. Recognition of the characteristic features of the above three distinct comparative orientations and their respective methodological approaches would help us discriminatively treat different stages or parts of a comparative project-complex. Traditionally, to my knowledge, comparative projects with the above third and second orientations (especially when resorting to contemporary development and resources of philosophy) have yet to receive due emphasis for some reasons. First, as far as comparative projects regarding Chinese and Western philosophies are concerned, a comparative project sometimes tends to be taken as a mere byproduct or extension of studies of the classical Chinese philosophy which itself sometimes tends to be taken largely as merely historical studies of the history of (the classical) Chinese philosophy. Second, on the other hand, comparative approach as a methodological approach has not yet been considered primarily as an effective approach to doing philosophy per se. Third, the aforementioned four ‘sins’ (especially, those of ‘oversimplification’, ‘over-use of external resources’ and ‘blurring-assimilation’) have been more or less considered as some taken-for-granted ‘sins’ and have thus discouraged reflective efforts in the direction of the third orientation (or even the second orientation) which would often unavoidably but appropriately commit those ‘sins’ in many cases. Fourth, most importantly, Chinese philosophy and Western philosophy (especially its mainstream traditions19) are sometimes taken as being essentially alien to one another; this kind of mentality would undermine or pre-empty any serious 19 I use the plural form of the term ‘mainstream tradition’ to indicate that the identity of a mainstream tradition is not exclusive but sensitive to regions and times throughout the history of Western philosophy. In the twentieth century, the analytic tradition is a mainstream tradition in the English speaking countries while the Continental tradition is a mainstream tradition in the European Continent.

a theme introduction

19

reflective efforts in comparative projects with the third orientation and, in my opinion, negatively contribute to prejudice Western philosophers as well as some scholars in studies of Chinese and comparative philosophy may assume that Chinese philosophy is not philosophy in the sense of the term ‘philosophy’ that is intrinsically related to a series of fundamental concerns and issues as addressed in Western philosophy (especially its mainstream traditions). Now, as more and more philosophers in the fields of Chinese and comparative philosophy have a holistic understanding of Western philosophy (both its past and its contemporary development, both its appearance and its deep concerns, and both its distinct working perspectives and its guiding principles at a deep level) and become constructively engaged with Western philosophy on a series of fundamental common concerns and issues, it is more widely agreed among philosophers who are familiar with both Chinese and Western philosophies, as already indicated in the first part, that they are not essentially alien to one another: they have common concerns with a series of fundamental issues in philosophy and have taken their characteristic approaches to them. They thus could learn from each other and jointly contribute to the common philosophical enterprise through constructive dialogue and engagement. Consequently, there is serious need to emphasize comparative projects of the third and second orientations, though this emphasis certainly would not deny the legitimacy and due value of the first orientation as one effective approach but stress its constructive compatibility with the other orientations. The current comparative project “Davidson’s Philosophy and Chinese Philosophy: Constructive Engagement,” as highlighted by the subtitle, is the one primarily with the aforementioned third orientation.

3 The main text of the anthology consists of twelve essays contributed by experts in relevant areas of study, most of whom are established scholars. With the foregoing constructive-engagement-concerned orientation aiming at joint contribution to some commonly concerned issues of philosophy, the main text of the book focuses on five issues or topics which are respectively the subjects of the following five parts into which the twelve contributed essay are organized:

a theme introduction

20 Part One:

Conceptual Schemes, Relativism, and Cross-cultural Understanding; Part Two: Principle of Charity and Chinese Philosophy; Part Three: Rationality, Normativity, and Inter-cultural Disagreement; Part Four: Meaning and Interpretation; Part Five: Truth Concern and Dao Concern.

Two notes are due. First, these topics are both reflectively interesting on their own and most conducive to constructive-engagementconcerned comparative treatment in view of Davidson’s philosophy and Chinese philosophy. Nevertheless, in so saying, that does not mean that the current five topics have ever exhausted all actual or potential aspects of such comparative engagement of Davidson’s philosophy and Chinese philosophy; the current selection of the topics more or less reflects the points of interest where a certain level of quality research results have been generated in the current scholarship in this connection. Second, because of the comprehensive nature of some of the contributed essays, the way to organize the entries is not exclusive. The reader who focuses on reading the entries on one topic (say, those in Part One) can thus have her cross reference to some other entries in another part (say, Part Two and Part Three). In the following, let me sketch the organizational strategy and outline how the entries are related to the thematic topics of the parts into which they are organized. One of the most philosophically interesting and significant points made by Davison that is closely relevant to the current enterprise of comparative Chinese-Western philosophy with the foregoing constructive-engagement orientation is Davidson’s point in his argument against the very notion of conceptual schemes, the dualism between scheme and content, and thus conceptual relativism, which was first systematically put forward in his well-known essay “On the Very Idea of a Conceptual Scheme” (1974). The point is reassured and sharpened in his “Foreword” to a recent volume on the relation between Chinese and analytic philosophical traditions in view of cross-cultural understanding and constructive engagement in philosophy: We seldom stop to think how much, as philosophers, we share with other philosophers from other ages, other countries, other traditions. We tend to discover our common problems and interests as we read, teach, and travel. The discovery surprises us for, to begin with, minds are best compared by finding as many points of similarity as every-

a theme introduction

21

day patterns of action and reaction afford. But once this fitting of pattern to pattern is accomplished, the remaining differences loom out of proportion. This perhaps explains why a first exposure to a new tradition seems to reveal an unbridgeable gap. What experience shows, though, is that, as in other areas, differences are to be understood only as seen against a background of underlying agreement. The underlying agreement may be largely unspoken and unnoticed, but it is always available. Sometimes we need help in appreciating how philosophy builds on what we all know. No world views or conceptual schemes are truly incommensurable.20

To Davidson, cross-cultural mutual understanding in philosophy is certainly possible; any disagreement between philosophers from different traditions can be understood only against a background of underlying agreement. Davidson does not intend to deny the diversity; rather he considers such diversity as the life of the philosophy. Davidson’s point is by no means to indiscriminately seek consensus and conformity but celebrates variety and differences as shown in distinctive perspectives to look at various aspects of the object of study for the sake of a comprehensive understand, as Davidson says: . . . we should not seek conformity in philosophy. On the contrary, in our intellectual work we should celebrate variety and do all we can to insure its survival. We should not make the mistake of supposing that under ideal circumstances our institutions and our philosophical proclivities would or ought to become more alike. If anything, we should expect the opposite, and we should welcome it. We welcome understanding, and with it tolerance, but variety and difference is at the heart of philosophy.21

In the citation above, when Davidson emphasizes that “No . . . conceptual schemes are truly incommensurable”, this saying seems to presuppose that there are conceptual schemes or it seems to be compatible with the notion of conceptual schemes. However, Davidson’s position is stronger than what the preceding statement appears to suggest: for the sake of arguing against relativism, Davidson opposes the very idea of a conceptual scheme, which is presupposed by relativism as standardly understood, in the sense that he renders this notion incoherent; another sort of argument against relativism which

20 Davidson (2001), “Foreword”, in Bo Mou ed. (2001), Two Roads to Wisdom?— Chinese and Analytic Philosophical Traditions, Open Court, p. v. 21 Davidson, op.cit., pp. v–vi.

22

a theme introduction

Davidson offers is that which opposes the dualism between scheme and content. However, is the notion of conceptual schemes really incoherent? How is relativism related to the notion of conceptual schemes and the dualism between scheme and content? What are the due implications of Davidson’s point here towards comparative philosophy? Can Davidson’s resources alone be sufficient for developing a sound foundation for a creative comparative philosophy which lies at the core of an emerging global philosophy? These questions will be examined in Part One of the volume, “Conceptual Schemes, Relativism, and Cross-Cultural Understanding”, which consists of three essays. In his essay “Relativism and Its Schemes”, Michael Krausz agrees with Davidson in disallowing the schemecontent dualism and thereby those relativisms that assume it. However, it seems to the author, not all relativisms need assume the schemecontent dualism; we need not understand schemes as giving form “to the data of sensation” but as matrixes in terms of which experience and data themselves are to be understood. The author continues to argue that the notion of conceptual schemes per se is coherent: persons of one scheme may understand persons of another, although the two schemes are not fully translatable; that is, disagreeing with Davidson, the author argues that non-translatability is not a sufficient condition for non-understandability (mutual understandability between schemes does not entail translatability). It seems to Krausz, the problem with those standardly understood relativisms that are tied to conceptual schemes concerns not their incoherence but their characteristic adumbrance (typically it is unclear just what truth or rightness is meant to be relative to). The author, Koji Tanaka, of the second essay, “Davidson and Chinese Conceptual Scheme”, focuses on elaborating Davidson’s unique insight into the incoherent nature of conceptual relativism and exploring the consequence of this insight for those comparative studies of Chinese and Western traditions of philosophy that presuppose their having built up mutually non-inclusive stocks of concepts and take this kind of conceptual relativism as the foundation of comparative philosophy. Through his case analysis of the early Confucian and Western approaches to morality, the author casts doubt on the legitimacy of the foregoing comparative projects because of their incoherent character. But the author also emphasizes that the way of carrying out comparative studies that he criticizes is not exhaustive and there may be other alternative ways. Indeed, as many

a theme introduction

23

essays in this volume effectively show, a worthy comparative project need not presuppose such a radical conceptual relativism as the foundation of its work. The next essay in this part can be seen as one constructive endeavor in that direction which tries to incorporate Davidson’s ideas and some other resources into a foundation for reaching cross-cultural understanding and carrying out creative comparative philosophy. As explained in the last section, most philosophically interesting projects in comparative philosophy do things more than, or proceed with the primary goal other than, descriptive comparison: based on cross-cultural understanding, they seek integration, mutual challenge, and joint contribution to the common philosophical enterprise, a kind of synthetic project. In his essay “Making Room for Comparative Philosophy: Davidson, Brandom, and Conceptual Distance”, Stephen C. Angle explores to what extent Davidson’s work can provide what is needed for developing a sound foundation for such a kind of comparative studies in philosophy as synthetic projects in his terms. Given how different philosophical traditions can be from one another, though, such ambitions can seem hopeless; the author argues that Davidson’s work provides grounds for comparative philosophers to overcome such worries. But, at the same time, the author maintains that in certain respects, Davidson makes overcoming difference seem easier than it really is. The author then draws on Robert Brandom’s inferential and pragmatic account of semantics to develop a foundation for comparative philosophy that, while still Davidsonian in important ways, shows us both why the synthetic projects of comparative philosophy are possible, and also why they can be difficult: conceptual differences can be robust and important, even if radical incommensurability is not in the offing. These synthetic projects lie at the core of what might more properly be called an emerging global philosophy, and they are well worth their trouble. Part Two, “Principle of Charity and Chinese Philosophy”, focuses on Davidson’s principle of charity which plays a crucial role in Davidon’s approach to the issue of how to understand and interpret other agents including how to reach cross-cultural understanding. The core idea of Davidson’s principle of charity is this: we must interpret others on the assumption that they are rational beings, navigating the same world as we are. But how to understand the nature, explanatory power, scope and limit of this principle? The two essays in this part explore the issue in view of Chinese philosophy.

24

a theme introduction

In his essay, “Where Charity Begins”, David B. Wong first examines Davidson’s own critical reflection and changes of the statements of the principle of charity, from maximization of agreement to optimization of agreement, pointing out that this is the right move but also that the requirement of optimization itself required interpretation; in so doing, the author criticizes overreaching attempts based on the principle of charity to restrict the variety of belief and desire, and then attempts to identify what real guidance the principle provides for the project of interpreting others. It seems to the author that charity is less a definite principle of interpretation but rather the assortment of the various ways we have of explaining others’ talk and actions. We do interpret others by analogy to our own beliefs and desires, but if the principle of charity bids us to interpret others as being like us, we need to interpret who “we” are. The author argues that the “we” is bound to be diverse in belief and desire over human culture, and in particular in the range of values that are central to particular cultures. It is argued that pragmatic factors play in a role in the decision as to who counts as one of “us” and that our very conception of what is rational results partly from the mechanism of natural selection through genetic permutations, and that our ideal of rationality must open to the possibility that a range of solutions to human problems may have equal claim to be called rational. The author’s strategy of argument consists in working backwards, from interpretations of the moral tradition of Confucianism from a contemporary American perspective; the question addressed is how analogy and models of rationality operate within these interpretations to increase our understanding of Confucianism. The other essay in this part, “Davidson’s Charity in the Context of Chinese Philosophy”, is intended to test and defend the explanatory power of Davidson’s principle of charity when facing some challenges in the context of Chinese philosophy. The author of the essay, Yiu-ming Fung, discusses three alleged problems with Davidson’s principle of charity. The first problem is A. C. Graham’s doubt about whether the principle can be survived in the context of comparison between Chinese and English language and philosophy. The author argues that Graham’s criticism is fundamentally self-defeating; it cannot be understood as a real challenge to Davidson’s principle of charity. The second problem is about the methodological character of the principle of charity which many tend to negatively render

a theme introduction

25

“transcendental”; the author argues that this is a misunderstanding due to the critics’ failing to take seriously some of the concrete examples provided by Davidson and not realizing a crucial distinction between the two kinds of transcendental argument in the literature. Suzuki’s interpretation of Zen Buddhism is considered to present another alleged challenge to Davidson’s principle. The author argue that Suzuki’s idea that giving up rational thinking and logic is a necessary condition of attaining Zen cannot escape from its “landing” problem, i.e., how to make sense the relation between two kinds of truth or two levels of world; the author’s strategy is to use this problem to force the Zen masters into a dilemma: either to explain the relation with our rational language and logic or to fall into an abyss of totally isolated from the real world. The foregoing discussions of Davidson’s rejection of the very idea of conceptual schemes and relativism and his principle of charity in view of Chinese philosophy all directly or indirectly point to Davidson’s idea of rationality. Furthermore, it is known that Davidson’s thought, though wide and complex, has a remarkably unitary character; what unifies Davidson’s views on various issues is his fundamental understanding and characterization of the very idea of rationality and normativity. The two essays in Part Three, “Ratinality, Normativity, and Inter-Cultural Disagreement”, explicitly address the issue and make their insightful elaborations of Davidson’s line especially in view of Davidson’s idea on the intrinsic connection between the normative and the factual (or the descriptive). They also explore how a Davidsonian concept of normativity and rationality can contribute to our understanding of disagreements between cultures. Both think that such disagreements do appear between different cultures or different paradigms; but such disagreements can make real sense only against the background or overall validity of Davidson’s approach to rationality or normativity. In his essay “Davidsonian Rationality and Ethical Disagreement between Cultures”, Samuel C. Wheeler extends several Davidsonian ideas in order to sketch a Davidsonian ethics. It is a reconstructed Kantianism holding that the normativity implicit in rational interpretation of another agent that maximizes agreement supplies the basis for understanding the normativity of ethical concepts and that the normative is not different in ontological or epistemic status from the factual; it is thus a kind of ethical objectivism holding that ethical

26

a theme introduction

sentences have truth value, even without a basis in pre-conceptualized desire-stuff. Nevertheless, Wheeler argues, this Davidsonian account is different from Kant’s account. In contrast to “obligation” as the primary concept in Kant’s account, “ought” is the primary ethical concept; and conditional “ought” sentences are formally akin to conditional probability sentences, and their categorical form has an implicit relativization (to various “senses” of “ought” and to backgrounds); this would render the connection between rationality and ethics different from Kant’s: ethical arguments and thinking are properly understood on the model of induction following “rules of thumb,” rather than deduction following Kantian universal moral principles; the rationality as the foundation of ethics would thus require more than formal consistency. Given that ethical actors are situated in cultures, this Davidsonian ethics would render a system of partialities, which shows cultural disagreements, objectively right for a culture and also render some of ethical disagreements between distinct cultures irresolvable, because there is no general algorithm for determining what is the better course of action. In his essay “A Davidsonian Approach to Normativity and the Limits of Cross-cultural Interpretation”, Yujian Zheng first examines Davidson’s distinctive approach to normativity. The author argues that it is Davidson’s distinctive sense of normativity, rather than something else, that unifies Davidson’s seemingly diverse arguments on various issues and that the crucial feature of Davidson’s approach to normativity lies in its holistic character in integrating various interdependent things and in rendering the connection between the descriptive and the normative intrinsic. The author then explores the possible implication of the Davidsonian view of normativity to the issue of interpretative ruptures in cross-cultural understanding and intends to argue for the following point. Although Davidson’s thesis for the impossibility of wholesale rupture (or incommensurability) between conceptual schemes (paradigms) is sound, it does not suffice to expel skepticism or relativism at more local levels of interpretation: temporary, local ruptures of interpretation are inevitable between language games; but the local failure of Davidson’s thesis can make real sense only against the background or overall validity of his approach to normativity. One can say that what the preceding authors treat are actually three distinct dimensions of the same issue and that the three themes

a theme introduction

27

can be viewed as three related issues in Davidson’s theory of meaning and interpretation. And it is true that, historically speaking, Davidson put forward his views on the three aforementioned themes in the due course of developing his theory of meaning and interpretation. It is natural to move on to this topic, and the discussion of it is not only significant on its own but will shed further light on the issues addressed before. Davidson’s important contribution to natural language semantics is his proposal, first made in “Truth and Meaning” (1967), to employ a Tarski-style theory of truth to supply what are expected for a theory of meaning. Here I would like to give a relatively more detailed account of Davidson’s approach to meaning for the sake of the reader’s understanding of how this topic is related to the preceding three themes as well as for the sake of the reader’s understanding of how Davidson’s theory of truth as a theory of meaning works by appealing Tarski’s Convention T, whose understanding is needed for understanding the central points of some of the essays in the following two parts (for example, Chung-ying Cheng’s essay). In his characteristic approach to meaning, Davidson first identifies the conditions that an adequate theory of meaning must meet: (1) it must enable us to give the meaning of every sentence in a natural language L under discussion: for each sentence of L, it must produce a sentence of the form ‘s means that p’, where ‘s’ is replaced by a structural description of any sentence of L, and ‘p’ is replaced by a sentence that is supposed to give the meaning of the sentence whose structural description is ‘s’; (2) it must show how any of sentences in L is constructed out of finite words and rules regarding the combination of words; (3) it must show that its presentation of the meaning of a sentence in L should be based upon those concepts that are not beyond the concepts that are used to express the sentence; (4) it must be empirically verifiable. With all those requirements in mind, Davidson’s proposal starts with this: the apparently non-extensional ‘means (that)’ is replaced by ‘is T if and only if ’: (T) s is T if and only if p;

we require of a theory of meaning for a language L that “without appeal to any (further) semantical notions it place enough restrictions on the predicate “is T” to entails all sentences got from schema T when ‘s’ is replaced by a structural description of a sentence of

28

a theme introduction

L and ‘p’ by that sentence.”22 This requirement, as Davidson sees it, is “in essence” Tarski’s Convention T. Tarski’s Convention T is Tarski’s criterion or test of the material adequacy for a formal truthdefinition candidate, which claims that a definition is an adequate truth definition if it has as its logical consequences all the following instances of the schema (T): (T) s is True-in-L if and only if p

where ‘s’ is replaced by a structurally descriptive name of any sentence of the language L for which Truth is being defined, and ‘p’ is replaced by the translation of the sentence in the meta-language ML of L. Davidson thus claims that “a Tarski-type truth definition supplies all we have asked so far of a theory of meaning”. One can say that Tarski’s Convention T is one stone ‘almost’ hitting four birds, i.e., meeting all the aforementioned four conditions. The reason I say ‘almost’ above is this: there is an important difference between Davidson and Tarski. Tarski’s Convention T appeals to the notion of translation, while Davidson cannot because of the aforementioned condition (3).23 Davidson cannot afford to buy in any notion of translation in Tarski’s way; for he thinks a satisfactory theory of meaning needs to meet conditions (3) and (4) in a strong sense and in a straightforward way: an interpretive theory of meaning for natural language must be empirically verifiable on the basis of evidence plausible available to the radical interpreter who has no antecedent understanding of the target language. Davidson’s solution is his account of radical interpretation in terms of the principle of charity which works by appealing to the interpreter’ own norms of rationality. That is where Davidson’s approach to meaning is intrinsically related to his principle of charity, his understanding rationality and normitivity, and his strategy against the very idea of a conceptual scheme and relativism.

22 Davidson (1967), “Truth and Meaning”, in Donald Davidson (2001), Inquiries into Truth and Interpretation (second edition), Oxford: Clarendon Press, p. 23. 23 In “Truth and Meaning” (1967), when saying the preceding requirement is “in essence” Tarski’s Convention T, Davidson actually also presupposed the notion of translation when bringing in a Tarski’s strategy to serve for his purpose of constructing a theory of meaning for natural language. In early 70’s, Davidson realized the problem and suggested his account of radical interpretation to solve the problem. Cf., Davidson (1973), “Radical Interpretation”, in Davidson, op. cit.

a theme introduction

29

The three essays in Part Three, “Meaning and Interpretation”, explore, directly or indirectly, various dimensions of Davidson’s approach to meaning and interpretation in view of Chinese linguistic practice and Chinese philosophy in various ways. In several publications, Davidson has argued that attributions of meaning to utterances have to be accompanied by attributions of belief to utterers. His view is entailed by a more general consideration of two senses of ‘meaning’, which are examined by A. P. Martinich in his essay “On Two Kinds of Meaning and Interpretation”. The sense of ‘meaning’ as applied to communication—the author calls it ‘c-meaning’—is standardly distinguished from meaning in the sense of importance or significance—the author calls it ‘s-meaning’. These two senses of the term ‘meaning’ correspond to two senses of the Chinese counterpart term, yi-si, of the term. Martinich argues that the two kinds of meaning, which appear to be semantically unconnected, are in fact semantically connected in ways that the author thinks have not been noticed in view of English and Chinese linguistic practices: namely, a similarity that appears in considering what is required for understanding or interpreting the two kinds of meaning. It seems to the author, an investigation of the judgments that an interpreter needs to make in order to understand the c-meaning of an utterance leads to the conclusion that such judgments also constitute an understanding of s-meaning; the absolute distinction between c-meaning and s-meaning is untenable. Any adequate theory of language meaning needs to account for a wide variety of language uses, not merely various literal utterances but also uses of figure of speech such as metaphor. Davidson’s approach to metaphor is an extension or application of his general theory of meaning to treatment of metaphor; it is a semantic approach which holds that metaphors function in virtue of their literal meanings, instead of a special kind of meaning, to make us see things about the world. Given that there are some distinctive characteristic uses of Chinese language as an ideographic language and the Western phonetic language like English, it would be a good test, and is reflectively interesting to see, whether Davidson’s approach can well account for metaphorical uses of Chinese language. In his essay “Metaphorical Use versus Metaphorical Essence: Examples from Chinese Philosophy”, in view of some interesting usage of metaphors in Chinese linguistic practice, Kim-chong Chong presents his evaluation of the scope, limit and potentiality of Davidson’s treatment

30

a theme introduction

of metaphor. According to the author, on the one hand, Davidson takes a narrow conception of metaphor when he contrasts it with simile, and describes as characteristic of metaphor that it cannot be wholly paraphrased without remainder. This conception is unnecessary, since it can be diagnosed as part of the motivation for thinking of metaphor as having an inherent cognitive content, and that he wishes to argue against. However, on the other hand, Davidson’s conception of metaphor as belonging to the domain of use liberates us from this narrow conception, including essentialist views of metaphor such as that of Lakoff and Johnson. Through various examples, this essay criticizes tendencies to posit an essence of metaphor in certain discussions of metaphor in Chinese philosophy. Though Davidson’s approach to meaning is a semantic one, instead of a pragmatic one, this by no means implies that Davidson renders the literal meaning of a sentence simply or typically conventional. Rather, as highlighted in “A Nice Derangement of Epitaphs”,24 Davidson’s point is this: for an adequate interpretation to occur, it is not simply to accept a conventional understanding of the speaker’s words; one must also understand what the speaker means by the words on that occasion; that is, interpretation is theoretically ad hoc. To this extent, one might as well characterize Davidson’s approach in terms of his own pragmatic strand. Along the line of Davidson in this regard, in his essay “Reading the Analects with Davidson: Mood, Force, and Communicative Practice in Early China”, Yang Xiao engages in an important debate in contemporary philosophy of language between Dummett and Davidson by drawing out its implications in the context of early Chinese philosophy and language. The debate concerns the relationship between the grammatical moods of a sentence and the pragmatic forces of the utterances of the sentence. Dummett insists that there is a strict correlation between mood and force, and that illocutionary force is always conventional. Davidson rejects the thesis, contending that neither force nor ulterior purpose of an utterance is governed by linguistic conventions. The author argues that, when studying the Analects, we should not limit ourselves to the study of grammatical and conventional features of sentences. Following Davidson, the author suggests that we should make the ‘pragmatic turn’, focusing on people’s linguistic or communicative 24 The essay was published in R. Grandy and R. Warner eds. (1986), Philosophical Grounds of Rationality, Oxford University Press.

a theme introduction

31

practice, namely the utterances of sentences in concrete occasions on which the sentences are put to work. Now, with all the foregoing important themes in Davidson’s philosophy (his thought on the very idea of conceptual schemes and his rejection of relativism, his principle of charity, his understanding of rationality and normativity, and his approach to meaning, etc.), one can see that one crucial concept emerges from behind: it is the concept of truth that plays a central explanatory role in Davidson’s philosophy; to this extent, the preceding themes can be somehow incorporated into what is called ‘Davidson’s theory of truth’. The central role of truth in Davidson’s account is certainly not because of Davidson’s mere preference but because, it seems to Davidson, the concept of truth as a matter of fact plays a central role as explanatory basis to explain other important things such as our propositional thought, our understanding of the world and other minds. It seems to Davidson, “Without a grasp of the concept of truth, not only language, but thought itself, is impossible.”25 Davidson treats the concept of truth, along the line of our pre-theoretic understanding of truth, as an axiom-like primitive that is already understood and is partially captured by such instances of the form (T) as “‘Snow is white’ is true if and only if snow is white”. Now, in view of Davidson’s philosophy and Chinese philosophy, some philosophically interesting, significant questions concerning the role of truth in philosophy can be raised. Among others, first, can Davidson’s theory of truth be validly used to examine, and enhance our understanding of, Chinese philosophy, on the one hand, and be enriched or even constructively developed by the recourses of Chinese philosophy, on the other hand? Second, given that the truth concern is a long-term, fundamental or dominant concern in Western philosophical tradition, as Davidson’s approach prominently, effectively and fruitfully illustrates, while the dao concern has been considered a long-term, fundamental or dominant concern in Chinese philosophical tradition, what is a due relation between the truth concern and the dao concern, generally speaking, and the truth pursuit and the dao pursuit, specifically speaking? Is the dao concern dramatically different from the truth concern so that Chinese philosophy and Western philosophy (say, Davidson’s philosophy) are alien to each other in this fundamental connection? Surely, one can directly and effectively discuss the first question while 25

Davidson (2005), Truth, Language, and History, Oxford: Clarendon Press, p. 16.

32

a theme introduction

simply presupposing a positive answer to the second question; on the other hand, a meta-philosophical examination of the second question is clearly reflectively worthy. The two essays in Part Five, “Truth Concern and Dao Concern”, focus respectively on the two issues. In his essay “From Donald Davidson’s Use of ‘Convention T’ to Meaning and Truth in Chinese Language”, Chung-ying Cheng examines the foregoing first issue from his onto-hermeneutical point of view. In the first part, Cheng first characterizes Davidson’s approach to interpretation of truth in terms of the Tarskian Convention T and then explains how the Davidsonian project can relate to understanding Chinese language and Chinese philosophy; in so doing, he also raises critical questions concerning some restrictions of Davidson’s approach in view of Chinese language and how to re-construct the Davidsonian theses in light of our understanding of Chinese language and Chinese philosophy. In the second part, author first suggests five principles to characterize the ontological and cosmological visions embodied in both Chinese language and Chinese philosophy: (1) the principle of open syntax and grammar; (2) the principle of natural creativity; (3) the principle of internal reality externalized; (4) the principle of historical reference (historicity) and life-world; (5) the principle of comprehensive totality and ultimate origin. The author then integrates the five principles in a system which modifies Davidson’s major theses in view of a philosophical reflection on Chinese language and for the sake of interpreting the dao-concerned Chinese philosophy. Addressing the aforementioned second issue in my essay “Truth Pursuit and Dao Pursuit: From Davidson’s Approach to Classical Daoist Approach in View of the Thesis of Truth as Strategic Normative Goal”, I focus on two related issues. First, I explore the due point of the thesis of truth as (strategic) normative goal (the TNG thesis) through analyzing Davidson’s approach and drawing some morals via three distinctions (between truth nature and truth criterion, between truth pursuit as strategic norm and truths pursuit as tactic norm, and between the semantic-ascent version and the paraphrase-explanatory-reduction version of the TNG thesis). Second, in view of the TNG thesis and of Davidson’s approach, I explore the relation between the truth pursuit and the dao pursuit in regard to their roles as normative goals respectively in the Western and Chinese (Daoist) philosophies. I argue that the dao pursuit of classical Daoism is essentially the truth pursuit in the way as capturing by the point of the

a theme introduction

33

TNG thesis. I further explain how the classical Daoism as presented in the Dao-De-Jing and the Zhuang-Zi, can make its substantial contribution to our understanding of the truth concern in philosophically interesting ways. The issue of how it is possible for Chinese philosophy and Western philosophy in the analytic tradition, generally speaking, and Davidson’s philosophy and Chinese philosophy, specifically speaking, can constructively engage each other is now not merely a theoretical issue at a meta-philosophical level but also an issue of how to evaluate the fruitful reflective practice in this connection. As evidenced by the contributors’ creative works in this volume, the current level of reflective inquiry into the issue has reached beyond a purely metaphilosophical discussion of the possibility of such engagement to the extent that many scholars have already effectively and fruitfully carried out comparative studies of Chinese philosophy and Western philosophy in the analytic tradition with the aforementioned constructiveengagement purpose. One advantage of the situation is this. If such reflective practice with the constructive engagement as its primary purpose does enhance and enrich the reader’s understanding of the involved common themes, issues or concerns of philosophy as addressed in the essays of this volume and does help the reader in her treatment of them, that would not merely provide a prima facie positive answer to the question of whether those perspectives from Chinese philosophy and Western philosophy (specifically here, from Davidson’s philosophy) under comparative engagement are really ‘comparable’, i.e., whether the constructive engagement as characterized above is possible; that would also constitute a best evidence and effective illustration of, generally speaking, how such constructive engagement is actually possible, instead of merely theoretically possible, and, specifically speaking, exactly how Chinese philosophy and Davidson’s philosophy in the Western analytic tradition can jointly contribute to the common philosophical enterprise. If so, the purpose of this volume would be fulfilled in one important connection. Now it is time for the reader to tell.26 26 I am grateful to Chad Hansen, Douglass Heenslee, Chenyang Li, Youzheng Li and Xianglong Zhang for their helpful comments and criticism of an early version of the main contents of Section 2. My thanks also go to A. P. Martinich for his helpful identification of a number of awkward expressions and typos in Sections 1 and 3.

PART ONE

CONCEPTUAL SCHEMES, RELATIVISM, AND CROSS-CULTURAL UNDERSTANDING

CHAPTER ONE

RELATIVISM AND ITS SCHEMES* Michael Krausz

Donald Davidson offers two sorts of arguments against relativism as standardly understood.1 One sort (with which I agree) opposes the dualism between scheme and content. The other sort (with which I disagree) opposes the coherence of the very idea of a conceptual scheme. I shall suggest that the notion of a conceptual scheme is coherent and that persons of one scheme may understand persons of another. Yet relativism need not assume the offending dualism between scheme and content. A relativist need not, though characteristically does, fall into the trap of assuming the offending dualism. Standardly, relativism holds that cognitive or value claims involving truth or rightness (or their cognates) are relative to the conceptual schemes in which they appear. Such schemes may include cultures, societies, civilizations, traditions, historical epochs, points of view, perspectives, standpoints, world views, paradigms, forms of life, practices, languages, linguistic frameworks, networks of categories, modes of discourse, systems of thought, disciplinary matrices, constellations of absolute presuppositions, symbol systems, or the like. Accordingly, the truth of a proposition or the rightness of a way is said to be relative to one or another of such schemes.2

* For their helpful comments and suggestions I extend thanks to Cheryl Chen, Catherine Elgin, Jay Garfield, Bo Mou, Sam Wheeler, and David Wong. 1 Donald Davidson (1982), “On the Very Idea of a Conceptual Scheme”, in Relativism: Cognitive and Moral, edited by Jack Meiland and Michael Krausz, Notre Dame: Notre Dame University Press, pp. 66–80. 2 When I use the term ‘scheme’ I have in mind (as does Davidson) something like Thomas Kuhn’s idea of a paradigm—without, though, Kuhn’s hint of a correspondence with an independent World or Nature. It is a matrix in terms of which modes of inquiry are pursued. The constituents of schemes come as holistic packages and are not reducible to isolated beliefs or theories.

38

chapter one 1

Before considering Davidson’s arguments, here are twelve preliminary points. (1) Schemes and their cognates are adumbrant in the sense that they have no clear boundary conditions; they are open; they are indeterminate or indistinct as to their limits. Where, for example, do the Inuit and Canadian cultures begin and end? Where do the Mexican and American cultures begin and end? Where do the Medieval, Renaissance and Baroque periods begin and end? Where do Marxist and feminist perspectives begin and end? For one who holds that truth or rightness is relative to these, it is difficult to specify clearly in relation to what such relativizing should operate. (I shall return to this point.) (2) In the above characterization of relativism, I used the word “standardly” to signal that some definitions of relativism do not invoke schemes as defining features. For example, Joseph Margolis defines his non-standard “robust relativism” in terms of “two essential doctrines: (i) that, in formal terms, truth values logically weaker than bipolar value (true or false) may be admitted to govern otherwise coherent forms of inquiry and constative acts, and (ii) that substantively, not merely for evidentiary or epistemic reasons, certain sectors of the real world open to constative inquiry may be shown to support only such weaker truth-values. That is all.”3 (3) Relativism is often motivated by the recognition of historical or cultural diversity. Yet that recognition does not amount to relativism, since cultural or historical diversity is logically compatible with either relativism or anti-relativism. An anti-relativist might react to the diversity of beliefs or practices by employing the notion of progress, according to which pertinent beliefs and practices which differ from one’s own are thought to be unenlightened, or backward, or something of the sort. Such an anti-relativist might assume that one’s views are or should be the end product of a process toward 3 Joseph Margolis (1989), “The Truth About Relativism”, in Relativism: Interpretation and Confrontation, edited by Michael Krausz, Notre Dame: Notre Dame University Press, 1989, p. 232. For a discussion of Margolis’s view, see Michael Krausz (1999), “Interpretation, Relativism, and Culture: Four Questions for Margolis,” in Interpretation, Relativism and the Metaphysics of Culture: Themes in the Philosophy of Joseph Margolis, edited by Michael Krausz and Richard Shusterman, Amherst, NY: Humanity Press, pp. 105–124.

relativism and its schemes

39

a non-relative truth. Of course, other people whose beliefs and practices differ from ours might reach a similar conclusion about our own beliefs and ways. What the anti-relativist demands is a non-relative criterion applicable to everyone. It would reveal truth and rightness as such. But the relativist holds that there is no such criterion. Any wouldbe neutral criterion would reflect the biases or prejudices of one’s home scheme. Accordingly, to say that some belief or practice is true or right relative to a scheme is to say more than that individual who embraces it happens to believe something, or happens to live according to certain practices. Relativism of truth or rightness is not just an ascription of a belief or an anthropological fact. The relativist holds that at the boundaries of schemes where standards of evaluation give out, one has no way to adjudicate between contending claims. One has no context-neutral objective way to appeal to a putative overarching nature, human nature, absolute principle or the like. Yet the relativist does claim to have resources necessary to discriminate between distributive claims that fall within a pertinent scheme. That is, the relativist has the resources to say, for example, that it is true that this sentence is composed on a computer, without recourse to extra-scheme considerations. (4) Anti-relativism does not entail absolutism, the view that there is a permanent or eternal foundation of meaningfulness, existence, truth or rightness. One may be an anti-relativist but not an absolutist. One may oppose relativism and remain agnostic or even deny absolutism. Davidson, for example, rejects relativism, but that does not commit him to absolutism. (5) Relativism is not skepticism. The skeptic holds that knowledge about matters of fact or value is impossible. Unlike skepticism, relativism affirms that such knowledge is possible, yet relative to a pertinent scheme. The relativist sees the skeptic as setting up an impossible goal—absolute truth—and then damning the relative truth that one might attain because absolute truth is impossible. It is as if an explorer, on a fruitless quest for a mythical treasure, were to toss aside the “lesser” treasures that he or she might acquire along the way. In contrast to the skeptic, the relativist holds that truth or rightness is attainable, even if it is not the absolute truth first desired by the skeptic. (6) Sometimes relativism is confused with fallibilism. Fallibilism is the view that, at any stage of one’s inquiry, one may be wrong. It

40

chapter one

serves as a tonic for those who might hold that truth is absolute, and that they in fact have attained it. But one could be a fallibilist and still endorse an absolutist notion of truth, as Sir Karl Popper does. He says: “A . . . doctrine of absolute truth, in fact a fallibilist doctrine . . . asserts that mistakes we make can be absolute mistakes, in the sense that our theories can be absolutely false, that they can fall short of the truth. Thus the notion of truth, and that of falling short of the truth, can represent absolute standards for the fallibilist.”4 That is, one could embrace the thought that truth or rightness are not relative to schemes, but that at any stage one could be wrong about one’s beliefs or ways. Relativism has no special claim on fallibilism. (7) An absolutist might worry that if there is no absolute truth to be discovered, then there is no worthwhile goal of inquiry. If we cannot aim for an absolute truth, for what can we aim? If there is no absolute truth for everyone, how can we ever say beliefs or practices are true or right? The absolutist worries that by ruling out absolute truth one rules out the possibility of progress in knowledge. The relativist replies that there is no reason to assume a global view of progress. Knowledge can be progressive, if only in a local way according to standards linked to designated schemes. (8) Relativism is characteristically defined as the view that, in the absence of overarching standards of adjudication between pertinent schemes, they are equally admissible. I have distinguished such a view from “multiplism” whereby, in the absence of such standards, not all admissible schemes or interpretations are equally preferable.5 The multiplist claim that several opposing interpretations may be admissible does not preclude our giving good reasons for preferring one interpretation over others. Often there are no univocal commensurating standards between admissible interpretations. I call cases in which there are no such standards “inconclusive.” Inconclusiveness does not entail arbitrariness. Rather, it allows for critical comparability and reasoned preferability of admissible interpretations. And it allows for reasoned preferability of a narrative in which a given interpretation is nested.

4 Karl Popper (1976), “The Myth of the Framework”, in The Abdication of Philosophy: Philosophy and the Public Good, First Edition, Open Court Publishing Company, p. 35. 5 See Michael Krausz (1993), Rightness and Reasons: Interpretation in Cultural Practices, Ithaca and London: Cornell University Press; and Michael Krausz (2000), Limits of Rightness, Lanham, MD: Rowman and Littlefield.

relativism and its schemes

41

(9) Relativism is sometimes understood to entail that mutual understanding between those of different schemes is impossible. But relativists need make no such assumption. As Alasdair MacIntyre says, we are not “condemned to or imprisoned within our own particular standpoint”6 And Popper concurs when he says: “Frameworks, like languages, may be barriers; but a foreign framework, just like a foreign language, is no absolute barrier. And just as breaking through a language barrier is difficult but very much worth our while . . . so it is with breaking through the barrier of a framework.”7 Popper affirms that untranslatability between two languages can be transcended when he says further: In the comparative study of these languages we use, as a rule, our own language as a metalanguage . . . in a critical way, as a set of rules and usages which may be somewhat narrow since they are unable completely to capture, or to describe, the kinds of entities which the other languages assume to exist. But this description of the limitations of English as an object language is carried out in English as a metalanguage. Thus we are forced, by this comparative study, to transcend precisely those limitations which we are studying. And the interesting point is that we succeed in this.8

(10) Self-referential arguments against relativism are well known. They concern the issue of how one should answer the question whether the thesis of relativism itself is true or right. If we say it is absolutely true, a contradiction results. If we say it is relatively true, its reach is limited to the scheme in which it appears. The latter of these alternatives is not vicious if we allow that the relativist’s purpose in arguing need not be to convince the non-relativist. The relativist may use arguments to present his or her view, only to better articulate the view. (11) If Davidson’s argument against conceptual schemes is sound, it unseats those standard forms of relativism which hold that cognitive or value claims are relative to schemes. And his argument associates schemes with languages in this way: We may accept the doctrine that associates having a language with having a conceptual scheme. The relation may be supposed to be this:

6

Alasdair MacIntyre (1989), “Relativism, Power, and Philosophy”, in Relativism: Interpretation and Confrontation, edited by Michael Krausz, p. 199. 7 Karl Popper, “The Myth of the Framework”, p. 46. 8 Karl Popper, “The Myth of the Framework”, p. 38, emphasis added.

chapter one

42

if conceptual schemes differ, so do languages. But speakers of different languages may share a conceptual scheme provided there is a way of translating one language into the other. Studying the criteria of translation is therefore a way of focusing on criteria of identity for conceptual schemes.9

(12) Let us consider two possible examples of pairs of schemes. I say these are “possible” examples because, should Davidson’s argument be sound, they would be disqualified as bona fide examples. Consider first shame and guilt cultures. Ruth Benedict reports: True shame cultures rely on external sanctions for good behavior, not, as true guilt cultures do, on an internalized conviction of sin. Shame is a reaction to other people’s criticism. A man is shamed either by being openly ridiculed and rejected or by fantasizing to himself that he has been made ridiculous. In either case it is a potential sanction. But it requires an audience or at least a man’s fantasy of an audience. Guilt does not. In a nation where honor means living up to one’s own picture of oneself, a man may suffer from guilt though no man knows of his misdeed and a man’s feeling of guilt may actually be relieved by confessing his sin.10

Notice that in Benedict’s characterization nothing precludes a person of the Japanese shame culture from understanding a person of the North American guilt culture. As a “bi-cultural” or “bi-lingual” anthropologist acquainted with Japanese shame culture and American guilt culture, Benedict succeeds in comparing and contrasting them. A person of one culture can understand a person of another. Yet there may be no non-relative standard according to which one culture is right and the other is wrong. Here is a second possible example of a pair of schemes: IndoEuropean languages and the Hopi language as understood by Benjamin Lee Whorf. He reports: We are constantly reading into nature fictional acting entities, simply because our verbs must have substantives in front of them. We have to say ‘It flashed’ or ‘A light flashed,’ setting up an actor, ‘it’ or ‘light,’ to perform what we call an action, “to flash.” Yet the flashing and the light are one and the same! . . . Hopi can and does have verbs without subjects, a fact which may give that tongue potentialities, probably never to be developed, as a logical system of understanding some 9

Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 67. Ruth Benedict (1974), The Chrysanthemum and the Sword: Patterns of Japanese Culture, New York and Scarborough: New American Library, p. 223. 10

relativism and its schemes

43

aspects of the universe. Undoubtedly, modern science, strongly reflecting western Indo-European tongues, often does as we all do, sees actions and forces where it sometimes might be better to see states . . . This is the trouble with schemes like Basic English, in which an eviscerated British English . . . is to be fobbed off on an unsuspecting world as the substance of Pure reason itself.11

As in Benedict, nothing in Whorf ’s characterization precludes an Indo-European from understanding a Hopi, and vice versa. Whorf, after all, is writing in English about Hopi. Nothing precludes mutual understanding. Benedict and Whorf compare and contrast their pairs of examples from the standpoint of a guilt culture in English. Each makes sense of the scheme alternative to hers or his. This does not mean, however, that the schemes are intertranslatable.

2 Let us now turn our attention to Davidson’s argument against the dualism between scheme and content. With Davidson, I reject the dualism. He says: “Conceptual schemes, we are told, are ways of organizing experience; they are systems of categories that give form to the data of sensation; they are points of view from which individuals, cultures, or periods survey the passing scene.”12 Davidson objects to the idea that there is data of sensation there (or some analog thereof ) to be organized. Accordingly he rehearses the dualism this way: The idea is then that something is a language, and associated with a conceptual scheme, whether we can translate it or not, if it stands in a certain relation (predicting, organizing, facing or fitting) to experience (nature, reality, sensory promptings). The problem is to say what the relation is, and to be clear about the entities related.13

Davidson says further: The images and metaphors fall into two main groups: conceptual schemes (languages) either organize something, or they fit it (as in “he warps his scientific heritage to fit his . . . sensory promptings.” The first 11

Benjamin Lee Whorf (1962), Language, Thought, and Realty, New York: Technology Press of MIT and John Wiley & Sons, Inc., pp. 243–4. 12 Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 66. 13 Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 74.

44

chapter one group contains also systematize, divide up (the stream of experience); further examples of the second group are predict, account for, face (the tribunal of experience). As for the entities that get organized, or which the scheme must fit, I think again we may detect two main ideas: either it is reality (the universe, the world, nature), or it is experience (the passing show, surface irritations, sensory promptings, sense data, the given).14

And John McDowell formulates the offending dualism between scheme and content in this way. “Scheme-content dualism is incoherent, because it combines the conviction that world views are rationally answerable to experience—the core thesis of empiricism—with a conception of experience that makes it incapable of passing verdicts, because it removes the deliverances of the senses from the domain of the conceptual. According to the dualism, experience both must and cannot serve as a tribunal.”15 Davidson links relativism with the offending dualism between scheme and content. But relativism need not be defined in this way. We need not understand schemes as “organizing experience” or as giving form “to the data of sensation.” Rather, schemes may be understood as matrixes in terms of which experience and data themselves are to be understood—that is, without assigning them a status preexisting to the schemes in question. Accordingly, no fact of the matter beyond language (as in the Indo-European and Hopi cases) would be presumed, and no fact of the matter about moral behavior (as in the shame and guilt cultural cases) would be presumed. Some relativists might link their relativism with the offending dualism. But they need not. As Davidson says, the dualism must be rejected. Yet if we do reject the dualism we do not forfeit objectivity. Davidson rightly says: In giving up dependence on the concept of an uninterpreted reality, something outside all schemes and science, we do not relinquish the notion of objective truth—quite the contrary. . . . In giving up the dualism of scheme and world, we do not give up the world, but reestablish unmediated touch with the familiar objects whose antics make our sentences and opinions true or false.16

14

Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 74. John McDowell (1999), “Scheme-Content Dualism and Empiricism”, in The Philosophy of Donald Davidson, edited by Lewis Edwin Hahn, Library of Living Philosophers, vol. XXVII, Chicago and La Salle Illinois: Open Court, p. 96. 16 Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 79. 15

relativism and its schemes

45

Again, versions of relativism to which Davidson’s critique of the scheme-content duality are vulnerable are those which assume the duality between what is given to the mind through sensation on the one hand, and concepts which the mind uses to organize this given on the other hand. Davidson’s critique aims at the idea of a reality apart from organizing concepts. He objects to the thought that there is an external reality on to which the relativist (or the absolutist) grafts schemes, for no coherent sense can be made of such an idea of an external reality. The point, therefore, is not whether there is one or more scheme to be grafted onto such a reality—as the traditional standoff between relativists and absolutists presents itself—but whether the very idea of grafting is wrong headed to start with. Davidson says: “Even those thinkers who are certain there is only one conceptual scheme are in the sway of the scheme concept; even monotheists have religion.”17 Davidson opposes those relativisms and absolutisms which presume conceptual schemes that are taken to “mirror” the world(s). He says further: “It would be . . . wrong to announce the glorious news that all mankind—all speakers of language, at least—share a common scheme and ontology. For if we cannot intelligibly say that schemes are different, neither can we intelligibly say that they are one.”18 Davidson’s argument against the scheme-content dualism applies to relativism and absolutism if both presuppose that there is a nature of things independent of schemes which “answer to”, or are “represented by” pertinent schemes. And if the scheme-content dualism were indeed rejected, then the question of relativism versus absolutism, as standardly posed, would not arise in the first place. Yet, since the idea of a conceptual scheme need not assume the scheme-content dualism, relativism without the dualism may still deploy the idea of a conceptual scheme. Rejecting the scheme-content dualism does not undercut all relativisms, only those relativisms that assume it. The resulting kind of relativism would be benign, if heterodox from Davidson’s light. Let us contrast the kind of relativism that assumes the offending scheme-content dualism with the more benign kind of relativism that does not. Consider again the cases of Benedict and Whorf. Recall

17 18

Donald Davidson, “The Myth of the Subjective”, p. 66. Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 79.

46

chapter one

that Benedict reports that “true shame cultures rely on external sanctions for good behavior, not, as true guilt cultures do, on an internalized conviction of sin.” One who links relativism with the scheme-content dualism would construe Benedict’s example this way. Good behavior denotes a fact of the matter, a phenomenon that is there antecedently waiting to be captured or accounted for or “faced.” Shame cultures do it in terms of external sanctions and guilt cultures do it in terms of internalized convictions of sin. In both cases good behavior is presumed to be pre-existently constituted before interpretive activity. In contrast, one who does not link relativism with the scheme-content dualism would drop the pretense of such pre-existence and would instead do what Benedict actually does, namely, speak of good behavior in terms given by the schemes of shame and guilt cultures. Good behavior would be just what is so constituted by pertinent practices in designated cultures. That’s all. Similarly, consider Whorf ’s contrast between the Indo-European who, when making sense of action phenomena, reads into nature “fictional acting entities” and the Hopi who does not. One who links relativism with the offending scheme-content dualism would hold that the nature of action phenomena is pre-existing, constituted before interpretive activity, that the Indo-European and the Hopi both are seeking a “match” between their respective schemes and action phenomena. In contrast, as in Benedict’s case, the relativist who does not link her relativism to the offending dualism, would hold that action would be just what is so constituted by pertinent practices in designated cultures. That’s all. My general point about relativism without the scheme-content dualism is captured by what I elsewhere call “constructive realism” or by what Patricia Hanna and Bernard Harrison call “relativist realism” when, in their new and important book, they say: All talk of what an expression ‘designates’ or ‘refers to’ is in the end merely a shorthand way of talking about the manner in which that expression engages with, or is involved in, some practice or other . . . The supposed entities in the case would then dissolve, not quite into thin air, but into modes of engagement. The mode of engagement of an expression with a practice, now, is clearly not part of the furniture of the natural, extralinguistic world. On the contrary, it is quintessentially a work of human invention, as much a fabrication of ingenuity in the forging of convention as, say, the Petrarchan sonnet form or the rules of golf . . . when we speak of the entities referred to or designated by

relativism and its schemes

47

expressions, we speak, so far as we speak of anything at all, of fabrications of the mind.19

Now, an absolutist might concede that the cases that I have adduced— namely, the guilt versus shame cultures and the Hopi versus IndoEuropean grammars—are conventional. But she might insist that the scheme-content dualism still applies to such cases as middle-sized objects or to molecules. Yet, first, I agree with Davidson that such an absolutist response is unsustainable because his sound arguments for the rejection of the scheme-content dualism apply globally and not in piecemeal fashion. Second, contra Davidson, my concern here is to suggest that the idea of a conceptual scheme is coherent. And that is demonstrated at least in the adduced examples. A more sustained treatment of such entities as middle sized objects and molecules would follow along lines suggested by Hanna and Harrison.

3 Now let us turn to Davidson’s argument against the idea of a conceptual scheme. He argues that the coherence of the idea of a conceptual scheme requires the coherence of the idea of an alternative conceptual scheme. Yet, he argues, the idea of an alternative conceptual scheme is incoherent, for if an alternative conceptual scheme is translatable into the first conceptual scheme, it is not “alternative.” And “the failure of intertranslatability is a necessary condition for difference of conceptual schemes.”20 So, if a putative alternative scheme is not thus translatable, nothing intelligent can be said about it to distinguish it from the first conceptual scheme. Accordingly, Davidson says, “. . . we could not be in a position to judge that others had concepts or beliefs radically different from our own.”21 Since grounds for distinguishing one conceptual scheme from an alternative scheme do not obtain, the distinction between them collapses. 19

Patricia Hanna and Bernard Harrison (2003), Word and World: Practice and the Foundations of Language, Cambridge: Cambridge University Press, chapter 3. See also Bernard Harrison and Patricia Hanna (2003), “Interpretation and Ontology: Two Queries for Krausz,” in Interpretation and Its Objects: Studies in the Philosophy of Michael Krausz, edited by Andreea Ritivoi, Amsterdam: Rodopi Publishers. 20 Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 72. 21 Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 79.

48

chapter one

And with that collapse, the coherence of the very idea of a conceptual scheme is unseated, as is the coherence of those forms of relativism that presume the coherence of the idea of conceptual schemes. Let us call this Davidson’s “alternativity” argument. Davidson’s alternativity argument applies to partial and complete translatability. It applies to portions of a language or a scheme that are non-translatable as well as language taken as a whole. If portions or the whole of a language are not translatable they could not be recognized as alternative, and if they are translatable they are not alternative. According to Davidson, we characteristically make sense of people by listening to what they say, and that means that they must speak a language that we understand. If we cannot understand a person’s utterances, we could not ascribe beliefs to them. We could not say whether they had a different conceptual scheme. And we could not tell whether they are speaking a language in the first place. Accordingly, Davidson advances his principle of charity; namely, we can make sense of what a person means only if we begin by assuming that their beliefs are largely in agreement with our own. Anomalies inevitably arise. We can deal with them in one of two ways. Either 1) we can say there is a failure of translation and therefore a difference in conceptual scheme, or 2) we can say that there is a difference in belief. But there is no hard and fast rule that forces us to deal with anomalies according to 1) or 2). Therefore, if we come across a putative partial failure of translation, we can always say that it arises from a difference of belief rather than a difference of conceptual scheme.22 But we must be careful here. From the fact that there is no hard and fast rule in this regard, it does not follow that we can always say that the partial failure of translation is a matter of belief rather than scheme. Further, there is a difficulty in the vocabulary of “partial” and “complete” translation, since it is hard to designate what is complete and what is partial in relation to what is complete. The limits of language are always emerging and open. It is partial all the way through. What, after all, could count as a complete natural language like Chinese or English?

22 Thanks are due to Cheryl Chen for her here paraphrased characterization of Davidson’s view on this matter, in private correspondence, August 8, 2003.

relativism and its schemes

49

One might retort to Davidson’s alternativity argument that all that it shows is that, while we may have no criterion to verify the presence of an alternative scheme, it may still exist. But Davidson might well reply that it is not enough to indicate the possibility of an alternative scheme. We must also give some reasons or grounds to believe that such an alternativity exists. Yet such reasons or grounds can be given. To start with, there is enough overlap between those of guilt and shame cultures for its inhabitants to compare and contrast their cultures with one another. Within that conversational space—which is improvisatory and emergent—they may show one another their disparities concerning the moral character of what they do or the grammatical character of what they say. One might be tempted to say that one can distinguish conceptual schemes by taking a neutral stance by divesting oneself of all schemes and then comparing those that present themselves. But Davidson disqualifies this possibility, because conceptual schemes are embedded in languages. And to divest oneself of all schemes would require giving up the use of language. Language is necessary for thought. So, if one gives up the use of language, one could not even compare. Davidson says: “Speaking a language is not a trait a man can lose while retaining the power of thought. So there is no chance that someone can take up a vantage point for comparing conceptual schemes by temporarily shedding his own.”23 According to Davidson’s alternativity argument—which associates translatability with understandability—the two pairs of examples adduced earlier in this paper, because of their “would-be” intertranslatability, cannot be bona fide. Each of the pair is not an alternative to the other. In contrast, I suggest that understanding between people of different schemes is possible despite the fact that there may be no complete translatability between them. Davidson might reply that the examples I offer are not bona fide as alternative schemes, since understanding is possible and therefore translation is possible. Such a reply would disallow that understood schemes can be nontranslatable. It would disallow the testimony of bilinguals that they understand different languages despite the fact that full translations are not forthcoming. Put otherwise, Davidson holds that nontranslatability is a sufficient condition for non-understandability. But I

23

Donald Davidson, “On the Very Idea of a Conceptual Scheme”, p. 68.

50

chapter one

suggest that a failure of translation does not imply a failure of understandability. Mutual understandability does not entail translatability, although translatability might entail understandability. Alasdair MacIntyre observes that translation is not always possible. Yet we can make sense of alternative cultures, that is, we can understand persons of alternative cultures. Translatability is not necessary for understandability. So the non-translatability of schemes does not bar one from saying that one is alternative to the other. Accordingly, even if shame cultures and guilt cultures were not translatable, a person of a guilt culture like Benedict could still understand persons of a shame culture, and, presumable, vice versa. I return to the question of partial translatability. As David Wong most helpfully suggests, one can gloss terms of a language even if we have no strict equivalent terms in one’s home language. He says: When translators of the Chinese word ren in the Analects render it as ‘goodness,’ or ‘benevolence,’ or ‘humanity,’ or ‘authoritative personhood’ we know these are just approximations. We look at what people say about the role of the concept as one for comprehensive virtue, and we may bring to our understanding of comprehensive virtue, perhaps, our understanding of Greek ethics, but we know that this is also just an approximation. We are told that Confucius at one point associates ren with loving persons (Analects 12:22), but we should also be careful about equating whatever notions of love we have with the one Confucius was using in this context . . . ‘Goodness’ or ‘humanity’ or ‘benevolence’ or ‘authoritative personhood’ serves more like a pointer.24

The “pointer” Wong speaks of should not be taken to point to a determinate equivalent term or phrase. The term in an object language (Chinese) need not have a ready-made counterpart term in the metalanguage (English). The gloss should not be taken to point to a determinate equivalent term. The idea that there must be a determinate equivalent term ignores a characteristic context in which questions of translation arise. When one translates from Chinese to English a word in a legal document pertaining to the transfer of

24 David Wong, in correspondence, July 30, 2003. For the glosses of ren as 1) goodness, 2) benevolence, 3) humanity, or 4) authoritative personhood, see respectively: 1) Arthur Waley (1989), The Analects of Confucius, New York: Vintage Books; 2) D.C. Lau (1979), Confucius: the Analects, London: Penguin; 3) Wing-tsit Chan (1969), A Sourcebook in Chinese Philosophy Princeton: Princeton University Press; and 4) Roger Ames and Henry Rosemont, Jr. (1999), The Analects of Confucius: A Philosophical Translation, New York: Ballantine Books.

relativism and its schemes

51

property, for example, one assumes acquaintance with various institutions of property rights. The translator might well seek terms neutral to the contending claims so as to avoid a verbal sleight. But such avoidance requires foreknowledge of the implicated institutions. And questions of determinate equivalent terms arise yet again in the “translation” between those institutions. The point is that praxial considerations systematically enter into the business of translation. One cannot translate independently of the practices and purposes to which translation is put. The multiplicity of pertinent practices and purposes undermines efforts that require lexical equivalencies. Accordingly, a given term (like ‘Dao’) may be glossed in many ways according to a multiplicity of schemes, but it may be lexically equivalent to none. Further, an interpreter may have more than one “home” scheme, without full translatability between them. Indeed, as MacIntyre observes: One of the marks of a genuinely adequate knowledge of two quite different languages by one and the same person is that person’s ability to discriminate between those parts of each language which are translatable into the other and those which are not. Some degree of partial untranslatability marks the relationship of every language to every other.25

Languages not fully translatable between one another need be no barrier to one’s understanding them. MacIntyre emphasizes the point: Notice that this recognition of untranslatability never entails an acknowledgment of some necessary limit to understanding. Conversely, that we can understand completely what is being said in some language other than our own never entails that we can translate what we understand. And it is this ability both to understand and to recognize the partial untranslatability of what is understood . . . [that creates] the predicament of the bilingual speaker.26

By distinguishing between translatability and understandability MacIntyre claims that one can understand two languages or appropriate portions thereof while not being able to translate between them. Indeed, bilingual speakers need to be able to do this in order

25

Alasdair MacIntyre, “Relativism, Power and Philosophy”, pp. 188–9. Alasdair MacIntyre, “Relativism, Power and Philosophy”, pp. 189, emphasis added. 26

52

chapter one

to determine—as he or she does—what is not translatable from one language to another. Just as untranslatability does not entail a limit on understanding, understanding does not entail translatability. MacIntyre’s point undercuts Davidson’s association of untranslatability with non-understandability. So even if one were to hold fixed non-translatability as a criterion of conceptual schemes, it would not follow that one would not be in a position to recognize an alternative conceptual scheme as a bona fide case. Without translatability we need not remain silent about an alternative conceptual scheme. The putative “alternative” need not be absorbed into the home conceptual scheme. One can still distinguish conceptual schemes. Hilary Putnam agrees when he says: “If one recognizes that the radical interpreter himself may have more than one “home” conceptual scheme, and that “translation practice” may be governed by more than one set of constraints, then one sees that conceptual relativity does not disappear when we inquire into the “meanings” of the various conceptual alternatives.”27 In short, mutual understandability does not entail translatability, and translatability is not necessary for understandability.

5 Here now are my own misgivings about those relativisms that are tied to conceptual schemes and their cognates. As I said in my preliminary remarks, the problem with such relativisms concerns not their incoherence but their adumbrance. The schemes relative to which truth or rightness is supposed to operate often cannot be clearly delineated. Where, after all, does a guilt culture and a shame culture begin and end? Where, after all, does an Indo-European language and a Hopi language begin and end? For that matter, where do Chinese and English begin and end? The difficulty concerns the fact that, in saying that truth or rightness is relative to such schemes, there are no ready application procedures. If I say that truth is relative to my scheme, which is mine? What if I have several schemes

27 Hilary Putnam (1989), “Truth and Convention: On Davidson’s Refutation of Conceptual Relativism”, in Relativism: Interpretation and Confrontation, edited by Michael Krausz, pp. 180–1.

relativism and its schemes

53

that blend into one another? What if I am a Chinese American New Yorker feminist? My misgiving concerns the applicability of the relativist rubric upon pertinent schemes. That does not mean that such schemes are incoherent. It only means that there is a problem in applying the notion of a scheme to do the job that relativists standardly demand of it. Schemes are characteristically adumbrant. And they may overlap in complicated and unstable ways. The adumbrance of schemes does not disqualify understanding between inhabitants of each. But typically it is unclear just what truth or rightness is meant to be relative to. Of course, as degrees of adumbrance may vary, the severity of the problem of applicability may vary. In sum, I join Davidson in disallowing the scheme-content dualism, and thereby those relativisms which assume it. However, not all relativisms need assume the offending dualism. Further, I allow the coherence of the idea of conceptual schemes and their mutual understandability, thus dislodging the link between non-translatability and non-understandability. Finally, in face of the characteristic adumbrance of schemes, those relativists who do not assume the offending scheme-content dualism still need to provide procedures under which their relativism can apply. Or perhaps we should turn to non-standard relativisms that define themselves along lines quite distinct from conceptual schemes altogether.

CHAPTER TWO

DAVIDSON AND CHINESE CONCEPTUAL SCHEME* Koji Tanaka

1. Comparative Philosophy Comparative philosophy is a branch of philosophy which examines and contrasts different traditions of philosophy. For example, a comparative philosopher may examine African, Buddhist, Chinese, Indian, Muslim, and Western traditions of philosophy in comparison with one another. Comparisons may also be made between sub-traditions within a tradition: one may compare Confucianism and Daoism within the tradition of Chinese philosophy, for example. The particular concern for us in this paper is the comparative study of Chinese and Western traditions of philosophy. One may question the legitimacy of examining Chinese philosophy and Western philosophy in general terms because it suggests a unified way of talking about these philosophical traditions. Now, it is certainly true that there are disagreements within a tradition. These disagreements are sometimes the very force behind the development of a philosophy. However, the fact that philosophers within a tradition are all engaged in a meaningful philosophical debate suggests that they share something in common. Otherwise, it would not be clear how they are able to communicate with each other and consequently contribute to the development of their tradition. Now, in comparative philosophy, as Larson points out, generalization, or what he calls ‘holistic characterizations’, of a tradition

* This paper was presented at the international conference “Philosophical Engagement: Davidson’s Philosophy and Chinese Philosophy” held in Beijing, China, in June 2004. Many thanks go to the members of the audience whose comments forced me to clarify the main point of the paper. Also I would like to thank Barry Taylor for clarifying the point of Davidson’s argument. I have also greatly benefited from the comments of two anonymous referees as well as those of Bo Mou. Many thanks also go to Bronwyn Finnigan for proof-reading the paper.

56

chapter two

may be misleading.1 However, without any generalization, it is not clear how we can even start an examination of a foreign tradition. As Kasulis argues, “[w]ithout the generalization, however, people cannot proceed in their quest for understanding anymore than they can use most databases without first defining the fields of entry”.2 In fact, the opposite seems to be more harmful. If there is no generalization that can guide an investigation, all we can do is superimpose our framework of thought onto others. As any comparative philosopher would agree, that is not the way that the activities of comparative philosophy should be carried out. For these reasons, some generalization seems necessary in comparative philosophy. However, for a generalization to be possible, it must be assumed that there is something common that is shared by the members of the tradition. If this is true then it seems reasonable to suggest that comparative philosophy is a way of understanding the philosophies of different traditions: a generalization of the commonalities which underlie a philosophical tradition is thought of as a way of understanding the tradition itself. In fact, Parkes suggests exactly this: The major concern of comparative philosophy, however, is an understanding of the philosophies themselves—and thereby, to a greater or lesser extent, of the world.3

Yet how can we discover the common features that can be attributed to a tradition? And how can the comparisons of traditions be made possible based on those commonalities? If comparative philosophy involves generalization, such questions seem to be fundamental to comparative philosophy. Unless we can answer these questions, it is not clear how one can comparatively philosophize about different traditions.

1 Larson, Gerald James (1988), “Introduction: The ‘Age-Old Distinction Between the Same and the Other’”, Interpreting Across Boundaries, G.J. Larson and E. Deutsch (eds.), Princeton: Princeton University Press. 2 Kasulis, Thomas P. (2002), Intimacy or Integrity, Honolulu: University of Hawaii, pp. 7–8. 3 Parkes, Graham (1987), “Introduction”, Heidegger and Asian Thought, Graham Parkes (ed.), Honolulu: University of Hawaii Press, p. 3.

davidson and chinese conceptual scheme

57

2. Davidson on Conceptual Scheme Whether one considers it with pleasure or sorrow, the answers to the above questions are hard to come by. The difficulty can be enunciated by referring to Davidson’s discussion of conceptual relativism.4 Davidson argues that the notion of conceptual relativism is incoherent. The notion that he rejects is based on what he calls conceptual schemes: Conceptual schemes, we are told, are ways of organizing experience; they are systems of categories that give form to the data of sensation; they are points of view from which individuals, cultures, or periods survey the passing scene.5

Conceptual relativism is then the doctrine that what is considered to be real is relative to a conceptual scheme: Reality itself is relative to a scheme: what counts as real in one system may not be in another.6

An example of this claim is made by Whorf who famously argued that the Hopi conceptualize reality differently from Westerners primarily due to a different conception of time. In arguing against conceptual relativism, Davidson associates a conceptual scheme with a language. We may accept the doctrine that associates having a language with having a conceptual scheme. The relation may be supposed to be this: where conceptual schemes differ, so do languages.7

However, a speaker of one language does not necessarily organize their experience differently from a speaker of another language. This is because a language may be translated into another by invoking only one conceptual scheme. And this is a conceptual scheme that they both share. For example, ‘blanc’ is a French word for ‘white’. One need not think that ‘blanc’ and ‘white’ express two different concepts; rather, they both express the one concept, i.e., white or whiteness. That is, the words ‘blanc’ and ‘white’ can be translated

4 Davidson, Donald (1984), “On the Very Idea of a Conceptual Scheme”, Inquiries into Truth & Interpretation, Oxford: Clarendon Press, pp. 183–98. 5 Davidson, 1984, p. 183. 6 Davidson, 1984, p. 183. 7 Davidson, 1984, p. 184.

chapter two

58

into each other. Hence, as Davidson writes, “Studying the criteria of translation is therefore a way of focusing on criteria of identity for conceptual schemes”.8 Conceptual relativism thus means the failure of the ‘intertranslatability’ of languages. This is because conceptual relativism denies that languages share a conceptual scheme, which is the criterion for translation. Showing the existence of languages which are not translatable is tantamount to making a case for conceptual relativism. We may then ask the question of whether there are any languages which are untranslatable. In addressing this question, Davidson considers two kinds of failures of translatability: complete and partial. A case of complete failure is when no significant portion of a language can be translated into another language. Davidson examines two metaphors that may be considered to show the complete failure of languages. The first is that languages organize experience or the data of sensation; another is that languages fit experience or the data of sensation. According to the metaphor of organization, two languages are (completely) untranslatable if they organize the same experience in such a way that no significant portion of the languages can be intertranslated. Here the two languages must be said to organize one experience. However, Davidson argues, “[t]he notion of organization applies only to pluralities”.9 We cannot speak of organizing one thing, such as a closet, without referring to the plurality of things in it. The problem with the case under consideration is that it involves only one thing: one and the same experience. Hence, there is no coherent sense in which two untranslatable languages can be said to organize this single experience. Thus the metaphor of organization does not make the idea of untranslatable languages coherent. The metaphor of fitting faces much the same difficulty. To say that a language fits experience or the data of sensation is to accept the sentences of the language as (largely) true. Hence, the criterion of untranslatable languages is that they are “largely true but not translatable”.10 The problem is that the notion of truth is not independent of translation. Two different languages may be (largely) true with respect to one experience. Yet the recognition of this assumes 8 9 10

Davidson, 1984, p. 184. Davidson, 1984, p. 192. Davidson, 1984, p. 194.

davidson and chinese conceptual scheme

59

translatability of the languages.11 Without this assumption, we could not demonstrate that these two languages truly express the same experience. Hence the metaphor of fitting does not establish untranslatable languages. Therefore, neither of the two metaphors establishes a case for (completely) untranslatable languages. Now the case of partial failure may be thought to be more plausible. It assumes that one language shares something in common with other languages. The suggestion is that some ‘common part’ acts as a neutral ground from which the failure of intertranslatability can be identified. Plausible candidates for such common parts are beliefs and other mental attitudes that produce certain speech acts. However, it is not clear how we can understand a person’s beliefs without understanding their language, which itself is assumed to be untranslatable. Furthermore, we cannot understand a person’s language if we do not know their beliefs and attitudes which serve as the neutral ground. Thus, it is not clear upon what basis we can understand the failure of intertranslatability of a person’s language. We may respond to the case of partial failure by arguing that two people can come to have a general agreement on beliefs. They may simply stipulate what is to be believed for the sake of communication. However, such agreement could only be reached in the same language. Such a situation would not invoke a language which had partially (or completely) failed to be intertranslated. If the differences can be explained and described at all, that would be because of the common language that was used to do so. Thus, again, an untranslatable language cannot be identified. Therefore we could not be in a position to judge that others had concepts or beliefs radically different from our own.12

Moreover, if we cannot intelligibly say that schemes are different, neither can we intelligibly say that they are one.13

11 Davidson explains this point by referring to Tarski’s T-scheme or what he calls “Tarski’s Convention T” (Davidson, 1984, p. 194). I have omitted such details for the benefit of those unfamiliar with this T-scheme. 12 Davidson, 1984, p. 197. 13 Davidson, 1984, p. 198. We can frame the problem that Davidson points out in terms of Hansen’s reading of Zhuang Zi. (Hansen, Chad (1992), A Daoist Theory

60

chapter two 3. . . . and Comparative Philosophy

Davidson’s thesis seems to have a consequence for comparative philosophy, in particular for a comparative study between Chinese and Western traditions of philosophy.14 It is often thought that these traditions differ conceptually. A comparative study between them is considered to be based on comparisons between two different conceptualizations. Consider, for instance, the notion of personal identity. As is pointed out by many, for example Rosemont, (contemporary) Western philosophers largely view persons as atomic and autonomous individuals.15 On the other hand, the Chinese (at least early Confucians), as argued by Rosemont, think of persons as ‘roles’, particularly family and social roles such as child, parent, teacher, neighbor and so on. Hence, for the Chinese, to be a person is conceptualized to be part of a network of relationships that are defined by roles. Despite the fact that the Chinese concept of personal identity seems radically different from the Western concept, Western philosophers seem to be able to understand the Chinese concept of personal identity and vice versa as above. This may indicate that the above two concepts are, in fact, not radically different. However, for Western philosophers to learn this new Chinese concept of personal

of Chinese Thought, Oxford: Oxford University Press.) If Hansen is correct in his interpretation, Zhuang Zi was concerned with the basis from which we accept and reject social ‘conventions’. Zhuang Zi argues that there is no perspective which we can accept to settle the issue of which convention to accept and which to reject because the basis upon which we accept and reject the convention must also be in question. All we can do is accept or reject a discrimination pattern based on a certain perspective, and this perspective is one we happened to have accepted for whatever reason. Thus, there is no way to get to the common basis from which we can settle the issue. However, given that this paper is written in English, addressing issues to English-speaking philosophers, whether Zhuang Zi’s perspective can be presented and understood at all is precisely what is at issue. 14 Davidson’s primary target is in fact the very idea of a conceptual scheme. His rejection of the idea is often thought to be based on a rejection of the content/scheme distinction. Granted that he does reject the content/scheme distinction, Davidson’s thesis is more than the rejection which prefigured in Quine. (W.V.O. Quine (1951) “Two Dogmas of Empiricism”, Philosophical Review, Vol. 60, pp. 20–43.) Davidson rejects the very idea of a conceptual scheme by rejecting conceptual relativism. My primary focus in this paper is Davidson’s unique insight into the incoherent nature of conceptual relativism, as it is conceptual relativism that has some bearing on comparative philosophy. 15 Rosemont, Henry Jr. (1991) A Chinese Mirror—Moral Reflections on Political Economy and Society, La Salle: Open Court.

davidson and chinese conceptual scheme

61

identity, they must share something in common with the Chinese. As Davidson explains in another place: We tend to discover our common problems and interests as we read, teach, and travel. The discovery surprises us for, to begin with, minds are best compared by finding as many points of similarity as everyday patterns of action and reaction afford. But once this fitting of pattern to pattern is accomplished, the remaining differences loom out of proportion. This perhaps explains why a first exposure to a new tradition seems to reveal an unbridgeable gap. What experience shows, though, is that, as in other areas, differences are to be understood only as seen against a background of underlying agreement. The underlying agreement may be largely unspoken and unnoticed, but it is always available. Sometimes we need help in appreciating how philosophy builds on what we all know. No world views or conceptual schemes are truly incommensurable.16

If Davidson is correct, what strikes us at first glance as different may in fact share a ‘background of underlying agreement’. This seems to be the reason why Western philosophers can come to understand the Chinese concept of personal identity and vice versa.17 Perhaps it is the case that Western philosophers charitably interpret the Chinese concept of personal identity based on what they know about personal identity. Moreover, the fact that Chinese and Western philosophers can communicate with each other may show that they share a ‘background of underlying agreement’: the possibility of communication between them is constituted by a background of underlying agreement (to give it a Kantian ring). However, if the background of underlying agreement is thought of in terms of the ‘sameness’ of the ways in which Chinese and Western philosophers conceptualize personal identity, as is often done in a comparative

16 Davidson, Donald (2001) “Foreword”, Two Roads to Wisdom?—Chinese and Analytic Philosophical Tradition, B. Mou (ed.), Chicago and La Salle: Open Court, p. v. 17 It is difficult to talk about Davidson’s philosophy without talking about language. In this paper, I take for granted that (classical) Chinese differs from English to the extent that the issue of translation arises. But comparative philosophers often take the issue of translation as a rather trivial matter given the superficial difference between the two languages. They often shift the focus from languages to concepts (or conceptualizations) by placing a primary importance on concepts that can be materialized in languages. (This seems to be the way that comparative philosophers think of the relationship between languages and concepts.) For this reason, I am primarily concerned with concepts in this paper, while acknowledging that a properly Davidsonian analysis of comparative philosophy should take place in the consideration of intertranslations of languages.

62

chapter two

study of Chinese and Western traditions of philosophy,18 then it is not clear that they could be in a position to judge that they are in fact conceptualizing the same thing. If this is correct, then Western philosophers (or Chinese philosophers) could not be in a position to judge that the Chinese concept (or Western concept) is a different concept of personal identity. Let us narrow our focus on the problem. As we saw above, we seem to be able to compare Chinese and Western conceptions of personal identity, whether the difference is radical or not. However, for a comparative study of the conceptions of personal identity to make sense, both conceptions must be recognized as conceptions of personal identity. Yet, what will guarantee that they are both conceptions of personal identity? What makes the above Chinese conception a conception of personal identity? Given that it appeals to family and social roles, why isn’t that considered to be a conception of family and social identity? How can Western philosophers know that the Chinese philosophers have built their conception of ‘personal identity’ on what Western philosophers all know about personal identity? Western philosophers cannot refer to their own experience of familial and other social interactions to answer this question since they do not necessarily conceive of such interactions as matters of personal identity. They cannot even refer to experiences of interacting with other individuals, since the Chinese may not view these interactions as interactions between persons. Furthermore, to say that the Chinese conception truly expresses the experience which is viewed as a matter of personal identity by Western philosophers is to assume that the Chinese and Western conceptions are both conceptions of personal identity and that, therefore, they can be translated into each other as two different conceptions of personal identity. However, this begs the question. Thus Western philosophers cannot make a case for the idea that the Chinese conception expressed above organizes or fits the experience they conceive of as being a matter of personal identity. Analogous considerations show that Chinese philosophers cannot demonstrate that the Western conception organizes or fits the experience which they view in terms of personal identity. Therefore, in the same way that Davidson argues against the idea of two ways 18 I do not claim that this is the only way in which comparative studies of Chinese and Western traditions of philosophy are carried out. However, this seems to be the assumption that is made in much literature on comparative philosophy.

davidson and chinese conceptual scheme

63

of conceptualizing the same experience, we can cast doubt on the possibility of comparing two different ways of conceptualizing personal identity. At the very least, neither Chinese nor Western philosophers seem to be in a position to judge that they have two different conceptions of personal identity. This seems to suggest that a comparative study between Chinese and Western traditions of philosophy cannot be based on the examination of different ways of conceptualization. How, then, is comparative philosophy possible? In the remainder of this paper I will consider two approaches to comparative philosophy that seem to have been adopted by many comparative philosophers.19 I will examine them to see whether they provide any plausible answers to the above question. In particular, I will examine whether or not those two approaches fall pray to Davidson’s argument.

4. Special Logic Resort One popular suggestion for explaining the foundations of comparisons is (or was once upon a time) that each philosophical tradition has a ‘special logic’ of its own. According to this suggestion, as Hansen puts it, “What makes sense to Chinese does not make sense to Westerners”.20 For example, in the case of personal identity, what makes sense to the Chinese is that to be a person is to be in a network of relationships defined by family and social roles. However, this notion of personal identity does not make sense to Western philosophers. For them, a person is an atomic and autonomous individual who exists in some way prior to the society. In this way, it is argued that comparisons of Chinese and Western philosophies can

19

I do not claim that there have been only two approaches to comparative philosophy. Whether or not there are other approaches, I leave for another occasion. 20 Hansen, Chad (1983) Language and Logic in Ancient China, Ann Arbor: University of Michigan Press, p. 20. There has been a vast amount of literature presenting non-Western philosophies in this way, for example D.T. Suzuki in his presentations of Zen Buddhism. See also Hansen, 1983, ch.1 in the context of Chinese philosophy. One thing which is never clear in such literature is their use of ‘logic’. Logic, for a contemporary Western logician, is about validity of arguments. It is about what proposition or sentence, i.e., conclusion, follows from what propositions or sentences, i.e., premises; it is not about what propositions or sentences make sense. The approach considered in this section does not seem to conform to this standard account of logic.

64

chapter two

be made based on the differences in ‘logics’ which dictate what ‘makes sense’. However, the special logic resort, as Hansen identifies it, seems to make comparisons untenable. If the Chinese have a logic which is not shared by Westerners and if a certain logic dictates what makes sense, Western philosophers, with their own logic, would be unable to make sense of the philosophical discourses of the Chinese. Yet, if the Chinese indeed had a special logic which prevented Western philosophers from understanding their discourses, the Westerners would be unable to recognize Chinese philosophy as meaningful at all, let alone as being a philosophy. In such case, it is not clear how Western philosophers can even have the idea that the Chinese have a special logic in the first place. Similarly, Chinese philosophers cannot discover anything about the Western special logic either.21 If we understand ‘logic’ according to the special logic resort, it seems implausible to suggest that a special Chinese logic underlies Chinese philosophy and that it is the special logic that makes Chinese philosophy differ from Western philosophy and vice versa. Thus, the special logic resort falls prey to Davidson’s discussion of the complete failure of translatability. Since it cannot be identified as anything meaningful, Chinese ‘philosophy’, let alone the Chinese conception of ‘personal identity’, cannot be compared to Western philosophy. Indeed, according to the special logic resort, there is

21 See Hansen (1983, p. 14) who also argues in the same way: “If Western minds are incommensurably different from Chinese minds then we could not discover anything at all about Chinese thought—including the actual details of the “logic” of Chinese.” Note also that the recent psychological studies such as Nisbett, Peng, Choi and Norenzayan seem to face the same difficulty. (Nisbett, Richard, Kaiping Peng, Incheol Choi and Ara Norenzayan (2001) “Culture and Systems of Thought: Holistic Versus Analytic Cognition”, Psychological Review, Vol. 108, pp. 291–310.) Nisbett et al. argue that cognition of East Asian people is different from Western people. If this is true, it is not clear how anyone can cognize that the cognitive processes of two peoples are different in the first place. This difficulty is also applicable to the philosophical discussion on ‘ethno-epistemology’ by Nichols, Stich and Weinberg, and Weinberg, Nichols and Stich. (Nichols, S., S. Stich and J.M. Weinberg (2003) “Metaskepticism: Meditations in Ethno-Epistemology”, The Skeptics: Contemporary Essays, Steven Luper (ed.), Aldershot: Ashgate, pp. 227–247. Weinberg, J.M., S. Nichols and S. Stich (2001) “Normativity and Epistemic Intuitions”, Philosophical Topics, Vol. 29, pp. 429–460.) They argue for the ethnicity of intuitions that underlie philosophical arguments by claiming that intuitions are culturally bounded. If this is true, it is not clear how we can intuit that our intuitions are culturally bounded.

davidson and chinese conceptual scheme

65

nothing that Chinese philosophy can be compared to. The same must be said about Western philosophy.

5. Comparisons of Presuppositions In recent years, a more sophisticated approach to comparative philosophy has developed. For example, such an approach can be found in the works of David Hall and Roger Ames. Hall and Ames,22 while primarily focusing on comparative philosophy rather than Chinese philosophy per se, analyze Chinese philosophy by illuminating its underlying presuppositions in contrast to those which underlie Western philosophy. For Hall and Ames, presuppositions are “usually unannounced premises held by the members of an intellectual culture or tradition that make communication possible by constituting a ground from which philosophic discourse proceeds”.23 Based on these presuppositions, it is claimed, the Chinese tradition of philosophy has built up a stock of concepts, which form their beliefs, knowledge, arguments and so on. Hall and Ames24 approach Chinese and Western presuppositions with an attention to the historical development of the philosophical system of each tradition. They provide a historical study describing how the first problematic, or analogical/correlative, thinking came to dominate Chinese philosophizing and how the second problematic, or causal, thinking came to dominate Western philosophizing. Thus, English speakers associate the expression ‘Heaven’ with a spiritually transcendent realm that is associated with God who is responsible for the causal order of the universe. Chinese speakers, on the other hand, associate ‘tian’ (often translated as heaven) with an ancestral structure which analogically adumbrates their family structure. Hall and Ames explain this difference in terms of different ways of thinking: causal and analogical. They seem to think of these two ways of thinking as grounding the ways in which concepts are related to each other. Because of the dominance of causal thinking in Western

22 Hall, David L. and Roger T. Ames (1995), Anticipating China, Albany: State University of New York Press. 23 Hall, David L. and Roger T. Ames (1987), Thinking Through Confucius, Albany: State University of New York Press, p. 11. 24 Hall and Ames, 1995.

66

chapter two

philosophy, Westerners associate ‘Heaven’ with a spiritually transcendent realm and God. The Chinese, on the other hand, associate ‘tian’ with an ancestral and family structure because of their analogical thinking. Hence, Hall and Ames illuminate different sets of associated concepts and the nature of their association. It is in this way that they contrast Chinese and Western traditions of philosophy.25 However, analyzing the foundation of comparative philosophy in this way seems to face the problem that Davidson points out. As many comparative philosophers have noticed, the stock of concepts within each tradition seems to be the product of the specific historical and cultural circumstances of the tradition. Each set of presuppositions is a product of the historical and cultural development of each tradition, as Hall and Ames themselves demonstrate.26 Kasulis succinctly describes the situation thus: [Philosophy] necessarily draws on a cultural legacy for its terminology, conceptualization of problems, and even relevance. Philosophy develops not in total isolation but within a community of discourse.27

However, that a tradition is indeed based upon certain historical and cultural circumstances seems to create a difficulty for accessing concepts that are foreign to the tradition. Hence, it is not clear whether or not we can judge that foreign concepts are the same or different from our own concepts, as the methodology for and the concepts used in our judgments will be historically and culturally sensitive. Hence, we still face the problem that we are trying to solve.

6. . . . and the Possibility of Comparative Philosophy Despite the above problem, Hall and Ames’ approach may be thought of as providing the foundation of comparative philosophy. This may 25 Whether or not they would agree with my description of their study, we seem to be able to present their study in this way without doing any injustice to them. 26 Whether or not this thought has been inherited from Heidegger and Gadamer’s account of hermeneutics, I let the reader to be the judge. 27 Kasulis, 2002, p. 14. See also Rosemont who argues that “our basic cognitive framework—which ranges from our unreflective conception of what it is to be a human being to our assumptions, beliefs, and presuppositions about the general features of the physical universe—is overwhelmingly determined for us by a set of highly specific environmental circumstances ranging from social relations accompanying stages of history and of culture, to the syntactical particularities of our native

davidson and chinese conceptual scheme

67

be argued for in the following way. Consider, for example, the early Confucians who are thought of having a concept cluster of ‘ethics’ that does not involve “the “rights-based” concept cluster of morals developed in modern western culture”.28 Based on their view of persons as autonomous individuals, Western philosophers seem to base their ethical concepts on a notion of rights that bears upon atomic and free individuals. They consider murder, for instance, to be ethically wrong because such act violates individual rights. The early Confucians, on the other hand, base their ethical concepts on the notion of piety, in particular filial piety, that adumbrates a social harmony.29 Thus, the early Confucians would also think of murder as ‘ethically’ wrong, but not for the reason that it is a violation of individual rights but because of an impiety that suggests a social disharmony or discordance in the intricate network of relationships. Hence, by explaining the bases from which the two traditions have built their different ways of conceptualization, i.e., presuppositions, we seem to be able to carry out a comparative study of ethics. Indeed, to claim that the early Confucian ethics is not based on the modern Western rights-based concept cluster seems to be to compare the early Confucian and Western conceptions of ethics, so the argument goes. Hall and Ames30 (and Kasulis)31 seem to explain the above activity of comparative philosophy as follows. Consider one of the gestalt pictures, such as the duck-rabbit picture. When the picture is presented for the first time, it appears to some as a picture of a duck and others as a picture of a rabbit, despite the fact that they are all sighting the same thing. Depending on the visual orientation a person has at the time, the picture may appear and be ‘seen’ differently. Only with the same, or at least similar, visual orientation, is the picture ‘seen’ as the same thing. In order to be able to explain what a person ‘sees’, then, we need to forge an elucidation of the person’s orientation in such a way that others can adopt it and thereby tongue. To be is to be the value of a pronoun form—and an indexical at that.” (Rosemont, Henry Jr. (1988), “Against Relativism”, Interpreting Across Boundaries, G.J. Larson and E. Deutsch (eds.), Princeton: Princeton University Press, p. 37.) 28 Rosemont, 1988, p. 63. 29 One may disagree with this description of early Confucians. However, the point of the paper remains the same even if someone presents a different picture of early Confucians. 30 Hall and Ames, 1995. 31 Kasulis, 2002.

68

chapter two

come to an understanding of what the other experiences. Similarly, the early Confucians’ notion of social harmony, adumbrated by the notion of piety, and the contemporary Western notion of rights may be said to point to the same thing despite appearing to the Chinese as piety and appearing as rights to Western philosophers. In order to see it differently, one would need an elucidation of the other’s presuppositions, for example, their presuppositions about personal identity. In this way, the early Confucian and modern Western philosophers would be able to compare their conceptions of ethics. What the above discussion suggests, so the argument goes, is that the issue of ethics cannot be conceptualized in an absolute manner. In the same way that what is seen in the duck-rabbit picture depends on one’s orientation, the demarcation of what counts as ethics seems to be made only in virtue of the conceptualization performed via a tradition of thought in which the philosophical thought has been cultivated. One may then be led to think that different traditions of philosophy can be compared by illuminating the different ways in which concepts form a coherent whole. For Westerners, the concept of ethics is associated with the concepts of rights and autonomous individuals, whereas for the Chinese it is associated with those of social harmony, piety, and family and social roles that define relationships. However, by associating the expression ‘ethics’ with the concept of social harmony and so on, Western philosophers can come to understand and compare the Chinese concept of ethics with their own Western concept. Similarly, Chinese philosophers can associate the expression ‘ethics’ with the concepts of autonomous individuals and rights. In other words, by exploiting the flexibility of the boundary demarcating what counts as ethics, the Chinese and Western conceptions of ethics can be identified and compared. This seems to mean that the boundaries of comparisons are set by the philosopher who is conducting a comparative study.32 Because 32 Potter seems to suggest exactly this when he writes: “difference in conceptual system is assumed by the investigator, and differences between persons, cultures, or philosophers is thus made to fit differences in meanings of segments, pieces of behavior, and specific categories. In truth, the choice of boundaries across which to compare is the same choice as that among categories of investigation; as we saw, it is the practical concerns of the investigator which ultimately dictate these boundaries, they are not absolutely there to be discovered in the material investigated”. (Potter, Karl H. (1988) “Metaphor as Key to Understanding the Thought of Other Speech Communities”, Interpreting Across Boundaries, G.J. Larson and E. Deutsch (eds.), Princeton: Princeton University Press, p. 24.)

davidson and chinese conceptual scheme

69

of the flexibility and non-absoluteness of boundaries, ‘ethics’ cannot be conceptualized in an absolute manner. Hence, Western philosophers may think that the expression ‘ethics’, for instance, can be associated with piety and so on, even though it is customarily associated with a radically different concept cluster. Though the choice of the boundaries of comparisons may be thought to suggest that a different tradition of philosophy is to be understood in an investigator’s individual way, it is the flexibility and non-absoluteness of the boundaries that give rise to the possibility of understanding other traditions and, thus, that of comparative philosophy. In this way, Hall and Ames’ approach may be thought to provide the foundation for comparative philosophy.

6. Despite all this . . . Despite the above prima facie reasonable consideration, comparative philosophers such as Hall and Ames have not completely overcome the difficulty that we enunciated earlier in the paper. Of course, Western philosophers can examine the Chinese concept of piety and roles. They can also compare it to their concept of rights and autonomous individuals. They can even associate the expression ‘ethics’ with their concept of social harmony and piety by calling it ethics, as we saw above. Similarly, Chinese philosophers can learn the Western concept of rights and autonomous individuals and compare it to their concept of piety and roles. They can even illuminate practical consequences of adopting the modern Western notion of rights-based ethics. However, what could not be done is that Western philosophers demonstrate that the Chinese concept of piety is an ethical concept comparative with their own conception of ethics. Neither could Chinese philosophers demonstrate the Western conception of rights to be an ethical concept. In other words, the problem we enunciated above with respect to personal identity has not been resolved. There does not seem to be any guarantee that Chinese and Western philosophers are talking about, or pointing to, the same thing. For Western philosophers, according to Hall and Ames,33 the discourse of ethics

33

At least, if my elucidation of their approach is accurate.

70

chapter two

involves the notion of free and autonomous individuals. However, our problem is that we cannot show that the notion of free and autonomous individuals is another way of conceptualizing what the Chinese would view in terms of family and social roles. Similarly, it is not clear that Western philosophers would conceive of the notion of piety, based on family and social roles, as a matter of ethics, since the Western concept of ethics is largely based on the notion of free and autonomous individuals. How can Western philosophers come to see the Chinese concept of piety as a concept of ethics while the Western concept of ethics forms a conceptual cluster that does not involve the notion of piety? What makes the association of ethics with the concept of piety legitimate? Analogously, how can Chinese philosophers come to understand the Western concept of rights as an ethical concept given that their conception of ethics is based on social harmony and so on? As is shown by Davidson, an appeal to experience itself does not answer these questions. To see this, consider, for instance, the situation of seeing someone who is murdered. This experience may be organized or fitted in two different ways; one in terms of the concept of impiety and another in terms of rights violation. For Western philosophers, for whom ethics is concerned with rights that bear upon autonomous individuals, this experience would be viewed in terms of ethics. For the Chinese, this experience would be a matter of impiety and social disharmony. However, for Western philosophers to infer from this that the Chinese would thus view the experience in terms of ethics is to assume that the notions of impiety and social disharmony are ethical notions. Yet this is the very assumption that we are trying to establish. As we saw above with respect to personal identity, the provability of this assumption remains to be seen. In fact, if Davidson is right, demonstration of the legitimacy of associating ‘ethics’ with what the Chinese would view in terms of impiety and social disharmony seems to escape Western philosophers’ judgment. Analogous considerations show that Chinese philosophers could not judge that rights violation is a matter of ethics either. One may argue that the reason why ‘piety’ and ‘rights’ are both concerned with ethics is because of their associated terms found in the texts. It seems true that contemporary Western philosophers use the concept ‘rights’ in the same discourse that they use ‘ethics’, and Chinese philosophers do the same with ‘piety’ and ‘social harmony’ (or perhaps ‘ren (humanity)’). However, to conclude from this that

davidson and chinese conceptual scheme

71

Chinese philosophers are also concerned with ‘ethics’ is to assume that the notion of social harmony (or ren) can be translated as ethics or that it has ethical contents, and hence begs the question. Therefore, by rephrasing Davidson’s conclusion, Western philosophers could not be in a position to judge that the Chinese had concepts of, or beliefs about, ethics radically different from their own, nor could they intelligibly say that they are the same. Chinese philosophers, with their own conception of ethics, could not be in an analogous position. The same can be said about personal identity and other concepts.34 This does not mean that comparative philosophy is a meaningless activity nor that comparative philosophy is impossible. Western philosophers can compare their notion of autonomous individuals with the Chinese notion of roles. Based on this comparison, they can compare their rights-based notion of ethics with the Chinese notions of piety or social harmony. If someone wishes to use the term ‘comparative philosophy’ in this sense, I will concede the term. There may also be alternative ways of conducting the activities of comparative philosophy than I have addressed in this paper.35 Unfortunately, however, there are comparative philosophers who seem to have thought that they were sincerely engaged in comparative studies of ethics, metaphysics, epistemology and so on based on the notion of different logics, different concepts or mutually noninclusive stocks of concepts given by different presuppositions through which we organize our experience. This paper casts doubt on the legitimacy of such activities and claims.36

34 This discussion seems to be applicable to philosophy itself. In the same way that there is no criterion to determine how ethics is conceptualized in two different traditions, there seems no criterion to determine how philosophy is conceptualized in two different traditions. What consequences this has for comparative philosophy and my discussion in this paper, I leave for another occasion. 35 I wish I had had an opportunity to pose the question of how comparative philosophy is possible to Davidson himself. 36 Larson complains that comparative philosophers have “favored philosophical boundaries of European thought since Descartes for identifying conceptual problems in general . . . Agenda items for comparative philosophizing have been selected from ethics, epistemology, philosophy of language, and so forth” (Larson, 1988, pp. 11–12). Larson’s complaint may be well justified. However, a more serious problem seems to be the fact that comparative philosophers have thought that they could select conceptual items from ethics, epistemology, philosophy of language, and so forth, and conduct comparative studies of those items based on the notion of different ways of conceptualizing them.

CHAPTER THREE

MAKING ROOM FOR COMPARATIVE PHILOSOPHY: DAVIDSON, BRANDOM, AND CONCEPTUAL DISTANCE Stephen C. Angle

The best comparative philosophy does more than compare. According to three recent accounts, it seeks to “integrate” or “challenge” or “seek truth.”1 Each of these ideas depends on being able to compare ideas, texts, and reasons across philosophical traditions, but each goes beyond comparison to urge that we engage in creative philosophy. The desire to synthesize aspects of “different philosophical traditions that developed in relative isolation from one another” (Yu and Bunnin, 2001, p. 296), however, can seem hopeless. The many differences among the cultures and languages in which these different philosophical traditions are imbedded have led to a set of worries about the coherence of comparative philosophy: doctrines like relativism or incommensurability claim that reasoned comparison or translation across cultural gulfs can be impossible. If comparative philosophy would stick to comparisons, then perhaps these worries could be side-stepped: any two things can be compared in one respect or another. To integrate or challenge or seek truth, though, we must tackle the worries head-on. Over the last three decades, one of the principal resources on which comparative philosophers could draw when they sought assurance has been the work of Donald Davidson. In his Foreword to Two Roads to Wisdom? Chinese and Analytic Philosophical Traditions, Davidson wrote: A first exposure to a new tradition seems to reveal an unbridgeable gap. What experience shows, though, is that, as in other areas, differences are to be understood only as seen against a background of underlying agreement. The underlying agreement may be largely unspoken and unnoticed, but it is always available. Sometimes we need help in

1

See (Allinson, 2001), (Van Norden, 2001), and (Yu and Bunnin, 2001), respectively.

74

chapter three appreciating how philosophy builds on what we all know. No world views or conceptual schemes are truly incommensurable. (Davidson, 2001, p. v)

Davidson has argued convincingly that anything we can recognize as a language must, in principle, be translatable into any other language. As I will elaborate below, this argument has sometimes been misunderstood, but it nonetheless stands as an important support for the enterprise of comparative philosophy. As significant as Davidson’s work has been, however, my thesis in this essay is that comparative philosophers need still more than Davidson’s theory is able to provide. It is not enough to know that translation is possible in principle: we need to be able to talk about conceptual differences with more subtlety, and to reason about what is at stake in overcoming them. Keeping in mind that Davidson’s theory has at its heart theorems that translate sentences of an “object language” into “metalanguage,” note the blasé attitude he expresses in this quote: “Even when the metalanguage is different from the object language, the theory exerts no pressure for improvement, clarification, or analysis of individual words, except when, by accident of vocabulary, straightforward translation fails” [Davidson 1984, 33]. The problem is that in an important sense, our vocabularies are often no accidents. Davidson is not wholly without resources to explain the range of conceptual differences and the dynamics of language change. I will point to some promising ideas in his later work, as well as to some elaborations of his basic theories proposed by others. But in the end, the best solution to these matters is to be found in the work of Robert Brandom. In many respects Brandom’s views are Davidsonian; turning to Brandom is not to abandon Davidson’s core insights, but to recast them in a framework that allows them fuller expression. My goal will be to show why the creative philosophical projects of contemporary comparative philosophy are possible, even though they are often difficult. My first step is to quickly sketch Davidson’s argument and then to distinguish it from two similar but mistaken alternatives. The essence of Davidson’s argument is that a language’s sentences can always be translated into sentences with the same propositional content in any other sufficiently rich language, and furthermore that there are no limits, at least in principle, on how languages can be

making room for comparative philosophy

75

enriched or revised. Let us start with the first half: sentences can always be translated. Davidson contends that anything we can identify as a sentence is amenable to translation, because the interpretation of linguistic behavior cannot get off the ground unless we find a way to identify both what our interlocutor believes, and what she means by her words. Since there are an infinite number of false things one might say in response to given situation, but a limited number of (salient) true ones, Davidson proposes that only a principle of charity can ground interpretation. That is, we start by provisionally interpreting our interlocutor as speaking truly. Interpretation proceeds on this basis: while no particular sentence uttered by our interlocutor must be true, we can only identify falsehoods against the background of broad agreement. This reasoning leads Davidson to conclude that anything we can identify as language can always be translated because there is no other means to identify something as a sentence other than by interpreting it. If all attempts at interpreting the noises or movements of some creatures fail, then we (so far) have no grounds for attributing language to them. And this means that we can have no grounds for attributing radical conceptual gulfs. To be sure, we can temporarily fail to understand someone, but incommensurability—understood as the doctrine that the concepts of two languages differ so radically that they cannot be translated—is impossible. Let me now briefly turn to the second half of Davidson’s argument, as I explained it above, namely that there are no limits, at least in principle, on how languages can be enriched or revised. I think this is clear, albeit almost always implicit, in Davidson’s writing. He comes closest to making this point explicit when he writes about the incredibly flexible ways in which we can understand others, and make ourselves understood, despite grammatical errors and malapropisms. Davidson says that “an interpreter has, at any moment of a speech transaction, what I persist in calling a theory. (I call it a theory . . . only because a description of the interpreter’s competence requires a recursive account.)” However, the theory in question is not unchanging: “The theory we actually use to interpret an utterance is geared to the occasion” (Davidson, 1986, p. 441). We adjust our interpretive theory on the spur-of-the-moment, taking into account new things the speakers says. This leads Davidson to distinguish between prior theory and passing theory:

76

chapter three For the hearer, the prior theory expresses how he is prepared in advance to interpret an utterance of the speaker, while the passing theory is how he does interpret the speaker. For the speaker, the prior theory is what he believes the interpreter’s prior theory to be, while his passing theory is the theory he intends the interpreter to use. (Ibid., p. 442)

What must be shared for communication to succeed, he concludes, is our passing theories. Davidson insists, though, that passing theories do not “correspond to an interpreter’s linguistic competence,” nor can they be said to be “learned, or to be governed by conventions. Of course things previously learned were essential to arriving at the passing theory, but what was learned could not have been the passing theory” (Ibid., p. 443). There are two sides to Davidson’s conclusion. On the one hand, we see that Davidson emphasizes a certain kind of linguistic change, namely our everpresent practical ability to adjust our (linguistic) theories to new (linguistic) evidence. On the other hand, in so doing he rejects a very common way of understanding our linguistic competence, namely the idea of a clearly defined, convention-governed, shared structure which language users acquire and then apply to cases. That is, as he puts it, he abandons “the ordinary notion of a language, [and thus erases] the boundary between knowing a language and knowing our way around in the world generally” (Ibid., pp. 445–6). Both aspects of his argument are relevant to my concerns here. As we proceed, I will argue that while much of what Davidson says here is correct, he goes too far in abandoning a central role for the norm-governed social practice that is language.2 Thinking about language in terms of conventions is problematic, but neglecting any role for norms makes linguistic change more easy, and conceptual difference more trivial, than they really are. By way of further fleshing out how Davidson’s argument works let me now turn to two alternative formulations. The first, from 2 There are two different levels at which norms might be relevant to language: at a general level, applicable to all language use; and at a specific level, applicable only to the use of normative words like “ought.” Sam Wheeler has developed some Davidsonian ideas to argue that the logic of conditional possibility fits how we use “ought” better than the logic of obligation. See (Davidson, 1980) and (Wheeler, 1974). I am not convinced that these two options exhaust the alternatives, but will not enter that debate here. Rather, I confine myself to the first, more general level. I will argue below that contra Wheeler, Brandom’s talk of the ways in which we “commit” ourselves via language use does make sense—and this because the commitment cannot be understood in probabilistic fashion, as he would have it.

making room for comparative philosophy

77

Martin Hollis and Steven Lukes, can appear to be the same as Davidson’s but is actually importantly different in ways that matter a great deal to comparative philosophy. The second is Alasdair MacIntyre’s version of, and response to, Davidson’s argument. I will show that while MacIntyre fails to refute Davidson, MacIntyre does help to highlight a weakness that I will exploit later in this essay. In a series of influential articles published from the 1960s through the 1980s, Hollis and Lukes argued for what they called a “bridgehead”: In the very identification of beliefs and a fortiori of belief systems we must presuppose comonly shared standards of truth and of inference, and . . . we must further presuppose a commonly shared core of beliefs whose content or meaning is fixed by application of these standards. (Lukes, 1982, p. 262)

Lukes adds that “practical everyday beliefs” are “prime candidates for the bridgehead. Truths of this kind cannot in general be assumed not to be shared, since that would be strictly unintelligible” (Ibid., p. 265). Lukes explicitly invokes Davidson to support his position, though he does realize that in arguing for a specific “bridgehead,” he is going beyond Davidson’s more general principle of charity. His reason is that Davidson’s argument is inadequate as a “method of translation. . . . The principle of charity gives us no guidance as to where agreements are to be assumed before disagreements can show up” (Ibid., p. 263). To a significant degree, Lukes is right. Davidson does not provide us with a method of translation, but only an understanding of the nature of linguistic behavior which guarantees that translation of another language will always be possible. In suggesting that Davidson’s argument has provided assurance to comparative philosophers, I have not gone so far as to say that Davidson provides us with all the tools we need. Lukes argues that Richard Grandy’s “Principle of Humanity” is a better guide to translation than Davidson’s Principle of Charity.3 Lukes also suggests that guidance from sociology and anthropology will be needed to settle questions about what counts

3

(Lukes, 1982, p. 264f ). See (Grandy, 1973, p. 443), wherein Grandy explains that the Principle of Humanity tells us that in interpretation, “the imputed pattern of relations among beliefs, desires, and the world [should] be as similar to our own as possible.”

78

chapter three

as good reasons for action or belief in a given context.4 It is likely that he is right on both of these counts, but what is important for our purposes is to see that he is wrong in seeing Davidson and Grandy (or Davidson and the anthropologists) as competitors. Davidson’s argument aims to establish possibility, and is based on very general features of languages; Grandy and the rest are best understood as aiming to help us actually translate in a particular context. Even when engaged in this practical activity, though, we must keep in mind Davidson’s dictum that interpretation is always provisional. I can grant many of Luke’s points, therefore, but must stop short of endorsing the idea of an inflexible “bridgehead.” Hollis supposes that an interpreter might “pin down the native counterparts of English sentences like ‘Yes, this is a brown cow’” in order to establish a bridgehead, and insists that the bridgehead is not an “hypothesis” that can be refuted or confirmed by evidence (Hollis, 1970, p. 215). Davidson has shown us, though, that there is no way other than via interpretation to “pin down” the beliefs or meanings of the natives, so even seemingly secure sentences about brown cows could turn out to be misinterpretations, if we uncovered enough evidence. Hollis seems to think that impossible, but counterexamples are easy to manufacture.5 Distinguishing Davidson from Lukes and Hollis has helped us to see what Davidson can (and cannot) do for comparative philosophy. If we turn to MacIntyre’s effort to rebut Davidson, we will uncover a different kind of limitation to Davidson’s approach. MacIntyre believes that Davidson’s argument has two premises: (1) all we have to do to assure understanding of another culture’s point of view is to translate their language, and (2) nothing that we can identify as a language could resist translation. MacIntyre contends that these premises rest upon: . . . a way of translating texts from alien and different cultures, and of responding to them, which is central to the cosmopolitan cultures of those modern internationalized languages-in-use, such as contemporary Trans-Atlantic and Trans-Pacific English, one of whose central fea-

4 (Lukes, 1982, p. 274n39). Lukes notes that he and Hollis part company at this point; Hollis insists that standards of rationality must be determined a priori, and also is more explicit about the inflexibility of the “bridgehead.” See below, and (Hollis, 1982). 5 For example, see the discussion below about fair-weather vs. foul-weather animals.

making room for comparative philosophy

79

tures is that utterance in them presupposes only the most minimal of shared beliefs. These are languages, so far as is possible, for anyone at all to use, for those who are equally at home everywhere and therefore nowhere. (MacIntyre, 1991, p. 114)

MacIntyre’s idea, which he develops more fully elsewhere, is that certain modern languages that are spoken around the world have been, in a sense, neutered: in order to be usable by people from widely-different cultural backgrounds, these languages have lost some important characteristics that all local languages once had.6 The chief features that he says internationalized languages have lost are: first, naming systems that presuppose certain beliefs on the part of the language’s speakers, and second, a tight relationship between canonical texts expressing “strong, substantive criteria of truth and rationality” and acceptable utterances (MacIntyre, 1988, pp. 377, 384). What we are left with are languages in which “the relationship of a name to what is named will have to be specifiable . . . independently of any particular scheme of identification embodying the beliefs of some particular community” (MacIntyre, 1989, p. 193), and in which formerly canonical texts now serve only as sources for literary allusions, not as standards of truth (Ibid., p. 194). How does MacIntyre think that the putative transition from local to internationalized languages might refute Davidson’s argument? Since the new languages have no tight connections to particular sets of beliefs, MacIntyre reasons, the sorts of obstacles that would stand in the way of translating from Aristotle’s Greek to Confucius’s Chinese, for example, are gone.7 The plausibility of Davidson’s contention that we can translate anything that we can identify as a language into our own language, that is, rests on taking our own language to be one of these neutered modern languages. But once Aristotle and Confucius have been translated into English, they have lost their essential ties to particular beliefs that helped to define their standpoints. Once rendered into English, MacIntyre concludes, they are no longer genuine Confucianism nor genuine Aristotelianism, but merely new menu items for the “modern individualism of aestheticized personal choice” (MacIntyre, 1991, p. 115). Nothing has been done, in other words, to dissolve the incommensurability between the actual languages of Confucianism and Aristotelianism. 6 7

See (MacIntyre, 1988, ch. 19) and (MacIntyre, 1989, especially section 4). See (MacIntyre, 1991).

80

chapter three

I am dissatisfied with MacIntyre’s response to Davidson for two reasons. First, MacIntyre has overstated the extent to which modern languages have been neutered. It seems true that the use of names may have changed somewhat along the lines he describes,8 but the blanket claim that we no longer have canonical texts is surely false, especially in ethics. Many speakers of internationalized English, by no means limited to so-called fundamentalists, still look to the Bible, for instance, as expressing “strong, substantive criteria” of right and wrong. Many others look to more recent classics, like the Constitution. Admittedly, appeal to the Bible or the Constitution does not always settle disputes. For one thing, contemporary America is composed of many overlapping communities with commitments to different sets of canonical texts. It is also true that canonical texts must be interpreted, and that they therefore supply standards of right and wrong only together with the styles of reasoning and traditions of interpretation that have grown up around them. But in this we are no different from our predecessors of any age: texts are never self-interpreting.9 Second, MacIntyre’s whole discussion of internationalized languages seems to me to misunderstand the nature of Davidson’s argument. The essence of MacIntyre’s response to Davidson, after all, is that although it appears that Confucian terms can be translated into English, this is in fact only a watered-down kind of quasi-translation. This response would only be harmful to Davidson, though, if Davidson’s argument had used our apparent ability to translate Confucian Chinese into English as evidence for his conclusion. As we have seen, though, Davidson’s argument does not depend on any specific instances of successful translation, but instead relies on very general features of languages which apply equally to modern and to pre-modern languages. Davidson could easily accept MacIntyre’s claim that current translations from Chinese into English are mere quasi-translations, 8 For those familiar with Saul Kripke’s argument that the meanings of proper names cannot contain descriptive content, MacIntyre makes the following response: “What this argument shows is, not that the names of persons do not or cannot have informational content, but that either they lack such content or it is true of them that their use presupposes commitment to a belief, such that were this belief discovered to be false, the name would not continue to be used in the same way” (MacIntyre, 1988, p. 377). For Kripke, see (Kripke, 1980, Lecture One, pp. 61–2 and passim.). 9 On “styles of reasoning,” see (Hacking, 1982); that texts do not interpret themselves is of course one of Wittgenstein’s most important lessons.

making room for comparative philosophy

81

in fact, because Davidson recognizes that it may be necessary to enrich or revise our language before true translation will be possible. Still, even if MacIntyre has not given us a rebuttal to Davidson’s argument, he has helped to point out a danger. I just said that Davidson’s argument does not depend on any specific instances of successful translation. This is true, but Davidson can sometimes sound like he thinks instances of ready translation across alleged conceptual gulfs abound. About Thomas Kuhn, one of the main proponents of the idea that different scientific languages (or paradigms) can be incommensurable, Davidson writes: “Kuhn is brilliant at saying what things were like before the revolution using—what else?— our post-revolutionary idiom” (Davidson, 1984, p. 184). But Kuhn has emphasized that his work is no simple translation into existing vocabulary, but rather the learning and then teaching of a new language (Kuhn, 1983). In a similar way, MacIntyre helps us to see that there’s a crucial difference between a breezy rendering of Confucian texts into “trans-Atlantic English” and a genuine translation. Even without endorsing either Kuhn’s or MacIntyre’s precise diagnosis of the difficulties that conceptual gulfs can cause, I think we can see that Davidson may have been too blasé. I now turn to efforts to build onto Davidson’s framework a sensitivity to these issues. In 1989, Bjørn Ramberg published a splendid book on Davidson’s philosophy of language in which he tried to show that Davidsonians can satisfactorily account for the phenomena that MacIntyre, Kuhn, and others call incommensurability. Indeed, Ramberg argues that Davidson’s framework is better-suited than other models of semantics to taking seriously talk of meaning change and conceptual gulfs, because it does not make essential use of the idea of reference. He realizes, though, that his goal might appear quixotic: Assimilating the concept of incommensurability to a Davidsonian semantics . . . would appear to be highly problematic. For the incommensurability thesis is not only a denial of the view that the continuity of reference is a necessary presumption of successful communication. It is also, perhaps primarily, intended to positively identify a certain kind of semantic obstruction between would-be communicators—a semantic obstruction which . . . is commonly taken by critics of the idea to be intranslatability. And on the radical-interpretation model of semantics, it is hard to conceive of any such obstruction. It is so hard, in fact, that the radical-interpretation model is generally presumed to yield a powerful argument against the very idea of incommensurability. (Ramberg, 1989, p. 119)

82

chapter three

The “powerful argument” to which Ramberg refers is precisely the Davidsonian argument I have been discussing from the beginning, and endorsing as an important foundation of comparative philosophy. The key to Ramberg’s effort to make Davidsonian semantics fit comfortably with the idea of incommensurability is a distinction he draws between abstract linguistic meaning, which is modeled by the synchronic truth-theories that radical interpretation is said to produce, and the actual “production of meaning,” which takes place via language and is (according to Ramberg) “constituted” by linguistic conventions.10 Davidson’s rejection of “convention” notwithstanding, Ramberg says that our conventions bind us in various ways—with “varying viscosity,” he says—and: We rely on conventions to understand and make ourselves understood. Incommensurability, as a communication breakdown, can be understood as a breakdown of linguistic conventions, caused by changes in use that are too abrupt to be absorbed smoothly, or changes that a particular set of conventions are too rigid to accommodate. Semantically, then, incommensurability is a disruption in the ongoing interpretationthrough-application of our linguistic conventions. (Ibid., p. 130)

With adequate time and changes to our conventions, Ramberg is saying, anything is translatable: that lies at the center of Davidson’s argument. But viscous conventions can cling to us, keeping us from seeing how different our conventions would have to be to correctly interpret some foreign language, such that we repeatedly misunderstand the foreigners.11 Ramberg concludes that “incommensurability is a diachronic relation, not a synchronic one; it is not a relation between structures, but a symptom of structural change” (Ibid., p. 131). I agree entirely with Ramberg’s core contention, namely that the phenomenon that has been labelled incommensurability is (1) no threat to translation or comparison in principle, because (2) it arises out of the diachronic relations between (communities of ) language users, but still (3) has important practical significance. Given the always/in principle possibility of proper interpretation, I believe it

10

(Ramberg, 1989, p. 130), and see also (Ibid., ch. 8). Ramberg writes that “If Azanda magicians and Western physicists were to discuss causality, incommensurability would arise. But is would arise because translation would too often be wrong, which is to say that the interpocutors would frequently believe they were using the same language when actually they were not” (Ramberg, 1989, p. 131). 11

making room for comparative philosophy

83

makes more sense to say that languages are “incommensurate” than “incommensurable,” but this is a minor point. Ramberg’s Davidsonian version of the doctrine of incommensurability promises to simultaneously ground the possibility of comparative philosophy while cautioning us about that enterprise’s difficulties, and explaining those difficulties in terms of the viscosity of our linguistic conventions. It is this last part, the explanation in terms of conventions, that I find problematic. My next goal in this essay is to argue not just that Ramberg cannot successfully appeal to conventions to serve his needs, but also that this is for reasons which come back to haunt Davidson’s picture itself. In the end, therefore, I will not offer here a revised Davidsonian account, but an account which pushes beyond Davidson in some crucial respects.12 I have already introduced something of Davidson’s opposition to basing our understanding of linguistic competence on a “language,” in the sense of a clearly defined, convention-governed, shared structure which language users acquire and then apply to cases. He repeatedly claims that while every-day interpretation is greatly facilitated by the convergence of speech behavior that we usually mark by saying that people speak the same language, “we do well to ignore this practical issue in constructing theories of meaning, of truth, and of linguistic communication” (Davidson, 1990, p. 311). He believes that: . . . what interpreter and speaker share, to the extent that communication succeeds, is not learned and so is not a language governed by rules or conventions known to speaker and interpreter in advance; but what the speaker and interpreter know in advance is not (necessarily) shared, and so is not a language governed by shared rules or conventions. What is shared is, as before, the passing theory; what is given in advance is the prior theory, or anything on which it may in turn be based. (Davidson, 1986, p. 445)

Davidson adds that “of course things previously learned were essential to arriving at the passing theory, but what was learned could not have been the passing theory” (Davidson, 1986, p. 443). Ramberg helps us to see how radical Davidson’s view is when he says that Davidson is arguing, “in effect, that even if we used nothing but malapropisms, communication would still be possible” (Ramberg, 1989, 12 In other writings, I have tried to follow a strategy more like Ramberg’s, albeit without reliance on conventions, but now find the current strategy more compelling. See (Angle, 1994) and (Angle, 2002).

84

chapter three

p. 101). Linguistic meaning is modeled by radical interpretation, which does not (Davidson and Ramberg believe) essentially depend on the notion of a language. While Ramberg follows Davidson in all these matters, he does believe that Davidson’s rejection of language and conventions is too complete, which keeps Davidson from seeing the proper significance of incommensurability. Ramberg writes that: In dismissing the body of conventions that constitute a language as a contingent feature of linguistic communication, as nothing more than a complex, and immensely valuable, practical aid, Davidson obscures the dialectical relation between meaning and what we might call the production of meaning. Meaning, understanding the meaning of an utterance, is what is modeled in radical interpretation. The production of meaning, on the other hand, is modeled by what we call a language. (Ibid., p. 110, emphasis added)

Radical interpretation involves the construction of a truth-theory for a speaker, but Ramberg says that “in a normal speech situation,” we do not advert to truth theories. “Here, conventional strategies, not the construction of truth-theories, determine what truth-conditions we attach to utterances.” In other words, “in so far as we are speakers of a language, the truth-conditions of the sentences of that language are conventionally taken for granted” (Ibid., p. 111). In at least most cases of the production of meaning—that is, actual speaking and interpreting—Ramberg thinks we are constrained by the conventions of our language. Since these conventions change slowly, while radical interpretations adjusts instantly to new evidence, a gap can appear between the radical interpretation of meaning and the conventional production of meaning. What Ramberg means by the “dialectical relation” between meaning and meaning production, therefore, is presumably that radical interpretation only sporadically impacts our production (and consumption) of meaning, such that a “diffusion of meaning [or] a blurring of linguistic understanding” (Ibid., p. 112) becomes possible. This blurring is incommensurability. By restoring conventions to the broader picture of linguistic understanding, Ramberg has made room for conceptual gulfs and persistent failures of understanding, even while preserving the idea that absolute intranslatability is impossible. Or so it might seem. In fact, Ramberg is trying to have his cake and eat it too, by relying on a fuzzy notion of convention. If we sharpen the focus, we will see that Ramberg’s so-called conventions

making room for comparative philosophy

85

cannot do what he wants them to, which in turn will help us see that the relation between meaning and the production of meaning must actually run deeper than Ramberg —or Davidson—realize. Conventions, according to Ramberg, are “heuristic devices,” “strategic shortcuts,” “diachronic generalizations,” and things that are “conventionally taken for granted” (Ibid., pp. 110–1). That is, what Ramberg calls “conventions” are ways that we tend to talk, unconscious habits that we persist in because they are useful. A first point that bears making is that this is not what most philosophers mean by convention. According to one standard account, for instance, a difference between conventions and other regularities in our behavior is that we follow conventions in part because we believe that others will also conform, and we prefer that they do so.13 On this view, driving on the right in the U.S. is not just an unconscious habit, convenient shortcut, or empirical generalization of people’s behavior over time, though it may also be those things: It is a convention established by the beliefs, desires, and intentions of U.S. drivers. So Ramberg’s understanding of convention is unusual. In one way, this is just as well, because as I discuss below, Robert Brandom has shown that the more standard view of convention is inconsistent with Davidson’s central tenet that belief and meaning emerge together. Still, the idiosyncrasy of Ramberg’s conventions suggests that we should ask whether they can be confined to the limited role he envisions for them. Ramberg’s conventions, to repeat, are convenient (but dispensable) shortcuts that facilitate the production of meaning. Our practical dependence on these shortcuts leads us to the blurring of linguistic understanding called incommensurability. But are conventions really dispensable? An analogy that Ramberg uses to flesh out his idea is extremely revealing. He says that “the individual speaker stands in the same relation to the conventions of his/her language as Aristotle’s phronimos does to the virtues.” That is, “If it is a convention to use ‘snow is white’ to say that snow is white, this convention is hammered out only in a series of assertions that snow is white, in just the way the meaning of a moral precept is hammered out for Aristotle in the actions of the phronimos” (Ibid., pp. 111–12). The problem with

13 See (Lewis, 1969). Ramberg says that his own account is consistent with Lewis (Ramberg, 1989, p. 113n1), but I see no evidence of this.

86

chapter three

this analogy is that according to Aristotle, one cannot understand virtue apart from the actions of the virtuous person: they are constitutive of virtue, not dispensable shortcuts. Ramberg makes it sound like our repeated assertions, like the consistent actions of the phronimos, simply imbue us with habits that become hard to shake. Of course habit does play an important role for Aristotle, but virtuous actions are not just any habits: they define the standard of virtue, and as such are normative: We should all emulate the actions of the phronimos. Language, too, has this normative quality, as when we correct ourselves when we realize we have misused an expression. Ramberg seems to think that self-correction can be understood on his pared-down model of convention (Ibid., p. 110), but I cannot see how. If I have come to habitually utter “umm” when pausing to think, but then for some reason fail to do so, should I feel the need to correct myself ? I have begun to suggest here that language must be understood in a more thorough-going normative fashion than Ramberg can allow for, whether or not that means that we have to employ a more fullbodied notion of convention. (In the end, I will agree with Brandom that we must do without conventions, and follow his alternative.) A broad notion of language is essential to understanding linguistic behavior, and not merely convenient. I believe that Ramberg’s and Davidson’s view that “even if we used nothing but malapropisms, communication would still be possible” is wrong. As a supplement to the considerations just offered about conventions, here is a more general argument that Ramberg and Davidson (at his most radical) are missing something crucial. A central feature of Davidson’s view of linguistic behavior is that we are to be understood as possessing (largely implicitly) theories. I mentioned the “prior” and “passing” theories briefly above, and quoted Davidson’s statement that: “I call it a theory . . . only because a description of the interpreter’s competence requires a recursive account.” In much of his work he refers to these theories as truth theories or T-theories, because he believes that truth conditions— that is, the conditions under which a given sentence is true—play critical roles in the theories. Davidson argues that a description of the speaker’s competence requires a theory for two main reasons. First, without a recursive theory able to generate infinitely many new theorems, we would be at a loss to explain our manifest ability to generate and understand new sentences. Second, unless our under-

making room for comparative philosophy

87

standing of a given sentences is systematically inter-related to other sentences (as well as sub-sentential units), interpretation will not be successful. Let me explain. In the imagined case where an interpreter uses radical interpretation to construct a T-theory from the ground up, we have no problem seeing that the theorems she attributes to her interlocutor will be systematic. Clearly, we cannot demand that, in order to have adequate evidence that a theorem is systematic, the interpreter test all possible other theorems and axioms that might bear on the truth of the theorem in question. On the other hand, more is needed than our interpreter hearing her subject utter “It rains every Tuesday,” and, observing that it is raining quite hard, attributing (A) The subject’s T-theory states that “‘It rains every Tuesday’ is truein-L at this time for her iff it rains hard today.”

to the speaker. This is clearly inadequate, since “It rains every Tuesday” does not mean that it rains hard today. Of course, Davidson never suggests that an interpreter should stop with (A). The provisional attribution of (A) is no more than an appropriate step in the long process of puzzling out a speaker’s T-theory, and it would be followed by attempts to determine whether the speaker would utter “It rains every Tuesday” on a clear day, on a day when it rains lightly, and perhaps on a Tuesday without rain. These tests—the very essence of radical interpretation—would lead our interpreter to reject (A) since that interpretation, while adequate to explain her initial evidence, manifestly fails to be systematic, since it is not part of a consistent theory. Exposure to a long series of the speaker’s utterances is therefore required for radical interpretation, as is the assumption that the meanings of the speaker’s words do not change greatly over the course of investigation. Without this assumption, we could have no guarantee that a theorem could be confirmed or disconfirmed by the evidence: If the second time our interpreter heard “It rains every Tuesday” she could not assume that it meant the same as the first time she heard it, she could not take the fact that it was not raining that day as tending to disconfirm her initial hypothesis. Davidson himself even suggests that it will be necessary to assume rough equivalence of meaning across speakers in order to avoid unacceptable level of indeterminacy:

88

chapter three A theory for interpreting the utterances of a single speaker, based on nothing but his attitudes towards sentences, would, we may be sure, have many equally eligible rivals, for differences in interpretation could be offset by differences in the beliefs attributed. Given a community of speakers with apparently the same linguistic repertoire, however, the theorist will strive for a single theory of interpretation: this will greatly narrow his practical choice of preliminary theories for each individual speaker. (In prolonged dialogue, one starts perforce with a socially applicable theory, and refines it as evidence peculiar to the other speaker accumulates.) (Davidson, 1984, p. 153)

The idealized process of constructing a T-theory for a speaker, we can conclude, requires taking a language as the primary object of one’s theorizing. In one essay, Davidson acknowledges that, in some of his early writings, he had neglected the possibility that someone might know a set of theorems without knowing them to follow from a T-theory (and thus without knowing them to be systematic) because he “imagined the theory to be known by someone who had constructed it from the evidence, and such a person could not fail to realize that his theory satisfied the constraints” (Davidson, 1984, p. 173). Davidson does not go on to tell us how someone in a less imaginary situation could be justified in attributing systematic theorems to a speaker, but this is just the issue that we must now face. We have seen that what justifies the radical interpreter in modeling her subject as possessing a T-theory is—among other things— both exposure to a long series of the speaker’s utterances, and assuming that the meanings of the speaker’s words (and perhaps even: the words of speakers in the relevant linguistic community) do not change greatly over the course of investigation. Our problem is to identify conditions that would justify an everyday interpreter in modeling her subject in the same fashion. The answer seems obvious: both conditions are met, roughly, in the course of speaking a language. We construct T-theories, that is, by relying on evidence not from individual utterances, but from a language. If the language has been returned to the center of our theorizing, we have come a long way from the notion that “even if we used nothing but malapropisms, communication would still be possible.” Communication would not be possible under such circumstances. It is worth noting, though, that a language-based approach can also handle the individual idiosyncrasies of T-theories which seem to have driven Davidson to speak of prior versus passing theories. Recall that Davidson did allow that “things previously learned

making room for comparative philosophy

89

were essential to arriving at the passing theory.” I think that a proper appreciation of the force of this admission—so-called passing theories are essentially parasitic on theories that take as their subject a whole language, rather than an occasion of utterance—should convince us to abandon talk of “passing theories” altogether. We begin any linguistic interaction with a theory for the language-in-use of our prospective interlocutor. As we converse, we may have to make some adjustments to that theory. Some will be unconscious, automatic, and passing, as when we take malapropisms into account. Some will be conscious, lasting, and perhaps even carried out with the interlocutor’s help, as when we need to ask “What do you mean by X?” So long as “X” is another word for a concept that we already possess, or stands for a simple concept that we can readily add to our conceptual scheme, such adjustments will pose no difficulties to our communication. Let me review. Comparative philosophers want two things from a philosopher of language. On the one hand, we look for an assurance that—sometimes only after hard work—communication, comparison, and challenge are possible across languages, cultures, and traditions. On the other hand, we do not want these things to seem misleadingly easy, as if there were no difficulties in arriving at correct translations or legitimate comparisons. Thusfar I have argued that Davidson succeeds on the first score but falls down on the second. I looked at Ramberg’s promising-seeming effort to combine Davidson’s argument against untranslatability with the possibility of incommensurability, but showed that both Ramberg and Davidson himself still run into trouble by not taking seriously enough the role of language in linguistic meaning. It is time to look at an alternative that strays farther from Davidson’s theory than did Ramberg, though still without giving up on the core orientation that provides us with the first of the assurances just mentioned. The alternative I have in mind is Robert Brandom’s inferentialist and pragmatist account of semantics. In the next several paragraphs, I aim to show that although Brandom has little to say about radical interpretation, he shares an approach with Davidson that grounds the ultimate possibility of translation of all languages. He carries this out quite differently from Davidson, taking “commitment” to be central instead of Davidson’s stress on truth, but these differences do not overshadow their shared starting point in what Brandom calls a “relational” theory of language and thinking. What is crucial for

90

chapter three

my purposes is that Brandom has found a way to do this that nonetheless has a fundamental place for social norms. While Brandom argues against talking of linguistic “conventions,” he shows that the norms implicit in our practices play central roles in making possible linguistic interactions. This role for norms will allow me to make good on the second need of comparative philosophy, thus completing my task. Recall that Davidson has argued that neither beliefs nor meanings can be established independently of the other. This is illustrated by the plight of a radical interpreter who knows neither what his interlocutor’s words mean, nor what she believes. Davidson’s solution is that we must provisionally assign true beliefs to the speaker: only against a background of agreement can we come to identify things on which we disagree—things that we think are true and she, or they, think are false. In fact, Davidson goes father than this, arguing that only in the context of interpretation can we make sense of the notion of something’s being objectively true, as opposed to our just thinking things are a certain way. Only when we can say something like “She thinks that rabbits are bigger than hares, but I know differently” do we have access to the difference between thinking that things are a certain way, and their actually being that way. (This is independent of whether we are right about the way they are, of course.) By putting interpretation at the center of his theory, therefore, Davidson is able to accomplish two things: (1) show that we must generally interpret people as speaking truly, and (2) show how we come to appreciate the difference between speaking truly and speaking falsely. Brandom follows Davidson’s lead in arguing that “the concepts of objective truth and error necessarily emerge in the context of interpretation.”14 Brandom also puts the interpretive interaction between speaker and hearer at the center of his theory. Like Davidson, he believes that “intentional states” (cf. Davidson’s “beliefs”) and “speech acts” (cf. Davidson’s “meanings”) are “fundamentally of coeval conceptual status, neither being explicable except in an account that includes the other” (Brandom, 1994, p. 152). It will not be lost on readers, though, that “meanings” and “speech acts” are quite different from one another; speech acts (like asserting) are typically thought

14

(Davidson, 1984, p. 169), cited in (Brandom, 1994, p. 152).

making room for comparative philosophy

91

to express meanings. Davidson says: start with someone’s behavior that you think might be linguistic. The only way to arrive at the person’s beliefs and meanings simultaneously is to provisionally fix one, so you apply the Principle of Charity and assume that her beliefs are true. Then try to build a theory of her language. If (in principle) you cannot, then you conclude that she is not speaking a language—her behavior was not linguistic, after all. In contrast, Brandom says: start with that same behavior, which you think might be linguistic.15 To interpret someone as speaking is to treat her as having taken part in a certain kind of practice, the paradigm for which is assertion.16 To perform an assertion is to take on a certain kind of commitment: one becomes socially answerable for one’s performance.17 If I say “There is a red ball on that table,” and there is no ball on the table, or only a blue one, then—subject to correction by further evidence—it seems that I am not playing the assertion game. In the typical case, though, my interpreter will understand me as having taken on various commitments through my assertion, and will understand my utterance as an expression of these commitments. Brandom argues that we should view the (semantic) contents of my utterance in terms of the inferential relations I have licensed by expressing these commitments: my saying that there is a red ball on the table licenses the inference that there is a ball on the table, that there is not nothing on the table, and so on. These 15

This paragraph is based on (Brandom, 1994, ch. 3); see in particular p. 142. Brandom tends to use “interpretation” in the narrow sense, following Wittgenstein, of explicit hypothesis formation. He notes that Davidson has been criticized for thinking of ordinary intralinguistic understanding as this sort of interpretation, though Brandom himself remains silent on whether this criticism is apt. Brandom does allow that his deontic scorekeeping is a “kind of interpreting,” but it is “implicit, practical interpretation” (Brandom, 1994, pp. 508–9). 17 Sam Wheeler has argued (in a personal communication) that since (1) “commitment” carries with it the logic of obligation (according to which the addition of a new premise to a valid argument cannot invalidate the argument), and yet (2) what we “ought” to believe based on our existing beliefs is better understood according to the logic of conditional probability, so (3) Brandom is wrong to talk of commitment. Wheeler writes: “The ‘rules’ of successful language use are not like rules of games, but more like rules of thumb, that is, generalized conditional probabilities.” I think he is wrong about this, and we can see this if we think about one of Wheeler’s examples. The person who believes that “chien” means “dog,” but does not think that a chien is a mammal (because he does not believe that dogs are mammals) is not just rare, but wrong. If we were to encounter such a person, we would expect that when we presented him with the appropriate reason or evidence, he would acknowledge a mistake—that is, the violation of something the use of “dog” committed him to. 16

92

chapter three

various inferences make up the “meaning” of the utterance. It seems clear enough, then, that for Brandom, attribution of intentional states (taking on the commitments which undergird inferences, which Brandom calls “doxastic commitments”) and attribution of the particular sort of performance called a speech act (assertion or whichever) must go hand-in-hand. Brandom says that when we do this, we are acting as “deontic scorekeepers,” by which he means that we keep track of the commitments and entitlements that people we are interpreting as speakers take on. If you utter “There is a red ball on the table,” I score you as committed to a variety of inferences. If you then say “It’s the only thing on the table,” I update my deontic scorebook, since there are now new commitments you have taken on. If you then push the ball off the table, I update the scorebook again: since our shared situation has changed, your “deontic status” has changed without your needing to say, for instance, “Now there’s nothing on the table.” Instead of Davidson’s approach, according to which we attribute sentences “held true,” then, we start with the practice of assertion.18 Brandom is quite explicit about inverting a central principle of Davidson’s, even while preserving the insight it embodies: The attitude of taking-true is just that of acknowledging an assertional commitment. . . . Evidently this principle can be exploited according to two different orders of explanation: moving from a prior notion of truth to an understanding of asserting (or judging) as taking, treating, or putting forward as true, or moving from a notion of asserting to a notion of truth as what one is taking, treating, or putting forward a claim as. (Brandom, 1994, p. 202)

Davidson has made it very clear that he thinks we should start from a prior notion of truth (Davidson, 1990, p. 314), whereas for Brandom, truth has an “expressive” role rather than an “explanatory” role. That is, using the concept of truth permits us to say various things about assertion—to make explicit the connections between some assertions and others, for instance. It is not something that “can be understood in advance of assertion” and used to help us understand assertion itself (Brandom, 1994, p. 202).

18 For instance: “Specifically linguistic practices are distinguished as just the social practices according to which some performances have the significance of undertakings of assertional commitment” (Brandom, 1994, p. 168).

making room for comparative philosophy

93

If Brandom eschews an explanatory role for truth, though, he cannot appeal to Davidson’s Principle of Charity to explain how radical interpretation gets off the ground: We cannot look to an antecendent, shared notion of truth to (provisionally) fix the beliefs of interlocutors, in order to work out what their words mean. Instead, I will argue that Brandom can look to at least three things to explain how communication might get off the ground: shared circumstances, shared inferences, and the default attribution of assertion. Before looking at each of these, let me note where we stand in the argument. Since Brandom agrees with Davidson that understanding someone to be engaging in linguistic behavior—that the noises she or he is making constitute a language—can only be accomplished through successful interpretation, Brandom is on firm ground to reject the idea that there might be an untranslatable language. By way of clinching my account of Brandom as sharing this basic orientation with Davidson, I now propose to explicate how Brandom would motivate the idea that successful interpretation could be possible. To begin with, we share circumstances with those whom we would interpret. There are things in our world with which we both interact. At least in most cases, assertional practice gets its empirical content via what Brandom calls “language entry” and “language exit” transitions: that is, assertions we make upon perceiving something (rather than upon hearing or reading or thinking of something), and things we do upon hearing (or reading or thinking of ) something (Brandom, 1994, p. 222). To borrow Quine’s famous example, a rabbit runs by and our interlocutor says “Gavagai.” Does this mean “There’s a rabbit”? Initially, of course, any interpretation is dramatically underdetermined by the evidence, but if we presume an assertion has been performed, we can begin to try out assigning different sets of “deontic scores.” The process will be a familiar one, and I need not dwell on its details. Two points are important to make, though. First of all, what about Quine’s insistence that his Gavagai example shows how translation is indeterminate? Wouldn’t all situations in which we might translate “Gavagai” as “There’s a rabbit” also be situations in which we could translate it as “There’s an undetatched rabbit part”? Brandom has an ingenuous, though quite technical, answer to this worry, based on the strategy that he has developed to deal with singular terms.19 19

See (Brandom, 1994, pp. 409–12). The basic idea is that the natives’ linguis-

94

chapter three

Second, are we just assuming that our interlocutor picks out objects the same way we do? If so, this starts to sound like Hollis’s “bridgehead” that we rejected above. Mightn’t “Gavagai” refer only to rabbits seen on sunny days, or to some sort of rabbit-like god? Certainly it could. Suppose that we spend some time with our subject, and begin to see that the assertion—and concomitant commitments—we initially attributed was not entirely apt. We might come to understand, for instance, that she sorts animals into fair-weather and foulweather types (Ramberg, 1989, p. 84). This still provides the language-entry transition that Brandom has said is crucial to securing empirical content for our assertions. In addition, Brandom emphasizes that deontic scorekeeping always involves keeping separate track of what an interlocutor takes to follow from her or his commitments, on one hand, and what (as we see it) actually follows, on the other (Brandom, 1994, pp. 185 and 488). Suppose that we have come to see that our subject does take “Gavagai” to signal the presence of a rabbit-shaped deity. We know how to score such an assertion both for her, and for ourselves. That is, when she says “Gavagai,” we attribute to her a commitment to there being a rabbit-god nearby, while we are entitled to inherit a commitment to a mere mortal rabbit being present. This is so even if our backs were turned and we did not see the rabbit.20 One thing that is shared and which undergirds interpretation, in short, is our circumstances—even though, as we have just seen, one can come to see that an interlocutor may interact with those circumstances differently than one does oneself. A second thing that needs to be shared is what Brandom calls “sapience,” which he explains in terms of a responsiveness to reasons (Ibid., p. 5). This does not mean that for us to successfully interpret someone as speaking a language, he must find convincing all of our reasons, but rather that the practice of assertion itself essentially involves committing oneself to the propriety of various inferences, and the notion of inferring is to be understood, says Brandom, as “a certain kind of move tic and other behavior provides us with no reason to attribute complex sortal categories like “undetached part” to them. For Davidson on related worries, see (Davidson, 1984, p. 26n10). For discussion, see (Angle, 1994, pp. 117–19). 20 Brandom gives examples and discusses related issues at (Brandom, 1994, pp. 480–90). Particularly important is the way that pronouns and other anaphoric expressions help us to communicate, even when our differing commitments lead us to mean very different things by our words.

making room for comparative philosophy

95

in the game of giving and asking for reasons” (Ibid., p. 158). Let us return to someone’s having uttered “There is a red ball on the table.” Unless the speaker is thereby committed to a whole range of inferences—and will be responsive to reasoning about them—then she has not, in fact, spoken, but just made some noises that sounded like words. She must deny “There is nothing on the table” and affirm “There is a ball on the table.” In general, our starting point in radical interpretation will be our whole inferential apparatus, that which is made explicit through the use of logical vocabulary but which is implicit in our everyday linguistic practice. So far my discussion of Brandom has only aimed to show that Brandom provides as solid a grounding for comparative philosophy as Davidson does, by showing that Brandom shares Davidson’s ability to rule out untranslatable languages. To bring home my larger argument, it now remains to show that the fundamental role played by social norms in Brandom’s account enables him to make good on what I have been calling the second need of comparative philosophy, namely to show how conceptual differences can be robust and important, even if radical incommensurability is not in the offing. While Brandom argues against talking of linguistic “conventions,” he shows that the norms implicit in our practices play a central function in making possible linguistic interactions. The simplest place to start is with Brandom’s rejection of convention. The most influential account of conventions—which is endorsed by Ramberg (Ramberg, 1989, p. 113)—is that of David Lewis (Lewis, 1969). According to Lewis, conventions are social regularities that are sustained by various beliefs, intentions, and desires of the parties to the convention. In addition to conforming to the convention, they must believe that others will do so, to prefer that everyone so conform, and so on. According to Brandom and Davidson, though, intentions and meanings arrive together: neither can be prior to the other. Brandom cites Davidson as follows: “Philosophers who make convention a necessary element in language have the matter backwards. The truth is rather that language is a condition for having conventions.”21 Without conventions, though, we seem to be without 21

(Brandom, 1994, p. 232), citing (Davidson, 1984, p. 280). Davidson does sometimes say that correct interpretation involves hearer matching how the speaker intended to be understood, which might seem to make intention prior to interpretation; see, e.g, (Davidson, 1986, p. 442). Brandom points out, though, that Davidson

96

chapter three

resources to explain how social norms might structure and constrain our meanings. Brandom’s solution is to look to norms implicit in our practices. Rather than looking to conventions that we can define in terms of prior intentions, look to proprieties (that is, norms or rules) that we acknowledge in practice. Brandom shows that Ludwig Wittgenstien and Wilfrid Sellars both advanced this idea; he cites Sellars as follows: We saw that a rule, properly speaking, isn’t a rule unless it lives in behavior, rule-regulated behavior, even rule-violating behavior. Linguistically we always operate within a framework of living rules. (The snake which sheds one skin lives within another.) In attempting to grasp rules as rules from without, we are trying to have our cake and eat it. To describe rules is to describe the skeletons of rules. A rule is lived, not described.22

Language, too, is lived rather than described. Dictionaries describe what we meant by our words yesterday. Most of the time we still mean the same things today, but if our usage has evolved, then the lexicographers need to catch up. This is not to say that an individual can mean anything she or he wants with a given word: usually, idiosyncratic usages are malapropisms and, if it is socially appropriate to point out the error, the speaker will acknowledge his or her mistake.23 The point, though, is that we authorize our language’s norms by what we do: what we say, the commitments we attribute and acknowledge, and so on. With an understanding of using language as one among the many things that we do, it is straight-forward to see that specifically lin-

“does not take it that the contents of these communicative intentions can be made sense of antecedently, in abstraction from interlocutors’ interpretation of one another” (Brandom, 1994, p. 670n6). 22 (Sellars, 1980, p. 135), quoted in (Brandom, 1994, p. 25). 23 Davidson reflects insightfully on the tension between speaker intention and hearer knowledge in (Davidson, 1991). As his discussion of Joyce’s Finnegans Wake shows, radically idiosyncratic uses are not always mistakes, but can be efforts to “provoke the reader into an involuntary collaboration.” Davidson notes that “coopted into Joyce’s world of verbal exile, we are forced to share in the annihilation of old meanings and the creation—not really ex nihilo, but on the basis of our stock of common lore—of a new language. All communication involves such joint effort to some degree, but Joyce is unusual in first warning us of this, and then making the effort so extreme” (Ibid., p. 11). This seems exactly correct to me: our “stock of common lore” is, in Brandom’s terms, our stock of common commitments, some of which we will come to violate as we interpret Joyce’s language. Thanks to Sam Wheeler for this reference.

making room for comparative philosophy

97

guistic practice must be bound up with many other practices. Brandom in fact emphasizes this when he talks about “language entry” and “language exit” transitions: our words are intimately bound up with what we perceive and do in our world. Like linguistic practice, our many other forms of practice also have norms implicit in them: proprieties and improprieties that shape how we interact with others and with our shared world. We understand ourselves in and through these practices, both linguistic and otherwise. When we seek to make explicit these self-understandings, we often advert to notions like “community.” A community is a group with whom we share (argue, play, eat, shop, reason, etc.). We belong to many communities, and they often overlap and have fuzzy borders. Brandom puts this thought in terms of how, and with whom, we say “we” (Brandom, 1994, p. 3). Sociologists, anthropologists, and historians have explored many of ways in which our practices define us. Or rather, they have explored the ways that we define ourselves through our practices, for, once again, we authorize our practices and their proprieties by engaging in them. If linguistic practices cannot be neatly separated out from the other practices through which we define ourselves, then what “we” can appropriately say becomes a complicated matter. Contrary to Davidson’s blasé remark quoted above, failures of straight-forward, word-for-word translations are often not “accidents.”24 Brandom writes that: “When the prosecutor at Oscar Wilde’s trial asked him to say under oath whether a particular passage in one of his works did or did not constitute blasphemy, Wilde replied ‘Blasphemy is not one of my words’” (Ibid., p. 126). Wilde recognized, that is, that using the word “blasphemy” brought with it certain commitments that he—and other like-minded individuals—rejected, even if he were to deny that a particular passage was blasphemous. Communication, seen here as the shared effort to understand one of Wilde’s writings, temporarily breaks down. Comparative philosophy rarely involves dramatic, face-to-face encounters between alternative communities, but such encounters do help to sharpen the issues involved in comparison and communication. Mario Biagioli gives us such an instance in his discussion of 24 Two particularly relevant studies—relevant because of the ways they examine the inter-dependence of linguistic and non-linguistic practices—are (Bourdieu, 1974) and (Biagioli, 1990).

98

chapter three

communicative breakdowns between Galileo and his Aristotelian rivals; Alasdair MacIntyre imagines difficulties of comparison when Confucians encounter Aristotelians, or when a single individual is torn between two communities.25 According to Brandom, we are to understand the resistance to accommodation and the difficulty in finding common ground experienced by parties to these encounters in terms of the norms implicit in their (linguistic and other) practices. At the same time, Brandom assures us that where there is a will to overcome differences, we can communicate, for there are no untranslatable languages. The courtroom scene I described above was not well-suited to communication about the nature of Wilde’s writings, just as both Galileo and his rivals, according to Biagioli, had something at risk (their “socio-professional identities”) if they successfully communicated. Comparative philosophy, though, seeks to get beyond these barriers. This is not the place to evaluate the specific recent proposals, mentioned at the outset of my essay, in terms of which several comparative philosophers have proposed we can “integrate” or “challenge” or “seek truth” across traditions. My goal has instead been to demonstrate how Donald Davidson and, more completely, Robert Brandom have shown us both why the synthetic projects of comparative philosophy are possible, and why they can be difficult. These synthetic projects lie at the core of what might more properly be called an emerging global philosophy, and they are well worth their trouble.26

25

(Biagioli, 1990), (MacIntyre, 1991), (MacIntyre, 1989). My sincere thanks to my colleague Joe Rouse, for discussions on these themes and for his very helpful comments on an earlier draft. Thanks also to all the participants in the conference on Davidson and Chinese Philosophy, especially Sam Wheeler, Ye Chuang, and Xiao Yang. Sam’s two-barrelled argument that Davidson has a satisfactory account of “norms,” and that Brandom’s is hopelessly mired in the inappropriate logic of “obligation,” is ingeneous and challenging, though in the end I am not convinced, as I have tried to explain. 26

BIBLIOGRAPHY

Allinson, Robert E. (2001), “The Myth of Comparative Philosophyor the Comparative Philosophy Malgré Lui”, in Bo Mou. ed., Two Roads to Wisdom? Chinese and Analytic Philosophical Traditions, Chicago: Open Court. Angle, Stephen C. (1994), Concepts in Context: A Study of Ethical Incommensurability, Ann Arbor: University Microfilms. —— (2002), “Pluralism in Practice: Incommensurability and Constraints on Change in Ethical Discourses”, in Michael Barnhart, ed., Varieties of Ethical Reflection: New Directions for Ethics in a Global Context. Lanham, Maryland: Lexington Books. Biagioli, Mario (1990), “The Anthropology of Incommensurability”, Studies in the History and Philosophy of Science 21, pp. 183–209. Bourdieu, Pierre (1974), “The Economics of Linguistic Exchanges”, Social Science Information 16 (1974), pp. 645–668. Brandom, Robert (1994), Making It Explicit: Reasoning, Representing, and Discursive Commitment, Cambridge: Harvard University Press. Davidson, Donald (1980), Essays on Actions & Events, Oxford: Clarendon Press. —— (1984), Inquiries Into Truth and Interpretation, Oxford: Clarendon Press, 1984. —— (1986), “A Nice Derangement of Epitaphs”, in Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, Oxford: Basil Blackwell. —— (1991), “James Joyce and Humpty Dumpty”, in Peter A. French, Theodore E. Uehling, Jr., and Howard K. Wettstein, eds., Philosophy and the Arts (Midwest Studies in Philosophy, vol. XVI), Notre Dame: University of Notre Dame Press. —— (1990), “The Structure and Content of Truth”, Journal of Philosophy LXXXVII, pp. 279–328. —— (2001), “Foreword,” in Two Roads to Wisdom? Chinese and Analytic Philosophical Traditions, Chicago: Open Court. Grandy, Richard E. (1973), “Reference, Meaning, and Belief ”, Journal of Philosophy (1973), pp. 439–452. Hacking, Ian (1982), “Language, Truth, and Reason”, in Rationality and Relativism, Cambridge: MIT Press. Hollis, Martin (1970), “The Limits of Irrationality”, in Rationality, Oxford: Basil Blackwell. —— (1982), “The Social Destruction of Reality”, in Rationality and Relativism, Cambridge: M.I.T. Press. Kripke, Saul (1980), Naming and Necessity, Cambridge: Harvard University Press. Kuhn, Thomas S. (1983), “Commensurability, Comparability, Communicability”, PSA 1982 2, pp. 669–688. Lewis, David (1969), Convention: A Philosophical Study, Cambridge, MA: Harvard University Press. Lukes, Steven (1970), “Some Problems About Rationality”, in Wilson, Bryan R., ed., Rationality, Oxford: Basil Blackwell. —— (1982), “Relativism in its Place”, in Rationality and Relativism, Cambridge: MIT Press. MacIntyre, Alasdair (1988), Whose Justice? Which Rationality? Notre Dame: University of Notre Dame Press. —— (1989), “Relativism, Power, and Philosophy”, in Michael Krausz, ed., Relativism: Interpretation and Confrontation, Notre Dame: University of Notre Dame Press, 1989.

100

chapter three

—— (1991), “Incommensurability, Truth, and the Conversation between Confucians and Aristotelians about the Virtues”, in Eliot Deutsch, ed., Culture and Modernity, Honolulu: University of Hawaii Press. Ramberg, Bjorn T. (1989), Donald Davidson’s Philosophy of Language, Oxford: Basil Blackwell. Sellars, Wilfrid (1980), “Language, Rules, Behavior”, in J. Sicha, ed., Pure Pragmatics and Possible Worlds: The Early Essays of Wilfrid Sellars, Reseda, California: Ridgeview Publishing. Van Norden, Bryan (2001), “Mencius and Augustine on Evil: A Test Case for Comparative Philosophy”, in Bo Mou, ed., Two Roads to Wisdom? Chinese and Analytic Philosophical Traditions, Chicago: Open Court. Yu, Ji-yuan, and Nicholas Bunnin (2001), “Saving the Phenomena: An Aristotelian Method in Comparative Philosophy”, in Bo Mou, ed., Two Roads to Wisdom? Chinese and Analytic Philosophical Traditions, Chicago: Open Court.

PART TWO

PRINCIPLE OF CHARITY AND CHINESE PHILOSOPHY

CHAPTER FOUR

WHERE CHARITY BEGINS David B. Wong

1. Interpreting the principle of charity The principle of charity, under Donald Davidson’s influential construal, says that we must interpret others on the assumption that they are rational beings, talking about and navigating the same world as we are. Otherwise, we shall not be able to interpret them at all as holding beliefs or making intelligible utterances. Charity directs us to “optimize” agreement between them and ourselves where ever it is plausible to do so. The idea is to make them “right, as far we can tell, as often as possible.”1 Charity, Davidson explained, does not enjoin us from attributing intelligible error. As our interpretation of others takes shape, we might find that it makes better sense to attribute mistakes to them, given our emerging conception of how they are interacting with the objects of their beliefs. However, there is a limit to such attribution. To attribute massive error to them is to undermine a crucial assumption of interpretation: that they are forming beliefs about the same world as we are. If we were to attribute to ancients the belief that the earth is flat, and what is more, virtually none of the other beliefs we have about the earth, we would undermine the assumption that they have beliefs at all about the earth.2 In his recent writings on interpretation, Davidson noted that he previously tended to construe charity in terms of “maximizing” agreement in belief and that a more perspicuous statement of what he had in mind all along is that agreement in beliefs should be

1

Donald Davidson (2001), “Radical Interpretation”, in Inquiries into Truth and Interpretation, 2nd edition, Oxford: Clarendon Press, p. 136. 2 Donald Davidson (2001), “Thought and Talk”, in Inquiries into Truth and Interpretation, 2nd edition, p. 168.

104

chapter four

“optimized.”3 Rather than the “most” agreement, we need the “right sort” of agreement that enables understanding of others. We should try to reach agreement “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.”4 Matters become further complicated with the recognition that it is not just agreement in belief that must be optimized, but also in the desires, values,5 and intentions we attribute to others, since we must understand their behaviors not only in relation to what they believe about the world but also in relation to what they want of it, value in it, and intend to do in it. To make sense of the actions of others, we construe these actions as stemming from intentions, which in turn stem from certain patterns of beliefs, desires, and values. Some patterns make others intelligible to us and others don’t. To see why Davidson’s move from talk of “maximizing” to “optimizing” agreement seems the right move to make but also gives rise to many questions about how to interpret others, consider some uses in ethical theory of the earlier “maximizing” version of charity. David Cooper holds that “We can only identify another’s beliefs as moral beliefs about X if there is a massive degree of agreement between his and our beliefs.” His conclusion is that the principle of charity refutes any significant form of moral relativism. It is difficult to disagree with Cooper’s point that a moral belief must have for its subject matter something connected with “welfare, happiness, suffering, security, and the good life.”6 We would have strong reason to suspect our interpretation of another beliefs as moral beliefs if we were to construe them as having nothing to do with welfare, happiness, suffering, security, and the good life. However, this is not the same as refuting a moral relativism that asserts significant differences are over what is believed about these subjects. Quite a lot depends on 3 Donald Davidson (2001), “Introduction” to the 2nd edition of Inquiries into Truth and Interpretation, p. xix; “Radical Interpretation”, p. 136. 4 Donald Davidson (2001), “On the Very Idea of a Conceptual Scheme”, in Inquiries into Truth and Interpretation, 2nd edition, p. 196. 5 I am going to remain neutral, for the purposes of this paper, on the question of how to understand a person’s values in relation to her beliefs and desires, whether they are beliefs, desires, or some combination thereof. 6 David E. Cooper (1978), “Moral Relativism”, in Midwest Studies in Philosophy 3, pp. 101, 104. I discuss this kind of argument in David B. Wong (1984), Moral Relativity, Berkeley: University of California Press, pp. 114–116.

where charity begins

105

how one applies Cooper’s requirement that there be a “massive degree of agreement” between our moral beliefs and those of another person. It seems arbitrary to say that among competing interpretations of another person’s beliefs, the best interpretation is the one that simply produces the greatest number of overlapping beliefs, even if one had confidence in one’s ability to count beliefs, which I do not. This is one of the correct reasons, I believe, why Davidson has corrected his earlier statements of the principle of charity, from maximization to optimization. Michele Moody-Adams gives a more recent formulation of Cooper’s argument, starting with the premise that understanding others requires that there be “quite substantial agreement about many of the basic concepts that are relevant to moral reflection.”7 She then leaps to the conclusion that “ultimate” or “fundamental” moral disagreement is not possible.8 To validate this conclusion, she must hold that disagreement over concepts relevant to moral reflection is limited to “nonultimate” or “nonfundamental” concepts. This seems a coherent stance to take, but cannot be derived in a priori fashion from a principle of charitable interpretation of others, as Moody-Adams attempts to do. Why must groups have moral beliefs only if they have precisely the same stock of “basic” moral concepts as we do? Why is it not sufficient to have agreement on some critical mass, however the threshold is defined? And with respect to any one basic concept relevant to moral reflection, such as justice,9 what counts as having the same concept? When does one count two very different conceptions of, say, justice, as two different interpretations of the same concept, and when does one say that one has two overlapping but different concepts? And when basic concepts such as justice and compassion represent values, must other people set precisely the same priority as we do on these values when they come into conflict? If they don’t set the same priority, could their disagreement qualify as a fundamental disagreement under some circumstances?

7 Michele M. Moody-Adams (1997), Fieldwork in Familiar Places: Morality, Culture, & Philosophy, Cambridge, MA: Harvard University Press, p. 55. 8 Moody-Adams, 1997, p. 56. 9 Justice and compassion were example of some basic concepts given by MoodyAdams in a paper read by her, entitled “The Idea of Moral Progress”, at the Eastern Meetings of the American Philosophical Association, December 1998, in Washington D.C.

106

chapter four

A Davidsonian principle of charity, properly interpreted as calling for optimization rather than maximization, cannot possibly resolve these questions by itself because the requirement of optimization itself requires interpretation. In illustrating this point, Henry Richardson has pointed out interpreting a philosophical text requires taking account of the cognitive aims the authors had in writing what they did. Is it more charitable, Richardson asks, for a translator of Machiavelli’s The Prince to resolve ambiguities and seek to maximize agreement between Machiavelli and the relevant audience? Or is it more charitable to set him out as intentionally provocative and deliberately cryptic?10 In this paper I accept that charity rules out the possibility of others having beliefs and desires that are different from ours in radical and sweeping ways. Aside from imposing this constraint against radical difference, charity is less a definite principle of interpretation but rather the assortment of the various ways we have of explaining the talk and actions of others. I shall discuss a couple of strategies we employ in interpreting others—analogy and the attribution of rationality—with special reference to the understanding of Chinese thought and culture from an American perspective. To make others intelligible by likening them to us can frequently involve accepting that the analogies we use are extended and rough, consistent with accepting significant differences between them and us. Often, this is good enough, and investigating why this is good enough will lead to the question of who ‘we’ are, when we interpret others to have beliefs and desires similar to ours. I argue that the ‘we’ and the ‘ours’ harbor significant diversity in belief, and that any plausible interpretive approach must presuppose a certain range of diversity in belief and desire over human culture, and in particular in the range of values that are central to particular cultures. My strategy of argument consists in working backwards, as it were, from interpretations of the moral tradition of Confucianism from a contemporary American perspective. Such interpretations are by no means uncontroversial, but they are least eligible as illuminating and plausible interpretations, and that is all I need. The question I want to address is how analogy and models of rationality operate within

10 Henry S. Richardson (1997), Practical Reasoning about Final Ends, Cambridge: Cambridge University Press, pp. 268–269.

where charity begins

107

these interpretations to increase our understanding of Confucianism, both by likening the Confucians to us and by expanding our own sense of human possibility.

2. Interpretation of Confucianism through analogy Let me begin by identifying three significant and distinctive features of the Confucian ethic as it is represented in classical works of Chinese philosophy, primarily the Analects. We understand these features through noting their similarity with themes we find to be familiar and present in our own culture, but as with all analogies, the similarities co-exist with significant differences. In all three cases, I shall claim, we have little reluctance to accept the differences. The first feature is the centrality of xiao ( usually translated as “filial piety.”) It is a common feature of many cultures that one should honor thy father and mother, of course, and it is not difficult to find analogies within American society to Confucian filial piety. At the same time, the Confucian tradition is unusual in the stringency of its duties to parents. The Analects 2.7 identifies the requirements of xiao as going beyond providing parents material support when they are elderly, but more fundamentally showing them it’s not just about giving them food (supporting them when elderly), but jing ( ). Jing originally applied to the attitude one should have when sacrificing to ancestors, an attitude of devotion to carrying out great responsibilities to one’s ancestral spirits. Analects 2.8 amplifies the nature of jing in saying that the young should take on the burden when there’s work to be done and let the old enjoy the wine and food, but that hardly deserves to be called filial. It’s the expression on one’s face or demeanor that is difficult to manage. The scope of duties to parents includes taking care of what they alone could have given one—one’s body. Zeng Zi, one of Confucius’ students, is portrayed in 8.3 of the Analects as gravely ill and near death. He bids his students to look at his hands and feet, and quotes lines from the Book of Poetry to convey the idea that all his life he has been keeping his body intact as part of his duty to his parents. It is only now near death, he says, that he can be sure of having been spared and thus fulfilling this duty to parents. This very idea, that one must keep one’s body intact as a duty of gratitude to one’s parents, has remained a central idea in Chinese culture.

108

chapter four

Why is xiao so central a virtue in the Confucian ethic? Part of the reason seems to be a view about its centrality to the development of ethical character. In Analects 1.2, one of Confucius’ most prominent students, You Zi, says that being good as a son and obedient as a young man (perhaps reading that one is obedient to one’s elder brothers) is perhaps the root of character, the basis of respect for authority outside the family. One learns respect for others first for those within the family. Another part of the reason for the centrality given to filial piety is the need to express gratitude to those who have given one life and nurture. In Analects 17.21, Zai Wo objects to the traditional length of mourning for one’s deceased parents. One year is enough to disrupt one’s normal life in those ways, but the traditional period of three years is too long, he says. Confucius comments on Zai Wo’s lack of feeling in this regard. All children are completely dependent on their parents for the first three years of life, he observes. Did not Zai Wo receive three years worth of love from his parents? The virtue and its rationale have analogues in American culture. We can certainly recognize the themes of gratitude, the need to reciprocate in some fashion for great gifts received, and the conception of family relationships as pivotal in the development of character. Such similarities of theme, however, seem to underdetermine the centrality of filial piety and the stringency of its duties in Confucianism and in the larger traditional culture. While we recognize such rationales for filial piety, we generally do not accord it nearly as central a place in the catalogue of moral virtues, nor do we conceive its duties to be so stringent. And the theme that one owes ones body to ones parents and that it is deep ingratitude not to take care of such a great gift is something that can be understood from an American perspective, but is not generally accepted. Do such differences suggest that we have not correctly understood Confucianism or traditional Chinese culture? I submit that we accept such differences as part of the normal range of human possibility, perhaps because we can imagine ourselves having taken a path we have not taken. The themes of gratitude, reciprocity, and the importance of the family in moral development are familiar to Americans and at the same time have the potential for justifying a value having to do with respecting parents that is far more stringent than the one many Americans accept. And indeed, taking the Confucian per-

where charity begins

109

spective, an American might come to have the strange feeling that that perspective makes more sense. Consider another significant feature of Confucian ethics: its inclusion of an aesthetic dimension in its conception of a good and worthwhile life. Right action in Confucian is fitting action. It expresses appropriate care and respect for others in a manner that befits the nature of one’s social relationship to them and to other, more particular features of their situation and one’s own. There are often conventional forms for the expression of these ethical attitudes, ritual forms called the li ( ). To fashion oneself into a better person is to become practiced in the performance of such li such that they become second nature, but they must always express the appropriate attitudes. To attain the proper balance between form and feeling is to ennoble and beautify human nature. Antonio Cua’s translation of a passage from Xun Zi puts the point nicely: human nature provides raw material, and constructive human effort is responsible for the glorification and flourishing of elegant form and orderly expression. Without constructive human effort, human nature cannot beautify itself.11

It is instructive that Cua draws analogies to the perception of qualities in works of art, likening the grace or joy that can be seen an accomplished ritual activity to the grace of a curve in a painting or joy in a piece of music.12 David Hall and Roger Ames observe that the Confucian notion of the ‘right’ action has much in common with the artist’s choice of the ‘right’ brush or the ‘right’ color in the execution of a painting.13 For the Confucian, doing the right thing means not only doing one’s duty, not just for the right reason, and not just with the right feeling, with the proper grace and elegance that is both an aesthetic end in itself but also bespeaks the ease and contentment of one who has attained the virtues and realized one’s humanity. These uses of analogy help to make intelligible the aesthetic dimension of Confucian ethics, but they also illustrate how 11 Antonio S. Cua (2002), “The Ethical and Religious Dimensions of Li ”, The Review of Metaphysics 55, p. 481, translating from Ti-sheng Li, Hsün Tzu chi-shih, Taipei: Hsüeh-sheng, p. 439. 12 Cua, 2002, p. 483. 13 David L. Hall and Roger T. Ames (2003), “Chinese Philosophy”, in E. Craig (ed.) Routledge Encyclopedia of Philosophy, London: Routledge. Retrieved August 28, 2003, from http://www.rep.routledge.com/article/G001SECT4].

110

chapter four

analogy can help. They take what is familiar to us in one context (in this case, painting or music) and point to its occurrence in a different context. The analogy illuminates if we can conceive of the familiar taking place in that different context. The third feature of Confucian virtue ethics is its emphasis on harmonious relationships as a central part of the ethical life, as illustrated by the Analects on an adult’s relation to his parents. Notice that in 1.2, the crucial dimension of moral development that is started in the family is respect for authority. The strong Chinese preference for harmony emerges in 2.6, where the Master says to give parents no cause for anxiety except for illness. Consider 4.18, where Confucius considers occasions on which one’s own opinions as to what is right or best can conflict with one’s parents’ wishes. One should remonstrate with one’s parents gently, he says. What he says next about one’s course of action if they are not persuaded is ambiguous: you jing bu wei ( ), but whether one translates it as D.C. Lau does, “one should not become disobedient and remain reverent,” or as James Legge does, “Do not abandon one’s purpose to respectfully persuading them,” the value placed on harmony is apparent. The value placed on it, as indicated by 4.18, does not require silence in the face of real disagreement with one’s superiors. Indeed, the Confucian tradition celebrates the scholar-intellectual who says what he thinks about the ruler’s methods and ends, often to the ruler’s face. However, the ends served by such moral courage include the end of harmony. Rulers who fail to govern for the good of the community, the state, and the nation must be called to account precisely for the good of all. Someone must have the authority to settle conflicts, if only in the sense of saying whose view prevails this time. Human beings have yet to invent a society without having to designate such authority and to inculcate some degree of respect for it. The reasons for preferring harmony are quite intelligible, but here again, the reasons underdetermine the degree of preference for it manifested in the culture: e.g., informal negotiation involving interaction and reconciliation between the contending parties is still the traditional way of resolving business disputes in China; informal mediation committees operate to resolve disputes at the grassroots rural village and urban neighborhood levels; and Chinese courts encourage mediation between contending parties even after litigation proceeds have begun. That Chinese culture should show this high degree of preference

where charity begins

111

for harmony does not seem to threaten its intelligibility to those on the outside. Why? Some preference for harmony exists on the American side of the comparison, to which analogy can be made, however major the differences the analogy leaves in place. The American side, after all, embraces a significantly diverse range of subcultures in which a high degree of preference for harmony is shown. These subcultures include, of course, Chinese-American and other AsianAmerican subcultures, as well as Latino and Mexican-American subcultures. Moreover, the various European-descended subcultures of American society have in the past demonstrated a stronger preference for family harmony and cooperation within various levels of community than they do now. It is in part this internal diversity that helps to make Confucian values intelligible as a path we ourselves could and in some cases have taken. So far, I have been assuming that the relevant reference point for understanding Confucianism is the contemporary American perspective, and up until the previous paragraph I have assumed that this perspective is more or less unified. Of course, that is false. In reality, we treat such perspectives as unified only for the sake of certain comparisons, for the sake of certain comparative purposes. In other contexts, and for other purposes, we make much ado about the differences. The unification is at best relative to the purpose of understanding a presumably distant culture that is more difficult to comprehend. The ‘us’ in the comparison between them and us is diverse, and such diversity provides some of the analogies we use to make sense of ‘them.’ This raises the question of how this diverse group became ‘us’ in the first place. If we are limited to beginning from our individual selves as models for understanding others, it seems quite unlikely that we could get the range of beliefs, desires and values that we take for granted even within relatively small circles. I accept that some people are attracted to holding power and exercising it over others, even though my own psychology does not bear much resemblance to theirs in this respect. I accept that some of my students believe in the extreme libertarianism of Ayn Rand, even though there is an inevitable point where I fail to follow their thought processes when they explain them to me. I accept that some people believe that they have been abducted by alien beings, even though there is very little from my own experience that I could use to illuminate whatever experience and thought processes could have led to such a belief.

112

chapter four

I suggest that our self-understandings comprehend significant diversity because the very concepts we use to interpret both ourselves and others embody diversity. Consider the concept of rationality.

3. The model of rationality and the problem of diversity within ourselves The attribution of rationality to others is another dimension of the meaning of charity as an interpretive principle. Not only must be interpret others to be interacting with the same world as ours, but we must understand them as processing their interaction with the world in basically the same way as we do, through the formation of beliefs, desires, values and intentions. Because certain inferential patterns among these elements make sense to us and others do not, as we navigate ourselves through the world, we can make sense of other people’s actions only if we can attribute to them similar enough patterns. We can certainly be justified in attributing errors of reasoning to others. We are familiar with errors of reasoning in our own cases, and more fundamentally, we learn what good reasoning is not only by being given exemplars of good reasoning but also of typical bad reasoning—overgeneralization from an insufficient number of instances and wishful thinking that conforms beliefs too directly according to what we desire to be true. It is effective to provide exemplars of bad reasoning precisely because human beings have shown themselves prone to engage in it. Indeed, it would not surprising if the ratio of bad inferences to good ones made by the human race at any given time would be a fairly high number. This is not to deny that our interpretations of each other presuppose a normative ideal, but we are often best interpreted as deviating from the ideal quite often for intelligible reasons. The extent to which human beings do deviate from the ideal of rationality is an open, empirical question because we are capable of taking a third-person scientific perspective on ourselves as well as projecting ourselves into each other’s places and imagining what patterns of beliefs and desires and values would make sense of each other’s actions. A scientific understanding of human beings might indeed require revision of the model of human agency we presuppose in commonsense interpretation. Empirical studies indicate that character traits, dispositions to action, and values might be a lot more

where charity begins

113

situationally sensitive than we have supposed them to be and in ways that escape our self-understandings.14 Consider the study of seminary students who were presented with a coughing man slumped in an alley. The most powerful predictor of whether a student helped was whether he was in a hurry for an appointment. By comparison, whether a student had just before been preparing a sermon on the Good Samaritan (!) or not had no predictive power.15 In another experiment, being in a good mood, prompted by finding a stray dime, renders most people much more likely to help a stranger.16 Reflections in the previous section suggest, moreover, that not all diversity comes under the heading of error and failures of intelligibility. I think there is a good deal of truth in Allan Gibbard’s suggestion that we are guided in conceiving what makes sense for human beings to think and to feel by the imperatives of social coordination.17 We exert pressure on each other when we propose what makes sense in the way of thinking and feeling and acting, and this pressure can result in social coordination that might not otherwise be possible. However, there is no reason why there can’t be a variety of models of what makes sense, encompassing a variety of behaviors, patterns of relationships between beliefs and desires, and configurations of values. Natural selection, after all, works through genetic variation over time, and it is likely that human beings within a cooperating group are likely to show such variation in psychological drives and temperaments. The idea that our conceptions of what makes sense has the pragmatic function of social coordination can help to explain why our conceptions of “we” and “us” embody significant diversity. Not only is it true that human beings are likely to show a variety of psychological drives and temperaments, but this variety also can prove

14

See John Doris (2002), Lack of Character: Personality and Moral Behavior, Cambridge: Cambridge University Press; Gilbert Harman (1998–99), “Moral philosophy meets social psychology: virtue ethics and the fundamental attribution error”, Proceedings of the Aristotelian Society 99, pp. 315–31; and Gilbert Harman (1999–2000), “The nonexistence of character traits”, Proceedings of the Aristotelian Society 100, pp. 223–6. 15 J.M. Darley, and C.D. Batson (1973), “‘From Jerusalem to Jericho’: A study of Situational and Dispositional Variables in Helping Behavior”, Journal of Personality and Social Psychology 27, pp. 100–108. 16 A.M. Isen and P.F. Levin (1972), “Effect of Feeling Good on Helping: Cookies and Kindness”, Journal of Personality and Social Psychology 21, pp. 384–8. 17 Allan Gibbard (1992), Wise Choices, Apt Feelings: A Theory of Normative Judgment, Cambridge, MA: Harvard University Press.

114

chapter four

useful to the group. The more differentiated social roles become, the more useful it would be to have different kinds of people who are suited for different roles through their distinctive abilities, styles of social interaction, and preferences. Our conceptions of what makes sense, then, may be deeply influenced by the pragmatic need to include within the scope of the normal and acceptable the variety of kinds of people that can make social cooperation flourish. Another way to put the point is that saying to another person that her thoughts and actions “make sense” is not only to build a bridge of communication to that person but lay the groundwork for cooperation with her, and the need to do these things can be a powerful force for making the concept of “we” and “us” significantly diverse.18 A very common variation in natural temperament with which it is very important for human beings to deal is precisely variation in preference for harmony within groups. Arguably, a good part of the importance of adhering to rites (li ) in the Confucian ethic lies in taking whatever natural preference for harmony that individuals possess and strengthening it in relation to competing drives by engaging in customary forms of behavior that express attitudes of respect and concern for others. 12.1 says, “To return to the observance of the rites through overcoming the self (ke ji constitutes ren ). An ethic that places a strong emphasis on harmony is one type of solution to the problem of fostering social coordination among human beings. However, it is not at all surprising that other ethics would not place as much emphasis on the value of harmony but rather, say, on permitting conflicts in certain domains and keeping them under control through restrictions on the manner in which they are conducted and the measures taken to win. Our models of rational resolutions to problems of social coordination, then, should plausibly encompass a variety that is in keeping with the variety in temperaments, preferences, and values that we have every reason to expect to find in human beings given what we have been taught to expect from those in our immediate vicinity and given the best sciences we can apply to the project of understanding ourselves. Moreover, we can expect to revise our models of rational deliberation upon further scientific understanding of our capacities and 18 The line of thought in this paragraph was provoked by stimulating comments from Michael Krausz and Yang Xiao for comments in response to the presented version of this paper in Beijing.

where charity begins

115

limits. Models of maximizing expected utility have typically been held up as ideals for rational deliberation, but doubts about such models have emerged given the typical limitations on our knowledge of the consequences of the actions we consider, our ignorance regarding what value we will ultimately attach to those consequences we do foresee, and our capacity to take into account only some of the alternatives available in any choice situation.19 Alternative models better suited to human limitations embody the idea of selecting satisfactory means to our ends, rather than the best means, where a satisfactory means might be the first option considered that is “good enough”20 or the option that is not clearly inferior to any other option.21 Perhaps alternative models of rationality are more or less suited to particular problems and circumstances. Maximizing models are better suited to problems where the time frame that must be considered is relatively short and where the relevant information is largely available and preferably in quantifiable form.

4. Conclusion Understanding others is partly making them like us, but it is more like finding ways in which we overlap with them rather than finding identities between them and us. That understanding often turns on loose analogies should not be surprising when we consider how internally diverse we ourselves are, and how much we can surprise ourselves as well as others. We expect that others and ourselves may sometimes become opaque to us in the sense that their actions do not make rational sense, and we incorporate some of these patterns of irrationality into our conception of what it is to be human. Our very conception of what is rational includes a significant range of desires and values, partly because of the mechanism of natural selection through genetic permutations. And if determining what is rational is partly a matter discovering what processes of deliberation 19 H.A. Simon (1976), Administrative Behavior, 3rd edition, New York: The Free Press, p. 81. 20 Philip Pettit (1984), “Satisficing Consequentialism”, Proceedings of the Aristotelian Society supplementary volume 58, pp. 165–176. 21 Gary Klein (2001), “The Fiction of Optimization”, in Gerd Gigerenzer and Reinhard Selten (eds), Bounded Rationality: The Adaptive Toolbox, Cambridge, MA: MIT Press, pp. 101–121.

116

chapter four

and inference are best adapted to limited creatures such as us in our particular environments, our ideals of rationality must to some extent always be defeasible and open to the possibility that a range of solutions to human problems may have equal claim to be called rational.

CHAPTER FIVE

DAVIDSON’S CHARITY IN THE CONTEXT OF CHINESE PHILOSOPHY Yiu-ming Fung

1. A.C. Graham’s Sinologist’s Criticism and the Myth of ‘Pre-logical Thinking’ A.C. Graham, a widely respected Sinologist, may be the first scholar in the context of Chinese philosophy to express opinions counter to Donald Davidson’s principle of charity and to his view on the very idea of a conceptual scheme. As a Sinologist with a comparative perspective based on a strong British theological background and on a long-term experience through energetic work in Chinese Studies, Graham has made significant contributions to the field of Chinese philosophy, especially in his interpretations of Chinese texts and his explanations of the problems in the field. Graham’s contributions seem inseparable from his special status in terms of his reading meanings from or into Chinese texts with a comparative perspective from a double eye—a ‘British eye’ together with a ‘Chinese eye’. It seems to Graham a basic faith that his and his colleagues’ comparative studies with ‘bilingual’ capabilities (or to use my metaphor, with a ‘double eye’) are significant and that the comparisons between Chinese and Western thoughts are understandable or intelligible though they are based on very different conceptual schemes which fundamentally have very little in common. In this regard, Davidson’s challenge to the very idea of a conceptual scheme, I think, is also a challenge to the Sinologist’s basic faith. In his rebuttal to Davidson’s thesis, Graham declares: “For inquirers into the thought and language of other cultures, the issue is inescapable. That very idea [of a conceptual scheme] is one of their indispensable tools, to which Davidson’s objections do not directly apply, since their own tendency is to think of it in terms, not of propositions, but of classification by naming, and perhaps of syntactic

chapter five

118

structures.”1 One of the main reasons for Graham to reject Davidson’s objections is that “[a]t the roots of the systems of propositions called ‘conceptual schemes’ by philosophers there are patterns of naming, pre-logical in the same sense as patterns of perception are prelogical.”2 To illustrate this point more specifically, Graham mentions Le Gall’s and Bruce’s failure in using ‘forme’ (or ‘law’) and ‘matière’ (or ‘matter’) to translate ‘li ’ ( ) and ‘qi ’ ( ) in the texts of SongMing Confucianism as a starting point of his journey to search for a pre-logical or even pre-linguistic realm.3 He claims that “to think of Le Gall and Bruce as making mistakes which we now avoid would miss the whole point. There are no exact equivalents for li and qi among our concepts, and there is no way of approaching them except by breaking out from or awakening to one analogy after another.”4 Because he believes that “all thinking is grounded in analogization,” that the ‘metaphorical root’ behind Westerners’ ‘matter’ and ‘law’ are different from that behind Chinese ‘qi’ and ‘li’, and, unfortunately, that the ‘outsider’, unlike the ‘insider’ who habitually thinks with their concepts, are much less conscious of the differences at the bottom.5 What are ‘the differences at the bottom’? Borrowing Roman Jakobson’s ideas of ‘paradigm/syntagm’ and ‘metaphor/metonym’, Graham argues that, at the bottom of each language or thinking, there is some kind of pre-logical patterning of names that is “a stock of paradigms already grouping syntagmatically in chains of oppositions which at their simplest are binary.”6 He thinks that, for example, before entering into sentences, we can have the compound words as ‘daylight’ and such formulas as ‘the light of knowledge’ syntagmatically grouping from a stock of paradigms which consists of binary oppositions such as ‘day’ and ‘night’, ‘light’ and ‘darkness’, and ‘knowledge’ and ‘ignorance’, etc. When comparing ‘king’ with ‘lion’ as ‘men’ with ‘beasts’, by metaphor we can have the lion as king of the beasts and the king as a lion among men. King connects with throne as chairman with chair, so by metonymy the monarchy is

1 2 3 4 5 6

A.C. Ibid. A.C. A.C. A.C. A.C.

Graham, 1992, p. 59. Graham, Graham, Graham, Graham,

1992, p. 207. op. cit., p. 61. op. cit., pp. 61–62. op. cit., p. 62.

davidson’s charity

119

called the throne and the chairmanship the chair.7 Based on these ideas coming from what Graham calls ‘Semiology’, he believes that we can find the beginning of a conceptual scheme in these chains of oppositions. He calls the thinking in these chains ‘correlative’ in contrast with ‘analytic’ in the sense that the former is conceived as spontaneous, pre-logical, and operating at the level of the nonsentential combinations of words, while the latter is discursive, logical, and operating at the level of propositions. Based on the idea that a conceptual scheme is not a system of logically related propositions but a pre-logical pattern of names, Graham argues that a comparison of Western and Chinese conceptual schemes should begin at the level of non-sentential units. Through his ‘double eye’, he seems to be able to see that the structures of Chinese conceptual scheme “are exposed nakedly by the tendency to parallelism in the classical language, and are overly formulated in the Yin-Yang ( ) cosmological scheme.”8 In comparison with the Western scheme which tends to center on conflicting opposites (truth/falsehood, good/evil), he points out that the Yin-Yang scheme is focused on complementary polarities. Some of the English chains of oppositions such as ‘day/night’ and ‘light/darkness’ seem to fit neatly into a Yang/Yin scheme, but, as stressed by Graham, in the latter A and B are interdependent with A only relatively superior, and the chain does not lead to conflicting duality such as ‘good/evil’. So, he concludes, “[o]ur conceptual schemes differ, not in assuming the truth of contradictory propositions, but in including or excluding different pairs of words.”9 In contrast to the complementary characteristic of Chinese thinking, Graham thinks that David Hall and Roger Ames are right in demonstrating that “the West habitually treat[s] A as ‘transcendent’ in the sense that A is conceivable without B but not B without A; for Westerners there could be God without world, reality without appearance, good without evil.”10 Graham’s opinions mentioned above are mainly based on his deviant notion of ‘conceptual scheme’ and his interpretations of Chinese concepts used in ancient Chinese texts, especially in philosophical texts. His arguments, if valid, assume that there are two levels of 7 8 9 10

A.C. Graham, 1992, pp. 62–63. A.C. Graham, op. cit., p. 64. Ibid. A.C. Graham, op. cit., p. 65.

120

chapter five

thinking (analytic and correlative) and, most importantly, that the latter, in contrast with the former, is spontaneous, pre-logical, and even pre-linguistic though it is presented as a pattern of oppositions. Besides the mystical characteristic of the so-called ‘pre-logical’ and ‘pre-linguistic’ thinking, he also assumes that the meaning of a sentence is dependent on the meaning of the words which occur in the sentence though the truth of the sentence is independent of its component words. In this regard, Graham seems to adopt an atomistic or ‘building-block’ theory of meaning which is opposite to Davidson’s holism. In the following, I will argue that Graham’s criticism is not accurate both in the sense that his mystical idea of pre-logical thinking together with his atomistic theory of meaning is not well argued and in the sense that his interpretations of the Chinese concepts are not well grounded in Chinese sources. I will also point out that, even if we accept, for the sake of argument, his idea of the two levels of thinking, his explanation of the differences of Chinese and Western conceptual schemes is more consistent with Davidson’s principle of charity than his idea of bilingual but distinct understanding would suggest. Why do I think Graham’s mystical idea of pre-logical thinking is not sustainable? One of the major reasons is that the so-called ‘prelogical chain of oppositions’ is a self-contradictory description (How can we make sense of oppositions without logic?). Although Graham borrows his idea of binary oppositions from Jakobson, unlike Graham who understands them as ‘pre-logical’, Jakobson stresses their ‘logical structure’.11 Jakobson’s idea of ‘markedness’ is based on the logical nature of oppositions applied both at the level of the signifier and at the level of the signified.12 In Semiology or Semiotics, we know that there is a ‘semiotic square’, which is adapted from the ‘logical square of oppositions’, to characterize different kinds of oppositions such as contradiction, contrariety, and complementarity or implication. It appears unintelligible to say that these binary relations are oppositions without any logical sense. How can Graham 11 Roman Jakobson, 1978, p. 115. He and his student Morris Halle also remark that “the binary opposition is a child’s first logical operation.” See Jakobson and Halle, 1956, p. 60. 12 Jakobson mentions that “[e]very single constituent of any linguistic system is built on an opposition of two logical contradictories: the present of an attribute (markedness) in contraposition to its absence (unmarkedness).” See Jakobson, 1985, p. 85, cited in John Lechte, 1994, p. 62.

davidson’s charity

121

claim that they are ‘pre-logical’ without any supporting argument? If Graham’s idea is not mere speculation, there is still a question whether this chain of oppositions in people’s mind is innate or obtained by learning. If it is innate, people from different cultures should have the same chain reflected in their spontaneous reactions unless there is some kind of biological ground or genetic evidence to explain why two different races or two different groups of people from different cultures innately have different chains of oppositions. If different chains of oppositions reflected in different cultures are basically grasped by learning, it is impossible for us to understand a chain of oppositions as ‘pre-logical’. Because what items selected and correlated as opposites from learning cannot be understood in a space without any logical relations. How can we know, from learning, if two items (linguistic or ontological entities), such as truth/falsehood and good/evil (in English), are conflicting, or other two items, such as Yin/Yang and day/night (in Chinese), complementary? If all these are grasped by learning and not innate, a native speaker’s seemingly spontaneous reaction in using his or her words in a chain of oppositions should not be considered as different in nature from a skillful swimmer’s or a mature driver’s reaction. Some Westerners’ viewing day and night as conflicting may be based on a perspective emphasizing the different characteristics between them, while some Chinese perception of them as complementary is probably based on a different perspective focusing on their alternation. I think both people can easily understand each other’s opinion without presupposing the necessity of going back to Graham’s ‘pre-logical’ underground. While the idea of ‘pre-logical’ discussed above can be considered as self-refuting, the idea of ‘pre-linguistic’ can be understood as mystical. It is interesting that Graham interprets human beings’ reaction of correlation as pre-linguistic on the one hand, and treats this reaction as similar to Pavlov’s dog’s conditioning on the other.13 In discussing the problem whether animals have thought as humans have, Davidson also mentions the dog, in addition to his example of the fly. In this regard, Graham seems to be totally ignoring Davidson’s arguments about ‘the concept of belief ’ or ‘the belief of belief ’. It is obvious that the idea of ‘pre-linguistic correlation’ is inconsistent

13

A.C. Graham, 1992, p. 207.

122

chapter five

with Graham’s other description of the same thing, ‘a pre-logical patterning of names’ (How can we have names by pre-linguistic correlation?). However, let’s put aside the problem of inconsistency and focus on this dog-like correlation. According to Davidson’s view in his “Thought and Talk”, a dog probably knows that its master is home but does not know that Mr. Smith (its master) is home and the president of a bank is home.14 A dog, unlike a rational creature, cannot have a thought in such an intensional context. So, a speechless creature may be able to discriminate something from other and may be conditioned to make similar reaction to similar stimulus, but it does not mean that it entertains a distinction between what is believed and what is the case. In other words, a dog does not have a belief which is known to be either true or false. In order to have a belief known to be either true or false, a creature must have the belief of a belief, i.e. the concept of a belief. In Davidson’s words, “[t]he reason neither a dog nor any other creature can have a single belief, such as that it is seeing a cat, is that what identifies a belief is what we loosely call its propositional content. Thus, to have a belief about a cat, one must have mastery of the concepts that are involved in this judgement of belief. A creature does not have a concept of a cat merely because it can discriminate cats from other things in its environment. Mice are very good at telling cats apart from trees, lions, and snakes. But being able to discriminate cats is not the same thing as having the concept of a cat. You have the concept of a cat only if you can make sense of the idea of misapplying the concept, of believing or judging that something is a cat which is not a cat. To have the concept of a cat, you must have the concept of an animal, or at least of a continuing physical object, the concept of an object that moves in certain ways, something that can move freely in this environment, something that has sensations. There is no fixed list of things you have to know about, or associate with, felinity; but unless you have a lot of beliefs about what a cat is, you don’t have the concept of a cat.”15 A dog doesn’t have a single belief (a first-order belief ), because it doesn’t have the belief of a belief (a second-order belief ). If Westerners’ thinking or Chinese people’s thinking at the so-called ‘correlative’ level is dog-like as Graham describes, it would be reasonable to treat the thinking as 14 15

Donald Davidson, 2001, p. 163. Donald Davidson, 2002, p. 124.

davidson’s charity

123

‘pre-logical’ without being true or false. Nevertheless, is it possible for Chinese people to have a chain of oppositions, such as ‘day/night’ and ‘Yin/Yang’, without some beliefs about daytime and sunshine and some other beliefs about nighttime and the Moon? Is it possible for Westerners to have another pattern of naming, such as ‘good/evil’ and ‘true/false’, without some intentions about behaviors and some beliefs about sentences? If our conceptual thinking really operates at the very beginning on Graham’s underground level, Chinese thinking (in ancient times) on this level, for example, would have ‘day/night’ opposition without backing up by any beliefs about day and night. In other words, they would think of ‘day/night’ or ‘Yin/Yang’ as complementary without believing that “There is sunshine at daytime”, “The moon appears at night”, and so on. Can we imagine that Chinese people (in ancient times) are thinking of ‘day/night’ correlatively first without any beliefs and then later thinking about them analytically with propositional attitudes? What is the rationale for this learning process? How can Graham identify these correlative concepts or ideas without assigning some background knowledge to the speakers? If we accept, for the sake of argument, that the schemes themselves are patterns of names which are neither true nor false, and that factual statements depend on them for their meaning but not for their truth, Graham’s building-block theory would still have more trouble than the ordinary version of the theory. His words or names used at the ‘pre-logical’ level (of course, it is impossible for them to be used at the ‘pre-linguistic’ level) must be context-free, otherwise they would appear in a context that connects them to other words for identifying their meaning or sense, and this context must be sentential. He thinks that learning Chinese words through guess from the entries of Mathews’s dictionary would never be at home; instead, to understand the meanings of these words, such as ‘an’ ( ) and its opposite ‘wei ’ ( ), is not to analyze them in a logical way but to correlate them within different contextual patterns.16 However, these ‘contextual patterns’ are not patterns in sentential context, but patterns of naming at ‘pre-logical’ level. This may be one of the reasons why he considers his ‘pre-logical’ terms as having “no other content than the oppositions themselves”.17 But, when he uses some 16 17

A.C. Graham, 1992, p. 67. A.C. Graham, op. cit., p. 209.

124

chapter five

examples to illustrate the meanings of these sentential context-free terms or names, it indicates clearly that the terms’ contents are more than the oppositions themselves. He says, for example, “to be yang not yin is nothing else but to be light not dark, or male not female.”18 It is obvious that the example is not consistent with his ‘no more content’ thesis, because ‘to be yang not yin’ is not the same as ‘to be A not B’ or ‘to be A not ~A’. In order to know the similarity between yang and light or male and the difference between each pair, we have to know some contents more than just oppositions. If the terms have no other content than oppositions, then ‘to be yang not yin’ would have no difference from ‘to be A not B’ or ‘to be A not ~A’. If so, how can this abstract and logical idea be understood at the ‘pre-logical’ level? Since the differences in meaning between (ancient) Chinese and English words and sentences, for Graham, are ‘to the bottom’, at the ‘pre-logical’ level, it seems natural for him to claim that “Maowo-zai-xi-zi-shang” ( ) cannot be translated into “The cat sat on the mat”, “Cao-qing” ( ) into “Grass is green,” and ‘yang ’ ( ) into ‘sheep’. To the bottom, it seems that Chinese and English correlate things or divide up the world differently. But Graham’s explanations for these examples are not digging into the bottom if there is one. Although he mentions that ‘wo’ ( ) does not express the same posture as ‘sit’ and that the classification of floor coverings for ‘xi-zi ’ ( ) and that for ‘mats’ are different, this explanation is definitely not about something underground but about contents, more than just correlations and oppositions. Graham’s spontaneous thinking seems to exclude the possibility of having background beliefs about xi-zi in naming xi-zi. However, as he describes, in Chinese xi-zi is used of [sic] straw mats. In other words, it is something not made of cloths but straw only. This indicates that, when ancient Chinese people using the word ‘xi-zi’, it is impossible for them not to know that “xi-zi is made of cloths” is false and “xi-zi is made of straw” is true. Graham’s ‘bottom-claim’ is also not supported by his example of the unintertranslatability between “Cao-qing” and “Grass is green.” He mentions that the concept of cao ( ) has a wider scope than that of grass, qing ( ) as blue-green is based on Chinese primary Five Colors which are different in division or classification from

18

A.C. Graham, 1992, p. 209.

davidson’s charity

125

English primary Seven Colors. However, his term ‘wider scope’ means ‘wider extension’ and to identify the scope or extension of a concept is to identify the applicability of the concept, to test whether it is true or false in applying a concept to some object. Although Graham stresses that “[n]aming is by contrast within the scheme rather than by adequacy to the object”19 and that to correlate Chinese and English words need not assume coincide in extensions,20 it is obvious, in the explanation of this example, that he has to use the idea of scope or extension to identify the differences. Graham’s example about the unintertranslatability of ‘yang’ and ‘sheep’ is also not consistent with his ‘bottom-claim’. When reading the explanation in two languages of a vocabulary difference between them, Graham thinks that “one is positively grateful that they do not say exactly the same thing, much as when collecting information about an incident one wants photographs taken from different angles at different moments.” It means that from a Chinese angle we can see that the scope of ‘yang’ includes both things English people call ‘sheep’ and ‘goat’; but from an English angle we can see that the scope of ‘sheep’ is exclusively different from that of ‘goat’. Like the case that seeing or photographing an accident from different angles presupposes there is one and the same accident, this case also requires that there is one and the same object in plain sight; otherwise, we cannot have different perspectives or different descriptions of the (same) object in plain sight. Graham identifies his idea of correlative thinking or naming as ‘pre-logical’, ‘pre-linguistic’, ‘spontaneous’, and also ‘mystical’. It is mystical in the first sense that Graham’s idea presupposes some kind of mentalist meaning which is independent of analytic thinking; it is mystical in the second sense that Graham treats his ‘pre-logical pattern of names’ as the products of a classifying act of naming without understanding them as the singular terms of logic. Nevertheless, in what sense could this kind of ‘naming’ be understood as a classifying act? Isn’t it necessary to have some sense in which something can be named rightly under a classifying act while others cannot? If Chinese classifying yang is different from English classifying sheep in the sense that the English word ‘sheep’ cannot be used to refer to a goat but the Chinese word ‘yang’ can be used to refer to both 19 20

A.C. Graham, 1992, p. 212. A.C. Graham, op. cit., p. 68.

126

chapter five

a sheep and a goat, Chinese people should know at the same time that it is right to say ‘yang’ to cover both sheep and goats and wrong to refer to sheep only. It seems to be the same situation as indicated by Davidson that we can master the distinction between erroneously applying the concept cow to bulls when faced by a bull and correctly applying a concept that covers both cows and bulls through a test of learning to explain errors.21 To compare a foreign language with home language, I think we can also use this kind of test for learning to explain the differences. For the same reason, if someone says “That’s yang” when faced by a sheep consistently, and says the same thing when faced by a goat at only one time, a Westerner may not know at that time whether she is erroneously applying the word ‘yang’ (if he has the impression that it is equivalent to ‘sheep’) or correctly applying the word that refers to both sheep and goats. Until some day he knows more about her background beliefs through a triangulated interaction as described by Davidson, he could be sure that her concept yang is not the same as his concept sheep. It is unintelligible that without the concept of applicability of names we can have an act of classification. The job of classifying names cannot be done merely at the level of names, especially not at the socalled ‘pre-logical’ level, which is separated from or independent of the context of sentence or proposition. Graham is not only neglecting Davidson’s holism, but also arguing for his mystical version of meaning atomism in a self-refuting way. According to Davidson, “Because of the fact that beliefs are individuated and identified by their relation to other beliefs, one must have a large number of beliefs if one is to have any. Beliefs support one another, and give each other content. Beliefs also have logical relations to one another. As a result, unless one’s beliefs are roughly consistent with each other, there is no identifying the contents of beliefs. A degree of rationality or consistency is therefore a condition for having beliefs.”22 In other words, to identify the content of a belief, a large number of other beliefs with a high degree of consistency are necessary: the principle of charity with the holistic characteristic. This point is not only true for a single belief, but also true for a single word. Davidson says, “[W]ords, like thoughts, have a familiar meaning, a propositional content, only if they occur in a 21 22

Donald Davidson, 1998, p. 7. Donald Davidson, 2002, p. 124.

davidson’s charity

127

rich context, for such a context is required to give the words or thoughts a location and a meaningful function.”23 For example, Graham’s binary oppositions of ‘day’ and ‘night’ for a child cannot be understood as no other content than the oppositions themselves, unless the child could know something so abstract as the formal relationship between ‘A’ and ‘~A’ at the very beginning of learning. To understand the word ‘night’ requires someone to know that there is no sunlight when the word is used in a context related to what he or she sees. He or she is also required to have the belief that sometimes he or she can see the moon and stars at night. These relevant knowledge or beliefs about the general features of the event (or object) are the criteria for people to apply their concepts correctly, and thus to identify the content of the word. Davidson has an example with a similar explanation of the holistic character of having a single concept. He says that “we would deny that someone had the concept of a man who did not know something about what distinguishes a man from a woman, who did not know that fathers are men, that every man has a father and a mother, and that normal adults have thoughts.”24 So, generally speaking, according to Davidson, “the meanings of the words that refer to these features, and the contents of the concepts the words express, depend as much on the natural history of how the words and concepts were acquired as was the case for ‘porcupine’ and ‘echidna’. There are no words, or concepts tied to words, that are not understood and interpreted, directly or indirectly, in terms of causal relations between people and the world (and, of course, the relations among words and other words, concepts and other concepts).”25 He also mentions that, “in the simplest cases words and thoughts refer to what causes them, it is clear that it cannot happen that most of our plainest beliefs about what exists in the world are false. The reason is that we do not first form concepts and then discover what they apply to; rather, in the basic cases, the application determines the content of the concept.”26 Or put in another way, “[w]e can give the meaning of any sentence (or word) only by giving the meaning of every sentence (and word) in the language.”27 23 24 25 26 27

Donald Donald Donald Donald Donald

Davidson, Davidson, Davidson, Davidson, Davidson,

2002, p. 127. 1998, p. 2. 2002, pp. 50–51. op. cit., pp. 196–197. 1984, p. 22.

128

chapter five

Graham agrees with Chad Hansen that classical Chinese nouns in general are closer to English mass than to English count nouns, and thus accepts Hansen’s argument that Western thought is predisposed by number termination to conceive the world as an aggregate of distinct objects while Chinese by the mass noun to conceive it as a whole variously divisible into parts. He also declares that Chinese thought being conditioned to divide down rather than add up is in any case suggested by other features of the language and that Chinese thinking is in terms of process rather than of static entities—individual objects. However, some serious difficulties of Hansen’s explanations based on his ‘mass noun hypothesis’ and his misinterpretations of the Chinese texts seem to be totally unknown to Graham. Among other criticisms, I have argued about Hansen’s issue elsewhere that, based on this hypothesis, his interpretations and explanations of the problem in Chinese texts are basically focused on very few paragraphs mainly in Mohist Canon ( ) and Gong-SunLong-Zi ( ). But unfortunately, even limited to these paragraphs, some terms used in Mohist Canon obviously referring to individual object or abstract entity seem to be totally ignored by Hansen’s treatment. More unfortunately, even though he claims to use Davidson’s principle of charity to interpret Gong-Sun-Long-Zi, he has to interpret Gong-Sun Long as committing inconsistency, particularly in the case that he interprets Gong-Sun Long as using the term ‘fei-ma’ ( ) in two different senses: one meaning as English ‘not (identical with) horse’ and the other as ‘non-horse’ (in the context of mentioning ‘cow-horse’ collection, it means cow).28 Let’s put aside the problem whether ancient Chinese has count terms or not (it is clear that there are a lot of evidences provided by Chinese linguists to indicate that count terms do exist in ancient China), there is still a question of how the ancient Chinese people could have a language which lacks count terms to express individual objects. Is it possible that there is a language used by them whose words can only be used to refer to mass stuff instead of individual object? I don’t think so. I don’t think that there is a language without expressions referring to different individual objects. Claiming that Chinese thinking is in terms of process rather than of static entities is not only distorting Chinese thinking, but also misleading the real issue

28

Yiu-ming Fung, 2000, ch. 8.

davidson’s charity

129

in Chinese texts by such a groundless hypothesis. Even if we accept, for the sake of argument, the hypothesis, it is not necessary for us to accept Graham’s (or Hansen’s) pseudo (or real) linguistic determinism of thinking and ontology. Graham seems to be totally ignoring Davidson’s view on convention (and on the prior theory and the passing theory). According to Davidson’s view, “what interpreter and speaker share, to the extent that communication succeeds, is not learned and so is not a language governed by rules or conventions known to speaker and interpreter in advance; but what the speaker and interpreter know in advance is not (necessarily) shared, and so is not a language governed by shared rules or conventions.” One of the reasons is that “a sufficiently explicit framework could be discredited by a single malapropism. There is some evidence of a more impressive sort that internal grammars do differ among speakers of ‘the same language’.” He concludes that “[w]e must give up the idea of a clearly defined shared structure which language-users acquire and then apply to cases.” Because “we have discovered no learnable common core of consistent behaviour, no shared grammar or rules, no portable interpreting machine set to grind out the meaning of an arbitrary utterance. We may say that linguistic ability is the ability to converge on a passing theory from time to time.”29 It is quite strange that Graham recognizes Wittgenstein, Ryle, and Derrida as his heroes in fighting for his correlative thinking covered by analytic thinking. Nevertheless, it is evident that what these important figures of contemporary philosophy have done is not struggle and search for something at the bottom as Graham’s mentalist and mystical ‘root’. Wittgenstein’s or Ryle’s disenchantment work is to try to explain away any mystical element which is supposedly believed by the mentalists as other than something identified in a social context; and one of the aims of Derrida’s deconstruction is to subvert the ‘binarism’ of Structuralism by demonstrating the instability of binary oppositions. It is clear that these heroes cannot be understood as supporting Graham’s idea; on the contrary, they clean up his ‘bottom’. Let us imagine that a person, who is retarded and fails to recognize the words ‘sheep’ and ‘goat’, invents by himself or herself a word ‘shoat’ (not meaning a young pig just after weaning) to refer

29

Donald Davidson, 1986, pp. 444–446.

130

chapter five

to either sheep or goats based on his or her perception of some of their shared characteristics. With respect to his or her new word, I don’t think people using normal English as their home language cannot understand and interpret what he or she says about ‘shoat’; and I also don’t think people using ancient Chinese as their home language cannot understand ‘yang’ as the equivalent of ‘shoat’. It is unnecessary to treat the triple sides as having different conceptual schemes. They share a lot even though there are differences in vocabulary and grammar among different cultures or at home. Semantic and syntactic flexibility and expandability are not only possible for interpretation at home, but also possible for translation of words and sentences in a foreign language. As indicated by Davidson, “Different speakers have different stocks of proper names, different vocabularies, and attach somewhat different meanings to words. In some cases this reduces the level of mutual understanding; but not necessarily, for as interpreters we are very good at arriving at a correct interpretation of words we have not heard before, or of words we have not heard before with meanings a speaker is giving them.”30 As to a foreign language, for example, we know that, before silk or china transported from China to Europe several thousand years ago, there was no equivalent word for each of these things in English. But it is not reasonable to say that the word china was not translatable between Chinese and English at that time but is translatable now. Literally speaking, there was no exact word in English to translate the Chinese word of china at that time; but in principle, it was not untranslatable. Even though the English word ‘dragon’ in a Westerner’s mind may have different mental image, emotional color, or opinion from the Chinese word ‘long’ ( ) in a Chinese mind, if they both use their words in most of the propositional contexts with the same truth condition and with the same content-fixing cause, we would have no ground to say that most of the sentences about ‘long’ in Chinese and ‘dragon’ in English are not intertranslatable. As indicated by Davidson, there is no need to have a ‘word-by-word’ translation.31 His criticism of Thomas Kuhn’s idea is based on ‘the fact of reason’, if we could use the Kantian description in this context, that incommensurability, if making sense, presupposes a co-ordinate to identify a difference; and the difference “can be explained and 30 31

Donald Davidson, 2001, p. 277. Donald Davidson, op. cit., p. 72.

davidson’s charity

131

described using the equipment of a single language.”32 It means that Davidson’s notion of ‘translation’ does not mean literal translation but interpretative translation which allows explanation and description in addition to semantic and syntactic expansions. In the past the English words ‘mutton’ and ‘sheep’ defined one another by contrast of their being cooked vs. on the hoof. French word ‘mouton’ does not have this contrast; so, we cannot find its English equivalent in translation. However, when subtracting ‘mutton’ from English, it would lead ‘sheep’ to expand its extension into something like French ‘mouton’. I do not think there is any person who would be so unthinking as to claim that the literal translation among different languages is always workable without semantic and syntactic expansions. If there are people in this list, Davidson is definitely not.

2. A Methodological Reflection on the Principle of Charity If I am right in the above discussion, Graham’s Sinologist’s criticism of Davidson’s principle of charity together with his view on the very idea of a conceptual scheme can be judged as groundless and selfdefeating. I believe Davidson’s principle of charity can survive in its application to the Chinese or Sinologist context. However, there are quite a few criticisms of this principle which are not based on concrete contexts in examining its applicability, but are focused on the controversial issue of its methodological character. Most of the criticisms in this regard are targeted on its seemingly Kantian nature of transcendental argument or transcendental deduction (hereafter, simplified as ‘TA’ and ‘TD’ respectively). However, Davidson seems to be reluctant to reject this label in the beginning and unwilling to accept it later. In his “In Defense of Convention T”, Davidson writes: “Tarski is right, I think, in proposing that we think of natural languages as essentially intertranslatable (although I don’t see why this requires word-by-word translation). The proposal idealizes the flexibility and expandability of natural languages, but can be justified by a transcendental argument (which I will not give here).”33 In a footnote Davidson refers the reader to his articles “On the Very Idea of a Conceptual Scheme” and “The Method of Truth in Metaphysics” 32 33

Donald Davidson, 2001, p. 184. Donald Davidson, op. cit., p. 72.

132

chapter five

for this TA. But later when responding to Andrew Cutrofello’s criticism, he replies that “I don’t know if my arguments for the principle of charity are transcendental or not. Andrew Cutrofello does not quote me as saying so. In any case, the arguments he says are mine are not transcendental, good, or mine.”34 In response to A.C. Genova’s interpretation of his idea, he mentions that “[p]eople suggested that what I had hit on was a transcendental argument, and I didn’t reject the idea. But was it?” and gives a more concrete explanation that “[i]f you accept the steps that lead to my version of externalism, what Genova calls ‘semantic realism’, then you cannot, I think, be a skeptic about the existence of an external world much like the one we all believe we share, nor about the existence of other people with minds like ours. But the considerations in favor of semantic realism seem to depend in part not on purely a priori considerations but rather on a view of the way people are.”35 And in response to Barry Stroud’s discussion, he also stresses that “[w]hat is not trivial is to show that we know enough about the world to be able to say that it is pretty much as we think it is.”36 On another occasion, Davidson mentions that the principle of charity in interpretation is not a policy: “we might do better to think of it as a way of expressing the fact [my italic] that creatures with thoughts, values, and speech must be rational creatures, are necessarily inhabitants of the same objective world as ourselves, and necessarily share their leading values with us.”37 Although Davidson agrees that whether a creature ‘subscribes’ to the basic principles of rationality, including the principle of continence, the basic principles of logic, and the principle of total evidence for inductive reasoning, “is not an empirical question”,38 it seems he is only referring to the coherence part of the principle of charity; as to the correspondence part, it is obvious that the question is of a factual nature. It seems he is expressing the same point when he replies to Thomas Nagel in emphasizing that “[t]he conclusion [of the reasoning of charity] that I know that the world, both in general and in many particular ways, is as I think 34

Lewis H. Hahn, ed., 1999, p. 342. Donald Davidson, “Reply to A.C. Genova”, in Lewis H. Hahn, ed., 1999, pp. 192, 194. 36 Donald Davidson, “Reply to Barry Stroud”, in Lewis H. Hahn, ed., 1999, p. 163. 37 Donald Davidson, 1984, p. 18. 38 Donald Davidson, 1985, p. 352. 35

davidson’s charity

133

it is, depends on the fact [my italic] that I have just the beliefs I do.”39 In other words, the argument for the impossibility of massive errors of our beliefs or for the necessity of massive truth shared by the speaker and the interpreter is based in part on empirical evidence; in this sense we cannot label Davidson’s argument for the principle of charity ‘transcendental’. With respect to the arguments for the condition of attribution of thought and for the condition of language-hood, it is obvious that Davidson uses many similar key terms or phrases which are the indicators of using TA or TD in a Kantian sense or a popular sense since P.F. Strawson’s use. In Davidson’s third book which deals with three kinds of knowledge, we can easily find a pattern of terms or phrases used in his sentences, such as “X is possible only if Y”, “X is impossible unless (there is a presumption that) Y”, “X requires Y”, “X is essential to the possibility of Y”, “The very possibility of X demands Y”, “X is intelligible only on the supposition that Y”, “X takes its content from a background of Y”, “In order to have X, we must entertain Y”, “X is a necessary condition of Y”, and so on. These expressions are undoubtedly also used in the discourse of TA or TD. So, I think the popular perception of ‘transcendental’ attributed to Davidson’s arguments is not groundless. That Davidson lacks detailed explanation may be in part responsible for the resultant misunderstanding. In order to give a supplementary explanation for this problem, I will make a distinction based on Kant’s two kinds of TD or TA in the following. We know that in Kant’s three Critiques there are at least two kinds of TD or TA. In Kant’s first Critique there is a paradigmatic use of the TD, i.e. the transcendental deduction of the categories. It starts from a question “How is experience (or empirical knowledge) possible?” and answers with the following chain of expressions: “From the premises that there is experience and that the categories are necessary conditions of its possibility, it proves or legitimizes the objective validity of the categories.” In his second Critique Kant seems to use the same kind of TD. Its question is: “How is (autonomy) morality (or moral judgment and action) possible?” Its argument runs like this: “From the premises that there is

39 Donald Davidson, “Reply to Thomas Nagel”, in Lewis H. Hahn, ed., 1999, p. 209.

134

chapter five

(autonomy) morality and that its possibility is based on the postulate of the freedom of will, it proves or legitimizes the objective validity (or practical necessity) of the freedom of will.” In this regard, the first premise in each deduction seems to Kant to be empirically true, but the second premise in each deduction obviously includes some transcendental concept or transcendental entity, i.e. a priori category or the freedom of will, and thus it cannot be considered as empirically true. To assert that having some transcendental concept or transcendental entity is a necessary condition of, or is presupposed by, having an empirical fact is clearly not an empirical claim, but some kind of a priori claim or stipulation. In this sense we are legitimated to say that this kind of argument is a TD or TA in its proper sense or paradigmatic use. Another sense of TD or TA used in the first Critique is about the objective validity of (empirical) objects. As for the same question as above, “How is experience possible?” Kant’s reply is this: “From the premises that there is experience and that (physical) objects are necessary conditions of its possibility, it follows that there are (physical) objects.” We can see in this case that the second premise does not have any transcendental concept or transcendental entity; it seems not to be an a priori stipulation but an empirical claim of a necessary condition, just like the claim that “having food is essential to being healthy”, “good teaching requires good learning”, “stopping smoking is a necessary condition of preventing lung cancer”, and “having an event of bell ringing presupposes having a bell”. All these are based on some empirical fact. In comparison with the first kind of TA, a paradigmatic use of TA, this second kind of argument has also been classified as TA by many philosophers but is obviously lacking a ‘transcendental’ sense. The other difference between these two kinds of TA is related to the problem of objective validity. With respect to the first kind of TA, if there are different TAs based on different a priori stipulations for the possibility of the same empirical fact, what can be justified or legitimized would be different transcendental concepts or transcendental entities; and the so-called ‘objective validity’ of the target conclusion would be relativized and thus the argument could not be sound. On the other hand, each example of the second kind of TA is an empirical claim, each argument can be valid or invalid, sound or unsound, as long as we can have the relevant empirical evidence to confirm or disconfirm (or to verify or falsify) each sentence in the argument. This kind of TA may have Stroud’s problem with

davidson’s charity

135

‘superfluous’ or ‘redundant’ in relation to verification;40 but the first kind is definitely irrelevant to the problem. Although these two kinds of TA are very different, I think it is unfortunate that some people like to put these two together under the same label. To make them devoiced may be good for our understanding of many related but distinct arguments in the literature, and especially for our understanding of Davidson’s principle of charity. The impression of the ‘transcendental’ sense given to people by Davidson’s principle of charity may be due to its expressing in a very general way and its seemingly a priori characteristic. However, if we pay ample attention to some of the concrete examples which are usually used by Davidson to illustrate the applicability of the principle, this impression may be different. For example, a different impression would emerge if we read carefully the paragraph about the concept of a cat I quoted from Davidson’s article in the previous section.41 In this particular case, but not in the general form of the principle, it is obvious that this claim for the condition of having a concept and a belief, or attributing a concept or a belief, is supported by empirical evidence; this is not an a priori stipulation. If we use Graham’s seemingly counter-example as our example, it also confirms with empirical evidence that massive true beliefs shared by Chinese and Westerners are essential for interpreting and understanding the differences between their languages. Someone who fails to interpret or to translate the Chinese word ‘yang’ into the English word ‘sheep’ at the beginning would eventually discover that the scope or extension of the former is larger than that of the latter through a test of Davidson’s triangulation. In this case, even though they classify things differently and each side may have different mental images, emotive colors, and opinions attached to their words, they do have a lot to share. They both recognize that there are two sorts of objects, though English, unlike Chinese, does not put them together as belonging to the same kind. They both know that the objects either referred to by the Chinese ‘yang’ or referred to by the English ‘sheep’ or ‘goat’ are not plants but animals, that these animals are physically visible, freely moving, do not eat other animals, and so on. The list of shared beliefs and thoughts is too long to

40 41

Barry Stroud, 1968, p. 256. See the quotation in footnote 15.

136

chapter five

enumerate; but it forms an empirical background for each to interpret and to understand the other’s language. Based on this and other examples, I think Graham’s argument against Davidson’s view is actually self-refuting; Graham’s explanation of the differences between the two languages, in order to make sense, is also necessarily based on some kind of background beliefs shared by both sides. If a teacher of biology wants to persuade his or her students to stop smoking, he or she may ask the students a question like this: “How is preventing lung cancer possible?” Some smart students would give an answer from a health magazine that “It is the case that many people have no lung cancer, and (we know from statistical evidence) that a necessary condition of preventing lung cancer is to stop smoking; therefore, stop smoking.” This Modus Ponens argument is clearly based on empirical evidence. Let us take another example to illustrate the same point. We can ask an experienced teacher, like Davidson, a question “How is it possible to be a good teacher?” I think he would answer like this: “There are many good teachers in the world, and (we know from experience that) to be a good teacher requires someone to be a good learner; therefore, to be a good learner.” So, based on this valid argument, we should try our best to be a good learner if we want to be a good teacher. This argument starts from a question of possibility, its premises do not involve any transcendental concept or transcendental entity, and its assertion of the necessary condition is based on empirical evidence. Hence, it is not an a priori claim. The general form of Davidson’s principle of charity seems to suggest an a priori claim, but its particular examples are not. I think this principle is in accord with our intuition of the ordinary use of language (speaking or interpreting) and, more importantly, is generalized from a lot of particular cases with empirical evidence as its inductive base. This base is very stable and seems unshakable, because we have not yet found any concrete and obvious counter-example to the principle. In this sense, I conclude that the principle of charity, “Maximizing true beliefs is a condition of having thought,” is based on a huge amount of empirical evidence, as is the principle of nutrition, “Maximizing nutriment is a condition of having energy.”

davidson’s charity

137

3. Zen Buddhism Interpreted as Deconstructive Skepticism Davidson, like other contemporary philosophers, wants to kick out skepticism. What Davidson can do in this respect, I think, is to kick out some version of skepticism in a constructive or explicit form, but not a deconstructive or implicit form of skepticism. As demonstrated by Putnam’s model-theoretical argument, metaphysical realism and skepticism are two sides of the same coin.42 In this sense we may say that metaphysical realism or Platonism is a constructive form of skepticism. This kind of skepticism, if expressed explicitly, is not only claiming that our existing empirical knowledge of the world is an illusion or delusion, but also establishing its own position and asserting a view of reality. Since it is not only negatively rejecting what we have known as illusive or delusive appearance, but also positively affirming an ultimate truth of reality, there is trouble for this kind of skepticism to justify itself. It has trouble both because the burden of proof is now on its shoulder and because its non-empirical stance is obviously no better than our empirical stance even though it would try to legitimize or justify itself on its own stance. In this regard, Davidson’s arguments based on triangulation are obviously more convincing than the transcendental or metaphysical arguments of skepticism such as Platonic realism; and in this sense, it is fair to say that Davidson can kick out this kind of constructive skepticism. However, Davidson believes that “to show we know enough about the world to be able to say that it is pretty much as we think it is” is “to show that it [i.e. skepticism] is false”, though he agrees that “we cannot prove it false in particular cases.”43 Or more moderately speaking, “I set out not to ‘refute’ the skeptic, but to give a sketch of what I think to be a correct account of the foundations of linguistic communication and its implications for truth, belief, and knowledge. If one grants the correctness of this account, one can tell the skeptic to get lost.”44 It seems he is claiming that he can kick out skepticism in general. But I do not think this is true; I think the case is reversed. There are many brands of skepticism, constructive or deconstructive, metaphysical or mystical, or in particular, the thesis of Plato’s cave or Descartes’ demon. In general, it is 42 43 44

Hilary Putnam, 1984, ch. 1. Donald Davidson, “Reply to Stroud”, in Lewis H. Hahn ed., 1999, p. 163. Donald Davidson, 2002, p. 157.

138

chapter five

not easy to kick out all of them; but when facing some particular cases of constructive skepticism, it is quite easy to compare its power of justification with that from our empirical stance and thus not difficult to kick it out. Although all brands of skepticism cannot justify themselves, some of them cannot be falsified by empirical evidence. This is just like the case that although all brands of theism cannot justify there is God, some of them cannot be falsified by empirical evidence. Let us imagine that, if a theist claims that in reality God is in the world though in appearance you do not know that it is the case, what could an anti-theist say? She could say that “Please tell me how to identify or locate God in this world?” Then, the burden of proof would be put on the theist’s shoulder, because what he claims is more than a negative statement that the world is not what it appears, a mere logical possibility, but he also has a positive claim about something (i.e. God) which does not appear in the world but is really in the world. Just like the case of constructive skepticism, this can be easily judged as false unless the theist can identify or locate the non-empirical entity in the empirical world in a non-mysterious way. Nevertheless, if some ‘slippery’ theist claims that God does exist but ‘He’ and ‘His Action’ are ineffable and unintelligible in a rational way, it seems that we cannot directly prove the claim is false. When you try to prove its falsity, the theist would reply that “Hey, it cannot be thought in a rational way, so, it cannot be falsified in a rational way either.” My point is that, when facing this deconstructive, implicit or hidden kind of skepticism, we cannot kick it out directly, because it rejects the whole idea of rationality, either used for proving or for disproving the issue: it refuses to prove itself in this way and also refuses any disproof in the same way. It is totally out of reason. But, is it irrational? Davidson is right in saying that “we could not understand someone whom we were forced to treat as departing radically and predominantly from all such [rational] norms. This would not be an example of irrationality, or of an alien set of standards: it would be an absence of rationality, something that could not be reckoned as thought.”45 He is also right in making his conclusion that “If what we share provides a common standard of truth and objectivity,

45

Donald Davidson, 1993, p. 297.

davidson’s charity

139

difference of opinion makes sense. But relativism about standards requires what there cannot be, a position beyond all standard.”46 So, the deconstructive skepticism or slippery relativism mentioned above, according to Davidson, can be understood as not irrational, but nonrational (except its criticism of the rational view of the world). This means that it is not about thought, therefore it cannot be either true or false. Irrational thinking is false because it is qualified to be false; but non-rational thinking (if it can be called ‘thinking’) is not false because it is not qualified to be false. From Davidson’s point of view, non-rational thinking is just not making sense. Here, I totally agree with Davidson on this point. However, I do not think the story ends here. Because the deconstructive skepticism would reply that “based on your rational standards, you have your own right to judge that my thinking is not making sense; but based on my non-rational or super-rational standards, your knowledge of the world is illusive or delusive. My thinking not making sense to you is tantamount to my thinking not understood by you, because it cannot be understood by rational thinking. Why do I have to give up my own standards and then to share your rational standards and thoughts? Could I ask you to do the reverse? Your request of my giving up my standards and my request of your giving up your standards seem to be of the same weight of legitimacy.” At this point, I do not think Davidson and any other can kick out this deconstructive skepticism without more energy. Davidson seems to have smelt this kind of skepticism; so he mentions that “there is the idea that any language distorts reality, which implies that it is only wordlessly if at all that the mind comes to grips with things as they really are . . . Yet if the mind can grapple without distortion with the real, the mind itself must be without categories and concepts. This featureless self is familiar from theories in quite different parts of the philosophical landscape.”47 The thesis based on the idea of the ‘featureless self ’ in general, or the idea of Zen (or Chan ) Buddhism’s ‘no-self ’ in particular (which will be discussed later), is definitely rejected by Davidson, because he thinks that “the mind is divorced from the traits that constitute it.” However, the reason of Davidson’s rejection is still based on rational standards

46 47

Donald Davidson, 1993, p. 307. Donald Davidson, 2001, p. 185.

140

chapter five

which are not welcome to this kind of skepticism; so, I do not think it can persuade the skepticism to get lost. In Asian philosophy, especially in Chinese philosophy, the idea of ‘featureless self ’, ‘conceptless subject’, ‘contentless consciousness’, ‘absolute mind’, or ‘no-self ’ was very popular a few thousand years ago. Ancient Daoism, Zen Buddhism, and New Confucianism (including Song-Ming and contemporary Confucianism) are the main representatives with this characteristic in the field. In the following, our discussion will mainly focus on Zen Buddhism, with special reference to D.T. Suzuki’s interpretation. We know that Suzuki was very successful in promoting Zen Buddhism in the West. In order to attract Western people’s attention by some salient differences, sometimes Suzuki exaggerates the contrast between Western and Asian thinking. One of the major differences emphasized by him is the contrast between rationality and irrationality, or logical and illogical thinking. For example, Suzuki sometimes says that “Zen is the most irrational, inconceivable thing in the world”, that it “defies all conceptmaking”, and that the essence of Zen is satori ( ), the experience of ‘sudden enlightenment’, which is irrational, inexplicable, and incommunicable.48 In his popular book, entitled Living by Zen, he stresses that “If we are to judge Zen from our common-sense view of things, we shall find the ground sinking away under our feet. Our so-called rationalistic way of thinking has apparently no use in evaluating the truth or untruth of Zen. It is altogether beyond the ken of human understanding. All that we can therefore state about Zen is that its uniqueness lies in its irrationality or its passing beyond our logical comprehension.”49 In response to the Chinese historian Hu Shih’s ( ) criticism, he says, “Zen is not explainable by mere intellectual analysis. As long as the intellect is concerned with words and ideas, it can never reach Zen.”50 Therefore, to know Zen one must give up his or her rational thinking and dualistic logic, and then he or she could be enlightened with prajñà-intuition ( ), an unknowable knowledge. Why do you have to give up rational thinking and dualistic logic? It is because people without Zen enlightenment are living in the world of samsàra ( ) with the sufferings issuing from dualistic thinking. If you want to be emancipated from 48 49 50

D.T. Suzuki, 1996, pp. 13, 103. D.T. Suzuki, 1994, p. 20. D.T. Suzuki, 1953, p. 26.

davidson’s charity

141

these sufferings and to enter Zen’s non-dualistic world, you have to go beyond rational thinking. To be free from the dualistic cage and enter this beautiful world, one must know nothing; because to fall into the dualistic abyss, one is forced to know something conceptualized. Zen or the insight of sunyata ( emptiness) is nothingness, because it is the ‘undifferentiated totality’, there is nothing in it which can be conceptualized. Suzuki thinks that “the dualist view of reality has been a great stumbling block to our right understanding of spiritual truth,”51 and thus “Zen is decidedly not a system founded upon logic and analysis. If anything, it is the antipode to logic, by which I mean the dualistic mode of thinking.”52 He even condemns: “According to the philosophy of Zen, we are too much of a slave to the conventional way of thinking, which is dualistic through and through. No ‘interpenetration’ is allowed, there takes place no fusing of opposites in our everyday logic.” “Zen, however, upsets this scheme of thought and substitutes a new one in which there exists no logic, no dualistic arrangement of ideas.”53 In order to deconstruct dualistic logic, Suzuki sometimes stresses the significance or necessity of Zen masters’ using incoherent or paradoxical statements to express their insight. He explains the reason “why they [i.e. the masters] make those apparently incoherent statements. Their inclination is to set the minds of their disciples or of scholars free from being oppressed by any fixed opinion or prejudices or so-called logical interpretations.”54 More theoretically speaking, “Paradoxical statements are . . . characteristic of prajñà-intuition. As it transcends vijnana ( ) or logic it does not mind contradicting itself; it knows that a contradiction is the outcome of differentiation, which is the work of vijnana.”55 One of the paradoxical statements frequently used by Suzuki is the following example: “We generally reason: ‘A’ is ‘A’ because ‘A’ is ‘A’; or ‘A’ is ‘A’, therefore, ‘A’ is ‘A’. Zen agrees or accepts this way of reasoning, but Zen has its own way which is ordinarily not at all acceptable. Zen would say: ‘A’ is ‘A’ because ‘A’ is not ‘A’; or ‘A’ is not ‘A’, therefore, ‘A’ is ‘A’.”56 It seems that the way of Zen that Suzuki describes is the way 51 52 53 54 55 56

D.T. D.T. D.T. D.T. D.T. D.T.

Suzuki, Suzuki, Suzuki, Suzuki, Suzuki, Suzuki,

1994, p. 28. 1991, p. 38. 1996, p. 112. [1991], pp. 78–79. 1955, pp. 94–95, also in Charles A. Moore, ed., 1951, p. 24. 1953, p. 37.

142

chapter five

to subvert, generally, the duality of ‘A’ and ‘~A’; and specifically, the dichotomy of subject and object. He believes that “in prajñà this dichotomy no longer exists,” because, “[ p]rajñà is not concerned with finite objects as such; it is the totality of things becoming conscious of itself as such. And this totality is not at all limited. An infinite totality is beyond our ordinary human comprehension.”57 It means that not only the extension but also the content of Zen wisdom is beyond ordinary human comprehension. So he concludes, “Satori (emptiness) may be defined as an intuitive looking into the nature of things in contradistinction to the analytical or logical understanding of it.”58 Although Suzuki puts too much emphasis on the irrational and illogical character of Zen, it is basically a strategy. In reality, he is not rejecting rational thinking and logic for its own sake; in order to demonstrate their irrelevance to or distortion of the ultimate reality, he has to use this strategy to help people to go beyond them. In other words, Zen is a realm of ‘pure land’ inhabited by people with pure consciousness without any concept and logic. For Suzuki, to fight against rational thinking and logic is not to justify their falsity, but to throw them away altogether in order to enter Zen enlightenment. In this sense we can say that Suzuki’s Zen is not irrational but non-rational. To understand irrational thinking, as demonstrated by Davidson, we have to put it into the framework of rational thinking. I think Suzuki would agree with this point. However, to understand non-rational or super-rational thinking, for Suzuki, we have to throw away rational thinking altogether, there is no place for it to stand. In other words, giving up rationality is essential to attaining Zen enlightenment. Based on this point, I argue that, if Zen Buddhism’s view on the secular world can be interpreted as a form of deconstructive skepticism, its ‘non-rational web’ can be considered as very defensive to any attack from the ‘rational army’.

4. Locating Problem and Landing Problem Zen as a form of deconstructive skepticism is both like and unlike the relativism defined by Davidson and others. It is like relativism 57 58

D.T. Suzuki, 1960, p. 57. D.T. Suzuki, 1996, p. 84; and 1991, p. 88.

davidson’s charity

143

in the sense that it takes an unorganized reality for granted; it is unlike it in the sense that it throws away the so-called conventional scheme and defines its understanding or enlightenment as conceptless and speechless. As interpreted by Suzuki, Zen masters can grasp their ultimate reality through a special and private access, while ordinary people still in the ignorant state distort Zen’s truth and misunderstand reality as appearance. This picture of the two kinds of truth (Zen’s truth and ordinary truth) or of the two levels of world (Zen’s world and ordinary world) may be understood as a special sort of relativism; but it seems not appropriate to apply Davidson’s argument against the very idea of a conceptual scheme in this context, because, as indicated by Suzuki, Zen has no conceptual scheme. Suzuki would probably address the question in this way: “Zen masters would reject applying Davidson’s identifying or locating criterion by saying that there is no conceptual content, clear or unclear, in my state of enlightenment, and one can reach this state only if he or she gives up rational thinking and thus is free from the conceptual cage. We have no language at all when enlightened. We are not to use language to organize or to fix anything. For Zen, there is nothing to be organized or to be fixed. What we have is neither an object nor a subject, because the dualistic contrast between subject and object has been merged into an undistinguishing unity, an absolute subjectivity. If we have to describe what it is, it is the subject without opposite to object, the subject in itself, or more appropriately, no-self or emptiness (sunyata).” Since Suzuki stresses that only through each individual’s private access can one enter Zen’s world, we may consider Zen’s world as solipsistic and, according to Davidson, judge it as having no size or as not a world at all.59 But Suzuki would not agree with this accusation, because he believes that one who enters Zen’s world with enlightenment “is absolutely convinced of its universality in spite of its privacy.”60 It is private because it cannot be grasped by concepts and can be entertained or enjoyed only in each individual’s mind. It is not solipsistic because it is enjoyable by all people though not enjoyed by those who are still in the realm of illusion. Furthermore, even using Davidson’s idea of the ‘compartmentalization of the mind’ would be unhelpful, because

59 60

Donald Davidson, 2002, p. 119. D.T. Suzuki, 1953, p. 33.

144

chapter five

Zen’s problem is not the division of the mind, but the deconstruction of the mind. This deconstructive or hidden skepticism implies a special sort of relativism which is very slippery and is not easy to get rid of. Based on the principle of charity, of course, Davidson would ask Suzuki to identify or locate Zen’s thinking in a human language. As mentioned by Davidson, “We do not understand the idea of such a really foreign scheme. We know what states of mind are like, and how they are correctly identified; they are just those states whose contents can be discovered in well-known ways. If other people or creatures are in states not discoverable by these methods, it cannot be because our methods fail us, but because those states are not correctly called states of mind—they are not beliefs, desires, wishes, or intentions. The meaninglessness of the idea of a conceptual scheme forever beyond our grasp is due not to our inability to understand such a scheme, nor to our other human limitations; it is due simply to what we mean by a system of concepts.”61 I think Suzuki’s Zen or other skepticism painted with a similar color of mysticism would respond to Davidson’s request of identification of their inner state in this way: “Yes, they are not usual or ordinary mental states. You have to be trained or to practice by yourself (in a special religious or moral way of practice or gong-fu ) in order to transform your ‘self ’ from illusion to enlightenment.” And Davidson might answer in this way: “Oh, that’s all I can say of your so-called ‘inner states’. If they cannot be discovered and identified by these methods, I don’t know what they are. I only know that they are not what we call ‘mental states’ in the common usage, period.” Nevertheless, I do not think Suzuki and his comrades are satisfied with Davidson’s answer on this point, and there would be no further dialogue between Davidson and them to follow. However, I prefer a strategy of putting aside the locating problem and ask them another question: “We agree, for argument’s sake, that you can be enlightened without rational thinking, but, how can you transform your ‘self ’ and know that you are crossing the gap after you have been trained or practiced in your special way and then arrived at the realm of the so-called ‘inner experience’ of enlightenment? How can you know, without any ordinary concept, the very secular concepts are all wrong in

61

Donald Davidson, 2002, p. 40.

davidson’s charity

145

grasping your ultimate concern? You only say that your ‘inner world’ cannot be expressed in discursive language or even in any language (i.e. ineffability thesis), and sometimes say that your transcendental entity (i.e. absolute subjectivity) is not an ordinary object, such as table, tree, apple, or an event in the physical world; but that is not enough for us and even for you to understand what it is.” I think this is not to ask for an answer of the locating problem, but another problem. Let us suppose there is a game of twin boxes. Both boxes are similar in the cover, but one of them is empty and the other either has something inside or has not. We do not know which one is empty and whether both are empty. However, the master of the game gives us some hints. He tells us that inside the boxes both include no ordinary object, such as apple, pencil, or cake, and that the only difference between these two boxes is that one of them is empty and the other one has a non-ordinary object inside. I think what Suzuki tells us in his game is similar to this box game: Based on our standard of identifying ordinary object, we are not only unable to identify the so-called non-ordinary object inside the transcendental box, we also cannot understand whether there is any difference between having this very thing in the transcendental box and having nothing in the other ordinary box. It is no use to characterize it based only on some negative descriptions, such as ‘non-ordinary object’, ‘emptiness’ or ‘no-self ’, because these can also be applied to the case when there is nothing inside the transcendental box (or applied to the ordinary empty box), the truth condition for both cases is the same. Suzuki’s example of tea drinking also indicates a similar game. He says: “You and I sip a cup of tea. That act is apparently alike to us both, but who can tell what a wide gap there is subjectively between your drinking and my drinking? In your drinking there may be no Zen, while my [sic] is brim-full of it. The reason for it is: you move in a logical circle and I am out of it.”62 This time the twin boxes seem to have the same kind of object inside, but the game master stresses that only he who moves beyond a logical circle can know they are different. But I do not think Suzuki can make the distinction in a significant way. When Suzuki declares that the aim of Zen is “the unfolding of a new world hitherto

62

D.T. Suzuki, 1991, p. 88.

chapter five

146

unperceived in the confusion of a dualistically trained mind”,63 what is his standpoint? If he declares from the standpoint of the old world, this declaration is self-defeating; if he declares from the standpoint of the new world, it would be nonsense. It is self-defeating because the dualistically trained mind rejects itself in the former case; it is nonsense because there is no point in contrast in the latter. I do not think he can explain his standpoint. He sometimes stresses “no other method than that of casting away this intellectual weapon and in all nakedness plunging into ‘sunyata’ itself,”64 because he thinks that, “[a]s long as conceptualization goes on, there will be no discovery of the real self ”65 and “[t]he self escapes from all these meshes of conceptualization or objectification.”66 However, on the other hand, in order to explain the foundational meaning of ‘sunyata’, he on another occasion emphasizes that “sunyata is not a negative term but a positive concept [my italic], and is not arrived at by abstraction or postulation, for it is ‘what makes the existence of anything possible’.”67 He is wandering in between the conceptual and the conceptless world. My conclusion is that not only can we not understand that, but also the Zen masters are unable to distinguish their own ‘inner experience’ from ordinary things, because their Zen wisdom is totally beyond logical and rational thinking and thus they cannot think of any difference between Zen and other things. In other words, they give up their right to make any distinction and they become the slave of their non-dualistic thinking. The problem of locating or identifying is to ask the question of how to locate the position of an entity under investigation in a rational or public space. If we consider the idea of Zen as having this problem, it would be a question about the location of Zen’s transcendental self (or no-self ). However, as mentioned above, Suzuki does not take this problem seriously, because he rejects the ‘rational doctor’ to treat his problem and thinks that Zen has its ‘nonrational immutability’ which is free from the ‘disease’ cured by the rational doctor. Although Suzuki’s interpretation of Zen could be recognized as immune from the locating problem, it still has another

63 64 65 66 67

D.T. D.T. D.T. D.T. D.T.

Suzuki, Suzuki, Suzuki, Suzuki, Suzuki,

1996, p. 84; and 1991, p. 88. 1951, p. 5. 1954, p. 174. op. cit., p. 168. 1951, p. 4.

davidson’s charity

147

problem which may be thought as troubling him. This other problem, which I call the ‘landing problem’, is about ‘bridging’ the Zen’s inner world with the secular outer world, or about ‘landing’ the Zen’s transcendent or transcendental self into empirical life. More particularly speaking, it is the problem of how to transform from the illusive life of attachment to the secular world into the enlightened life of freedom from such attachment. As indicated by Suzuki, “what makes Zen unique as it is practiced in Japan is its systematic training of the mind. Ordinary mysticism has been too erratic a product and apart from one’s ordinary life; this has Zen revolutionized. What was up in the heavens, Zen has brought down to earth.”68 However, he also believes that “[w]hen conceptually understood, the lifting of a finger is one of the most ordinary incidents in everybody’s life. But when it is viewed from the Zen point of view it vibrates with the divine meaning and creative vitality.” This would appear to be difficult to land on earth. Our question is: how could one shift from the ordinary view to the Zen view? He mentions that “[t]hose who have only read the foregoing treatment of Zen as illogical, or of Zen as a higher affirmation [i.e. affirmation without negation as antithesis], may conclude that Zen is something unapproachable, something far apart from our ordinary life, something very alluring but very elusive; and we cannot blame them for so thinking. Zen ought, therefore, [to] be presented also from its easy, familiar and approachable side. Life is the basis of all things; apart from it nothing can stand. With all our philosophy, with all our grand and enhancing ideas, we cannot escape life from as we live it. Star-gazers are still walking on the solid earth.”69 But, how can Zen-gazers land on the solid earth? If Zen cannot land on the earth of our ordinary life, it is irrelevant to our ordinary life and thus is unable to criticize our ordinary view of the world as illusion and delusion, or as rational attachment and distortion of reality. If I am right on this point, I think Zen’s situation would be no different from that of a speechless dog in its inability to criticize the illusion and delusion of humans’ attachment to the secular world. However, Suzuki thinks that “Zen must never be confused with naturalism or libertinism, which means to follow

68 69

D.T. Suzuki, 1991, p. 45. D.T. Suzuki, op. cit., pp. 80–81.

148

chapter five

one’s natural bent without questioning its origin and value. There is a great difference between human action and that of the animals, which are lacking in moral intuition and religious consciousness. The animals do not know anything about exerting themselves in order to improve their conditions or to progress in the way to higher virtues.”70 Even though he shows us a demarcation between humans’ intentional action and animals’ natural reaction, this picture is oversimplified. If humans’ knowing something about exerting themselves in order to improve their condition or to progress to higher virtues is based on rational thinking, it is definitely not the characteristic of Zen; if it is based on Zen wisdom, Zen wisdom would be the reason of an intentional action and thus could not be separated from rational thinking. As indicated by Suzuki, “all that we can do in Zen in the way of instruction is to indicate, or to suggest, or to show the way so that one’s attention may be directed towards the goal.”71 If the transcendental goal is not separated from the empirical world in the sense that what is done in our ordinary life is intentionally related to the goal, then the verbal and nonverbal actions done for Zen can be understood and interpreted in our rational language. If Zen masters’ verbal and nonverbal actions are intentional, they cannot be either totally irrational or totally non-rational. My question is that, without rational thinking, how can a Zen master show his way towards the goal of enlightenment? Should he and his disciples not be conscious of whether it is a right or wrong direction when they intend to go ahead? It is clear that, as an intentional action, making effort to attain Zen enlightenment cannot be separated from rational thinking. The locating problem seems inescapable for any religious thought which claims that their world of enlightenment or transcendence is radically or rigidly different from the secular world of common sense. But some special religious thought, such as Zen Buddhism as interpreted by Suzuki, which stresses its practical wisdom as necessarily supervened on the ‘Aufheben’ of rational thinking and logic, may have a good excuse for rejecting the request of locating. However, the locating problem is only performed as the first check-point of rationality, even though putting aside this problem, there is still a

70 71

D.T. Suzuki, 1991, pp. 85–86. D.T. Suzuki, op. cit., p. 92.

davidson’s charity

149

problem of landing, the second check-point of rationality, to which this kind of skepticism is required to respond. In other words, the masters may have some excuse to reject or to ignore the request of identifying or locating its non-conceptual and speechless entity; but they cannot escape the request of explaining the possibility of bridging the gap between the two worlds and the feasibility of transforming the ordinary self into the transcendental self (or no-self ). It is obvious that, without landing the transcendental mystery on empirical life, entertaining the private experience of the mystery would be non-sense or irrelevant to actual life. If it has to land on actual life, the Zen master has to explain how his ‘qi-zhi ’ ( natural disposition) can be transformed into the ideal state of Zen living, how he can know, from a view point without rational thinking and logic, that the ordinary understanding of the world from a view point of rational thinking and logic is illusive and delusive, and how he can know what he has attained is not the wrong thing. These are all questions of landing problem they have to answer. Suzuki’s Zen Buddhism shares with New Confucianism and other schools of Buddhism one of the major characteristics of mysticism and pantheism, i.e. the ultimate reality is both transcendent and immanent. Similarly, he also recognizes there are two levels of world (satori’s own world and a world of multitudes), two kinds of experience (Zen experience and ordinary experience), and two sorts of truth (Zen truth and rational truth). Furthermore, he also claims that Zen’s suchness, experience, or truth is both transcendent of and immanent in the ordinary world. However, we know that ‘transcendent immanence’ or ‘immanent transcendence’ is not a coherent concept,72 how can Suzuki bridge these two contrary terms in a coherent way? It is obvious that Suzuki encounters the same kind of difficulty as most of the mysticists and pantheists confront. Although Suzuki thinks that these two contrary terms are ‘essentially incommensurable’73, he still strongly believes that “[i]n satori what is immanent is transcendent and what is transcendent is immanent.”74 But it is merely a claim without argument. Of course, Suzuki would say that if we provide argument for the claim, satori would be murdered 72

See my article on the three dogmas of New Confucianism, Yiu-ming Fung, 2001, pp. 245–266. 73 D.T. Suzuki, 1994, p. 119. 74 D.T. Suzuki, op. cit., p. 47.

150

chapter five

by this dualistic or rational thinking and it would not be genuine satori. But how can Suzuki bridge these two extreme terms and thus solve the landing problem if no rational thinking is permitted? It seems that there is no other way to the end except appealing to satori itself. Actually, Suzuki’s strategy of persuasion is appealing to satori’s absolute and private authority. He thinks that “[t]he gap between satori and rationality could never be bridged by conceptmaking and postulation, but by an absolute negation of the reason itself, which means ‘an existential leap’.”75 According to Suzuki, an ‘absolute negation’ or ‘absolute affirmation’ can only be made by someone who has a ‘satori-eye’. It is therefore clear that appealing to satori’s authority can solve the landing problem; so, eventually, “Satori is ‘an existential leap’.”76 Nevertheless, mysticism and pantheism also claim the absolute authority of their inner, private, absolute, transcendental, mystic, and holy experience, and appealing to private access is obviously not a privilege of Suzuki’s Zen Buddhism. I think the landing problem is not easy for him to solve. Even if the Zen masters are able to escape the locating problem, they still have to solve their landing problem, otherwise they would face the following dilemma: On the first horn, if they really argue against conceptualization and duality, it would presuppose the conceptualization of its anti-thesis and make a duality between the thesis under attack and their own anti-thesis. It means that their anti-thesis would be self-refuting. On the second horn, if they want to transcend but not reject conceptualization and duality, their transcendental journey would be irrelevant to the daily world, a journey which is not only logically impossible to be identified but also substantially unrelated to actual life. It means that they cannot make any thesis or anti-thesis except sleeping on their private bed of mystery.

5. Conclusion In the previous sections, I have tried to discuss some problems about Davidson’s principle of charity. The first problem is the question of whether the principle can survive in the context of comparison between Chinese and English language and philosophy. In this regard, 75 76

D.T. Suzuki, 1994, p. 69. D.T. Suzuki, op. cit., p. 125.

davidson’s charity

151

A.C. Graham’s criticism has been sorted out as a salient example of the reaction issued from most Sinologists’ basic faith. It seems that one of the major academic interests of most Sinologists is motivated by their expectation of finding some alien characteristics in Chinese culture and language, or by the goal of so-called ‘Chineseness’. Even though there might be something called ‘Chinese-ness’, such as Graham’s idea about a distinct ‘root thinking’ in Chinese language and philosophy, this ‘Chinese-ness’ cannot be merely and basically described and explained as ‘Chinese-ness’, i.e. understood only in a ‘Chinese’ way of thinking and interpreted only in a ‘Chinese’ way of expression. So Graham’s criticism is fundamentally selfdefeating; it cannot be understood as a real challenge to Davidson’s principle of charity. The second problem is the methodological character of the principle of charity. Although most of the discussants on the problem are inclined to identify Davidson’s arguments as ‘transcendental’, I have argued that this is not accurate and that this misunderstanding is due to their not taking seriously some of the concrete examples provided by Davidson and not realizing a crucial distinction between two kinds of TA in the literature. The general form of Davidson’s principle may give people a perception that it is an a priori claim; however, there are a lot of concrete examples which constitute a strong inductive base of empirical evidence to support the principle. If Davidson’s principle is defendable either in a concrete context or through a methodological examination, and thus qualified as indicating a necessary condition of understanding and interpretation, can it be used to kick out all kinds of skepticism and relativism? As to this third problem, Suzuki’s interpretation of Zen Buddhism has been selected as a challenge to Davidson’s principle. I have argued that the idea that giving up rational thinking and logic as a necessary condition of attaining Zen could escape the locating problem required by Davidson’s principle. However, there is still a problem of landing. For the transcendental Zen to be relevant to the empirical world, it should solve its landing problem, for example, the problem of explaining how people can transform their ‘self ’ from an empirical into a transcendental state and how the Zen masters can criticize dualistic thinking without committing to dualistic thinking. While the locating problem is a request for identifying a location in a rational space, the landing problem, on the other hand, is a request for

152

chapter five

making sense of the relation between two kinds of truth or two levels of world. My strategy is not to use Davidson’s locating request to force the Zen masters to give up their non-dualistic thinking for identifying Zen’s location in a rational space, but to use the landing request to force them into a dilemma: either to explain the relation with our rational language and logic or to fall into an abyss of total isolation from the real world.

APPENDIX 1: A.C. GRAHAM ON CONCEPTUAL SCHEMES IN THE CONTEXT OF CHINESE LANGUAGE AND PHILOSOPHY

In order to illustrate the detailed differences to the bottom of thinking in different cultures, Graham gives some examples to explain why English and Chinese sentences are not fully intertranslatable. In his first example, he says that if “an English speaker says, ‘The cat sat on the mat’, and a Chinese, Mao wo tsai hsi-tzu-shang [Mao-wozai-xi-zi-shang ], only the cat is satisfying these conditions. For the English its posture is similar to a man sitting in a chair, for the Chinese to a man lying (wo) whether face forward or on his back. As for the mat, we cannot expect an unrelated language to share precisely our classification of floor coverings as mats, rugs, carpets; hsi-tzu [xi-zi ] is used of straw mats. In addition, the verb is tensed in English but not in Chinese.”77 Therefore, he thinks that the sentences do not express the same proposition, because “Mao wo tsai hsi-tzu-shang is true even if cat has never before now sat on the mat, false if it sat on a cloth mat.” More generally speaking, the reason is that “Chinese and English divide up and organise the world differently.”78 The second example mentioned by Graham is that when using “Grass is green” to match with “Ts’ao ch’ing” [Cao-qing ], we would find that “the meaning of ts’ao depends on a division of vegetation into ts’ao mu [Cao-mu ] ‘grass and tree’, implying a wider scope than our ‘grass’. Ch’ing is one of the Five Colours, the blue-green which contrasts equally with red, yellow, white and black. If grass were blue ‘Grass is green’ would be false but Ts’ao ch’ing would be true.”79 So, he considers these two sentences not translatable to each other. Based on these examples, Graham pushes to the point that the unintertranslatability between these two languages fundamentally lies in word level, not in sentence level; in correlative thinking, not in analytic thinking; in pre-logical process, not in logical operation. In 77 78 79

A.C. Graham, 1992, p. 65. Ibid. A.C. Graham, op. cit., pp. 65–66.

chapter five

154

this regard, to explain the ‘differences to the bottom’ is tantamount to understand the words by correlation within the scheme, a prelogical process which analysis assists but cannot replace. He believes that a native speaker is in command of his or her home language only when he or she stopped analyzing and applying grammatical rules to the chain of oppositions; and the gap of the chain, for example, ‘dog/dogs’, tree/trees, house/ . . .’, can be spontaneously filled with ‘houses’. So, he thinks, “if I learn the words primarily by correlating them, with analysis secondary even if employed at all, I understand the Chinese as I understand the English, and can confirm the truth of either Ts’ao ch’ing or ‘Grass is green’ by looking at grass and other herbs without bothering about translatability. If in a particular context a Chinese reports what he saw by Mao wo tsai hsitzu-shang, I am oriented towards what he saw as towards things I have seen myself, possibly but not necessarily by visually imagining as in my own case I visually remember. I can then say, ‘The cat sat on the mat’, as I might say, ‘You still have that cat then’, responding to the event which he observed without concern for whether I am saying what he said. I do have to co-ordinate the Chinese and English sentences, but will do so most accurately by correlation sensitive to more difference and similarity than I can analyse; there is no need to relate them logically because if I want to infer from one of them it will be in the same language.”80 Names, for Graham, cannot be understood as singular terms of logic; he treats names as the products of the classifying act of naming in a pre-logical pattern. He thinks that some of the Chinese and English names seem to match each other but actually differ in their different patterns of classification. An interesting example is the translation of the Chinese word ‘yang’. He says, “there is a single goat in plain sight, and X says Yu yang ‘There is a yang (conventionally translated ‘sheep’)’ and Y ‘There is no sheep’, I may be startled if I fail to appreciate that yang include goats as shan yang ‘mountain yang’ [ ]; but for anyone who has fully correlated the Chinese and English words the observation confirms both sentences.” Even though Davidson offers a similar case of agreement disguised by different usages of ‘yawl’ and ‘ketch’, Graham thinks that his account differs from Davidson’s in “not having to assume (even if it is indeed the

80

A.C. Graham, 1992, p. 67.

davidson’s charity

155

case) that the extensions of ‘yang’ and ‘sheep and goats’ precisely coincide, that they are intertranslatable like the dates of the two calendars.” The distinct picture of his example is that “[w]hen reading explanation in two languages of a vocabulary difference between them one is positively grateful that they do not say exactly the same thing, much as when collecting information about an incident one wants photographs taken from different angles at different moments.” On this basis, therefore, he judges Davidson’s criticism of Whorf ’s relativism unsound, because “there is no paradox here; Whorf would hardly have denied that bilingual readers would be clearer about the divergence with an equally sophisticated Hopi account to compare with his.”81 Unlike most of the views of justification held by the people in the field of epistemology, Graham thinks that “to confirm or to refute requires not only logic and observation but also checking whether words have the same sense.”82 In contrast to the justification by logic and observation, this ‘checking’ is not on the level of proposition, but on the level of pre-logical patterns of names. Hence, he continues, “[t]o escape the conclusion that all truth is relative to incommensurable conceptual schemes it is enough to show that the schemes themselves are patterns of names neither true nor false, and that factual statements depend on them for their meaning but not for their truth; we need not bother about what lies between these extremes.”83 In addition to the syntagmatic connextions of pairs in a chain, Graham’s notion of ‘conceptual scheme’ also includes the syntactic structures which organize words in sentences. To compare English with Chinese grammar, he believes the differences in conceptual schemes are also obvious. With respect to this point, Graham recommends Hansen’s claim that classical Chinese nouns in general are closer to English mass than to English count nouns. He says, “Hansen argues that Western thought is predisposed by number termination to conceive the world as an aggregate of distinct objects, Chinese by the mass noun to conceive it as a whole variously divisible into parts. Le Gall’s inability to understand the ch’i [qi ] except as a collection of atoms would be a good illustration. The hypothesis survives Harbsmeier’s classification; most or all philosophical terms would 81 82 83

A.C. Graham, 1992, p. 68. A.C. Graham, op. cit., p. 69. A.C. Graham, op. cit., p. 72.

156

chapter five

presumably be not count but generic nouns. This is plainly the case with ch’i; the yi ch’i ‘one ch’i ’ [ yi-qi ] divides into the erh ch’i ‘two (sorts of ) ch’i ’ [er-qi ] down to the wan wù ‘myriad (sorts of ) thing’ [wan-wu ]. That Chinese thought would be conditioned to divide down rather than add up is in any case suggested by other features of the language.”84 Graham, therefore, thinks that “[t]here would be no such compulsion to assume the primacy of individuals if English, on the analogy of Classical Chinese, lacked number termination, and we said ‘the closet’ and ‘its shoe and shirt’ as we say ‘its dust’ or ‘its smell’. The effect of number termination is such that we cannot even make the simple statement that language classifies things as similar or different without implying in advance that the ‘thing-s’ are different.”85 Graham believes that Chinese thinking is in terms of process rather than of static entities. Although Davidson recognizes that different languages may individuate differently over a certain range of words, Graham does not agree that this is only a local difficulty for translation. The impossibility for Davidson to explore such differences without sharing ‘concepts that individuate the same objects’, for Graham, is now magically transformed into a situation of possibility. Graham mentions, for example, “that a Chinese student of English has been assuming that yang and ‘sheep’ are synonymous but begins to doubt it. He points out a sheep and a goat, asks of both ‘Is that a sheep?’, and in the second case I answer ‘No, a goat’. He has no need to guard against the danger that I might take him to be pointing at the horn; I cannot answer ‘No, a horn’ because ‘horn’, unlike ‘goat’, is not on the same paradigmatic level as ‘sheep’. It would be less appropriate to his problem to ask the [sic] ‘What is that?’ which allows me to answer ‘A horn’, forcing him to introduce a shared concept by narrowing his question to something like ‘What is that animal?’.”86 Graham thinks that introducing syntactic structure into the conceptual scheme does not bring us nearer to epistemological relativism, because truths of fact are independent of the scheme. Here the word ‘independent’, he stresses, does not mean that factually true statements are translatable into any natural language, but that to 84 85 86

A.C. Graham, 1992, p. 4. A.C. Graham, op. cit., p. 75. Ibid.

davidson’s charity

157

confirm or refute a factual statement by reason and observation you have only to understand its place in the appropriate conceptual scheme, you do not have to share the scheme. Nor does it suggest that if schemes could be perfectly corrected by logic and observation they would all become the same.87 Graham interprets Wittgenstein, Ryle, Kuhn, and Derrida as his heroes who can make an important move to dig below the surface of our supposedly exact knowledge to find the correlative at its foundations. He thinks that this recognition is the same as the Sinologist’s when in searching for metaphorical roots of a Chinese concept he discovers that to compare and contrast it with Western concepts he has to explore their roots as well. In the end, the solution, he suggests, is to accept and come to terms with the thought that analysis starts from the results of spontaneous correlation.88

87 88

A.C. Graham, 1992, p. 78. A.C. Graham, op. cit., p. 81.

APPENDIX 2: THE LANDING PROBLEM IN SUZUKI’S IDEA OF ZEN

The landing problem in Suzuki’s idea of Zen can be focused on two questions: (1) How can a Zen master and his disciples transform their self from a delusive state into an enlightened state, how can they communicate with each other, and how can they know that they are enlightened? (2) How can a Zen master criticize the dualistic thinking from his non-dualistic thinking and how can he make sense of his anti-logical thesis without logic? One of the Zen masters’ dialogues (mondo ) with their disciples quoted by Suzuki can be used to illustrate his idea of giving up the duality of logic for entering into Zen enlightenment. This can also be used for explaining our first question about landing problem. Wei-kuan ( ), a Zen master of Tang dynasty, was asked by one of his disciples about the question “Where is Dao?” The dialogue continues as follows:89 Kuan: Monk: Kuan: Monk:

Right before us. Why don’t I see it? Because of your egoism you cannot see it. If I cannot see it because of my egoism, does your Reverence see it? Kuan: As long as there is ‘I and thou’, this complicates the situation and there is no seeing Dao. Monk: When there is neither ‘I’ nor ‘thou’ is it seen? Kuan: When there is neither ‘I’ nor ‘thou’, who is here to see it?

In comparison with Robinson Crousoe’s physical isolation from other people, the mental state of the Zen master stated above is absolutely and logically isolated from the other minds. We can call this state ‘Robinson Crousoe in an isolated mental state’, instead of ‘Robinson Crousoe in an isolated island’: the person entering into a mental state which is impossible to communicate with others. The terms ‘I’ and ‘thou’ mentioned above may have two alternative interpreta-

89 The translation is taken from Suzuki, 1996, p. 209; similar dialogue appeared in Suzuki, 2000, p. 342.

davidson’s charity

159

tions: one is the conventional use in ordinary language; the other a special use in Zen language. If it is the first option, i.e., Wei-kuan uses the term ‘I’ to refer to I and ‘thou’ to refer to thou as no different from his disciple’s ordinary use, then he is still in the complicated situation in which “there is no seeing Dao” as mentioned by himself. In other words, the requirement he states for seeing Dao is self-defeating, and his dialogue cannot help his disciple’s enlightenment. If it is the second option, i.e., Wei-kuan uses the terms ‘I’ and ‘thou’ not in the conventional way, but in Zen’s way, it is clear that his speaking is not against the conventional use of ‘I’ and ‘thou’, and thus not against the duality of the rational discourse, though we do not know what they really mean. In this regard, we can say that Wei-kuan’s mind falls into a lonely place; as a master with private Zen experience, he can be called ‘Robinson Crousoe in an isolated mental state’. Although he reaches the level of enlightenment, it is logically impossible for him to communicate with others. Furthermore, it is also logically impossible to explain how he could transform himself from the state of puzzlement to enlightenment, because he would not know what’s wrong with the conventional thinking for attaining enlightenment without an analysis of the conventional thinking. As regard to the second question of the landing problem, Suzuki’s idea is also inescapable from the predicament of self-defeat. If the negative sentence “The flower is not red, nor is the willow green” can be regarded by Suzuki as the same as its affirmative “The flower is red and the willow is green” when they are understood from two kinds of perspectives or seen by two kinds of ‘eyes’: the ‘Zen’s eye’ and the ordinary eye, then what Suzuki declares is definitely not really anti-logical or a violation of logic. Furthermore, if we agree, for argument’s sake, that there is ‘Zen’s eye’ and Zen’s experience, the Zen masters or Suzuki himself still cannot escape from the ‘trap’ of duality. In order to illustrate the self-refuting characteristic of Suzuki’s thesis of anti-dualism, specifically, and anti-logic, generally, let me pose the following hypothetical dialogue between Suzuki and me: Suzuki: The absolute affirmation must not be the one accompanied or conditioned by a negation.90 Y M: Could you give us an example?

90

D.T. Suzuki, 1991, p. 68.

160

chapter five Suzuki: Certainly! For example, Zen would say: “A is not A, therefore A is A” and “A is A, therefore A is not A.”91 Or more concretely, “I is not I, therefore I is I” and “I is I, therefore I is not I.”92 Y M: The two examples you have just said can be formulated as “~Q , therefore Q” and “Q , therefore ~Q.” If so, does your sentence “I is not I, therefore I is I” mean or imply the sentence “It is not the case that I is not I therefore I is I”? Suzuki: Absolutely, it doesn’t mean that! Y M: If we simplify the above sentence “~Q , therefore Q” as ‘P’, and “~(~Q , therefore Q )” as ‘~P’, would you reject the sentence “P means or implies ~P” or the sentence “P, therefore ~P”? Suzuki: If my answer is ‘yes’, so what? Y M: If you reject the sentence “P, therefore ~P,” how can you assert the sentence “Q , therefore ~Q”? Here they share the same form, and ‘P’ and ‘Q’ can be substituted by each other in the above sentences. So, I think your absolute affirmation or great affirmation is self-refuting. Suzuki: If so, shouldn’t I . . . I assert the paradoxes “Q , therefore ~Q” and “~Q , therefore Q”? Eventually, do I have nothing to assert? Y M: Absolutely, you do have something to assert! The Buddha or Zen teaches us no attachment. So long as you are not attaching to the idea of ‘anti-logic’ or ‘the transcending of logic’, and not excluding the duality in our ordinary way of thinking and in your Zen thinking, you will naturally and happily live in a world without attachment.

If someone wants to hold a thesis of anti-logic or the transcending of logic for freedom from the attachment of (logical) dualism, he or she would be inevitably trapped in the attachment to another dualism: the logical and the anti-logical. In reality, what Suzuki claims is eventually self-refuting. This is his predicament related to the second question of the landing problem.93

91

D.T. Suzuki, 1991, pp. 59–60; and 1996, pp. 115, 269. D.T. Suzuki, 1971, pp. 30–31; and 1961, pp. 272–273. 93 I would like to express my gratitude to my colleague Professor Angelina Yee for her comments on the English writing of this paper. 92

WORKS CITED

Davidson, Donald (1984), Expressing Evaluations, the Lindley Lectures (monograph), University of Kansas. —— (1985), “Incoherence and Irrationality”, Dialectica, 39:4. —— (1986), “A Nice Derangement of Epitaphs”, in E. LePore and B. McLaughlin (eds.), Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, Oxford: Blackwell. —— (1993), “Locating Literary Language”, in Reed W. Dasenbrock ed., Literary Theory after Davidson, University Park, PA: Pennsylvania State University Press, pp. 295–308. —— (1989),“The Conditions of Thought”, in J. Brandl and W. Gombocz eds., The Mind of Donald Davidson, Amsterdam: Editions Rodopi, Grazer Philosophische Studien Band 36, pp. 193–200. —— (2001), Inquires into Truth and Meaning (second edition), Oxford University Press. —— (2002), Subjective, Objective, and Intersubjective, Oxford University Press. Fung, Yiu-ming (2000), Gong-Sun-Long-Zi: A Perspective of Analytic Philosophy, Taipei: Tung Tai Book Company. —— (2001), “Three Dogmas of New Confucianism: A Perspective of Analytic Philosophy”, in Bo Mou ed., Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions, Chicago and La Salle: Open Court. Graham, A.C. (1992), Unreason within Reason: Essays on the Outskirts of Rationality, LaSalle: Open Court. Hahn, Lewis H. ed. (1999), The Philosophy of Donald Davidson, Chicago: Open Court Publishing Company. Jakobson, Roman and Halle, Morris (1956), Fundamentals of Language, Hague: Mouton. Jakobson, Roman (1978), Six Lectures on Sound and Meaning, Cambridge Mass.: MIT Press. —— (1985), Selected Writings, ed. by Stephen Rudy, Hague, Paris: Mouton. Lechte, John (1994), Fifty Key Contemporary Thinkers: From Structuralism to Postmodernity, London: Routledge. Moore, Charles A. ed. (1951), Essays in East-West Philosophy: An Attempt at World Philosophical Synthesis, Honolulu: University of Hawaii Press. Putnam, Hilary (1984), Reason, Truth and History, Cambridge, Mass.: Cambridge University Press. Stroud, Barry (1968), “Transcendental Arguments”, The Journal of Philosophy, 65, pp. 241–256. Suzuki, D.T. (1951), “The Philosophy of Zen”, Philosophy East and West, 1:2, pp. 3–15. —— (1953), “Zen: A Reply to Hu Shih”, Philosophy East and West, 3:1, pp. 25–46 —— (1954), “Zen and Pragmatism: A Reply (Comment and Discussion)”, Philosophy East and West, 4:2, pp. 167–174. —— (1955), Studies in Zen, New York: Dell Publishing Co., Inc. —— (1960), “Lectures on Zen Buddhism”, in D.T. Suzuki, Erich Fromm, and Richard De Martino eds., Zen Buddhism and Psychoanalysis, New York: Harper and Row, Publishers. —— (1961), Essays in Zen Buddhism (First Series), New York: Grove Press, Inc. —— (1971), What Is Zen? London: Buddhist Society.

162

chapter five

—— (1991), An Introduction to Zen Buddhism (revised edition), London: Rider & Co., An Imprint of Random House Group Ltd., 1969. —— (1994), Living by Zen, York Beach, Maine: Samuel Weiser, Inc. —— (1996), Zen Buddhism: Selected Writings of D.T. Suzuki, ed. by William Barrett, New York: Doubleday & Company, Inc. —— (2000), Essays in Zen Buddhism (Third Series), New Delhi: Munshiram Manoharlal Publishers Pvt. Ltd.

PART THREE

RATIONALITY, NORMATIVITY, AND INTER-CULTURAL DISAGREEMENT

CHAPTER SIX

DAVIDSONIAN RATIONALITY AND ETHICAL DISAGREEMENT BETWEEN CULTURES Samuel C. Wheeler

Introduction Chinese ethical thought has long been a paradigm for the West of a way of thinking that is sophisticated, highly developed, advanced on any reasonable scale of advancement, but very different in many judgments from the ethical thinking of the West. Stock examples include differing degrees of respect for authority, different views about the responsibility of groups for the actions of individuals, and the like. This essay investigates whether such divergence shows that ethics is less ‘objective’ than natural science. This essay concerns the abstract question of the objective truth-values of ethical sentences, rather than any particular difference that might be claimed to hold between these two cultures’ ethical thinking.

1. Kant and Founding Ethics on Rationality Many ethical theories have attempted to show that it is rational to do the good. The idea is that the relatively uncontroversial normativity of correct reasoning can found the normativity of morality. The most ambitious of these theories was that of Kant. Kant attempted to show that it was irrational to do anything other than the right thing. The argument for the categorical imperative is an argument that the very concept of acting on purpose requires that a perfectly rational agent do only what has a coherent general principle. The least a perfectly rational agent could do, given that he was acting for reasons, would be to act so that he was consistent, because acting for reasons requires acting according to considerations that apply generally. If the considerations are universally applicable, it ought to be possible that everyone follow them. Thus acting for reasons consistently requires acting in a way that anyone could act.

166

chapter six

Kant hoped that this criterion of consistency would suffice to give both necessary and sufficient conditions for moral actions. Kant’s program was quite ambitious. He not only tried to supply a rational ground for doing good when a person knows what the good is, he also required that there be a rational basis by which one could always know what the good is. Rationality would not only provide a reason for doing the good, but would also provide a criterion for the good that would enable a careful person always to do the right thing. Kant’s conception of ‘doing the right thing’ is ‘doing one’s duty’ or ‘fulfilling one’s obligations’. An ethics based on obligations and duties is essentially deductive, since principles of obligation yield arguments that can be known to be sound on the basis of limited information. Thus, a Kantian theory requires that ethics have the structure of a system of principles, so that ethical reasoning is primarily deductive. Given adequate principles, a person is equipped to know the right thing to do in every circumstance. Thus the person must be able to deduce the correct thing to do on limited information.

2. Davidsonian Kantianism in Outline A combination of several ideas in Donald Davidson’s philosophy suggests a way to reconstitute something like Kant’s project. For Davidson the normativity implicit in reasonable interpretation of another as an agent supplies the basis for understanding the normativity of ethical concepts. Interpretation, though, uses a much richer notion of rationality, one that includes ‘inductive’ logic. The conception of ethics that results is not deductive. No algorithm will allow even a perfect rational agent to always do the right thing. Davidson’s ideas connect the normativity of rational interpretation to the normativity in the truth-conditions of ethical language. On this account, the normative is not different in ontological or epistemic status from the factual. Let me sketch some of these Davidsonian ideas: a) Interpretation as Maximization of Agreement: Davidson, following the suggestions of Quine, uses the relatively testable and empirical notion of what we do in understanding another

davidsonian rationality and ethical disagreement

167

to analyze the concept of ‘rational agent’. He determines the content of ‘rational agent’ or ‘entity acting for reasons’ by seeing what constraints we apply in interpreting another. Davidson’s account of interpretation yields an account of rationality that is rich enough to give hope that one could show that doing good is reasonable. Interpretation maximizes agreement, so that interpretation is constrained by the necessity for treating the other as mostly believing the true and wanting the good. ‘Maximize agreement’ is a formula for a set of probabilistic constraints whose content is, roughly, “Interpret on the supposition that the entity being interpreted is an agent.” “Maximize agreement” says, “Interpret the agent by using the features of a paradigm agent, namely oneself.” To interpret according to a probabilistic fit with our own case of rational agency is to apply the term ‘agent’ on the basis of a hypothesis that the other is like us, an agent. Thus, since this is the general procedure for applying predicates, the reasonableness of such constraints is a priori. Familiar examples of the operation of maximization of agreement discuss maximization of agreement in beliefs. But action interpretation must also maximize agreement in desires, that is, conceptions of the good, as well. When someone drops a rock on his foot, the interpretation that the person wanted pain and believed this was an effective way of bringing it about is reasonably rejected, barring very special circumstances,1 and interpreted as unintentional. Action interpretation always maximizes agreement in both desire and belief. Given only that the other entity is an agent, and given only that x is a belief or desire, if we have that belief or desire x, so, probably, does the other entity.2 A broadened conception of the rational that constrains the content of desires is required for action interpretation. Without such constraints on content, interpretation could not get started, since any behavior is consistent with any beliefs, if that any desire whatsoever

1 For instance: Aliens have landed on campus and are choosing whom to abduct for their experiments. They wish to take only those without already existing pain, so as to have an accurate baseline. In this situation, seeking pain would be an explanation. 2 From this thesis alone, some degree of agreement on ethical matters follows. Other things being equal, a value one individual holds is a value the individual being interpreted holds as well. Some agreement on values, conceptions of the good, follows from another entity being a rational agent.

168

chapter six

is as likely to be present as any other.3 So, such constraints as that a person ought not to want pain are part of the concept ‘is a rational agent’. Speech interpretation is a special case of action interpretation. What a person is doing is, for example, asserting that the cat is on the mat. Speech interpretation, though, requires hypotheses about the speaker’s intention. For instance, in interpreting an utterance as a sincere assertion, one applies ‘maximize agreement in beliefs’ in the light of the hypothesis that the person desires to say the truth. Thus we seek a truth-definition for the language the person is speaking that will make the assertion true. Given that the utterance is a sincere assertion, i.e. that the intention is to tell me what is the case, and given that the speaker is by and large a believer of truths, the utterance ought to be true. The ‘ought’, which is close to a ‘probably’, falls out of the very structure of the constraints on interpretation—that they are maximization constraints and not absolute constraints. Briefly, principles of interpretation are ‘ought’ principles. b) Ought Sentences Davidson’s discussion of weakness of the will4 presents a brief account of conditional ‘ought’-sentences (hypothetical imperatives) that suggests a connection of ethics with interpretation-theory. Roughly, conditional ‘ought’ sentences are formally akin to conditional probability sentences, in that they do not ‘detach’ when the antecedent is true. In logical form, conditional ‘ought’-sentences are relative to something like an consideration-base, akin to an evidence-base to which a conditional probability claim is relative. That is, the apparent ‘antecedent’ of a conditional probability claim, an ‘If A, then probably B’ claim, is really the evidence relative to which the ‘consequent’ is probable.5,6 By analogy, conditional

3 Rather than try to give a general proof of this, I will give an example. Consider a person who is tying his shoe. Given an overwhelming desire that a comet crash into Connecticut and an utter lack of concern about shoe-wearing safety, this could reflect the belief that tying this shoe now will bring that about. 4 In Davidson, “How is weakness of the will possible?”, Essays on Actions and Events, Oxford 1980. 5 This point about conditional probability is from Hempel (1960), “Inductive Inconsistencies”, Synthese 12, pp. 439–469. 6 ‘Is probable’ is related to ‘the probability of A, given B is 0.5’ in something

davidsonian rationality and ethical disagreement

169

‘ought’ sentences give the bearing of a consideration on whether a person ought to do something. ‘Ought’, like ‘probably’, is fundamentally inductive. In contrast, ‘obligation’ sentences, the focus of Kant’s ethical theory, have the logic of deductive argument, as outlined in various deontic logics. But ‘ought’, not ‘obligation’, is the primary ethical concept, since one can always ask whether one ought to fulfill an obligation. If ‘ought’ is the primary ethical concept, and Davidson is right about the form of ‘ought’-sentences, then we get a very different picture from Kant’s of the connection between rationality and ethics. Ethical arguments and ethical thinking are properly understood on the model of induction rather than deduction. In an inductive argument, a set of true premises {p1 . . . . . pn} that strongly leads to the conclusion B is compatible with the existence of a true premise pn+1 which, together with {p1 . . . pn}, strongly leads to the conclusion not-B. In an ethical consideration, a set of truths {p1 . . . . . pn} may strongly lead to the conclusion that you ought to do B, while there can be another truth pn+1 which, together with {p1 . . . pn}, strongly leads to the conclusion you ought not to do B. If ethical argument is like induction, then ethical reasoning does not proceed according to Kantian universal moral principles. Briefly, an ethical actor will not in general know that he is doing the right thing, since he is always operating on limited information, using guidelines that allow for the possibility that further information can undermine the conclusion so far reached. The ‘principles’ that can be applied are not universally quantified commands, but rather ‘rules of thumb’. No algorithm will determine, given a description of a situation, what in fact a person should do in that situation. Furthermore, the rationality that might be the foundation of ethics would require much more than formal consistency. Ethical conclusions are more like scientific conclusions than like logical consequences. Just as there is no algorithm for the construction of a new theory in the light of novel experience, so there will be no algorithm for the correct ethical judgment, given that every situation is novel in some respects.

like the way that ‘tall’ is related to ‘Is 2 meters in height’. ‘Is probable’ and ‘probably’, the adverbial form, are attributives on a dimension of ‘the degree of support the “antecedent” gives to the “consequent”.’ ‘If A, then probably B’ means, roughly, ‘the probability of B given A is pretty high’.

chapter six

170

Davidson’s account on which ‘ought’-sentences have a form similar to that of conditional probability sentences suggests a theory of ‘ought’-sentences. If ‘ought’-sentences are essentially like conditional probability sentences, then their categorical form has an implicit relativization. Just as you may say “It will probably rain”, so one may say, “We ought to help those people.” In both cases, there is an implicit relativization, represented more explicitly by an ‘All things considered’ clause. Such relativizations suggest a unified account of the distinctions between the various ‘senses’ of ‘ought’. Consider the general principle that one ought to believe the logical consequences of what one believes. This principle has obvious counter-examples. If Fred believes that Susan is honest and that Susan has been embezzling his funds, he should not conclude that honest women sometimes embezzle, but rather ought to give up one of his premises. The various ‘senses’ of ‘ought’ are to be understood as further relativizations to backgrounds. For ‘ought’’s applied to human intentional actions, various implicit ‘consideration’ relativizations yield different ‘ought’’s. The ‘prudential’ ‘ought’ might be relativized to an implicit ‘set’ of a person’s self-concerned interests, the ‘logical’ ‘ought’ might be relativized to considerations of consistency, and so on. A Davidsonian account treats all ‘ought’ sentences as having the same basic structure and semantics. The methodological principle operative is that, other things being equal, a theory should minimize homonyms. Could the ‘ought’ in “You ought not to inflict pain” and in “If you turn the key, the car ought to start” be the same word? The two occurrences may be different words. However, a plausible theory that makes the two ‘oughts’ the same word is preferable to a theory that multiplies homonyms.

3. Relativizations a) Two Kinds of Relativization A conditional ‘ought’ utterance, just like a conditional probability utterance, has two general kinds of relativities. These are further complicated by the fact that they are sometimes implicit rather than overt:

davidsonian rationality and ethical disagreement

171

(1) The first kind of relativity is often overtly specified in the ‘antecedent’ of the utterance. “If you want a nice dinner, you should go to Cavey’s” recommends an action relative to a desire. On the normal understanding of such an utterance, the utterance is a ‘prudential’ ought. The ‘prudence’ indicated is relative to having the desire specified in the antecedent. Given that desire, and given information that you have, and so that the other should have, the action that the other should (prudentially) take is to go to Cavey’s. When the utterance is a prudential ‘ought’, the antecedent gives a special circumstance which, given ‘prudential’ considerations, make it reasonable for a person to choose Cavey’s. These ‘conditional’ relativizations describe what, relative to the ‘sense’ of ‘ought’ (the “considerations” described below), a person should do given all the surrounding circumstances. That is, in the light of all the information available,7 if you want a good dinner, you should go to Cavey’s. The bearing of the information on what you should do is by and large inductive. Features of Cavey’s, information about your tastes, information about alternatives, and so forth reasonably lead to that conclusion, albeit not by an algorithm. In effect, such reasoning is a ‘practical syllogism’ with the ‘syllogism’ expanded to include more adequate forms of reasoning. (2) The other relativity distinguishes the various ‘senses’ of ‘ought’. This is relativity to considerations intended to be taken to bear on the choice. To illustrate, consider a player in a chess game, in a position where she can force mate in four with a bishop sacrifice at f7, while every other move loses. In one ‘sense’ of ‘ought’, it is clear that she should go bishop f7 check. But her opponent is her boss, who will be angry if this employee defeats him in chess. So, prudentially she ought to make some other move than bishop to f7 check. But her loss to the boss may demoralize her colleagues so that their lives are worse if she loses, even though her own life is worse if she wins. Or, it might be dishonorable to lose a game on

7 A question here is whether the ‘ought’ is relative to all the information that’s available or to all the truths that are relevant. If, unbeknownst to all, a mediumsized meteor is in fact headed for Cavey’s, and meteor-impacts greatly diminish dining pleasures, then what is reasonable to do given all available information may differ from what it is reasonable to do objectively.

172

chapter six

purpose. The noble thing for her to do, and perhaps what she ought to do, is win.8 These different understandings are different restrictions on the considerations relevant to what to do. In the first case, the restriction might be called ‘chessic’. Only the goal of chess, mating the opponent’s king, is taken to be relevant. What the person ought to do is taken relative to just those considerations. Considerations that concern only her personal welfare would yield a prudential ‘ought’. If the recommendation that she move bishop to f7 were interpreted prudentially, the claim would be that, taking into account only considerations relevant to the ‘self-interested’ desires of the agent, she should go bishop to f7. The sentence itself does not determine this relativization. Rather the intent of the utterance does. So, even though the unqualified recommendation in a book of chess problems that bishop to f7 is the best move, it may not be, given another intended set of considerations. Such restricted background considerations can be regarded as a set of interpreted sentences,9 but the conditions of set membership are too vague to make that a precise proposal. In particular, the notion of ‘personal welfare’ and ‘self-interest’ will turn out to be irredeemably vague. The background considerations will have to remain an unanalyzed primitive, akin to the backgrounds that are presupposed in remarks that it will probably rain, or in counterfactuals.

8

Analogous relativization occurs with antecedents in the ‘logical’ ‘ought’. “If you believe it’s raining and you believe it’s cold, you ought to believe it’s cold and rainy.” Nothing recommends the belief apart from the antecedent, combined with the limitation of considerations at play to the ‘purely logical’. In the case of the ‘logical ought’ other considerations that over-ride are even more apparent. If the evidence is very much against rain and against cold, then a person who believes it is raining and believes it is cold should abandon both beliefs rather than believe their conjunction. 9 A Davidsonian will need a theory of propositions that will fit the roles of ‘proposition’. The details of how one does this by application of Davidson’s “On Saying That” (in Inquiries into Truth and Interpretation, Oxford UP, 1980, pp. 93–108) are not clear.

davidsonian rationality and ethical disagreement

173

4. Questions of Logical Form: How Do These Two Kinds of Relativizations Work? As Davidson showed long ago, the relativizations cannot be thought of as having the form of conditionals that detach. Thus a conditional ‘ought’-sentence cannot have the form of a connective joining two independent clauses, one of which has an ‘ought’. This implies that the conditional ‘ought’ is basic, and that the absolute, categorical ‘ought’ is something like “ought, all things considered. This accords with recent results in probability theory that argue that conditional probability, rather than absolute probability, is the fundamental probability notion.10 So, an ‘ought’-sentence must be a twoplace modality, a construction that depends on more than the truth-values of the two clauses. Modality brings in a complicated set of difficulties about logical form and ontology on its own. A difficulty, then, is that a serious Davidsonian account of the form of ‘ought’-sentences will have to include a general account of modalities, propositions (or their surrogates), and conditionals generally. This is a daunting task, and one that by no means complete. Let me sketch some of the features of a completed Davidsonian account, which features I will not address in any detail here: First, the account should treat ‘if . . then’ so that ‘if ’ must be doing the same thing in ‘if . . then . . ought’ sentences as in other occurrences. Second, some account must be given of the objects that take the place of A, B and C. If ‘ought’ is a predicate, the considerations C, the ‘antecedent’ A, and the ‘consequent’ B must be construed as some kind of entities. A Davidsonian would assimilate such objects to the general type of demonstrable linguistic objects, along the lines of “On Saying That”. Third, some general account of modalities will be required. For a Davidsonian, it is arguable that modality must be primitive.11 The 10 See Alan Hajek’s “What Conditional Probability is Not”, Synthese, Vol. 137, December 20, 2003. 11 In radical interpretation, having a ‘theory’ of another’s speech is having an account of what alternative utterances would mean. Counterfactuals are admitted at the very core of the theory. Since an account of what a person means on an occasion of utterance is central to the Davidsonian account of meaning, there is no hope for a non-circular reduction of the modal to the linguistic, since the linguistic presupposes the modal.

174

chapter six

different ‘senses’ of the usual modal ‘operators’12 will be different relativizations to considerations, as sketched above for ‘ought’. ‘Considerations’, on this account, are construed as sets of ‘things said’ that, for the case of ‘can’t’, function as limitations on the principles from which the negation of the sentence is a consequence. The modalities typically treated, ‘necessary’, ‘possible’, and the like, are based on logical consequence, and so would have deductive underpinnings, relative to different considerations. Given that there is no hope of reducing induction to deduction, analogous modalities that depend on inductive connections would use a primitive ‘consequence’ relation about which there is only the beginning of a theory. ‘If . . then . . probably’ and ‘if . . . then . . ought’ express such modalities. So, a working proto-theory would be that the form of conditional ‘ought’-sentences is Ought (C, A, B) rather than (A->Oughtc B), where ‘C’ stands for ‘considerations’, and where ‘ought’ is a modality whose rough content is that ‘Ought (C, A, B)’ is true if and only if, given A, B is reasonable on considerations C. Rather than being some kind of conditional with a consequent containing a ‘sense’ of ‘ought’, a conditional ‘ought’ sentence is a three-place modal predication among an ‘antecedent’, a set of considerations, and a ‘consequent’. In order to fill places in a predication, A and B and C must be surrogates for propositions or sets of propositions, rather than names of truth-values. Thus ‘ought’ is a predicate, with the “relativizations” specified by A and C and the action by B.

5. Connections between ‘ought’ and “conditional probability” ‘Ought’ often amounts to ‘probably’. ‘If you turn the key, your car ought to start’ is usually true if and only if ‘If you turn the key, your car will probably start’ is true. The relativization in a probability sentence is not merely to the contents of the ‘if ’-clause, but to ‘background conditions’ as well. The assumed overall conditions 12 For instance, the ‘senses’ of ‘can’t’ in which I can’t be at the meeting (because I have a dentist’s appointment), in which I can’t get to China in one billionth of a second, and in which I can’t find a ratio of integers whose square is two. I am supposing that the modal ‘operators’ can ultimately be treated as predicates of ‘things said’ or sets of ‘things said’.

davidsonian rationality and ethical disagreement

175

that the laws of nature obtain, the material in the tank is gasoline, and so forth would also be implicit in the ‘antecedent’ or relativization of the “probability” sentence. While ‘if then probably’ sentences and ‘if then ought’ sentences are connected, a sharp definition of one in terms of the other is neither likely nor necessary. ‘ought’ and ‘probably’ are each semantically independent modal predicates. A rough account of ‘ought’’s connection with ‘probably’ would rely on the probabilistic, maximization character of agreement. Given any feature that correct interpretation maximizes, if an individual being interpreted is indeed an agent, then that individual, relative to being an agent, probably has the feature. For any truth, then, relative just to an entity being an agent and the truth being a truth, the agent probably believes it. ‘ought’ is then roughly a chain of such probabilistic connections. A person who believes as he ought, given his information, makes every inference that he would probably make.13 The picture of ethical reasoning is that any complex logical or practical inference can be broken down into a series of smaller steps. Each of the beginning premises is a desire or belief the person probably has. Each of the smaller steps is one that a person would probably make, (and so ought to make), just in virtue of being a rational agent. So, ethical reasoning would arrive at ‘ought’-sentences by constructing what a person would do if that person always did what, on principles of reasonable interpretation, the person would probably do. For most real-life cases of an ethical decision, the chain of calculations and inductive judgments in the ideal determination of what a person ought to do will be enormously complex. Although some actual ethical reasoning is something like ideal reasoning, most of the time various short-cuts are appealed to. The short-cuts, or ethical principles, are rules of thumb that by and large give the right results. Given the information-costs of thoroughly calculating every decision, it pays to have devices that give the right result much of the time. Such short-cuts on occasion give the wrong result. In that respect they are analogous to the irrationalities that Twersky14 and 13

See my “Inference and the Logical ‘Ought’”, Nous, Volume VIII, No. 3, September 1974, pp. 233–258. 14 See Twersky, Amos and Kahneman, Daniel (1974), “Judgment under Uncertainty: Heuristics and Biases”, Science, 185, pp. 1124–1131.

176

chapter six

others have researched. So, the ethical principles that Kant took to constitute the core of ethical reasoning, and that allow deductive determination of what a person ought to do on a given occasion, are not, from a Davidsonian perspective, central at all, even though they are important and largely true.15

6. The Moral ‘ought’ a) Outline According to this theory-sketch, the “moral” ‘ought’ is the minimal restriction of considerations, the relativization of relevant considerations just to ‘is a person’ or ‘is a rational agent’. The ‘moral’ ‘ought’ is ‘ought, all considerations considered’ (considered in the light of what it is to be an agent). The rationality that is built into interpretation is not just formal, but substantive: What a person desires at bottom is interpreted by ‘maximize agreement’. The thought is then that, given rational desires (in the content as well as structural sense) and given rational practical inferences, what a person ought to do will turn out to be the output. In effect, this yields the familiar idea that, given that people at bottom want the same things and are reasonable, there is a basic moral agreement. Disputes, one might hope, can be settled by reasoned discussion rather than by brute force. For this idea to yield morality as we understand it, the rational desires will have to be strongly shaped by an argument that other peoples’ desires rationally count, just as a person’s own desires do, that it is irrational to show exclusive concern for one’s own interests. Kant tried to show this by arguing from the very notion of purposive action, and its connection with universalizability. More recently, Thomas Nagel’s book, The Possibility of Altruism16 tried to establish this, likewise by reflecting on what it is to do something for a reason. Unless a connection between rationality and counting others’ desires as rationally motivating holds, the ‘moral ought’ is a chimera. I have 15

There is something called “obligation” that has to do with contracts and promises. The long-standing tradition of identifying morality with obedience to law, specifically the covenants with a divine being, have misled Western moral thought. 16 Nagel, Thomas, The Possibility of Altruism, Princeton UP, 1970.

davidsonian rationality and ethical disagreement

177

no arguments better than Kant’s or Nagel’s, although my discussion of preference below will try to make the Kant-Nagel thesis seem natural. Even given the soundness of the Kant-Nagel argument that others’ desires count, can an adequate moral theory resolve ethical issues and arrive at ethical truth? I will argue that the answer is, “Sometimes, to exactly the extent that physical discussions can resolve disagreements about ‘fact’.” This weaker result leaves us with a substantive morality, but without the idea that every ethical question has a truthvalue we can determine by calculation or discussion. Sincere people, perfectly rational from their own points of view, can have irresolvable moral disagreements. But this does not mean that all moral questions lack truth-value. The presumption that irresolvability means relativity for ethics but not for physics rests on a mistaken view of preferences. The lack of resolvability will be most apparent between cultures with different histories and different traditions, as we will discuss below. However, we should expect the same irresolvability, perhaps on a smaller scale, between any two individuals. Just as for a Davidsonian the fundamental language is the idiolect, so in cultural terms the elementary culture is the individual. ‘Same culture’, like ‘same language’, has no strict sense. Nevertheless, of course, I can loosely say that I speak English and am a Westerner. The difference from Kant’s hoped-for result derives from the nature of the constraints on interpretation that give a Davidsonian conception of rationality. One important aspect of these constraints, and the rationality that is constructed on their basis, indicates that irresolvable differences will not disappear: The constraints are maximization constraints. The other shares one’s beliefs and values only probabilistically. Even when the values are shared, the ranking of any pair of goods is only probabilistically the same. Furthermore, our inductive inference-practices are only probabilistically identical. So, in effect, just as each of us has his own idiolect, each of us has his own rationality. This feature is part of the cost of a notion of rationality that goes beyond mere structure, but rejects the analytic-synthetic distinction. Allowing that there can be rational constraints on content, so that an interpretation that ascribes a desire for pain is probably rejected, would still give us a sharp notion of the rational if there were a sharp line between desires that are essential to being rational and

178

chapter six

desires that are not. However, for a Davidsonian, there is no sharp line between a part of the theory that is essential and a part that is contingent. Of course, there are important differences between parts of theories, and important differences between desires. Some desires can only be ascribed given overwhelming evidence; others are ascribed failing strong evidence that the person lacks the desire. But there are neither desires every rational agent has to have nor desires that no rational agent could have. b) A Problem with Objectivity? Both of the above considerations, though, apply to physical theories as well as to opinions about what one ought to do. Why do questions of ethics appear more subject to unresolvable disagreement than questions about truth-values of non-ethical claims? The usual answer is that, whereas agreement in physical theories involves only getting beliefs in agreement, ethics involves beliefs and desires. Desires, it is claimed, unlike beliefs, do not ‘correspond’ to anything objective, so that people differ in preferences about which there is nothing like a ‘true’ or ‘false’ to supply objectivity. Unlike differences in belief, differences in desires, it is held, reflect nothing but internal differences in persons. There is nothing like triangulation for desires. There appears to be no common world of value to which both an interpreter and the interpretee are related to when an individual expresses a preference. So the interpretation of actions lacks one of the groundings in a common world that interpretation of belief has. There appears to be no ‘common world of correct desirability’ by which people learn what is better than what. I argue below that this apparent lack of a common ground that functions as a common world is illusory. c) A Quinean Hedonistic Solution to Objectivity One might hope that physiological states such as pleasure and pain could be the stimulations that ground all desires by supplying something analogous to the common world in which belief-triangulation takes place. A Quinean version of hedonism could then say that the relation between the good and pleasure and pain is like that between external stimulations and a physical theory. Just as there is no sentence-by-sentence reduction of physical sentences to patterns of exter-

davidsonian rationality and ethical disagreement

179

nal stimulation, even though the theory does nothing but organize stimulations, so there is no good-by-good reduction of valuations to pain and pleasure. An account of valuations would thus be grounded in internal stimulations, but holistically. A ‘web of belief and desire’ would distribute the ‘sensory meanings’ of sentences as degrees of belief and valuation among the sentences of the language. This would be a kind of pragmatism, since the organism would desire whatever maximized its welfare. Differences in preference would be explained by differences in physiology, leading to an innocuous relativism. If the arguments that it is rational to value other agents’ welfare are sound, the theory would be a grounding for morality as well as for prudential thinking. Such a theory would say that, just as it is unreasonable to take one’s own perceptions as the only reliable ones, so it is unreasonable not to take the preferences of others as counting in the determination of the objectively valuable. Quinean hedonism could thus be a universal utilitarianism of a sort. The difficulty with such a theory is analogous to the difficulty with the Quinean Web of Belief. The theory supposes that sensations are given unconceptualized desires, the periphery of desire-stuff that is organized via beliefs into particular preferences for, e.g., volleyball over badminton. Absent such a given, the special tie to pleasure and pain as physiological phenomena would drop out. But something akin to the above theory could be a Davidsonian account of ethical truth.

7. Davidsonian Ethical Objectivism A Davidsonian can treat ethics as objective, and hold that ethical sentences have truth-values, even without a basis in pre-conceptualized desire-stuff. Consider the apparent difference between beliefs and preferences that marked a difference between ethics and physics above. We need not regard differences in preferences as brute, just as we need not treat idio-rationalities as making “rationality” relative to persons. Just as we do not generally identify beliefs with truth, so there is no reason to identify preference with our good or with the good. Given our past experience, we know that some of our beliefs are likely to be mistaken. In the same way we find that some of our desires were ill-considered. Thus, just as we do not identify the true

180

chapter six

with our beliefs, so we do neither identify the good with what we want nor identify the rational with our total theory. We can realize that we are less than perfectly reasonable, even though we hold, of each particular view, calculation, and valuation, that it is reasonable. The decision-theoretic tradition has taken preferences as just brute facts about which nothing is rationally required but coherence, whereas beliefs are true or false. But in interpretation, we clearly take some preferences to be irrational, such as the simple preference for pain over lack of pain. Some preferences can only be assigned to an agent on the basis of very strong evidence. “Pain is worse than no pain, other things being equal,” is true. So, prima facie, some ‘better than’ sentences are objectively true. If some ‘better than’ sentences are true, the reasonable supposition is that “better than” sentences usually have truth-values. The conception of preferences that suggests itself is that a preference that A rather than B is a belief that A is better than B. The causes that give rise to preferences would then be construed as akin to the causes of the involuntary utterances17 that are sensory judgments. On this conception, pain and pleasure would, in motivating action, be involuntary utterances ‘This is bad’ and ‘This is good’. Such involuntary utterances, of the speaker to himself, as it were, are incorporated into the web of belief and desire. Sometimes they are over-ridden by other considerations, just as sudden apparent flashes of light are over-ridden by the consideration that you may have an eye problem. That is, we voluntarily get our separated shoulder replaced in its socket by a very painful procedure. Such overriding is not the discovery that the pain is good, but rather that, while the pain is bad, the conjunction of the pain and the cure is good. Given that the person is reasonable, and knows that the same event will be both a pain and a cure, the person takes steps to bring about the event.18 But this does not mean that the person’s evaluation of the pain itself was mistaken. A kind of practical failure of reason can explain how that a person cannot bring himself to bring about the event, even though it is better and, ‘intellectually’, known

17 On an account on which there is no given, sensory beliefs can be construed as utterances that occur involuntarily, and have to be interpreted. In adapted Quinean terms, such beliefs have to be incorporated into our web of belief and desire. 18 This is essentially Aristotle’s account in the Nichomachean Ethics.

davidsonian rationality and ethical disagreement

181

to be better. In that event, the immediate evil, the involuntary utterance that is the pain, is over-rated relative to the future good. I argue below that we can regard the wide-spread disposition to choose for actions favoring one’s own interests as another example of the same kind of failure of reason.

8. ‘Better for Me’, ‘Self-Interest’, and Objectivity If agreement on the Good is indeed what we maximize in interpretation, we should not take into account only our own involuntary utterances about the Good and the Bad. If triangulation is to apply to the Good, the objectively good must be prior to both what I think is good and what is good for me. Only if that is the case could we claim that everyone’s desires reasonably ought to count as reasonable for me to take into account. I will make this plausible in two stages: First, I will argue that ‘self-interest’ is by no means as clear a notion as it has often been taken to be. Second, I will explain how it is that most valuations can be correct while almost everyone fails to take the moral point of view. I will then conclude the section by examining what sort of ethics a Davidsonian will have. I will argue that we should not expect that a Davidsonian ethics will be a utilitarianism. Whether and to what extent ethical sentences have non-relative truth-values will depend on what is the case about partiality. a) What is ‘Self-interest’? In outline, the argument of this sub-section will be as follows: The particular interests of any normal agent go far beyond anything identifiable as particular states of the particular organism that is the agent, or ‘self-interest’ as usually conceived. Usually the identification of one’s interests with the interests of others is limited to a subset of the others, and so would be termed a ‘partiality’, a discrimination. Ascribing such partialities is part of interpreting action and speech, and so part of the rationality that is part of interpretation. Interest and lack of interest in the welfare of others is thus subject to rational evaluation. ‘Reasonable’ attachments are maximized in interpretation. But, if Kant and Nagel are right that it is rational to extend one’s concerns to the interests of every entity that has interests, then

182

chapter six

it is reasonable to extend the scope of one’s concerns with the interests of others to everyone. If ‘X’s self-interest’ is ‘what is better for X’, the question is how to understand the ‘for’ in ‘better for X’. The relativization ‘for’ is ‘relative to the interests of ’. But ‘better relative to the interests of ’ does not correspond to the notion of ‘self-interested’ as usually conceived. The pains and pleasures of an individual are not usually all that an individual seeks. The interests a person may have can include things like world peace or the success of the Red Sox. Without a reductive tie to given sensations, self-interest is not separate from interests in very many things intuitively separate from the self. There is more to people’s preferences than just partiality to oneself, narrowly conceived. Not only do people often take their interests in their own welfare more thoroughly to heart than they take the interests of others, but their interests include interests in the welfare of a select group of others. Parents have special bonds to children, friends to friends, and many Americans to other Americans.19 Such special bonds that A has to B include the interests of B, (where B can be other people or groups), in the interests of A. There is no sharp line between the prudential interests of a father and the interests of his child. The father’s happiness and well-being is as closely tied to the daughter’s sensations and well-being as it is to any of the sensations that originate in his own internal reports of pleasure and pain. The line between self-interest and the interests of others is not sharp in any individual other than a sociopath. The identification with other peoples’ interests that flows from human relationships is part of a normal life and part of a good life. If every relationship were contingent on the continuing merit of the participants, life would be lonely indeed. More or less unconditional love is important. An interpretation of a person’s behavior that attributed no such attachments would require substantial background. It would be interpreting a pathological case.

19 Among the features that are maximized in interpretation are emotional responses and emotional attachments. For a Davidsonian, emotions are subject to evaluation by reason, and therefore part of the content of rationality, since we maximize appropriateness of emotional response in interpretation, just as we maximize appropriateness of distaste or positive preference in the case of sensations. A person who enters his burning house to rescue his Tupperware may be evaluated as having behaved irrationally. So Kant’s move of leaving emotion out of morality is not available.

davidsonian rationality and ethical disagreement

183

Every culture’s ‘ethical theory’ makes some relationships reasons for special concern. Interests in others’ interests, then, are partial identifications of the interests that motivate an individual with the interests of others. They are also a kind of deconstruction of the primacy and clarity of the ‘purely personal’ preferences that have bedeviled ethical theories since the beginning. People routinely care about a lot more than their own pleasures and pains, crudely speaking. Partialities can be evaluated as rational or irrational, and we do so routinely in interpretation. A partiality can be excessive, as some allegiances to states and sports teams often are. Degrees of partiality are also subject to reasoned evaluation. A man who is just as attached to the Red Sox as to his children is being unreasonably loyal to the Red Sox. The basis for such judgments is that there is no reason for the partiality. Partiality can be inadequate. A lack of favoritism towards one’s children, for example, calls for explanation by pointing out that the mother is den mother of the group of Cub Scouts of whom her child is a member. So interests in the welfare of others can be rational or irrational. The maximal scope of concern is concern for the interests of every entity that has interests. Is this maximal partiality rational? If Nagel and Kant are right, it is. If we are motivated by partialities that it is reasonable to have, then this one, ‘the moral point of view,’ is rational, and reasonably ought to motivate any agent with interests.20 The question is why it motivates so rarely, and so much in disproportion to judgments of ‘what is better for me and mine.’ b) Why We Make This Mis-Judgment Given that there is no well-defined subject matter for self-interested desires, and given that people in fact take into account the desires of at least a subset of others, the disposition to focus on limited perceptions of the good must be akin to limited perceptions of the true. That is, just as many people over-rate their own experience in judging

20 This is a very limited result, since there is no argument that concern with others should be equal. It is implausible that a mother rationally should care about all children to the degree that she cares about her own, for instance. And this implausibility is built into interpretation. So, large questions about genuine conflicts of interests, when you want something for your child that I want for my child, are being ignored.

184

chapter six

what is true, so many people over-rate their own perceptions of the Good in judging what is better than what. How is this compatible with their valuations being mostly correct? Here is an analogy: Most perceptual judgments are correct; error tends to be in the inferences drawn from those judgments. In the same way, most judgments of what is better than what are correct, but the inferences are mistaken. The fundamental ‘selfish’ error is to judge that A and B is better than not-A and not-B because A is good, where A is an involuntary judgment that something is good. That is, while it is correct that A is good, and better than not A, it may not be that A and B is worse than not-A. In effect, people over-rate the goods that are immediate to them or that are experientially connected to goods that are immediate. In the case of immediate judgments of value, involuntary utterances ‘This is good’ ‘That is bad’ carry undue weight in calculating the value of compounds of which this and that are components. ‘A is better than B for C’21 can be treated as a perspectival notion, that is, relative to C’s local valuational beliefs. The analogy with points of view in perception should be apparent. Consider a pyramid with a square base. The pyramid from my angle is square; from another angle it is triangular.22 So we can coherently take preferences to be beliefs about what is objectively better, and reasonably take ‘narrow’ views on what is better to be due to failures of reason.

21 ‘Good’, ‘better than’, and similar words have many complexities. ‘Good polevaulter’ is not a conjunction, since a good pole-vaulter who is also a shot-putter need not be a good shot-putter. A theory of ‘good’ and ‘better than’ also needs to take into account ‘good at’ and ‘good as’ and well as two ways of using ‘good for’. The theory that most attracts me is that of Richard Larsen, begun in his essay, “Olga is a beautiful dancer”. For purposes of the present essay, I will suppose that some theory of ‘good’ and “better than” is in place, since little hangs on what theory is applied. 22 I may make a mistake about what aspect the pyramid presents from my angle, by careless observation. In the same way, C’s interests are not just C’s desires. C’s interests are his objective goods, about which he can be mistaken, if he is misinformed about what the objective situation is. There is a difference between what C thinks is better for him and what really is. A correct perspective on the good requires correct information about the true.

davidsonian rationality and ethical disagreement

185

c) Davidsonian Ethics c1) Davidsonian ethics need not be utilitarian What is better, all things considered, might or might not be what is better for everybody, and might or might not be a summation of everyone’s preferences. While an identification of what is better, all things considered, with what is better for everyone is a candidate Davidsonian ethical theory, it is not the only theory that could arise from a Davidsonian approach. It could be that what is really better requires something more. Roughly, the relation between what is objectively better and the preferences of agents is analogous to the relation between the truths and the summation of beliefs of agents. While the beliefs have to be largely true, truth is not reducible to consensus. In the same way, while most preferences have to correspond largely to what is really better, the Good is not necessarily the maximal satisfaction of preferences. c2) Partiality and relativism Partiality is perhaps the deepest puzzle about morality and its rationality and objectivity. Call a system of partialities a list of partialities that a person ought to have and can have, with a ranking of which partiality should trump which. One troubling fact about partialities is that different cultures have different systems. While (almost) everyone takes certain partialities to be normal and reasonable, there is no way to argue for one system of partialities over another. Partialities, by their very nature, are relative in the sense that they provide reasons for preferential treatment of your son for you but not for me, but not based on any intrinsic properties of your son. One source of irresolvable differences among cultures is the differences in the partialities that are normal in the culture. What, for instance, should be the relation between one’s partiality towards one’s parents and one’s partiality towards one’s children? Which should be greater? A Davidsonian can take one of four positions about partiality: (1) Kant is right, and all partialities are irrational, and therefore morally mistaken. In effect this says that the moral point of view requires not only taking everyone’s interests into account, but taking everyone’s interests into account equally. Perhaps something of

186

chapter six

parent-child relationships could be salvaged on utilitarian grounds, perhaps not. (2) One system of partialities, with their rankings of relative importance, is objectively correct. So, given that an American, Fred has promised both his daughter and his father to spend the afternoon with them, and he cannot spend the afternoon with both, there will be an objective answer as to which partiality takes precedence. As observed above, cultures differ in the partialities they regard as normal or obligatory. This position would be that, wherever that situation occurs, like-placed individuals should do the same thing. (3) A system of partialities is correct only relative to a culture. This option would be a form of relativism that allowed that there were some objective truths about what was better than what, but a whole range of states of affairs that were only good or bad relative to the cultures in which individuals lived. Partialities would be valuations that were essentially subjective, about which there was no right answer as to whether the given state of affairs, say that Fred chooses to spend the afternoon with his father, was good or not. It is hard to see how this could be compatible with preferences being judgments that a state of affairs is good. (4) A system of partialities is objectively right for a culture. Given that Fred is an American, an application of partiality may be objectively right even though for a Chinese in precisely the same situation, doing what Fred did would be objectively wrong. This option takes into account that actors are situated in cultures. This fourth position, to which I am most drawn, is not a relativism. Except in rare cases in which it is indeterminate which culture an individual inhabits, there is an objective answer to what action is good. All four positions allow that there is a moral point of view in which every being that has interests is taken into account, and that one should take that view. The options that allow partiality to be rational, however, allow that people ought not to, or that it is morally acceptable not to, treat everyone’s interests as of equal weight.

9. The Ethics of Cultures There are several grounds that should lead us to suspect that there will be irresolvable differences between distinct cultures. Such irre-

davidsonian rationality and ethical disagreement

187

solvable differences are compatible with a great deal of agreement on ethical matters. Some such differences may be cases of one culture getting something wrong, others may be indeterminacies, and others may be case of option (4) above. Some of the sources of irresolvable disagreement are the following: (1) Difference in systems of partiality is one reason that there can be profound irresolvable differences between cultures, as discussed above. (2) Even though two cultures agree that, for instance piety and courage are virtues, they may disagree on their relative ranking. In such disagreement, one may be right, or the question may be indeterminate. (3) Ethical mistakes may become imbedded in cultural practices. If ethics is the science of what is really good, and of what is really better, then ethics is very difficult in application to many concrete cases in which many factors are relevant. The familiar analogy might be with weather-prediction. Some weather predications are correct and easy. If dark clouds are to the West, it is likely to rain. If it’s January in Connecticut, it will be less than 90 degrees Fahrenheit. But determining in detail whether and how much it will snow in Storrs is beyond even dedicated and sophisticated specialists. However, what the weather in detail will actually be in Storrs on a given day is almost certainly in principle predictable. In the same way, if something like the above account is right, very many choices about what ought to be done are likewise decidable in principle but practically very difficult to determine. The difference in the case of ethical questions is that questions in ethics demand answers in a way that questions about the weather do not. Since we have to do something, we decide.23 Since we have some rules of thumb, and some intuitions that must be right, we use those rules and intuitions. Much of the time, just as a farmer’s weather predictions are, we are right. Another difference is that, whereas everyone can determine when the weather-prediction is wrong, no

23 A moral skeptic may claim that the better analogy is with astrology, where the King demands an astrological forecast, since the stars must be presaging something. The skeptic will say that just as the right response there is to deny that the stars control destiny, so here the right response is that there is no objectively better in general.

188

chapter six

such clear-cut evidence is forthcoming when the ethical theory embodied in a culture’s practices yields wrong results. The necessity to have answers to ethical questions such that many people can agree on the answers generates the ‘ethical theories’ that are embodied in cultures and which are part of the socialization of members of a culture. When the answers generated by the necessity to have an answer are mistaken, entire cultures can be wrong. What should we expect when two different cultures have an ethical disagreement on a kind of hard case? No matter how sincere and rational the negotiators are, there is no general algorithm for determining what is the better course of action. Each culture will perforce be using its own conception of the rational. That does not mean that there is no right answer. Sometimes both are wrong, sometimes one is right, but some cases may just be indeterminate.

CHAPTER SEVEN

A DAVIDSONIAN APPROACH TO NORMATIVITY AND THE LIMITS OF CROSS-CULTURAL INTERPRETATION Yujian Zheng

1. Varieties of Normativity People are familiar with notions of rules and norms at the group or societal level, such as laws, moralities, religious imperatives, cultural conventions, and more local regulations such as the dress code in a school or corporation. Many of us are also used to certain personal rules or policies, some of which are self-imposed, or voluntarily adopted (and adjusted) as a result of one’s own deliberation at a particular time over questions such as how to solve the dynamic conflicts between one’s short and long-term interests. Most people, however, are perhaps unaware of other important and fundamental types of norms, such as logical or rational principles involved or deeply embedded in our ordinary practice of reasoning, of understanding or interpreting the speech or behavior of others, of forming beliefs and judgments of one’s own, etc. There seem to be natural enough reasons for such unawareness of the normative nature of these principles: on the one hand, there is no clear consensus as to what kind of thing is possibly involved in those forms of practice. In other words, whatever underlies them is largely implicit, and so is not easily recognizable as a matter of explicit principles. On the other hand, even those explicit logical rules we sometimes apply consciously tend to be associated with empirical laws of nature, which seem more readily recognizable as descriptive rather than normative. This consideration leads to our last type of normativity, i.e., that necessarily involved in our understanding or explaining natural events or phenomena. In an obvious sense, the hypotheses in science, no matter how well established they are, are products of, or results filtered by, rational procedure, which ultimately is nothing but human decisions in a collective and evolutionary form. This procedure is explicable in terms of a certain set of rules and norms usually regarded

190

chapter seven

as epistemic or methodological in nature (e.g. Ockham’s razor). Moreover, the function of a scientific hypothesis is to explain (and predict) certain natural phenomena by subsuming them under its nomological structure and, to a no less degree, to serve as a (though provisional or tentative, perhaps) normative constraint for other related hypotheses, as well as for our ordinary beliefs or inferences concerning those phenomena. In this respect, humanly discoverable and presentable (thus understandable) laws of nature share the normative status of methodological rules and principles including the most basic ones, such as those found in formal logic. The distinction between the descriptive and the normative appears to be so fundamental to our conceptualizing the world that the relative scarcity of direct philosophical discourse about its use as well as its foundation is striking. Corresponding to each level or type of normativity mentioned above, for instance, there is an issue concerning its source, which issue is distinct from the question of what possible roles or manifestations this type of normativity may have. A rough, yet very helpful expression of the distinction might be formed in terms of ‘direction of fit’: the direction of fit for the descriptive is from the mind to the world, whereas for the normative it is from the world to the mind. In other words, the criterion for descriptivity lies in external reality, while that for normativity is internal. Anything that satisfies such minimalist expressions could count as descriptive or normative (at least in the context of this paper), no matter how remote it may appear from ordinary perspectives.

2. An Integrated Approach to Normativity Underlying and Unifying Davidson’s Philosophical Contributions It is well recognized that Davidson’s thought has a remarkably unitary character, even though it ranges over a wide array of problems concerning knowledge, action, language, and mind. Although what best characterizes this unitary character is perhaps something short of agreement, I’d like to demonstrate here that it is a distinctive sense of normativity that can serve as a unifying thread to Davidson’s whole set of apparently interconnected and mutually reinforcing ideas which have been developed separately in disparate contexts traditionally belonging to distinct domains.

a davidsonian approach to normativity

191

Let me start with Davidson’s well-known principle (adopted from Quine) in his theory of interpretation, i.e., the principle of charity. What the principle says is that you should optimize the overall truth and consistency of beliefs of the person being interpreted when you try to understand his or her speech or behavior. The normativity of this principle is obvious. But the question is how to justify it. On what basis can you be so sure that an interpretee’s beliefs are not false or inconsistent, as most of us could surely find ourselves in possession of such unfortunate beliefs from time to time? Davidson’s answer is this: although you can be never sure that any particular belief(s) of the interpretee must be true or coherent with his other beliefs, his belief system as a whole cannot deviate largely from rationality and truth. For otherwise you as an interpreter would not even be able to recognize or identify him as a being, like yourself, capable of believing, speaking or thinking. Any diagnosis of false or irrational beliefs of his, therefore, could only take place against the background of granting him many more true and rational beliefs. The very possibility of interpretation, as one crucial form of human interaction, and just like any other form, depends upon agents’ mutual recognition of intentions or beliefs. The distinctive social existence of human beings would be impossible without such recognition.1 Moreover, interpretability governed by charity is not merely a methodological issue of understanding others, but also an ontological or genetic issue concerning how our own intentionality (either as the first person interpreter or as a self-interpreted agent) is possibly formed or constituted. Roughly, human intentionality, typically in the form of reasons consisting of beliefs and desires, emerges from the same processes of interpersonal interaction centrally involving interpretations. Such interactions, be they a matter of individual human infants learning their mother tongues or of the social evolution of our ancestors, must, at least in their initial stages, involve some kind of triangulation of causal relationship among two human subjects and one shared object. The causal effects from the same object (whose location is identified by the intersection of ‘visual lines’ of the two subjects) on each subject are compared or coordinated through mutual 1 Davidson’s ideas on charity-governed interpretation can be found mainly in the section of “Radical Interpretation” in Davidson, Donald (2001), Inquiries into Truth and Interpretation, 2nd ed., Oxford: Clarendon Press.

192

chapter seven

responding both to the object and to each other’s responses. The basic causal connection between the object and whatever simple or primary linguistic response of a subject somehow guarantees the optimal condition of truth for the latter while the interpersonal (and instantaneous) comparing and checking provide for rudimentary normative force of either a (well-established) rule-following (in the case of an adult-baby or teacher-student interaction) or some norm-formation or meaning-determination (in the case of two equals facing a new or unprecedented situation). Davidson seems to rely increasingly on such a triangular nexus of causal-cum-normative relationship (as opposed to, say, his earlier notion of the ‘omniscient interpreter’) in explicating the application of the concept of truth in his later works.2 According to Davidson, were there no such a triangular relationship, not only would other people not be able to know my beliefs, but even I myself would not be able to know what beliefs I have— because the very basis on which any and all beliefs depend and become individuated and thus identifiable would not exist. Without such a normative basis, neither belief (and other propositional attitudes) nor meaning, truth, and rationality could possibly take place. Given these considerations it is not hard to see the holistic feature of Davidsonian approach.3 It does not stop at describing a static picture of interdependence among various beliefs, but further extends to explicating the dynamic structure in which the formation of beliefs is necessarily bound up with the emergence of truth, meaning (interpretability), and justification (rationality). More concretely, Davidson borrows Tarski’s Convention T, a certain recursive linguistic device in his semantic theory of the predicate ‘true’ for formalized languages, and converts its function into something like a foundation for a truth theory of meanings (of every sentence in an object language), i.e., taking truth as the most primary and self-evident concept for constructing a theory of meaning. On the other hand, Davidson relates

2 See Davidson, Donald (2001), Subjective, Intersubjective, Objective, Oxford: Clarendon Press. 3 Davidson explicitly claims that “[w]hat I think is certain is that holism, externalism, and the normative feature of the mental stand or fall together”. See Davidson, Donald (2004), Problems of Rationality, Oxford: Clarendon Press, p. 122. What he fails to address, however, is the further question about the logical relationship between the three, e.g., whether normativity has any priority status in some orderly account of the mental over the others.

a davidsonian approach to normativity

193

truth to the basic propositional attitude ‘hold true’ (or ‘believe’) in a broader theoretical framework, i.e., his doctrine of radical interpretation with the principle of charity at its core. The normativity of this doctrine, apart from its methodological and ontological aspects mentioned above, also lies in its obvious epistemological implications: Davidson’s criticism of so-called “the third dogma of empiricism” (i.e., the distinction between empirical content and conceptual schemes),4 his rejection of skepticism and relativism,5 and his emphasis on the interdependence of three kinds of knowledge (i.e., about my mind, other minds, and the external world)6 all manifest the theoretical power of a holistic approach in its efforts to dig up some deep and integrated sense of normativity. Aside from the integration of theories of knowledge, truth, and meaning, another very influential and integrated bunch of theories for which Davidson is famous is in the area of studies on action, events, psychological explanation, irrational behavior, and the metaphysical status of the mental.7 Due to space limitations, I shall only concentrate on two central Davidsonian themes in this area in order to exhibit their underlying commitment to normativity. The first theme concerns the relationship between events and actions. The distinction between actions and events does not lie in any ontological difference between the two, but in what kinds of description each is covered by. Purely physical events can only have physical descriptions, whereas actions (or mental events) can have both intentional descriptions and physical ones. Any event describable as intentional must be in principle also subject to physical (neural or physiological) description, but not vice versa. In brief, actions are events, but events are not necessarily actions. A necessary condition for an event to become action is for it to be under some particular kind of description.8 E.g., “The room was dark and I flipped the light switch”

4 See “On the Very Idea of a Conceptual Scheme”, Inquiries into Truth and Interpretation. 5 See ibid.; “A Coherence Theory of Truth and Knowledge” and “Afterthoughts”, Subjective, Intersubjective, Objective. 6 See “Three Varieties of Knowledge”, Subjective, Intersubjective, Objective. 7 This area is mainly covered in his first anthology, see Davidson, Donald (2001), Essays on Action and Events, Oxford: Clarendon Press. And more discussions on irrationality appear in the last section of his Problems of Rationality. 8 The phrase ‘under a description’, though frequently used by Davidson, originally appears in Anscombe, G.E.M. (1957), Intention, Oxford: Basil Blackwell.

194

chapter seven

describes an action of mine by suggesting an obvious reason why a physical event, i.e., my finger movement or flipping, occurred. By contrast, the description “I flipped the light switch and the prowler lurking in the bushes outside the window was alerted” does not correspond to any intentional act of mine if I did not know there was a prowler out there, even though the flipping seemed to be causally responsible for the alerting. Yet there was only one event (or one causally continuous process) here: the event of unintentionally alerting the prowler is the same event as my intentionally flipping the light switch which, again, is the same event as the physical movement of my finger (and any other relevant parts of my body) in a certain way. Alternatively, it might be argued, despite the fact that ontologically speaking there is only one unified causal chain or nexus within a certain spatiotemporal domain (or one ‘zone’ as Quine would say), its possible manifestation as different events or types of event, which correspond to different cognitive meanings, depends upon different descriptions from selectively adopted perspectives on the same nexus of physical elements. Whether a perspective is meaningful or significant depends in turn on whether the pattern identifiable from the perspective has some explanatory or justificatory role to play in the larger context, or fits with some larger patterns which nonetheless are only recognizable by ascending to higher-order perspectives. So, for instance, the life-world meaning of ‘my turning on the light’ will be lost if we replace it by shifting entirely to the lowerlevel descriptions of the minute neurological facts underlying my above action, even though the latter descriptions may still have certain cognitive meanings for all physiologists. I shall come back to the implication of this theme for normativity presently. The second Davidsonian theme concerns the relationship between reasons and causes. To qualify as an action, some reason rather than a mere cause of it must be shown. Generally speaking, causes and reasons belong to different logical spaces: i.e., the space of the descriptive vs. that of the normative. What explains an action is the socalled ‘primary reason’, which consists of a pair of propositional attitudes, i.e., a belief and a desire. As indicated above, the constitution and recognition of a belief must rely on the holistic application of normative principles; similar arguments apply to the motivational part of a primary reason (and, moreover, many desires in real contexts are mediated or activated by beliefs of various kinds). A unique contribution of Davidson’s here lies in his both acknowledging the

a davidsonian approach to normativity

195

above distinction of logical spaces and emphasizing the fact that reasons and causes refer to the same ontological entities or events. In short, the (primary) reason for an action must be its cause (i.e., what actually mobilizes it). Therefore, the explanation of action by appeal to its reason must also be a causal explanation in the sense in physical science. If this Davidsonian thesis is tenable, the relationship between normative and descriptive elements will become much more interesting as well as more complex than may otherwise be supposed (as in some Wittgensteinian thesis). One may get a sense of the complexities such an approach may lead to by looking at Davidson’s discussions of how irrational behavior (such as weakness of will) is possible, especially with regard to his treatment of unavoidable paradoxes involved in the attempts to explain irrationalities. On the other hand, nonetheless, the first theme above, i.e., “description determines the meaning of an event (or action)” also reveals, from another aspect, the coupling between the natural event interpreted as a meaningful object on the one hand, and the normative constitution of its meaning by the subject who is doing the describing on the other. What is description? This is anything but a question merely for linguistics or rhetoric. I take it as a fundamental and far-reaching question for Davidson and his integrated project. Although he offers no direct and systematic account, one may still perceive the central status of this (potential) question in his various discussions of topics related to actions and events. What interests me most, however, is how description in the Davidsonian sense relates to normativity at this or that level. Here are just a few tentative observations of mine. First of all, description, as a kind of (linguistic) action issued from the subject, must have some kind of intentionality, or must be companied by a certain intensional context the subject is aware of. Thus it must already involve the aforementioned normativity connected with the individuation of beliefs. Secondly, the means of description can only be language, while the birth (and fixation) of linguistic meanings can only be effected via logical relations among a massive amount of sentences as well as the Davidsonian triangular relations against which certain basic sentences are situated. Both of these two sorts of relations have already embedded or embodied various normative principles. Thirdly, description qua propositional content that is supposed to fit the (external) object in the world, must obey the factual links or natural constraints in the space of causes. That is exactly the objectivated sense the term ‘the descriptive’ possesses.

196

chapter seven

Such a sense of objectivity makes a perfect dichotomy with the normative force or effort issued from the subjectivity of free will. Thus it can be seen that the notion of ‘description’, just like the notion of ‘truth’, is subject to pulls or constraints from two opposite directions (for Davidson, truth has two contrary features at the same time, i.e., external correspondence and internal coherence); so it can play some indispensable, crucial role between the space of causes and that of reasons, or between nature and human rationality. Following this line, one may easily understand why Davidson makes another seemingly strange combination in philosophy of mind: he subsumes mental items under the ontological category of physical items and yet insists that the mental descriptions can never follow the nomological model of the physical ones—hence the label ‘anomalous monism’.9 In summary, since mental events are endowed with normativity in their very genesis or constitution, mental description or interpretation can only be governed by rational norms such as the principle of charity, which find no exact counterparts in our descriptions of pure physical events. But, on the other hand, descriptions/interpretations of mental events also have truth values, and their truth conditions are rooted in the causal interactions between mental and physical event types. As all causal relations can be subsumed under strict forms of natural laws, each mental event must therefore be token-identical to a physical event. Otherwise, some non-physically-describable events would enter into strict forms of physical laws—which seems to be a contradiction. In other words, there cannot be strict laws connecting the mental domain to the physical one, or the space of reasons to that of causes; neither can there be strict causal laws in the space of reasons, i.e., among mental items themselves. Thus it seems clear that the idea about the normative space of reasons as well as the normative distinction between different spaces play some key role here.

9 See “Mental Events” and “Psychology as Philosophy”, Essays on Actions and Events.

a davidsonian approach to normativity

197

3. Can There be Interpretative Ruptures in Comparative Philosophy or Cross-Cultural Understanding from the Davidsonian Perspective? What is the possible implication of the Davidsonian approach to normativity for our understanding of the status of Chinese philosophy or culture at large in relation to world philosophy or (Western-dominated) global culture? Many people’s intuition seems to suggest that differences between Chinese and Western mind-sets cut much deeper than superficial comparisons or translation practices might reveal. But could we really pin down where these presumably deep differences lie? Could they, once identified, count as differences between two incommensurable paradigms in the Kuhnian sense?10 Obviously, Davidson rejects the relativist ‘incommensurability’ thesis.11 His reasons might be summarized as the following: Whichever culture or language you may be born into, the physical world (in its main, structural features as well as basic elements) is the same for you as is for anyone in any other culture; the type of triangular causal nexuses you must face in order to form beliefs, understand others and the world under various contexts is also the same for you or anyone who can manage to survive; and thus the overall optimization of truth and rationality, warranted by the triangle and expressed by the charity principle, is necessarily applicable to you. Therefore, there should be no room left for ‘incommensurable’ or untranslatable conceptual schemes or cultural models. If we accept such an argument, will that mean that cross-cultural comparative philosophy cannot have independent significance, as opposed to comparisons between personal opinions or theories of two individual philosophers in one culture? In other words, will it imply that Chinese philosophy, just like any other non-Western philosophy, cannot distinguish itself from the Western or world philosophy in any non-trivial sense, ‘trivial’ here meaning that any theoretical position can be regarded as distinctive from others? I guess many would tend to find the Davidsonian answer to be affirmative. In

10 For the classic explication of ‘paradigms’ in science and their ‘incommensurability’, see Kuhn, Thomas (1970), The Structure of Scientific Revolutions, Chicago: Chicago University Press. 11 Davidson’s main argument can be found in “On the Very Idea of a Conceptual Scheme”, Inquiries into Truth and Interpretation.

198

chapter seven

what follows, I attempt to sow a few seeds of doubt, without challenging the fundamental framework Davidson has laid out. The powerful Davidsonian weapon in attacking any relativist thesis about objective knowledge is the principle of charity. The applicability of such a principle is grounded in some fundamentally normative situation, to the effect that we have no choice but to follow certain rules intersubjectively generated and maintained in accommodating or coordinating each other’s causal reactions to shared stimuli so as to facilitate our social symbiotic existence. But a crucial point then is that such normative forces (enforcement and reinforcement through continuous interactions) occur on different levels and different scales without fixed limits, and manifest themselves to individual consciousness in either implicit or explicit forms (e.g., ranging from instinctive skills of knowing-how to sophisticated symbolic representations). Every utterance, whether it be a judgment or belief, a command or a hope, comes gradually to carry more and more corresponding relationship of commitment and entitlement to other utterances.12 The extent and strength of such a relationship depends not only on particular contexts in which this sentence may appear, but also on norms or rules of the language games in which the speaker may find himself or herself involved. Although the Davidsonian charity logically guarantees the optimality of truth and coherence for a majority of beliefs, it does not automatically resolve the operational issue, i.e., how to tell a particular true belief from nearby false (or uncertain) ones that may cohere with the rest of beliefs to different degrees and/or in various ways. In other words, charity is indispensable as a general starting point and perhaps an ultimate boundary condition, but it cannot replace the role of many concrete rules of the game which actually constitute and regulate the intentional relations and propositional contents of those beliefs involved. The diversity of such rules and the multiplicity of levels at which they may exist are simply obvious facts, which help account for the variety and proliferation of games in real life. Mother tongues or native cultures are normally taken as the grand language games that shape types of individual consciousness and forms of life. These games are isolated or independent from each 12 Such an inferentialist thesis is a natural extention of Davidsonian normative holism and has been systematically argued in Brandom, Robert (2000), Articulating Reasons, Cambridge, MA: Harvard University Press.

a davidsonian approach to normativity

199

other as far as their causal genesis (say, in terms of geographic environments) is concerned. Therefore we can’t exclude the possibility of “cultural gaps” between them. The problem seems to lie in the fact that there is no simple and universal method to prove or measure the depth or width of such “gaps.” Perhaps one may suggest that we could use certain statistical success rates of individual intercultural communications as a rough indicator. But what kinds of dialogues (e.g., how many rounds of Q&A) could count as successful communication? One recommendable way of understanding a strange language/ culture without relying on translation is anthropological participatory observations. If an anthropologist could merge himself deep enough into the various life games of a primitive tribe which so far had no exposure to the outside world, he would presumably emerge with understanding of its oral language as well as its belief systems associated with various customs within a limited time span. Contrastively, however, we may imagine that a member of this primitive tribe suddenly came to visit our civilized society. Could he make sense of the disparate natures or functions of all those essential games of our society, such as politics, finance, and academic tenure system, within a similar time span? Some asymmetry here seems to be obvious. The oral language of the tribe may not lack some potential or even explicit structure susceptible to first-order quantificational logic. Terms expressing negation, conjunction, and alternation are most likely available so that the basic inferential operator “if . . . then . . .” or its equivalent is likely to exist too. There are good reasons to believe that the linguistic difference between a primitive and an advanced culture is not categorical (i.e., between human language and a nonhuman or quasi-human ‘language’), but rather of differences between higher-order semantic relations or rational norms of the respective language games. For instance, ‘Ockham’s razor’ is such a high-order cognitive norm in the games involved in scientific explanation. If the interpretative failure caused by the above asymmetry between anthropological and ‘reverse-anthropological’ understanding is not accidental, we could easily conceive that a similar failure will occur between two primitive tribes of a similar advance level if their native geographic environments are sufficiently disparate (e.g., one located in tropical jungles near the equator while the other is situated in the tundra of Northern Canada). The rules of the games that have been developed through their unique responses to the challenges of

200

chapter seven

survival in such disparate environments, including those regulative norms accidentally created and yet socially ‘selected’ because of their fitness to the game’s appreciation or proliferation, may have shaped certain parts or structural aspects of their respective belief systems to such an extent, under the condition of mutual isolation, that when the two systems meet, the failure of interpretation is doomed to happen—they cannot make sense of everything (significant) in each other’s behavior, i.e., one cannot find any set of beliefs in one’s own belief system that suffice to accommodate or explain a targeted behavior of the other; or put alternatively, one cannot communicate everything about one’s intention to the other in the terms the latter fully understands. The interpretative asymmetry between a civilized society and a primitive one can be illuminated by a disparity of another kind. A civilized society normally grew out of the process of contact, clash/war, amalgamation among different primitive tribes, and outward expansion then into a similar process on a larger scale. So its language games must also have been under a dynamic process of interaction and adjustment among its originally heterogeneous components, often yielding ever higher-order integrations. When such integrations reach a certain level, the members whose mentality has been shaped by the norms of the new games will have interpretative resources rich enough to simulate, approximate, or grasp those highly unique or specialized parts of the language game of a total stranger from a primitive tribe. On the contrary, a primitive tribe member can hardly develop by himself sufficient doxastic or conceptual resources to understand a total stranger, no matter how smart or attentive he or she may be. By the same token, two initially isolated civilizations of a similar sophistication can in principle accommodate each other’s language games, at least covering enough aspects or levels of them, once they encounter one another. This is exactly the point Davidson would like, and I think correctly, to stress. But the interesting and more subtle question is whether or not such a cross-cultural interpretation is necessarily extendable to all aspects and levels of the games on the other side. That is clearly a question which can hardly be decided a priori by the principle of charity. Here are two reasons that make me skeptical about a positive answer to the above question. First, if there could be an interpretative rupture between two relatively simple or primitive language

a davidsonian approach to normativity

201

games, it would be equally hard to exclude the possibility of such a rupture between two greater language games each of which involves one “rupture element,” i.e., one primitive game, as its component, and never had contact with the other “rupture element” belonging to the other greater game. (The sense of “rupture” here is relative to the host greater game.) The relationship between components and the whole game could be so varying as to include the possibility that the role of one component can neither be replaced nor implied by any (combination of ) others, despite of the presumable large degree of integration among most components of the whole. A possible analogy in a different domain is this: an animal species whose genetic structure lacks certain DNA strings may not be able to digest some kind of food which is perfectly edible to other species having such DNAs. Here I certainly can’t get into a detailed, and probably fruitful, discussion about certain implicit questions such as how large a component has to be in order to have such an indispensable and irreplaceable role, especially in the domain of intentional items. One thing seems to be clear: correct interpretation requires something more than merely apparent coherence (as ‘coherence’ seems to allow degrees, e.g., one might “apparently swallow up the inedible food”), something in the direction of real causal history of engaging and integrating the object (or its components) in various ways. As I briefly suggested above, any successful integration of new parts into an old game would have changed or shaped the latter in some necessary, though often imperceptible, way. In this light, therefore, whenever we claim that, say, Chinese culture had assimilated some particular heterogeneous cultures (or cultural products, such as Buddhism from India), we can no longer claim, or even imagine, that the real, original Chinese culture is something like the present Chinese culture minus those particular exogenous ones. In other words, the interpretative power or resource of today’s Chinese culture is no longer the same as (or perhaps even hard to compare to) that of Chinese culture two thousand years or just two hundred years ago, confronted with a similar alien culture. What is, then, the implication of this point for the possibility of interpretative rupture between two grand cultures indicated above? The implication is the following. When one grand culture faces a ‘rupture element’ in the other, it does not literally face the pure element itself, as (‘the past self ’ of ) the other grand culture probably would when a primitive tribe was first encountered. For, on the one

202

chapter seven

hand, the original primitive game to which the rupture element corresponds had been merged into the larger game long before, and on the other hand, that element, either as an organic component of the grand culture or as some hidden deposit, is no longer independently recognizable without appropriate background interpretation of a much larger picture—this is exactly what is required by the Davidsonian holism. Now the complexity lies in the likely scenario that the interpreter from the external grand culture may not be able to find sufficient reasons to realize or demonstrate that a rupture is there even if in fact it is. For, as mentioned above, a grand culture usually has enough interpretive resources to warrant elbow room for tenable rationalization of almost any seemingly inconsistency in the object of interpretation.13 This complexity naturally leads to my second skeptical reason, which happens to involve a similar complexity. The construction process of the higher-order theories or superstructure of a civilization contains, generally speaking, two sorts of ingredients: the first sort includes those things that are necessary for explaining or accommodating people’s rudimentary or observational experiences, such as perceptions of natural regularities, while the second sort those contingent or accidental fluctuations or deviations which seem hardly avoidable, say, in personal creation or interpersonal transmission of ideas, visions, styles, etc. Now a complexity emerges from this second sort of ingredients for ultimately a grand language game or paradigm. Higher-order norms or principles of a language game may have easily absorbed or incorporated such accidental ingredients, some of which might be very subtle or mannerist, in unnoticeable ways and over a long period of time of cultural accumulation. But once such norms and principles exist or are explicated (in some distinct forms), they must take effect as respective wholes without possibly distinguishing those necessary ingredients from the contingent ones at their origin. Moreover, norms are gradually and constantly reinforced or reified over time and with heavy usage, and some coupling or mutually supportive effects among related norms may also increase or amplify. All these tendencies may help

13 Such a complex and even paradoxical situation is probably best reflected in Davidson’s own discussion of his so-called paradox of irrationality. See Davidson, “Paradoxes of Irrationality”, Problems of Rationality; also see Zheng, Yujian (2001), “A non-substantial approach to practical reason”, International Journal for Field Being (special edition on Whitehead).

a davidsonian approach to normativity

203

account for the fact that some originally accidental bifurcations or forks in personal ideas, thoughts, or imaginations ultimately turned out to be parts of the hard cores of certain ‘unshakable’ orthodox or paradigms in different cultures. And from our earlier discussion, we should add that these more or less integrated normative forces are causally responsible for constituting as well as regulating the community members’ webs of beliefs, or at least many structural aspects of such webs. No matter whether these initial thought bifurcations contain any errors or seeds for errors, the holistic paradigms partially brought about by them could become the obstacles, or worse still, the unconscious pitfalls for cross-cultural interpretations. I am quite convinced by Davidson of the impossibility of wholesale rupture (or incommensurability) between conceptual schemes or paradigms. However, if my skeptical reasons above hold any water, i.e., if some temporary, local ruptures of interpretation are inevitable between language games even of great scale and sophistication, then we should conclude that the Davidsonian argument against relativism cannot be applied to the local levels of interpretation. Or, following the Davidsonian style, one might say that the local failure or incompetence of Davidson’s theory can make real sense only against the background or overall validity of his approach to normativity.

PART FOUR

MEANING AND INTERPRETATION

CHAPTER EIGHT

ON TWO KINDS OF MEANING AND INTERPRETATION A.P. Martinich

I want to discuss two pairs of terms that are verbally identical, that is, are homonyms. But the two pairs seem to be semantically unconnected. I will suggest that they are in fact semantically connected, and in ways that I think have not been noticed before. The first pair of terms are the meaning (hereafter: c-meaning) of a text and the corresponding kind of interpretation appropriate to such meaning. Roughly, c-meaning is one half or aspect of a successful communication; and hearer interpretation is the other half. The second pair of terms are the meaning (hereafter: s-meaning) of some thing, state or event and the corresponding kind of interpretation appropriate to that kind of meaning.1 Roughly, the correct interpretation of some thing is the knowledge of that thing’s importance. The reason that these two pairs of meaning and interpretation do not seem logically connected with each other, notwithstanding their verbal similarity, is roughly the following: The c-meaning of a text expresses a peculiar kind of abstract, hypothetical entity,2 namely, the communicative meaning of the text; and it depends upon a distinctively invented entity, a language, which consists of a finite set of rules that generate (or describe) an infinite number of sentences. In contrast, s-meaning does not essentially express the communicative meaning of anything and does not depend on a language in the way that c-meaning does.3 It is not unusual for a nonlinguistic entity to have an s-meaning, for example, the Long March or the Cultural Revolution. Although the s-meaning can only be represented or

1 I don’t use the ‘c-’ and ‘s-’ prefixes for the two senses of ‘interpretation’ because I shall eventually suggest that clear cases of interpretations of c-meaning involve s-meaning. 2 For ease of exposition, I am pretending that meanings exist. 3 If Donald Davidson is right to hold that all thought requires language, there is a sense in which s-meaning requires language.

chapter eight

208

expressed with words, the words are not the thing that has the s-meaning.4 That there is a difference between the sense of ‘c-meaning’ and the sense of ‘s-meaning’ is indicated by the fact that for the former, statement (1): (1) By ‘It is raining’, the speaker meant that it is raining, but it was not raining.

far from being odd, can be true, because people sometimes say something false. In short, the fact that a person (or sentence) means something does not guarantee that the person or sentence is correctly describing that thing. In contrast, statement (2), (2) The meaning of the Long March is that Mao Tse-Tung’s army would survive the attacks of the Nationalist Army; but Mao Tse-Tung’s army would not survive the attacks of the Nationalist Army,5

is necessarily false, because either the first conjunct is false or the second conjunct is false. In short, the truth of an explicit statement of s-meaning (a sentence of the form ‘The meaning of x is that p’) guarantees that it is true that p. (Of course, the converse does not hold. The fact that it is true that p does not guarantee the truth of a statement of the form ‘The meaning of x is that p’.)6 The contrast I am drawing corresponds to the contrast in Chinese between: Case 1:

zhe-ge

ju-zi

(This one) (sentence) Case 2: wo xiang-yao

tao-lun

shi shen-me

yi-si

(is) (what) (meaning) lun-yu de

(I) (want to) (discuss) (the Analects) (of )

yi-si [or, yi-yi ] (meaning)

4 There’s a difference between having an interpretation and giving an interpretation. In giving an interpretation, the speaker must accommodate what he says to the beliefs of his audience. For the most part, I will be talking about giving an interpretation. 5 Alternatively: The meaning of the Long March is that Mao Tse-Tung’s army would not survive the attacks of the Nationalist Army; but Mao Tse-Tung’s army would survive the attacks of the Nationalist Army. 6 S-meaning is related to H.P. Grice’s n-meaning (“Meaning,” in The Philosophy of Language 4th ed., ed. A.P. Martinich, New York: Oxford University Press, 2001, pp. 92–97, but I am not sure exactly how. The central cases of s-meaning involve the significance or importance of something insofar as it relates to a complex of other things, and depends upon the interests of human beings. The central cases of n-meaning do not need to be matters of importance and may be independent of the interests of human beings, e.g., ‘Those clouds mean rain’.

on two kinds of meaning and interpretation

209

I Before beginning my primary project in this section, the explanation of what is involved in interpreting c-meaning, a preliminary matter needs to be discussed. Is there is any important difference between the understanding of c-meaning and an interpretation of it? Donald Davidson, I believe, holds that there is no difference; and that is my view as well. His view, as expressed in “A Nice Derangement of Epitaphs,” was criticized by Ian Hacking on the ground that understanding and interpretation are significantly different. According to Hacking, for most ordinary utterances, a hearer just understands them and does not interpret them: “The vast majority of things we say to our peers in ordinary conversation are not interpreted at all.”7 Now it is certainly true that there are one or more senses of ‘interpretation’, according to which interpreting is not merely understanding what was c-meant.8 But what Hacking needed to show was that Davidson was not using ‘interpretation’ in any reasonable sense. And that, I think, is false. There is a sense of ‘interpretation’ that means ‘linguistic understanding’. To think that there is not is to confuse a difference in usage with a difference of meaning (a difference in the cognitive content of the words). The words, ‘officer’ and ‘cop’ (in one of their senses) mean the same thing, namely, policeman: x is an officer if and only if x is a cop. But they are used in different situations. ‘Officer’ tends to be used when addressing a policeman and is slightly formal; and ‘cop’ tends to be used outside the hearing of a policeman and is informal.9 The difference between meaning and usage could be illustrated with numerous other examples, for example, ‘mother’ and ‘mama’ and ‘father’ and ‘papa’. As regards c-meaning, the word ‘understand’10 is typically used when the speaker11 is confident that he has understood a text correctly and believes that his audience will not disagree with him. The word ‘interpret’ is typically used when one of these conditions is

7 Ian Hacking (1986), “The Parody of Conversation”, in Truth and Interpretation, ed. Ernest Le Pore, Oxford: Blackwell, 1986, p. 451. 8 Hacking, p. 450. 9 A third word, ‘pig’ has the same meaning as ‘policeman’ but is colloquial and derogatory. 10 In this and other places, I might have added the phrase, ‘and its cognates’. For the sake of simplicity, I do not. 11 The speaker had been a reader of the text.

210

chapter eight

absent. There is a distinction between the (cognitive) meaning of a word and its usage. One might object that the difference between understanding and interpretation is that interpretation applies only to complex texts. But that is not the case. A familiar sign in many parks is ‘Keep off the grass’. But suppose at the bottom of the sign in small letters are the words: ‘By order of the health department’. One interpretation of this sign is that one should stay off the turf, perhaps because it has been sprayed with a pesticide. But another interpretation is that one should not use marijuana since ‘grass’ is a colloquial term for marijuana. In many poems, some of the sentences are easy and some of them are difficult to understand. One might urge that only the latter sentences require interpretation. But notice that in the first sentence of this paragraph, the word ‘understand’ applied as naturally to the sentences that supposedly require interpretation as to those that do not. That ‘understand’ and ‘interpret’ mean the same thing can be illustrated by actual examples of usage in which the author uses both words about the same text. That obscure phrase [‘high crimes and misdemeanors’] requires interpretation: we must ask which understanding of it fits best with the most persuasive overall structure of our constitutional arrangement.12

Hacking might object to the linguistic evidence I have been mentioning by observing that all that it shows is that some or all cases of interpretation are cases of understanding, but not the converse. That objection is strictly correct. But, notice that to make it is to concede that there is at least a large overlap between understanding and interpretation. So Davidson’s use of ‘interpretation’ to mean ‘(linguistic) understanding’ is not far off from ordinary usage, if it is off at all. In any case, Davidson could forestall such disputes by saying that he was using ‘interpretation’ as a technical term in a sense broad enough to encompass simple and complex, easy and difficult, cases of identifying the c-meaning of a text. In addition to rendering objections based upon some ordinary uses of the word ultimately irrelevant, what would be important about this position is that Davidson 12 Ronald Dworkin, “Philosophy and Monica Lewinsky”, New York Review of Books, March 9, 2000; my italics.

on two kinds of meaning and interpretation

211

is giving an account that applies equally to cases of (linguistic) understanding or interpretation. That is, he has revealed the underlying unity of the allegedly dual phenomena. It now remains for me to show that the same elements required for understanding or interpreting difficult texts are also required for understanding easy texts. Davidson in effect gave an elegant proof of this in “Belief and the Basis of Meaning.”13 I will give an informal proof by indicating, too briefly, that there is a continuum of cases of linguistic understanding (or interpretation), in which both kinds of text involve more than just a hearer’s judgment about the meaning of the speaker’s words or about what the speaker meant, and that the difference between the two kinds of cases involves the number or amount of nonlinguistic judgments that are required and not the kind of requirements for understanding it. To begin, let’s suppose that Mr. Speaker is talking to Ms. Hearer, who is standing in the doorway of her house. Speaker is standing on the walkway several feet away. Holding his hand out, Speaker says, “It’s raining.” Hearer sees some, though not a lot of, water falling from some distance above Speaker’s head. It is natural (but mistaken) to think that Hearer’s understanding consists of nothing more than the belief that Speaker said that it is raining. For one thing, Speaker might be joking or conceivably speaking a language other than English. So more is required than simply attributing a belief to the speaker. In the situation just described, Hearer plausibly needs at least the following beliefs, some of which she previously had and are merely activated for this occasion and some that are newly acquired for this occasion: Interpretation A: (A-1) ‘It’s raining’ means that it is raining. (A-2) In uttering, ‘It’s raining,’ Speaker means that it is raining. (A-3) Speaker believes that it is raining. (A-4) It is raining.

Notice that the interpretation of the situation involves (1), a belief about the linguistic part of the world, (2), a belief about what the speaker meant, (3), a belief about the psychological part of the world, and (4) a belief about the nonpsychological part of the world.

13

In The Philosophy of Language, ed. Martinich, pp. 464–72

212

chapter eight

It is natural to think that an interpreter always comes to these beliefs in some linear order. One might think that the order of understanding is: (A-1), (A-2), (A-3), and (A-4). Or one might think that they come independently of each other: (A-1) from Hearer’s knowledge of English, (A-2) from Hearer’s belief about Speaker’s knowledge, (A-3) from her psychology, and (A-4) coming from observation. I think both of these beliefs are incorrect.14 My guess is that there is no regular temporal ordering of all four kinds of beliefs. 15 In addition to the possible temporal sequence of (A-1) preceding (A-2), (A-2) preceding (A-3) and (A-3) preceding (A-4) in the process of coming to the interpretation, it is quite easy to see how Hearer’s seeing of the falling water may have led her (tentatively) to believe (A-4), and that may have helped her to understand and hence to accept (A-3) and (A-1) and (A-2) in that order. If the ordering of (A-3) and (A-1) prior to (A-2) seems odd, this may help make it plausible. Hearer’s belief in (A-3) and (A-1) may have led her to think that Speaker uttered, ‘It is raining’, and not what Hearer otherwise would have thought Speaker uttered, namely, ‘It’s Rainey [a person’s name]’,16 especially if the Speaker does not know anyone named ‘Rainey’. The psychological process of arriving at interpretation A might have included other propositions considered but rejected, such as these:

14 The following passage from Leo Strauss can be interpreted as asserting the temporal priority of understanding context, including the genre of the work, to interpretation: “The context in which a statement occurs, and the literary character of the whole work as well as its plan, must be perfectly understood before an interpretation of the statement can reasonably claim to be adequate or even correct” (Persecution and the Art of Writing, Glencoe, IL: The Free Press, p. 30). 15 Within biblical scholarship, some literary historians claimed that one must determine the original text of a book before it can be interpreted. A redaction theorist might hold that redactor’s contribution to a text needs to be identified before it can be interpreted. But, as Barton says, the “process of understanding is not so simply linear” ( John Barton, Reading the Old Testament, Philadelphia, The Westminster Press, 1984, p. 70; see also p. 65). 16 Since I am hard of hearing, I often rely on what I observe about the nonlinguistic context or know about the preceding linguistic context to largely determine what words I think I hear and what the speaker means and believes. People with better hearing and worse eye-sight might rely more on what they hear to determine what they think they see. Notice that I might have but did not include Hearer’s belief about the words Speaker uttered. It is possible that Speaker said, “Israney’s here” but meant ‘It’s raining’ and Hearer appropriately corrected Speaker’s words for the purpose of understanding what he meant.

on two kinds of meaning and interpretation

213

(B-4) It is not raining [the descending water has a shape not characteristic of rain],

or (B-3) Speaker believes that it is not raining (but Speaker is kidding Hearer),

or (B-2) In uttering, ‘It’s raining’, Speaker means that it is not raining.

Why none of (B-2), (B-3) or (B-4) were accepted is a complex question. Briefly and roughly, they fit less well with the complex Network of Beliefs with which Hearer came to the interpretive situation and the sensory input. Once a person is comfortable with a set of beliefs that serve as her understanding of the utterance (and the situation more broadly), the beliefs are accepted as a group (for practical purposes), because they are mutually supporting. While Hearer may come to believe (A-2) partially on the basis of believing (A-1), (A-3), and (A-4), it is equally true that Speaker may come to believe (A-4) partially on the basis of (A-1), (A-2), and (A-3). Hearer’s belief that Speaker believes that it is raining and means that it is raining gives Hearer good reason to believe that it is raining, ceteris paribus. Similar comments apply to (A-2) and (A-3). In a simple situation like the one being considered, the brain processes are so quick that there is no perceptible interval between hearing and understanding. In more complex situations, there is often a perceptible time lag, as the person tries to settle on a set of beliefs that adequately handles the facts. (Hacking thinks of the latter cases as the only genuine cases of interpretation.) One might think that information about the nonlinguistic and nonpsychological world are not relevant to an interpretation. To see the relevance of (A-4) or some other nonpsychological belief about the world to the interpretation of ‘It’s raining’, consider a slightly different situation. Suppose that Speaker had said, ‘It is not raining.’ A sensible understanding of this situation is the following: Interpretation B (B-1) ‘It is not raining’, means that it is not raining. (B-2) In uttering, ‘It is not raining,’ Speaker means that it is not raining. (B-3) Speaker believes that it is not raining. (B-4) It is not raining.

214

chapter eight

In this case, (B-4) is supported by (B-1), (B-2), and (B-3). As the situation for Interpretation B has been described so far, I have not said whether anything in Speaker’s environment suggests that it is raining or not. (Let’s not count Speaker’s words as part of the environment.) There are two basic cases to consider. One is that no water is or seems to be falling on Speaker. In this case, the environment gives further support to (B-4), not to mention (B-3). The other case is that Hearer sees water falling on Speaker. This gives strong though not conclusive evidence that (B-3) is false and that (A-4) is true. But in this case (B-2) and (B-3) become doubtful (more doubtful than (B-1) becomes). And Hearer has various courses of action available to her. She may change her beliefs with respect to (B-3) and (B-1). Alternatively, she may act to gather more information. By walking out of her doorway, Hearer may discover that Speaker is standing near a water sprinkler that Hearer could not see from the doorway. This would account for the falling water. Each of these changes would probably result in Hearer having a different understanding or interpretation of what Speaker meant. Let me change the text to indicate perhaps more clearly how judgments about the meaning of the text and judgments about the nonlinguistic facts interact. Suppose the text is ‘GODISNOWHERE’. Two possible interpretations are C and D. Interpretation C: (C-1) ‘GODISNOWHERE’ should be understood as consisting of the following four words: ‘God,’ ‘is,’ ‘now,’ and ‘here.’ (C-2) ‘GODISNOWHERE’ means that God is now here. (C-3) By ‘GODISNOWHERE,’ the author means that God is now here. (C-4) The author is a theist. Interpretation D: (D-1) ‘GODISNOWHERE’ should be understood as consisting of the following three words: ‘God,’ ‘is,’ ‘nowhere.’ (D-2) ‘GODISNOWHERE’ means that God is nowhere. (D-3) By ‘GODISNOWHERE,’ the author means that God is nowhere. (D-4) The author is an atheist.

It would be easy to change the facts slightly to get an interpretation that consisted of (D-1), (D-2), (D-3), and (C-4), that is, an interpretation that a theist seriously asserted that God is nowhere (because he exists outside of space and time); and another interpretation that included none of (C-1)–(D-4). (For the latter, suppose that the language is similar to English but does not use a copula, that a per-

on two kinds of meaning and interpretation

215

son is named “Godis’ is being referred to, that he is being said to be present at the time of speaking, and the speaker is an agnostic.) Examples like ‘GODISNOWHERE’ are relatively frequent in Chinese, for example, (a) jin-nian-hao-dao-mei-shao-bu-de-da-guan-si

which could mean (a-1) jin-nian-hao-dao-mei shao-bu-de-da-guan-si (This year has bad luck, and it is unavoidable to have a lawsuit.)

or (a-2) jin-nian-hao dao-mei-shao bu-de-da-guan-si (This has good luck, and there are no bad things, and there will be no lawsuit.)

If we were to consider an interpretation of a long poem, an interpretation would have to bring in many more nonlinguistic facts in order to give an interpretation of what the poet or the poem meant,17 because so much of the c-meaning of a poem is inferred from judgments about propositions that are not closely related to the meanings of the actual words used. Suppose the poet is Russian, or lived five hundred years ago and hence had very different beliefs from us. Nonetheless, the requirements for understanding the poem would be no different from understanding the sentence, ‘It is raining’. Let’s now consider an example that approximates to the complexity in scholarly interpretation. In such interpretations, it becomes necessary for the interpreter to make explicit propositions of a kind that might have been left implicit in simpler examples. I am now interested primarily in interpretations that are given or presented to another person, in contrast with interpretations that a hearer merely has. Often, the meaning of only a few words is made explicit. The rest, along with other information is merely presupposed. Interpretations that are given by scholars focus on the points of actual or most likely disagreement. This is illustrated by two examples of interpretations of a sonnet by Gerard Manley Hopkins:

17 It is relatively complicated to explain when one is concerned with what the poet meant and when one is concerned with what the poem meant. And I pass over that issue in the main text of this paper.

216

chapter eight Spring and Fall: To a Young Child Margaret, are you grieving Over golden grove unleaving? Leaves, like the things of man, you With your fresh thoughts care for, can you? Ah! as the heart grows older It will come to such sights colder By and by, nor spare a sigh Though worlds of wanwood leafmeal lie; And yet you will weep and know why. Now no matter, child, the name: Sorrow’s springs are the same. Nor mouth heard of, ghost guessed: It is the blight man was born for, It is Margaret you mourn for.

Interpretation I This poem, “Spring and Fall: To a Young Child,” is about the ability of a young child to feel emotions naturally generated by things of nature, and the inability of a middle aged person to feel those same emotions. In middle age, the emotional readiness of the child to react to nature is replaced by an intellectual response. In middle age, the adult will be able to understand the significance of Fall, but not be able to feel it. This interpretation is supported by the title of the poem. If the normal life span of a human being is 70, as it is proverbially held to be, and if it is divided into four roughly equal parts, then the Spring of a person spans the first eighteen years of a person’s life. The child Margaret addressed in this poem is certainly within that span, probably somewhere between the ages of five and ten; so she is in the “Spring” of her life. The Fall of a human life would then be between the ages of thirty-one and forty-five. This period, which is often referred to as “middle age,” is in fact part of the downward slope of a person’s life. However, the Fall of a human being is also the period of her intellectual height. Her undiminished intellectual power and her experience would give her the wherewithal to make wise judgments. Reason would predominate over emotion. The poem begins by indicating that the young girl is grieving over the falling leaves of autumn. What makes her sad is nothing that she can rationally comprehend; her “fresh thoughts” are incapable of rationally comprehending the significance of the falling leaves. When “her heart grows older,” that is, when she is old enough to be in the Fall of her life, she will not be able to react emotionally to the falling leaves—“It will come to such sights colder”—however, she will understand their significance, as she could not understand it when she was young.

on two kinds of meaning and interpretation

217

What she will understand is that the falling leaves signify her own death: “It is the blight man was born for, / It is Margaret you mourn for.”

It is not necessary for me to enumerate each proposition of the interpretation and to comment on its type. It is obvious which clauses or sentences say something about the meaning of the words or what the poet meant and which say something about the psychology of the poet or about the world more generally. The sentence ‘The poem begins by indicating that the young girl is grieving over the falling leaves of autumn’, is semantically similar to the sentences of our simple interpretations that stated utterance meaning. If greater explicitness were required, the interpreter could easily have written. (I-1) ‘Margaret, are you grieving over golden grove unleaving’ means that the speaker is asking the addressee Margaret whether she is grieving over the falling leaves of golden grove. or (I-2) By ‘Margaret, are you grieving over golden grove unleaving’ Hopkins means that the speaker is asking the addressee Margaret whether she is grieving over the falling leaves of golden grove.

The interpreter assumes that sentences like (I-1) and (I-2) are unproblematic and hence do not need to be set out explicitly. The sentences that the interpreter does use—what would naturally be called an ‘interpretation of the poem’—presupposes sentences like (I-1) and (I-2). The focus of Interpretation I are the facts that the interpreter believes Hopkins is using and that the reader may not know. The interpretation also clarifies implications of the poem: (I-3) What makes Margaret sad is something that she can feel but not rationally comprehend. (I-4) When Margaret is old enough, in the Fall of her life, she will not be able to react emotionally to the falling leaves, but will understand her sadness. (I-5) Margaret will realize that she is sad about her approaching death.

About half of the first paragraph provides factual information that the interpreter thinks is relevant to the interpretation. The last paragraph, for its part, combines explications of meaning, factual information, and further implications intended by Hopkins. The opening clause, ‘Margaret, are you grieving?’ and the final clause, “It is Margaret you mourn for,” are hardly more difficult than ‘It is raining’. The intervening clauses vary in difficulty. In short, the interpretation

218

chapter eight

or understanding of this sonnet is on a continuum with the short texts we analyzed earlier. Let’s now consider another interpretation of Hopkins’s poem in order to show that the factual information that may justifiably be used in interpretation may not be obvious from the language of the poem itself. The following interpretation builds on interpretation I, but then goes beyond it. In short, we are moving further along the continuum of interpretation in the direction of complexity and difficulty.

Interpretation J At one level, the poem, “Spring and Fall: To a Small Child,” is about the ability of a young person to feel emotions naturally generated by things of nature, and the inability of a middle aged person to feel those same emotions. In middle age, the emotional readiness of the child to react to nature is replaced by an intellectual response. In middle age, the child will be able to understand the significance of the fall, but not be able to feel it. One could correctly observe that the normal life span of a human being is 70, that it can be divided into four roughly equal parts, of which the Spring would consist of the first eighteen years of a person’s life and the Fall to the years between thirty-one and forty-five. This so-called “Fall” of a human life would coincide with the intellectual height of most people. Their undiminished mental powers and experience would enable them to make sober judgments. Reason would predominate over emotion. But there is much more to the poem than this. Gerard Manley Hopkins was a Jesuit priest, and we know from biographies and his letters that he was concerned about the consequences of original sin, the Christian belief that all human beings come into the world as sinners, and as such, are destined for Hell, unless something or somebody redeems them. Thus, the word “Fall” in the title refers to the “Fall of Mankind” or “the Fall of Adam and Eve.” And one of the principal consequences of original sin is death. The word “Spring” refers not so much to the season of the year as to the metaphorical jumping of a human being into the world. The idea of a somewhat uncontrolled entrance to the world is central to the thought of some modern religious thinkers. Soren Kierkegaard talked about a leap of faith; and. Martin Heidegger, though he ended life as an atheist, studied for the priesthood as a young man and never lost a kind of mystical understanding of human existence. If we take Hopkins’s interest in original sin seriously, then we will

on two kinds of meaning and interpretation

219

see that while the young girl’s “grieving” is emotionally appropriate, it is not intelligible to her. She does not understand it. The answer to the rhetorical question, Can you “with your fresh thoughts care” about “the things of man”?, is that of course she cannot. She lacks knowledge about original sin, the corrupted world, and about her own death. However, her “ghost [that is, soul] guessed.” As the years go by, she will lose her innocent emotional response and have them replaced by an understanding of these things, and also a realization of a selfcentered worry, that what is saddest to her is her own death, “It is Margaret you mourn for.”

In Interpretation J, the interpreter is providing information that she thinks the reader may not know. But this is information that Hopkins knew about himself at least to some extent and could have known completely with some effort. Hopkins wrote his poetry in English a century ago, and there is a lot of information about England at the time and about his life and attitudes.

II S-meaning At some point in the attempt to understand a poem or other text of some importance a hearer’s or a reader’s search for understanding or for an interpretation passes beyond a search for c-meaning to something else. Interpretation J contains something of this in virtue of the connection it makes between the poem, the psychology of Hopkins and Roman Catholic beliefs. But let’s consider an example where this is more explicit. The interpretation is a seventeenthcentury account of Mary Rolandson, an American woman kidnapped and held captive by Indians. I will first give a summary of the account with segments numbered for ease of reference. The italicized parts stand for elements that are obviously Rowlandson’s ‘opinion’ about the facts. These are discussed in the interpretation.

A Narrative of the Captivity and Restauration of Mrs. Mary Rowlandson (1) Indians attacked our settlement Lancaster, Massachusetts in February 1675. They burned several houses and killed many. Several settlers were stripped naked before being killed by the Indians. One who offered the Indians money for his life if they would spare him was

220

chapter eight stripped and disemboweled. Thus they went on burning and destroying before them. The Indians broke into my house and killed or injured most of the inhabitants, including one of my children, my sister, and her children. My dogs, which in other circumstances would have attacked the Indians hid on this occasion. (2) God did this to make us the more to acknowledge his hand and to see that our help is always in him. (3) The Indians laid hold of us and said, “Come go along with us; I told them they would kill me: they answered, that if I were willing to go along with them, they would not hurt me. (4) God by his almighty power preserved a number of us from death. (5) The Indians took me and twenty three other settlers away. Being totally exhausted, the next day I rode with my child on a horse behind an Indian. (6) God was punishing me for my sins but also showing his mercy towards me. (7) My daughter died nine days later. (8) The Indians buried her. An Indian offered me a Bible, and said the other Indians would let me read it. I often took out my Bible and got consolation from reading it (9) Tired of a pregnant European woman’s complaints, the Indians stripped her naked, sang and danced around her; then, smashed in her head and that of a child she held, and threw her in a fire. (10) The Indians scavenged in a field that the English settlers had harvested. I got two pieces of corn; one was stolen from me. An Indian gave me some horse liver to eat. I put it on the coals to roast. I ate half, and half was taken by other Indians. I broke down and cried in front of the Indians. One asked me, why I wept, I could hardly tell what to say; yet I answered, they would kill me: “No,” said he, “none will hurt you.” Then came one of them and gave me two spoonfuls of meal to comfort me, and another gave me half a pint of Peas, which was worth more than many bushels at another time. (11) I did various work for several Indians, each of whom paid me for my work. When I was hungry, a squaw showed herself very kindly to me and gave me a piece of bear; she also gave me some groundnuts. Later another squaw laid a skin for me and gave me some groundnuts and bade me come again. (12) The Indian party in which my son was kept captive was attacked by Mohawks, allies of the French. I am glad that my son was not captured by them, for it might have been worse for him had he been sold to the French than it proved to be in his remaining with the Indians. (13) The papoose of my Mistress died, and there was one benefit in it, that there was more room. They buried the papoose the next day, and afterward morning and evening there came a company to mourn and howl with my Mistress, though I confess, I could not much condole with them.

on two kinds of meaning and interpretation

221

(14) It rained and I stayed in a wigwam, while many Indians slept all night in the rain. Thus the Lord dealt mercifully with me many time, and I fared better than many of them. (15) After traveling all day, I laid down my load, and went into the wigwam, and there sat an Indian boiling of horse’s feet; they had nothing else. I asked him to give me a little of his broth, or water they were boiling in. He took a dish and gave me one spoonful of cornmeal porridge, and bid me take as much of the broth as I would. After three more days of traveling, I had little spirit left. Chief Philip, came up, took me by the hand, and said, “In two weeks, you’ll be free and with your husband.” I asked if what he said was true. He said it was. He let me wash and gave me a mirror to look at myself. Then he gave me food. Days later an Indian came to me, bid me come to his wigwam. He fed me many times, even though he had killed two Englishmen. Whenever I went to his wigwam, he would always give me something, even though they were strangers to me. Another squaw gave me fresh pork. I cannot but remember what a sweet, pleasant and delightful relish that bit had to me, to this day. (16) The English thought that if they cut down all the Indian corn plants, they would starve and die from hunger. And as much corn as the English found, they destroyed. Yet God preserved the Indians for his hold ends, and permitted the destruction of many of us. Strangely did the Lord provide for them, so that I did not see one Indian die from hunger. I can but stand in admiration to see the wonderful power of God in providing such a vast number of our enemies in the wilderness. (17) Another thing that I would observe is the strange providence of God, in turning things about when the Indian was at the highest and the English at the lowest. (18) They mourned (with their black faces) for their own losses, yet triumphed and rejoiced in their inhumane and devilish cruelty to the English. (19) When the Lord had brought his people to this, that they saw no help in any thing but himself, then he takes the quarrel into his own hand; and thought they [the Indians] had made a pit, in their own imaginations, as deep as hell for the Christians that Summer, yet the Lord hurled themselves into it. And the Lord had not so many ways to preserve them, but not he hath as many to destroy them. (20) At first the Indians were all against my going home.. But afterwards they assented to it and seemed much to rejoice in it; some asked me to send them some Bread, others some Tobacco, others shaking me by the hand, offering me a Hood and Scarf to ride in; not one moving hand or tongue against it. (21) Thus hath the Lord answered my poor desire. And now God hath granted me my desire. O the wonderful power of God that I have seen, and the experience that I have had. So I took my leave of the Indians. Thus hath the Lord brought me and mine out of that horrible pit and hath set us in the midst of tender-hearted and compassionate Christians.

222

chapter eight Interpretation of the S-Meaning of “The Narrative of the Captivity” “The Narrative of the Captivity and Restauration of Mrs. Mary Rowlandson” is the account of an ethnocentric seventeenth-century European woman, who is blind to the misery and humanity of the Indians because her mind is darkened by her beliefs in a providential God. Her statement of the facts about what happened to her in (3), (5), (8), and especially (8), (10), (11), and (15) clearly prove that the Indians were humane to her. Much of (20) is evidence that they felt kindly toward her. (11) is evidence that at least some of the Indians acted with justice towards her; and (5), (8), (11), and (15) indicate that they were often merciful. Notwithstanding all of this evidence, Rowlandson judges the Indians harshly and unfairly, as in (18). Oddly, in one way her perceptions of the Indians are not distorted. She accurately reports that the Indians fed, protected, and even comforted her. What is distorted are her judgments about their characters. She claims they are of the devil’s party. This dissonance between the way she was actually treated and her account of the nature and cause of that treatment is astounding. The best explanation of her distorted judgment is that her religious faith darkened her mind, and in two ways. One was her negative judgment of the Indians. The other was her belief that God was the genuine cause of all the good things that happened to her, for example, in (4), (14), (19), and (21), even though the Indians were usually the agents of them. On the occasions when she attributes some harmful action to God, she sees them as means to justifiable ends, e.g. (6). Especially strange is Rowlandson’s recognition that her son was probably better off with the Indians than with the French (12)! Certainly, the facts reported in (1), (5), (7), and (9) would justify some negative judgment about the Indians. But such a judgment would have to be balanced by the fact that the Europeans were as bad or worse. The English “cut down all the Indian corn plants,” so that the Indians would “starve and die from hunger” (16). This strategy of warfare is indiscriminate cruelty. A large part of the Indians she traveled with were noncombatants, elders, women, and children. In short, Rowlandson applies a double standard. In (18), she thinks it exceedingly strange that Indians mourned the deaths of their fellows and rejoiced at the death of the Europeans. But didn’t the Europeans behave in the same way? Moreover, Rowlandson’s indifference to the death of her mistress’s child described in (13) is callous. Her observation that the child’s death had the benefit of providing more room in the teepee is chilling.

This interpretation includes many statements of Rowlandson’s c-meaning, which presuppose statements about the word and sentence meaning of much of the text. If, contrary to fact, ‘attacked’ meant

on two kinds of meaning and interpretation

223

‘celebrated with’ in (1) or if Rowlandson meant this, then her hatred and disgust would be unintelligible. The interpretation also includes many statements about Rowlandson’s attitudes and beliefs. But the focus of the interpretation concerns the s-meaning of her account. One further point may be made. In addition to the interpretation in “Interpretation of the S-Meaning of ‘The Captivity’,” Rowlandson’s own text includes an interpretation of the s-meaning of the events that happened to her, many of which did not involve language, for example, the killing, the feeding, and much of the comforting. What caused Rowlandson to believe, say, (1), (3), and (5), the initial facts about the raid, were things that she saw, heard and smelled. In theory, she could have—it would have been crazy to do so, but she could have—thought that what she saw were actors dressed up like Indians and Englishman pretending to fight and kill each other. Instead, she came to believe or judge that what she saw were Indians raiding her settlement. The unitalicized portions of the account are uncontroversial. It would be natural to say on the basis of them that she understood that the Indians were attacking, and killing and so on. But the italicized portions are more controversial. Although some people would still accept them today, most people would not. And so although one might say that those segments express part of her understanding of what happened, one is more inclined to say that they were part of her interpretation of the events, because they are unlikely to be accepted by the reader. Far from being at variance with my theory, calling her understanding of the events her ‘interpretation’ is consonant with my view that there is no principled difference between understanding or interpreting language and understanding or interpreting the world. S-meaning is the significance or importance of something. Like cases of c-meaning, cases of s-meaning fall along a spectrum, from the very simple at one end to the highly complex at the other. At the simple end, interpretations of s-meaning indicate how one thing is connected to another thing.18 And the knowledge of these things and their connection makes them intelligible and indicates their significance. Suppose a parent walks into the kitchen and sees a broken glass of milk on the floor. She says,

18 For the rest of this chapter, I will often use ‘thing’ to abbreviate ‘thing, state, or event’.

224

chapter eight “What is the meaning of this?” [In Chinese: zhe she shen-me yi-si.]

She obviously is not asking for c-meaning; no one was trying to communicate anything. Rather, she wants to know more about the spill, about how the spilled milk occurred or why. She wants to know how the spill fits in with other facts. When the parent learns that her twins had been fighting and knocked over the glass of milk, she has the understanding (or interpretation) she was looking for: (S-1) The broken glass of spilled milk s-means that the twins have been fighting.

This example is of exactly the same kind, I believe, as these familiar philosophical ones. (S-2) That smoke means that there is a fire. (S-3) Those spots mean that the patient has measles. (S-4) The new budget means that inflation will remain low.19

The s-meaning indicates some importance. (S-2), for example, would be important to a forest ranger, or to people during a drought, because any fire poses the risk of a catastrophe. (S-3) is significant or important to the parent whose child has measles or to a pregnant woman if she has the spots. (S-4) is significant to members of the economy. Importance is always relative to a person or population, and it may be fleeting. (S-1) has some importance for a short time, namely, to the parent who has often lectured her children about not fighting, especially at the breakfast table. Although spilled milk is nothing to cry over, it can be source of exasperation. So it has some importance. In addition to being relative to a population, significance is also relative to the events it is connected with. The spilled milk is significant with respect to the twins’ scuffle, but not with respect to the budget deficit of Tonga. In simple cases, the answer to a question of the form, “What is the meaning of X?” often has ‘meaning’ in it: “The meaning of life is to love,” or “The meaning of life is to have as many satisfying experiences as possible.” In complex cases, something is significant because it is connected with many other events; consequently a statement of s-meaning requires many sentences to describe it. In such 19 H.P. Grice (1957), “Meaning”, in The Philosophy of Language 4th edition, ed. A.P. Martinich, New York: Oxford University Press, 2000, pp. 92–7.

on two kinds of meaning and interpretation

225

cases, it would be difficult to have a single sentence of the form, “The meaning of x is that . . . ”. Often, the word ‘meaning’ would not appear at all in the answer to the question. The answer to the question, “What is the meaning of Mao Tse-Tung’s Long March?”, might begin. In 1934, Mao’s army was encircled by the Nationalist Army led by Chiang Kai-Shek. Mao’s army broke through and marched first westward and then north during 1935. Mao’s army continued to be menaced by the nationalist Army and also by local bands of warriors. The successful march of 6,000 miles kept the Chinese Communist movement viable.

The word ‘meaning’ does not occur.

III I now want to point out two ways in which c-meaning and s-meaning and hence the two kinds of interpretation are connected. First, what I have been saying about s-meaning applies to the understanding or interpretation of c-meaning, as presented in section I. To interpret even a short text, a text of one word or one sentence requires relating it to other things, to propositions about the speaker’s belief, the meanings of the words, and the nonlinguistic world. That is, an interpretation of c-meaning at the same time is an interpretation of the s-meaning of the linguistic event. The degree of importance of this s-meaning is often low and fleeting, but it exists during the time that the interpretation is being prepared. If I am right about this, one might be able to find cases in which one speaker would ask a question of another speaker that might appear to be about the latter speaker’s c-meaning, but in fact is about the s-meaning. Here is an example: Helen-1: Pat-1: Helen-2: Pat-2:

Wendy came home from the hospital today. What do you mean? She had her baby on Monday. I didn’t know that she was pregnant.

Pat certainly knows what the speaker meaning of the sentence of Helen1 is, knows that Helen believes that Wendy came home from the hospital, and believes that Wendy did come home from the hospital today. But then what is Pat-1 asking about? It can’t be c-meaning.

226

chapter eight

One abstract possibility is that Pat-1 is asking about a kind of meaning that we have not talked about. That is, there is a sense of ‘mean’ that is equivalent to the sense of ‘intend’. But Pat-1 is not asking what Helen-1’s intentions are. He knows that her intention is to inform him that Wendy came home from the hospital today. What he does not know is how this information is related to other facts, so that it makes sense or is intelligible to him. Pat-1 is asking for is additional information that would allow him to connect what he knows about the c-meaning of Helen-1 with the rest of his beliefs. Helen does not respond to Pat’s question with the insulting question, “What part of ‘Wendy came home from the hospital’ don’t you understand?” She knows that Pat knows the meaning of those words and everything else we mentioned about its c-meaning. She knows that Pat needs background information in order to make sense of the utterance; and that is what she provides in Helen-2. The segment ends with Pat-2 indicating why he asked Pat-1. The correlation between c-meaning and s-meaning does not suggest that c-meaning is reducible to s-meaning. Rather, s-meaning is a necessary condition for understanding c-meaning, not identical with it. The second way in which c-meaning and s-meaning are connected is that one is often confused for the other or they are conflated. This second connection is more important than the first because what is at stake here is the clarity scholars have about what they are saying or doing. The examples that I could give are numerous. I will mention only a few here. One of the chief ideas of Quentin Skinner, the foremost historian of modern intellectual history and the philosophy of history, is that what the text of a great philosopher means is what his contemporary interpreters took it to mean. Thus, for example, he holds that Hobbes held neither a deontological nor divine command theory of natural law, because his contemporaries did not take him to hold this. This is a thesis about c-meaning. But he sometimes talks about this procedure in words that suggest that he is talking about s-meaning: “Some [historians] are instead concerned with the provision of interpretations, and thus with the process of placing texts and other such objects within the fields of meaning from which their own individual meanings can arguably be inferred.”20 It seems that placing 20 Quentin Skinner, Visions of Politics vol. I, Cambridge: Cambridge University Press, 2002, p. 10.

on two kinds of meaning and interpretation

227

one work into the field or context of other works both explains its c-meaning and shows its s-meaning. I want to end this lecture with a speculation about a possible confusion of c-meaning and s-meaning, expressed in terms of texts. Jacques Derrida is famous in some circles and infamous in others for saying, “There is nothing outside the text.” What can this sensibly mean? Or, how might he have come, unwittingly, to this position? One possibility is that it results from three aspects of his theory: (a) understanding a text requires knowing the meanings of words that are not part of the text; (b) it requires knowing facts about the world; and (c) there is no sharp line dividing what needs to be known from what does not need to be known. Therefore, there is nothing outside the text. This speculation is supported to some extent by this quotation: “What I call ‘text’ implies all the structures called ‘real’, . . . in short all possible referents.”21 It is also supported by one of Derrida’s sympathetic interpreters: ‘There is nothing outside the text’ means “There is nothing outside the context.”22 I think it is important to keep text and context distinct (even though they are relative terms), and important to keep c-meaning and s-meaning distinct, but also to understand how they are connected.23

21

Jacques Derrida (1988), Limited, Inc., Evanston: Northwestern University Press. Keith Jenkins (2000), “A Postmodern Reply to Perez Zagorin”, History and Theory 39, p. 189. 23 I want to thank Linton I-Chi Wang for his help on this paper. Since a principal motive for this collection of papers is to connect Chinese philosophy with Anglo-American philosophy, I want to express my regret that I have not been able to find in standard sources available to me treatments in Chinese philosophy of meaning and interpretation, for example, Chad Hansen’s Language and Logic in Ancient China (Ann Arbor, MI: The University of Michigan Press, 1983), A.C. Graham’s Disputers of the Tao: Philosophical Argument in China (La Salle, IL: Open Court, 1989), and Rudolf Wagner’s Language, Ontology, and Political Philosophy in China (Albany, NY: State University of New York Press, 2003). Also, there is no entry for interpretation in The Encyclopedia of Chinese Philosophy, ed. Antonio Cua (London: Routledge, 2003), and the few references for meaning are not relevant. I have not been able to make sense of the principles of interpretation for the Yi-Jing. I am open to being informed about these matters from Chinese philosophers. 22

CHAPTER NINE

METAPHORICAL USE VERSUS METAPHORICAL ESSENCE: EXAMPLES FROM CHINESE PHILOSOPHY Kim-chong Chong

The metaphor of the moon reflected in water, so often encountered in Buddhist writings, where it suggests something that cannot be grasped . . . may be used to refer to the perfection of the highest religious truths, which are so elusive and mysterious that, regrettably, they can never be fully apprehended. When, however, the same metaphor alludes to the ‘floating world’ or the impermanence of all things, there is implicit fault-finding with the world for being illusory and unreliable. Although the image is the same, the judgment embedded in it differs, one being praise and the other detraction. (Qian Zhongshu, “Metaphors Have Two Handles and Several Sides”).1

1. Introduction In an earlier essay, I have discussed the relation between Davidson’s view of metaphor and a strategic use of metaphor to be found in the Daoist text, the Zhuang-Zi.2 This essay came to somewhat tentative conclusions. In particular, I noted that Davidson’s view of metaphor as having no inherent cognitive content seems to be able to explain Zhuang Zi’s metaphorical strategy in steering clear of the espousal of any value. I also suggested that the liberating effect of this conscious use of metaphor does not seem to agree with the view of the cognitive linguists, George Lakoff and Mark Johnson, that metaphors are unconscious and embodied ways of perceiving the world. I am grateful to Cecilia Wee for her comments on an earlier version of this paper. 1 Qian Zhongshu (1998), “Metaphors Have Two Handles and Several Sides”, in his Limited Views—Essays on Ideas and Letters, Ronald Egan tr., Cambridge: Harvard University Asia Center, 1998, p. 122. 2 See Chong, Kim-chong (2006), “Zhuang Zi and the Nature of Metaphor”, Philosophy East and West 56:3.

230

chapter nine

In the present essay, I shall do the following. First, I develop further thoughts about Davidson’s view of metaphor. In one respect, Davidson’s conception of metaphor is too narrow. Davidson stresses that a metaphor cannot be paraphrased wholly without remainder. But this conception is unnecessary to Davidson’s position, because it easily leads to an essentialist view of metaphor, one that holds that there must be an inherent meaning to metaphor. At the same time, however, Davidson maintains that the way to understand metaphor is through its use, not through the idea of its meaning. This claim liberates us from the narrow conception, which includes what I shall refer to as essentialist conceptions of metaphor. Thus, the second thing I shall do in this essay is to argue against such essentialist conceptions and their application to the understanding of some aspects of Chinese philosophy. I have already discussed one such conception in my earlier essay on the Zhuang-Zi, namely, Robert Allinson’s. While I shall briefly repeat what I have said there, the main focus of the present paper will be a more extended discussion of the Lakoff and Johnson model. In particular, I shall discuss the application of the model by Edward Slingerland to philosophical texts such as the Zhuang-Zi, the Xun-Zi and the Analects.

2. Davidson on Metaphor Let us begin with an analysis of Davidon’s view of metaphor. Davidson argues that if a metaphor has some special cognitive meaning or content, then it can be paraphrased without remainder. However, with the exception of ‘dead metaphors’, this is not possible. The conclusion to be drawn from this is that a metaphor has no special cognitive meaning or content.3 The metaphorical image of the moon’s reflection on water by Qian Zhongshu in the quotation above helps to illustrate the point. The same image is used to refer to different things, and to make opposite judgments. On the one hand, it is used to refer to the teachings of the Buddha as an elusive highest religious truth. On the other hand, it is used to refer to the illusoriness of the world. If there is some special cognitive content to the metaphor-

3 Davidson, Donald (1984), “What Metaphors Mean”, in his Inquiries into Truth and Interpretation, Oxford: Clarendon Press.

metaphorical use versus metaphorical essence

231

ical image, this kind of dual and opposite reference would be contradictory. Given an understanding of Buddhist philosophy and the different contexts in which the same metaphor is used, however, there is no contradiction. The metaphor serves equally well to refer to an elusive truth that cannot be fully apprehended, and to the illusory world. On a strict literary definition of metaphor, it could be objected that an example like the one just mentioned may not be referred to as a ‘metaphor’. Thus if we are saying that the Buddha’s teaching is like the reflection of the moon in water, then this would be a simile, not a metaphor. A metaphor would intimate something from a statement of identity—the Buddha’s teaching is the reflection of the moon in water. To some extent, Davidson himself observes this distinction between a simile and a metaphor, and denies any simple explanation of metaphor in terms of simile. Thus, referring once more to the example of the image of the moon’s reflection, it could be held that the ‘metaphor’ is simply a figurative way of expressing certain likenesses between the Buddha’s teachings and the ungraspable quality of the moon’s reflection, or of the illusoriness of worldly phenomena. Given such an explanation, there would be nothing special about this ‘metaphor’, since what it intimates can be expressed clearly, in terms of certain similarities between two terms. In this usage, the so-called metaphor provides a figurative image that could be expressed in terms of a simile. However, I think that it would be a mistake to be bound by the picture of metaphor as necessarily something that cannot be paraphrased without remainder. Let us just say that sometimes, a full paraphrase may be possible, and sometimes not. As we shall see, there are various language-games of metaphor. Following Wittgenstein, we may say that there are family resemblances between various metaphors, including the ways in which they are used, that cannot be pinned down to any one essential description. Again, in this regard, the claim that there is always something left over from a paraphrase of a metaphor is unnecessary—I suspect that this claim is precisely what leads to the tendency to think that there is some special cognitive meaning to a metaphor, as I shall illustrate shortly. Davidson’s narrow conception of what a metaphor is differs somewhat from Wittgenstein’s view of language and meaning as depending on context and use. But the core of Davidson’s position can do without the narrow conception. Thus, Davidson states that “metaphor

232

chapter nine

belongs exclusively to the domain of use. It is something brought off by the imaginative employment of words and sentences and depends entirely on the ordinary meanings of those words and hence on the ordinary meanings of the sentences they comprise”.4 This emphasis on metaphor as belonging to the domain of use and the imaginative play of words and sentences not only opens the way to a broader usage of the term ‘metaphor’. It also puts us on guard against any tendency to view metaphor as having any inherent, special meaning. This is of a piece with the tendency to posit a particular essence to phenomena of language, a tendency that the later Wittgenstein argued against. In what follows, I shall examine some attempts to posit a particular essence to metaphor with reference to examples from Chinese philosophy.

3. Essentialist Views of Metaphor We begin with Robert Allinson’s claim that the butterfly dream in the Zhuang-Zi has an inherent meaning.5 In translation, the original account of the dream goes like this: Last night Chuang Chou [Zhuang Zi] dreamed he was a butterfly, spirits soaring he was a butterfly . . . and did not know about Chou. When all of a sudden he awoke, he was Chou with all his wits about him. He does not know whether he is Chou who dreams he is a butterfly or a butterfly who dreams he is Chou. Between Chou and the butterfly there was necessarily a dividing; just this is what is meant by the transformations of things.6

Like Davidson, Allinson is impressed by the fact that a metaphor cannot be paraphrased without remainder. However, he draws a different conclusion. He deduces, instead, that a metaphor must have an inherent, intuitive content. And this content can be grasped by an intuitive cognitive capacity of the human mind.7 Ignoring the latter claim, let us examine the claim that a metaphor must have

4

Davidson, 1984, p. 247. Allinson, Robert (1989), Chuang-Tzu for Spiritual Transformation: An Analysis of the Inner Chapters, Albany: State University of New York Press. 6 Graham, A.C. (1981), Chuang-tz5: The Seven Inner Chapters, tr., London: George Allen and Unwin, p. 61. 7 Allinson, 1989, p. 26. 5

metaphorical use versus metaphorical essence

233

an inherent, intuitive content. Thus, Allinson claims that the butterfly dream inherently presents the idea of beauty together with the highest positive value that is attached to it, and symbolizes metamorphosis from an inferior position to a superior one, from low to high, and so on.8 There is more than one reason for rejecting Allinson’s claim. First of all, it does not follow that if a metaphor cannot be paraphrased without remainder, then it must be the case that it has some inherent content that has not been propositionally captured. This is one instance of trying to posit an essence to a phenomenon of language where perhaps there is none. As we have seen, another conclusion is possible. That is, as Davidson argues, there is simply no such inherent or special content. Second, different interpretations of the butterfly dream are possible.9 Davidson holds as a corollary of rejecting the idea of a special cognitive meaning, that a metaphor can intimate any number of things.10 This should not, I think, be taken as a license to say that any interpretation—including Allinson’s—is as good as any other. The use of metaphor has its context. As we have seen, the image of the moon’s reflection in the water is used and understood in a certain way, within the context of Buddhist philosophy. But we can imagine the image as having another use and another understanding within, say, the context of the poetry of Western Romanticism. There is a third reason to reject Allinson’s attribution to the Zhuang-Zi as placing the highest value on beauty. That is, instead of imposing the highest value upon anything, Zhuang Zi steers clear of espousing any particular value. We shall see why he would want to do this and the strategy that he adopts, shortly. In looking at this strategy and its context, we would be looking at a particular imaginative use or a language-game of metaphor. Before embarking upon a description of this strategy, we should note another account of metaphor as an unconscious mapping or

8

Allinson, 1989, p. 73. Möller, Hans-Georg (1999), “Zhuang Zi’s ‘Dream of the Butterfly’—A Daoist Interpretation”, Philosophy East and West 49:4, p. 442, reads the dream in terms of one of the central themes in the text, the relation between life and death. The state of being in a dream and the state of being awake are equally authentic. Similarly, the states of being dead and of being alive. Being in one state, one is unaware of the other. 10 Davidson, 1984, p. 262. Davidson says that “there is no limit to what a metaphor calls to our attention . . .” 9

234

chapter nine

projection of embodied spatial relations onto the concepts that are used in our everyday lives. The projection is described in terms of ‘metaphorical schemas’. This is the theory of ‘conceptual metaphor’ expounded by George Lakoff and Mark Johnson in a series of jointly and separately authored works.11 This theory has recently been applied to Chinese philosophy by Edward Slingerland.12 With specific reference to the Zhuang-Zi, for instance, Slingerland notes that the concept of the ‘Self ’ takes the form of several metaphorical schemas. It is only necessary to cite a few of these.13 For instance, there is the general Subject-Self schema, consisting of a split between the subject and the self, and where the self can be represented by a person, object, location, and so on. Other schemas are said to be special cases of this basic, general schema. Thus, there is the “Locational Self ” schema, arising from interactions with bounded spaces and containers. Here, one talks of the self in terms of a container in which certain things can be stored. Or else the container can be one part of the self, such as the heart-mind. As an example, Slingerland mentions a passage in the Zhuang-Zi that talks of the fasting of the heart-mind, and where the heart-mind “is likened to a stomach that can be made tenuous or empty through metaphorical fasting. Once the fasting is complete, the only thing left will be the qi, which is in turn described as being so tenuous a substance that it has space to ‘receive things’ and serve as a reservoir for the Way to gather.”14 Another schema is the “Essential Self ” schema, where there is locution of what is essential to the self. This is combined with the container schema, where certain things are said to be properly internal to the self, and others are properly external to the self. Directly after this list, Slingerland states that “The power of this

11 Lakoff, G. and Johnson, M. (1980), Metaphors We Live By, Chicago: The University of Chicago Press; Johnson, Mark (1987), The Body in the Mind, Chicago: The University of Chicago Press; Lakoff, G. and Johnson, M. (1999), Philosophy in the Flesh, New York: Basic Books. 12 Slingerland, Edward (2003), Effortless Action—Wu-wei as Conceptual Metaphor and Spiritual Ideal in Early China, New York: Oxford University Press; Slingerland, Edward (2004), “Conceptions of the Self in the Zhuang-Zi: Conceptual Metaphor Analysis and Comparative Thought”, Philosophy East and West 54:3. 13 For the metaphorical schemas described in the next paragraph, see Slingerland, 2004, pp. 328–30. 14 Slingerland, 2004, p. 329. For the ancient Chinese, the heart-mind or xin (sometimes translated as ‘mind’ or ‘heart’) is the organ capable of both cognitive and affective functions.

metaphorical use versus metaphorical essence

235

metaphor schema is that it motivates a variety of entailments that have crucial soteriological significance and yet that can be accepted without need for justification or argument by anyone familiar with the use of containers.”15 Slingerland is right that these locutions of the ‘Self ’ are present in the Zhuang-Zi. However, this does not go far enough to help us to understand how Zhuang Zi uses metaphors in the text. Take the fasting of the heart-mind, for instance. This is correctly described as an aspect of the self that is referred to in terms of a container. Under this schema, the heart-mind can be said to be either empty or full. A further question can still be asked, namely, what motivates this locution in the Zhuang-Zi? According to the model of conceptual metaphor adopted by Slingerland, there is no conscious motivation. Instead, the ‘motivation’ is the result of an unconscious mapping of our sensori-motor experiences with locations and containers in everyday life (source domain) onto the linguistic and conceptual (target) domain. Under the theory, metaphors are non-propositional, in the sense of a causal mechanism of projection. This explanation, however, does not mention the creative employment of a metaphorical strategy in the Zhuang-Zi. Insofar as the theory of conceptual metaphor fails to mention this strategic use of metaphor in the Zhuang-Zi, it does not do justice to its philosophical aim and content. An appreciation of this strategy would make it hard to accept Slingerland’s claim that the significance of the heart-mind “can be accepted without need for justification or argument by anyone familiar with the use of containers”. Zhuang Zi refers to people who fail to “empty” their heart-minds of distinctions. And not everyone familiar with the use of containers would be able to come up with the creative use of metaphor in argument, in the way that Zhuang Zi does.

15 Slingerland, 2004, p. 331. The list of things said to be internal and external to the self is not essential to the present discussion. For the internal, we have Heaven, qi, spirit, Virtue, true self, the numinous. For the external, we have (the) Human, ‘full’ heart-mind, knowledge or scheming, fame or achievements, cultural standards, likes and dislikes, life and death, the (political) world, the physical form. The list as it stands may not be intelligible without some explanation, but any attempt to explain these terms would take us away from the issue at hand.

236

chapter nine 4. The Metaphorical Strategy of the Zhuang-Zi

In the text, Zhuang Zi mentions the endless disputes of philosophers such as the Confucians and the Mohists, and the restless states of their heart-minds (xin ).16 Zhuang Zi’s aim is to attain ‘clarity’ (ming ) of the heart-mind that involves being detached from these disputes. As part of this strategic aim, he never directly criticizes the Confucians and the Mohists. Instead, he uses the voice of characters like Nanguo Ziqi who describes the disputants as restlessly fighting with their heart-minds. The disputes are likened to sounds that erupt in the hollows when the wind blows, or to the peeps of birds. Although these sounds arise and stop spontaneously, the disputes of men cannot cease because they have a fixated heart-mind (cheng xin). In other words, they are dead set against allowing other perspectives. Despite this strategy of speaking through another voice, however, Zhuang Zi is conscious of the logical status of his own words. That is, if the words of the philosophers are no better than the sounds of the hollows or the peeps of birds, what is the status of Zhuang Zi’s own words? And isn’t Zhuang Zi similarly caught up in emotional entanglement? In response to this, Zhuang Zi adopts at least two metaphorical strategies: First, he conceives of the heart-mind as a mirror or still water that (ideally) reflects but does not store whatever comes before it. Second, he applies a certain ‘empty’ structure to his words by using what has been translated as ‘goblet words’ (zhi yan ). Both strategies are related, with the common goal of maintaining the ‘clarity’ of the heart-mind. This conception of the clarity of the heart-mind in terms of the metaphor of the mirror differentiates it from the clarity of discursive discourse. The aim of Zhuang Zi’s discussion is to clear the heartmind of disturbing ‘impurities’, namely, the storage of distinctions. It is not to heighten the heart-mind’s capacity to make distinctions, nor on that basis, to state any particular doctrine. The project, in other words, is not one of ‘critical inquiry’ but of ‘stilling the heart-mind’.17 16 In the following paragraphs that describe Zhuang Zi’s metaphorical strategy, I have borrowed from a more detailed and sinological account that I have given in (2006), “Zhuang Zi and the Nature of Metaphor”, Philosophy East and West 56:3. The account is mainly of the second chapter of the first seven chapters of the Zhuang-Zi, usually referred to as “The Inner Chapters”, and generally believed by scholars to be written by Zhuang Zi himself. 17 The ‘stillness’ or ‘clarity’ of the heart-mind is elsewhere described as being in

metaphorical use versus metaphorical essence

237

The second, related, metaphorical strategy consists in the use of ‘goblet words’. The zhi ( ) is a wine goblet that tips when it is full, and rights itself when empty. Shuen-fu Lin has described its metaphorical use in the Zhuang-Zi as follows: The basic meaning of chih [zhi ] is the “wine goblet.” Chih-yen [zhi yan] is used as a metaphor for the Taoist ideal use of the mind [xin, heartmind] in making speech. The point that is emphasized is that chih— a drinking vessel used as a metaphor for the mind—is originally empty and gets temporarily filled with liquid—a metaphor for words—which comes from a larger wine container only when the occasion requires one to do so. Chih-yen, then, is speech that is natural, spontaneous, unpremeditated, always responding to the changing situations in the flow of discourse. If one can engage only in this sort of verbal act, one can keep his mind perpetually in a state of naturalness, harmony, transparency, and emptiness. Chih has also been interpreted by scholars as a pun on the character chih [zhi ], meaning “uneven, irregular, and random.” Understood as such, chih-yen means “irregular and random words,” referring especially to the random comments made by the “implied author” . . . on the stories as well as by the characters within the stories themselves. In terms of the organization of ideas, each of the “Inner Chapters” does appear to have this “random and haphazard” quality.18

Let us see how the structure of Zhuang Zi’s words resembles the operation of the goblet. The strategy here is to take a particular distinction (good/bad, right/wrong, this/that, allowable/not allowable, self/other, being/nonbeing, beginning/no beginning, true/false, and so on) and through an outpouring of paradoxes and infinite regresses, ‘empty’ the heart-mind of the distinction. The following example comes immediately after Zhuang Zi says that he does not know whether his words fit into the category of other people’s:

an empty mode of ‘fasting’ and as ‘forgetting’. See Oshima, Harold (1983), “A Metaphorical Analysis of the Concept of Mind in the Chuang-tzu”, in Experimental Essays on Chuang-tzu, edited by Victor Mair, Honolulu: Center for Asian and Pacific Studies, University of Hawaii Press, p. 77; Cua, Antonio (1998), “Forgetting Morality”, in his Moral Vision and Tradition, Washington D.C.: The Catholic University Press of America. 18 Lin, Shuen-fu (1988), “Confucius in the ‘Inner Chapter’ of the Chuang Tzu”, Tamkang Review, Vol. 18, No. 1–4, p. 384. See also his (1994) “The Language of the ‘Inner Chapters’ of the Chuang Tzu’, in The Power of Culture—Studies in Cultural History, edited by W.J. Peterson, A.H. Plaks, and Ying-shih Yü, Hong Kong: The Chinese University Press.

238

chapter nine However, let me try making my statement. There is a beginning. There is not yet beginning to be a beginning. There is a not yet beginning to be a not yet beginning to be a beginning. There is being. There is nonbeing. There is not yet beginning to be nonbeing. There is a not yet beginning to be a not yet beginning to be nonbeing. Suddenly there is nonbeing. But I do not know, when it comes to nonbeing, which is really being and which is nonbeing. Now I have just said something. But I don’t know whether what I have said has really said something or whether it hasn’t said something.19

The infinite regresses have the effect of taking away what was initially a clear distinction, in this case, between beginning and no beginning on the one hand, and between being and nonbeing on the other. Zhuang Zi seems happy with the position that although he has said something, it could also be said that he has not said anything. Either he has said something or he has said nothing. But, paradoxically, in saying that he has said nothing, he has said something. In contemporary logic, this is known as a semantic paradox, of which a well known example is the liar’s paradox: the cognitive content of ‘This sentence is false’ (that is, what it says), cannot be determined. This strategy enables Zhuang Zi to steer clear of the espousal of any particular value. This is cashed out, for instance, in terms of the notion of ‘use’. In the Zhuang-Zi, we find stories about deformed men, gnarled trees and other objects which are conventionally designated as ‘useless’. In each case, Zhuang Zi posits some other perspective from which what has been deemed ‘useless’ may yet be seen to have some use. More broadly, what is conventionally regarded as deformed or ugly may still have the power to attract, or in some instances, enables one to remain free and to stay clear of harm’s way. However, this is not an absolute lesson, as we learn from the story of the goose that could not cackle (hence deemed useless because it could not warn its owner of intruders) and was chosen for dinner over another that could. This story comes directly after Zhuang Zi tells his disciples about the tree that is deemed worthless and is therefore able to live out its years.20

19

Watson, Burton (1968), The Complete Works of Chuang Tzu, Columbia University Press, p. 43. 20 Watson, 1968, p. 209. This story is discussed in some detail by Allinson, 1989, Chapter 12.

metaphorical use versus metaphorical essence

239

5. Conceptual Metaphor and Chinese Philosophy Let us resume the earlier discussion of the theory of conceptual metaphor, as applied to the Zhuang-Zi. I have agreed with Slingerland that there are certain metaphorical schemas under which the self is regarded as a ‘container’ in the Zhuang-Zi. Still, this does not explain the strategic use of metaphor in the text. The claim that the use of metaphor in the Zhuang-Zi is unconscious in an embodied way fails to do justice to the creative employment of metaphorical argument that is found there. As we have seen, the argument is consistently held together by the metaphors of the heart-mind as a mirror, and of ‘goblet words’, to logically avoid the particular espousal of any value in propositional terms. Whether Zhuang Zi succeeds in his attempt to avoid the statement of any proposition is another matter. But it should be clear that the sense in which Zhuang Zi’s words purport to be ‘non-propositional’ has to do with the strategy of these and other metaphorical devices.21 It has nothing to do with any unconscious causal mechanism of projection. As part of this strategy, it is important to note that the standard of coherence and consistency is an important issue for Zhuang Zi. In contrast, under the conceptual metaphor model, this does not seem to present any problem. Referring to some inconsistencies between different metaphorical schemas, Slingerland states that “Although these various schemas are at times literally inconsistent, they are generally not incompatible—that is, they serve to supplement one another and thereby fit together to form a coherent conception of the self.”22 Similarly, he states that “As we shall see, although the various schemas for the self in the Zhuang-Zi are literally inconsistent, they are not incompatible, and in fact the inference patterns that they provide fit together to motivate a coherent soteriological strategy.”23 Presumably, under the theory of conceptual metaphor, there is no problem with these inconsistencies, since the metaphorical schemas are held to be embodied and hence nonpropositional. Nevertheless, there remains talk of “inference patterns”

21

See Lin, 1988, for a description of other literary and metaphorical devices used in the Zhuang-Zi. 22 Slingerland, 2004, p. 328. 23 Slingerland, 2004, p. 328.

240

chapter nine

that are supposed to fit together to “motivate a coherent soteriological strategy”. If we follow the strategy of metaphorical argument of the use of “goblet words” that I have described for the ZhuangZi, and the logical context as well as the context of “clarity” in which it is situated, however, there is no need to resort to this forced talk about the compatibility of inconsistent schemas. That is, Zhuang Zi either succeeds in maintaining consistency, or he does not—he cannot have it both ways. Whatever may be the case, the strategy of ‘goblet words’, as we have seen, aims to avoid inconsistency. The same assertion about the coherence of inconsistent schemas is made with reference to another philosopher, Xun Zi. Referring to the “Human Nature is Bad” chapter of the Xun-Zi,24 Slingerland says: The fact that some of these metaphoric schemas are literally inconsistent does not present a problem for Xun Zi or the reader because they are conceptually coherent by virtue of their similar or complementary entailments. For instance, whether our inborn nature is a “place” that we leave and to which we do not return or a “thing” that we lose and cannot recover, the basic entailment is the same: that, as we might say in English, “there is no going back.” Similarly, although the portrayal of human nature as a substance always shared by everyone contradicts the metaphor of its being something that we irrevocably lose, these schemas do not come into direct conceptual conflict because they have very different targets (equal opportunity vs. cannot regress), which means that none of their entailments directly contradict each other. That is, we could understand human nature as something shared at birth while still realizing that it is “lost” as we mature. In this sense, the HUMAN NATURE AS SHARED MATERIAL metaphor makes explicit an entailment that is at least consistent with, and perhaps implied by, some of the other metaphors: we all “leave” from the same place or have the same “raw material” to work with.25

Contra Slingerland, it is important whether Xun Zi is talking of human nature as a “thing” that is lost, or as a “place” that is left behind. If there is “no going back”, this locution belongs to a place

24 For translations of the Xun-Zi, see Knoblock, John (1988, 1990, 1994), XunZi: A Translation and Study of the Complete Works, 3 vols., Stanford: Stanford University Press. The “Nature is Evil” (or “Nature is Bad”) chapter is in the 3rd volume, Chapter 23. Also, Watson, Burton (1963), Hsün Tzu: Basic Writings, New York: Columbia University Press. 25 Slingerland, 2003, p. 37.

metaphorical use versus metaphorical essence

241

and not a thing—Xun Zi (as Slingerland interprets him) would have to make up his mind whether he is talking of a place or a thing. Similarly for the notion of human nature as a shared substance or as a thing that is irrevocably lost. It is insufficient to point out that, happily, these have either complementary entailments or different targets. Debates over human nature, both ancient and modern, are notoriously controversial. It matters whether human nature is a substance, whether it is held to be good or bad, in what sense equal or unequal, and so on. And if it is held that this substance is irrevocably lost, what does this imply? If what is implied is that we cannot regress, then what does this in turn mean, and how are we to proceed in social and political terms? Does the impossibility of regress mean that we forge ahead and recreate ourselves, and if so, what are the implications for equality, and so on. Thus, the situation may not be as convenient as Slingerland makes it out to be. On the contrary, the possibility of conflict is ever present because of the social and political implications of the different schemas of human nature. These difficulties arise out of Slingerland’s interpretation of Xun Zi. As in the case of the Zhuang-Zi, this interpretation employs the theory of conceptual metaphor. In evaluating this interpretation, we should look at the context of Xun Zi’s arguments. A main target of Xun Zi’s critique in the “Human Nature is Bad” chapter is certain arguments put forward by Mencius. For example, Xun Zi provides a diagnosis of Mencius’s idea that human nature is good as being due to a belief in an original, pristine state that is thought to be inseparable from the resource of goodness, in the same way as certain sense organs are inseparably linked with their functions (the ears with hearing, the eyes with sight). Having provided this diagnosis, he proceeds to deflate this idea by saying that the moment one is born, one has already moved away from whatever original simplicity there may be—if indeed there was ever such a state in the first place. Thus, Xun Zi is not adhering to a position of human nature as a location, as in a starting point of a journey and to which one does not return. Instead, he is debunking what he sees as an assumption held by Mencius. That is, he is denying the existence of a pristine, moral state. Similarly, the idea that Xun Zi holds that human nature is “bad” or “evil” has to be qualified in various ways. First, there is no intrinsic evil or badness to human beings, in the way that it is the substance of human beings to be evil or bad. Instead, it is a fact that human beings have various desires, and under conditions

242

chapter nine

of scarcity and non-regulation, the circumstances in which human beings find themselves would be very bad indeed. Second, instead of any particular substance, Xun Zi talks of the cognitive and instrumental capacities that allow human beings in general to discover certain regularities in the world, as well as to create constitutive rules of ritual that will govern their proper interaction. As part of this, human beings are said to have the capacity to congregate and make distinctions. However, human beings have different characters and different abilities—not everyone is able to shape their capacities to the same extent. Thus, not everyone has the ability to become a sage. Everyone may have equal cognitive and instrumental capacities to begin with. But, according to Xun Zi, it does not follow that everyone has the same abilities.26 To take stock of my argument so far: I have argued that the theory of conceptual metaphor is unable to account for a metaphorical strategy within the Zhuang-Zi, one that attempts to avoid being stuck with the propositional adherence to any particular value. The non-propositional sense of Zhuang Zi’s words can be understood through the metaphorical device of the heart-mind as a mirror, and of ‘goblet words’. It does not lie in what is alleged by the theory of conceptual metaphor, that is, the unconscious mapping of the embodied spatial relations of everyday life onto concepts. Furthermore, it is important to the metaphorical strategy that I have described that there be no self-contradiction. The theory of conceptual metaphor, on the other hand, claims that the metaphorical schemas it describes are inconsistent, but not incompatible. Quite apart from the question of its coherence, this notion of ‘inconsistency without incompatibility’ raises the question of whether the interpretations offered in terms of conceptual metaphor are adequate, as I have illustrated in the case of Xun Zi. The point should be made more generally.

26 I have given a detailed account of Xun Zi’s arguments in relation to Mencius in Chong (2003), “Xunzi’s Systematic Critique of Mencius”, Philosophy East and West 53:2. Note that Slingerland does allow one instance of a metaphorical schema that entails not sitting well with the other schemas. This is the schema that “Human nature is a human agent that is bad, and we know that it is bad because it desires or wishes to be good.” The problem with this is that it entails saying that “human nature has internal tendencies toward being good, precisely because it is bad.” See Slingerland, 2003, pp. 37–38. For an explanation of what this point implies for Xun Zi’s position, see Cua, A.S. (1978), “The Quasi-empirical Aspect of Hsün Tzu’s [Xun Zi’s] Philosophy of Human Nature”, Philosophy East and West 28:1.

metaphorical use versus metaphorical essence

243

Sometimes, problems of interpretation can arise, quite independently of the schemas. Thus, whether a particular schema is legitimately imposed on a certain position depends on understanding that position. In other words, the schema cannot, by itself, establish the interpretation that it proposes. Let us illustrate this with reference to one more example, this time from the Analects. Consider the following example of metaphorical usage in this text. In the passage 3.8 of the Analects, Confucius’s disciple, Zixia, asks for an explanation of the the following lines from the Odes: “Her entrancing smile dimpling, Her beautiful eyes glancing, Patterns of colour upon plain silk”. The following dialogue between Confucius and Zixia ensues: The Master said, “The colours are put in after the white.” “Does the practice of the rites likewise come afterwards?” The Master said, “It is you, Shang [Zixia], who have thrown light on the text for me. Only with a man like you can one discuss the Odes.”27

In his reading of this passage, Slingerland says, “Just as all of the cosmetics in the world are of no avail if the basic lines of the face are not pleasing, so is the refinement provided by ritual practice of no help to one lacking in good native substance. It is this entailment that explains Confucius’s concern that cultural adornment be firmly rooted in its native substrate . . .”28 There is more than one description of this “native substrate” in Slingerland’s account. First, Slingerland refers to this as constituting some “basic emotions” that form the root of the ritual forms. For instance, that it is important not to lose touch with emotions such as grief when mourning, instead of being overly concerned with the formal details. I would agree with Slingerland on this aspect of the Analects, and this is something worth stressing in the ethics of Confucius, although in doing so, there is no need to mention any notion of a “native substrate”. However, Slingerland supplements this with a second description of what constitutes the “native substrate”. Thus, he cites passages that would indicate “some kind of innate tendency toward the good”29

27 28 29

Lau, D.C. (1979), Confucius: The Analects, tr., Harmondsworth: Penguin. Slingerland, 2003, p. 54. Slingerland, 2003, p. 72.

244

chapter nine

or even that “innate orientation toward the good is a universal quality.”30 Slingerland is concerned to show that there is a “paradox of virtue” in the Analects: one must already be virtuous in order to acquire virtue. Also, he finds a tension in the Analects between what he calls the ‘adornment’ metaphor, and the ‘craft’ metaphor. In the former, the rites do not alter the native substrate, but instead can only be said to depend for their effectiveness on the prior existence of the substrate. The craft metaphor, on the other hand, connotes a radical shaping of a material that has no particular prior mass or shape,31 or that requires “violent reshaping of the original material.”32 In describing these metaphors, Slingerland indicates the affinities with Mencius (human nature is originally good) and, at the same time, with Xun Zi (goodness is the result of accumulated effort and not the result of any originally good material). Slingerland is not wrong in pointing to these affinities. It is possible to read the Analects in either a Mencian way, or in a Xunzian way. Again, however, a further question arises as to whether it is legitimate to read the ethical philosophy of the Analects in either of these ways. For instance, an interpretation of the term zhi as ‘native substrate’ and the conception of the rites as an embellishment or adornment of this substrate is not self-evident. Thus, there is a reading that is quite the opposite of the reading (in passage 3.8) that it is the white that comes first. It could be held that the colors are first laid down, and white is used to enhance the colors instead. This reading means that although one has to have some character (color) in the first place, the rites (white) are still required to refine one’s character. Under this reading, the refinement is not an ‘adornment’ of anything, but involves a process of practicing and learning. In taking this reading, the Qing dynasty philosopher Dai Zhen firmly rejects any reference to the practice of the rites in terms of adornment. For him, adornment takes place when “a person’s feelings are wearing thin and yet he continues to work on the appearance of things”.33 As Dai Zhen sees it, the so-called ‘adornment’ metaphor would be inappropriate to a description of the purpose of the rites as an edu-

30

Slingerland, 2003, p. 72. Slingerland, 2003, p. 55. 32 Slingerland, 2003, p. 53. 33 Chin, Ann-ping and Freeman, Mansfield (1990), Tai Chen [Dai Zhen] on Mencius: Explorations in Words and Meaning, trs., New Haven: Yale University Press, p. 156. 31

metaphorical use versus metaphorical essence

245

cational and self-cultivational medium. For him, the rites play a more active and positive role in the formation of character. It would be a mistake to think that in rejecting the metaphor of adornment, we are thereby endorsing the ‘craft’ metaphor. The theory of conceptual metaphor offers these two alternatives: either one is talking of human nature as being based upon a ‘native substrate’, in which case one is saying that human nature rests upon a certain universally ‘good’ substance which needs to be adorned, or else one talks of human nature as being forged or crafted. However, it is fallacious to assume that we must read Confucius in either of these modes. Certainly, the metaphorical schemas of adornment and craft cannot be self-justifying, and the theory of conceptual metaphor cannot legitimize such a reading independently of other means of understanding the Analects. It should be stressed that the point is not that the interpretation that I put forward is correct, while Slingerland’s is not. Instead, the point is that conceptual metaphor cannot by itself adequately do the work of interpretation, or force the interpretation. No doubt Slingerland is right in pointing to certain locutions of, and affinities between, the Chinese philosophers. However, further investigation that goes beyond the parameters of conceptual metaphor is necessary, before any proper judgment can be made about the adequacy of an interpretation.

6. Conclusion It might seem that in doing comparative philosophy, one takes one’s pick: one interprets from some perspective, and there are different models to work with. Thus, it might be held that I am criticizing the application of the Lakoff and Johnson model of conceptual metaphor to Chinese philosophy from the perspective of Davidson’s theory of metaphor. However, so it might be held, the same game can be played from either side. But what I have said against the various interpretations of Allinson and Slingerland can be said independently of Davidson’s view of metaphor. I could have illustrated the problems with Allinson’s and Slingerland’s readings of the Zhuang-Zi, for instance, without any reference to Davidson (or even Wittgenstein). In any case, I have not agreed with everything that Davidson says about metaphor. For instance, that one must make a demarcation

246

chapter nine

between metaphor and simile, and that a metaphor cannot be paraphrased without remainder. It has seemed to me that this latter belief is unnecessary to the understanding of metaphor, and furthermore, I have diagnosed this belief as a cause of the tendency to claim that there is a special cognitive content to metaphor. In addition, Davidson’s corollary idea that a metaphor can intimate any number of things has to be qualified, if we are not to take this as meaning that ‘anything goes’ in the interpretation of metaphor. If it is the case that ‘anything goes’, we would have no reason to disagree with, say, Allinson’s reading of the butterfly dream. Despite these differences, however, I think that Davidson’s stress on metaphor as belonging to the realm of use strikes the right note, and this is where the significance of his view of metaphor lies. In this connection, I have highlighted the fact that Zhuang Zi has a highly imaginative metaphorical strategy that is missed or inadequately referred to by Allinson and Slingerland in their accounts of metaphor. Generally speaking, the reason why they miss or inadequately describe this strategy is that they share a tendency to believe that there must be an essence to metaphor. I have counterposed against this what I believe Zhuang Zi, Xun Zi and Confucius were doing, in the particular contexts within which they were situated. In other words, I have denied that the theory of conceptual metaphor is adequate to the understanding of these philosophers. This is not to say that the structure that the theory of conceptual metaphor attributes to the way we use metaphors in our daily lives is entirely wrong. Instead, I have argued against the tendency to apply what is regarded as essential to metaphor under this theory to all cases. In particular, I have argued that its application to the examples we have looked at has not helped their proper understanding. In short, there is no essence to metaphor. We have to look at the contexts and arguments in which a particular metaphor is used to appreciate its application and significance. As Wittgenstein says, instead of trying to grasp “the essence of the thing, one must always ask oneself: is the word ever actually used in this way in the language-game which is its original home?”34

34 Wittgenstein, Ludwig (1968), Philosophical Investigations, G.E.M. Anscombe tr., Oxford: Basil Blackwell, Part I, no. 116.

CHAPTER TEN

READING THE ANALECTS WITH DAVIDSON: MOOD, FORCE, AND COMMUNICATIVE PRACTICE IN EARLY CHINA1 Yang Xiao

The focus of this paper is on the word ‘and’ in the title of this volume Davidson’s Philosophy and Chinese Philosophy. I believe scholars of Chinese philosophy should engage with contemporary philosophy of language by drawing out its implications in the context of early Chinese philosophy and language; reading the Analects with Davidson can shed light on both the communicative practice in the Analects and Davidson’s philosophy of language. Reading classical Chinese texts with Davidson should also help us to see why we should not draw conclusions about the nature of communicative practice in early China based on observations about the grammatical and semantic features of classical Chinese. I shall call this style of reasoning the ‘grammatical approach’ to pragmatics. Some scholars have made arguments about what the Chinese can or cannot do with the classical Chinese language, and their arguments are based solely on their observations on the grammatical and semantic 1 An early version of this paper was presented at the conference “Davidson’s Philosophy and Chinese Philosophy: Constructive Engagement” in Beijing on June 8–9, 2004. I wish to thank my fellow participants at the conference for their very helpful comments, particularly A.P. Martinich, Michael Krausz, Koji Tanaka, Stephen Angle, David Wong, Yujian Zheng, and Samuel Wheeler. My special thanks go to Bo Mou, both for his admirable work as the organizer of the conference, and for his unfailing support and patience as the editor of this volume. I also wish to thank David Keightley, P.J. Ivanhoe, A.P. Martinich, Robert Ashmore, Carine Defoort, Michael Puett, and Anna Xiao Dong Sun; I am deeply indebted to their insightful comments on earlier versions of this paper. I would like to express special thanks to my wife Anna, from whom I have learned a great deal—perhaps more than I initially wanted!—about what an astonishingly wide range of things a seemingly innocent utterance in everyday life can mean. My wife is also the connection that brought me to Davidson. In the late 1990s I moved to Berkeley to be with her, who was a student there, and audited two of Davidson’s seminars. I didn’t realize until later that these seminars have changed my philosophical life. I dedicate this paper to the memory of Donald Davidson.

248

chapter ten

features of classical Chinese. Following Davidson, I suggest that we should make the ‘pragmatic turn’ by focusing directly on people’s linguistic or communicative practice, namely the utterances of sentences in concrete occasions on which the sentences are put to work. More specifically, I will argue against two assumptions in the grammatical approach. The first is what I shall call the empirical assumption, which asserts that, since classical Chinese is not an inflected language, it does not have any linguistic device to indicate grammatical moods. The second is what I shall call the mood-force correlation thesis, which claims that grammatical moods and pragmatic forces are closely correlated. In other words, the grammatical features, or any conventional features in general, determine how linguistic expressions can be used pragmatically. Obviously, the correlation thesis enables one to derive conclusions about pragmatic forces from observations about grammatical moods. It is through these two assumptions that I shall engage with an important debate between Dummett and Davidson in contemporary philosophy of language. The debate is regarding the relationship between the grammatical moods of a sentence and the pragmatic forces of the utterance of the sentence. Dummett endorses the moodforce correlation thesis that there is a strict correlation between mood and force, and that illocutionary force is always conventional. Davidson rejects the thesis, and argues that neither force nor ulterior purpose of an utterance is governed by linguistic conventions.2 Peter Strawson predicted in 1969 that the conflict between the communicationintention-based pragmatics and convention-based formal semantics has been, and would continue to be, the “Homeric struggle” at the heart of the philosophy of language.3 This debate between Dummett and Davidson can be seen as a continuation or unfolding of this struggle. 2 The Dummett-Davidson debate is anticipated by an earlier debate between Austin and Strawson concerning the issue of whether force is always conventional; please see Strawson, P.F. (1971a), “Intention and Convention in Speech Acts”, in Logico-Linguistic Papers, London: Methuen, pp. 149–69. Dummett has tried to defend Austin against Strawson; please see Dummett, Michael (1995), “Force and Convention”, in The Philosophy of P.F. Strawson, edited by Pranab Kumar Sen and Roop Rehha Verma, New Delhi: Indian Council of Philosophical Research, pp. 66–93, and Strawson’s reply in the same volume, pp. 403–7. 3 Strawson, P.F. (1971b), “Meaning and Truth”, in Logico-Linguistic Papers, London: Methuen, pp. 170–89. For an excellent brief history of contemporary philosophy of language from this perspective, please see Martinich, A.P. (1997), “Philosophy of

reading the

ANALECTS

with davidson

249

What I want to show in this paper is that this debate has important implications for the study of Chinese philosophy and language. If Davidson is right, we would have to say that, based on observations about the grammatical and semantic features of the Chinese language, one cannot draw any conclusion about what the Chinese speakers can or cannot do with Chinese sentences; one would have to base one’s arguments directly on observations about the pragmatic features of their communicative practice. In Section 1, I will introduce the basic concepts of mood and force, and I will introduce the ‘grammatical approach’, as well as the two basic assumptions of the approach, namely the empirical assumption and the mood-force correlation assumption. In Section 2, I will first respond to the empirical assumption by arguing that classical Chinese has its own linguistic devices (such as ending particles) to indicate moods; I will then discuss the limits of this response. In Section 3, I will discuss Davidson’s argument against Dummett’s mood-force correlation thesis and his conventionalist theory of force. In Section 4, I will use a passage from the Analects to illustrate Davidson’s point that there is no convention of sincerity, which is at the heart of his argument against Dummett’s conventionalist theory. Section 5 concludes the paper by briefly exploring the implications of what Davidson calls the principle of the ‘autonomy of linguistic meaning’.

1. In the English language we indicate grammatical mood by inflecting the form of the verb. For example, the verb phrase ‘to be immediately put into practice’ has at least four inflections, which signify indicative, interrogative, imperative, and subjunctive moods:

Language”, in Routledge History of Philosophy, Volume X Philosophy of Meaning, Knowledge and Value in the Twentieth Century, edited by John Canfield, London and New York: Routledge, pp. 11–38. Martinich convincingly shows that such a conflict between the two approaches to language starts with the debate between Russell and Strawson. As he puts it, “Behind Strawson’s objection [to Russell’s theory of description] is a view of language that is radically different from Russell. For Russell, words and sentences are the fountains of meaning. For Strawson, people using words and sentences are. For Russell, semantics is the primary object of linguistic study. For Strawson, it is pragmatics, how people use words” (p. 18).

chapter ten

250 (1) (2) (3) (4)

. . . is being immediately put into practice . . . . . . is [it] immediately being put into practice? . . . should be immediately put into practice . . . . . . were to be (could have been) immediately put into practice . . .

These inflected phrases are known as mood-indicators. When they appear in complete sentences, they indicate four grammatical moods of a sentence: (1a) Indicative sentence: What has just been learned is being immediately put into practice. (2a) Interrogative sentence: Is what has just been learned being immediately put into practice? (3a) Imperative sentence: What has just been learned should be immediately put into practice. (4a) Subjunctive (counterfactual) sentence: What has just been learned were to be (could have been) immediately put into practice.

As we can see, the mood of a sentence is a syntactic feature of the sentence, independent of any actual uses of the sentence. The mood of a sentence remains the same even when the sentence is being used to do different things in different situations; this is because the mood is a formal feature of the syntactic structure of a sentence. Now when a sentence is uttered by a speaker in a specific situation, the speaker is using it to do certain things. We need another term to refer to what the speaker is doing with the sentence; the term is the ‘illocutionary force’ of the utterance, or simply the ‘force’ of the utterance. For example, when the utterance of a sentence is being used to issue an order, we say that the force of the utterance is to issue an order. When one utters the above four sentences, (1a)–(4a), one can do at least four different things: (1b) (2b) (3b) (4b)

making an assertion asking a question or making a request offering advice (issuing an instruction, an order, or a command) expressing a wish (or regret)

The way I presented these examples might have given the impression that there is a strict correlation between the grammatical moods of a sentence (the interrogative, indicative, imperative, or subjunctive), and the forces of the utterances of the sentence (asking a question, describing a fact, offering a piece of advice, or expressing a wish).

reading the

ANALECTS

with davidson

251

It is indeed true that, when we ask a question, we often use an interrogative sentence; when we describe a fact, we often use an indicative sentence, and so on. Nevertheless, is it really the case that the interrogative sentences are always used to ask questions, just as imperative sentences are always used to issue an order? In other words, is there a strict correlation between mood and force? In “Moods and Performances”, which was first presented at a conference in 1976, Davidson tries to answer these questions. He thinks that the questions can be formulated a little differently, because the relationship between mood and force can also be seen as about the relationships of two ways of classifying utterance: The moods classify sentences, while uses classify utterances; but the moods indirectly classify utterances, since whatever distinguishes sentences can be used to distinguish utterances of them. So we may ask, what is the relation between these two ways of classifying utterances; how are assertions related to utterances of indicative sentences, for example, or commands to utterances of imperative sentences?4

What Davidson argues against is the mood-force correlation thesis, which claims that “the associated classes of utterances are identical: utterances of imperatives are commands, utterances of interrogatives are question-askings, etc”.5 If Davidson is right that there is no strict correlation between mood and force, then we should not try to determine what people can or cannot do with English sentences by looking at the grammatical features of the language; the fact that English is an inflected language with a variety of linguistic devices to indicate grammatical moods becomes unimportant and irrelevant. Let us now turn to some examples in Chinese. In order to illustrate what I call the grammatical approach, let us take a look at a classical Chinese sentence from 11.22 of the Analects:6 (C) Wen si xing zhi

.

4 Davidson, Donald (1984a), “Mood and Performance”, in Inquiries into Truth and Interpretation, Oxford: Oxford University Press, pp. 109–10. 5 Davidson, 1984a, p. 110. 6 All quotations from the Analects are to book and passage numbers in Yang, Bojun (1980), Lun-Yu-Yi-Zhu [The Analects with Translations and Comments], Beijing: Zhonghua shuju. All translations in this paper are my own, but I have been aided greatly by the existing English translations by Simon Leys and D.C. Lau.

chapter ten

252

The first character ‘wen’ means to hear; the second character ‘si’ means this; the third character ‘xing’ means to practice or to be put into practice; the last character ‘zhi’ means it—referring, in this case, what has just been heard. One may translate the sentence as something like this: ‘Having heard it, then immediately put it into practice.’ Or, ‘What has just been learned should be immediately put into practice.’ But there is a problem here: This is just one of many possible translations. Part of the problem comes from the fact that neither classical nor modern Chinese is an inflected language. If one believes that inflection is the only way to indicate the mood of a sentence, then there is no way to determine the mood of this sentence. Hence we can have at least four English translations for the original Chinese sentence:7 (1a) (2a) (3a) (4a)

What has just been learned is being immediately put into practice. Is what has just been learned being immediately put into practice? What has just been learned should be immediately put into practice. What has just been learned were to be (could have been) immediately put into practice.

That is to say, the Chinese sentence ‘wen si xing zhi’ in itself allows it to be translated into any of these English sentences, each with a different grammatical mood. How do we make sense of these grammatical differences between the Chinese and English languages? One may argue that, because there are no mood-indicators in classical Chinese, people must have been confused about illocutionary forces in ancient China. Or one may conclude that certain speech-acts (such as expressing a wish) cannot be done, due to the absence of

7 There would be more possible translations if we take into account time and number. The English language indicates time and number by inflection at every occurrence of a verb or noun. As A.C. Graham has pointed out, even though Chinese verbs and nouns have no inflection, this does not mean that the classical Chinese language does not have its own devices to indicate them. In fact, Chinese indicates time and number by particles only when time and number is relevant. As Graham argues, “we need to be told whether an event is past, present, or future no more often than is indicated by the temporal particles of Chinese. The idea that there are confusions in early Chinese thought due to the absence of tense and singular or plural seems to me quite untenable. At some places one has trouble rendering into English without committing oneself to tense or number, but this is merely a translator’s problem” (Graham, A.C. [1978], Later Mohist Logic, Ethics and Science, Hong Kong: The Chinese University Press, pp. 28–9). What I argue here is that what Graham says about time and number also applies to mood in classical Chinese.

reading the

ANALECTS

with davidson

253

the corresponding mood-indicators (such as the subjunctive mood indicator) in classical Chinese. Two scholars are representative of this kind of reasoning. Alfred Bloom has argued that, since the Chinese language does not have a linguistic device for conterfactuality, the Chinese do not have counterfactual thinking. Moreover, since argumentation needs counterfactuals, the Chinese are incapable of argumentation.8 Following this line of thinking, one can make a similar argument that the Chinese cannot express wishes, because expressing wishes also needs counterfactuals. Chad Hansen is a much more influential scholar; he has famously claimed that the ancient Chinese do not have concepts of sentence, belief, or truth, and that they never use sentences to describe facts or to express truths or beliefs; their words are only used to guide people’s behaviors. His arguments are based on observations about the syntactical differences between Chinese and English sentences. In his observations, he focuses on certain grammatical features of classical Chinese, which, in comparison to English, are obviously distinctive and unique. Here is a summary of some of these features: (1) The absence of sentence function marking in classical Chinese: The absence of sentence function marking, . . . and the use of predicate-only sentences contribute to viewing all words as having only a naming function and to the failure to distinguish the sentence as a functional composite linguistic form.9 (2) The lack of grammatical inflections in classical Chinese: Chinese does not have grammatical inflections, which in Western languages, draw attention to the sentence as a compositional unit. . . . Chinese theories of language did not concentrate on sentences because, simply, classical Chinese sentencehood is not syntactically important.10

Another manifestation of the lack of grammatical inflections in classical Chinese, according to Hansen, is that “Classical Chinese does not have explicit descriptive and prescriptive forms,” which can be

8 Bloom, Alfred H. (1981), The Linguistic Shaping of Thought: A Study in the Impact of Language on Thinking in China and the West, Hillsdale, New Jersey: L. Erlbaum. 9 Hansen, Chad (1985), “Chinese Language, Chinese Philosophy, and ‘Truth’”, Journal of Asian Studies 44 (3), p. 516. 10 Hansen, 1985, p. 500.

254

chapter ten

easily expressed by the inflections of the verbs in English.11 Based on these observations, Hansen concludes that the linguistic practices and theories in China and English-speaking countries are radically different: These observations about the differences between Chinese and English syntax explain (from a Chinese point of view) why we place so much emphasis on the sentence, or (from our point of view) why Chinese philosophers do not.12 Classical Chinese does not have explicit descriptive and prescriptive forms. Students of comparative translation, therefore, will find huge chunks of text that one translator renders in declarative English and another in imperative English. Behind this apparent ambiguity, I suggest, lies this assumption about the function of language. All language functions to guide behavior.13

One of the most striking characteristics of these arguments by Bloom and Hansen is that they never look directly at the linguistic and communicative practice; instead they focus on the grammatical features of Chinese sentences, and end up with a conclusion about the nature of Chinese linguistic practice.14 Let us now return to our earlier example from 11.22 of the Analects, (C) ‘wen si xing zhi ’, to illustrate this point. As we have shown, the Chinese verb ‘xing’ in (C) has no inflection, whereas the English verb phrase ‘to be put into practice’ has at least four inflections, which correspond to four grammatical moods. That is to say, for this one Chinese sentence (C), there can be at least four English translations: (1a), (2a), (3a), and (4a). Let me reiterate (1a) and (3a) as follows: (1a) What has just been learned is being immediately put into practice. (3a) What has just been learned should be immediately put into practice.

Note that (1a) is a descriptive, indicative English sentence, and (3a) is a prescriptive, imperative English sentence. Like Hansen, one may feel compelled to conclude that the English-speaking people can dis11 Hansen, Chad (1992), A Daoist Theory of Chinese Thought, Oxford: Oxford University Press, p. 51. 12 Hansen, 1985, p. 500. 13 Hansen, 1992, p. 51; the emphasis is Hansen’s. 14 When Hansen says, “My hypothesis is that real differences in the languages can explain differences in the popular theories of language” (Hansen, 1992, p. 25; emphasis added), he means the grammatical differences.

reading the

ANALECTS

with davidson

255

tinguish between two functions of language: stating facts and guiding behavior, whereas the Chinese see language as having only one function, .i.e., to guide behavior.15 In general, if one does take a grammatical approach, one might be tempted to argue that, since Chinese is not an inflected language, and since Chinese verbs do not have the grammatical moods such as indicative, interrogative, imperative, and counterfactual moods, the Chinese are either incapable of doing things such as making an assertion, asking a question, issuing an instruction, or expressing a wish, or they are incapable of telling them apart. Two assumptions in these arguments are problematic. The first is what I have called the empirical assumption, which is that the classical Chinese language does not have any linguistic device to indicate the grammatical moods. The second is what we have called the moodforce correlation thesis, which claims that grammatical moods and pragmatic forces are closely correlated; in other words, the grammatical features of linguistic expression determine how they can be used pragmatically. The correlation thesis enables one to derive conclusions about pragmatic forces from observations about grammatical moods. In the rest of the paper, I shall deal with these two assumptions in turn.

2. It is relatively easy to respond to the empirical assumption. Bloom and Hansen seem to presuppose that inflection is the only way for a language to have mood-indicators. However, empirical evidence shows that classical Chinese has different grammatical devices to indicate moods, one of which is through ‘ending particles’. These are words at the end of sentences that have no substantive meaning in

15 This argument has its logical problems. For example, one cannot derive homogeneity from ambiguity, i.e., from the fact that Chinese expressions can have ambiguous or multiple interpretations, one cannot draw the conclusion that there is only one interpretation. More concretely, if ‘wen si xing zhi’ can be read either as descriptive or prescriptive utterances, one cannot conclude that this means it is always prescriptive, i.e., to always guide behavior. If one sticks to the grammatical approach, then, in order to claim that all the uses and functions of sentences in a language L are to guide behavior, one would have to show that all sentences in L are unambiguously prescriptive ones.

chapter ten

256

themselves, and their only function is to indicate the moods of the sentences. Although the transcribers and editors of the Analects do not know the technical term of ‘ending particles’ or ‘particles’, they do have a systematic way of using a variety of ending words to indicate the moods of sentences. One example is the particle ‘hu ’. It has no substantive meaning when it appears at the end of a sentence, but it has an important grammatical function, which is to indicate that the sentence is an interrogative one. For example, ‘junzi shang yong (A gentleman prizes courage)’ is an indicative sentence, but if we add ‘hu ’ at the end of it, we get ‘junzi shang yong hu (Does a gentleman prize courage?)’, which is an interrogative sentence, used by Zilu to ask a question in 17.23. Another example is ‘Guan Zhong zhi li (Guan Zhong knows the rituals)’, which is indicative, whereas ‘Guan Zhong zhi li hu (Does Guan Zhong know the rituals?)’ is interrogative, and is used to ask a question in 3.22. Another interrogative ending particle is ‘zhu ’. Sentences with the ending particle ‘zhu’ are also often used to ask questions: 13.15. Duke Ding asked: “One single maxim that can lead a country to prosperity, is there such a thing (you zhu )?” Confucius replied: [. . .]. Duke Ding said: “One single maxim that can ruin a country, is there such a thing (you zhu )?” Confucius replied: [. . .].

As we can see, we can ask whether something exists when the particle ‘zhu’ is paired with the verb ‘you (there is)’ to form the following sentence: (Q )

?

You zhu

Is there such a thing?

To answer the question, one can give a positive reply by saying: (A) You zhi

.

There is such a thing.

We can find such a pattern in another passage: 7.35. The Master was gravely ill. Zilu asked permission to offer a prayer. The Master said: “Is there such a practice (you zhu )?” Zilu said: “Yes, there is ( you zhi ), and the prayer goes like this: ‘We

reading the

ANALECTS

with davidson

257

pray to you, spirits from above and spirits from below.’” The Master said: “In that case, I have been praying for a long time already.”

Now we have responded to the empirical assumption on its own ground by showing that there exist other linguistic devices to indicate grammatical moods in classical Chinese. However, this grammatical response is not satisfactory in many ways. I shall mention just two problems here. The first is that certain types of mood-indicators are absent in the Analects; for example, although we can find interrogative and indicative particles in the Analects, we cannot find any imperative particles. Now let us again take the sentence ‘wen si xing zhi ’ as an example. Earlier in the paper, we mentioned that there are four possible English translations, two of which are imperative and indicative sentences. However, there is no grammatical or semantic feature in the original sentence that gives us any information about whether it is indicative or imperative, because in the pre-Qin and Han periods there was no ending particle indicating the imperative mood. It is only in much later periods (the Tang and Song dynasties) that new ending particles such as ‘zhuo ’ and ‘hao ’ were invented to indicate the imperative mood.16 Had we found ‘wen si xing zhi zhuo ’ in the Analects, we would have been able to say that this is an imperative sentence. The second problem is that, with regard to the particles we do find in the Analects, even though they can indicate grammatical moods of the sentences, they do not always tell us what the pragmatic forces of the utterances are. In other words, there is no correlation between mood and force in classical Chinese. For example, if we look at all the sentences that end with the interrogative particle ‘hu ’ in the Analects, we would find out that these sentences are not always being used to ask questions or make requests. Confucius’s very first utterance in the first passage of the Analects has the ending particle ‘hu’. One translator correctly renders it as follows: “The Master said: ‘To learn something and then to put it into practice at the right time: is this not a joy?’”17 Although it is a grammatically interrogative sentence, 16 For a variety of examples of such imperative particles, see Luo, Ji (2003), BeiSong-Yu-Qi-Ci-Ji-Qi-Yuan-Liu [The Mood-Indicators in the Northern Song Dynasty and Their History], Chengdu: Bashu shushe, pp. 140–76 and pp. 230–8. The use of ‘zhuo’ as an imperative particle can still be found in many regional dialects today in China (pp. 144–52). 17 I am using Simon Leys’ translation here. The translation reflects faithfully the

258

chapter ten

it is obviously a rhetoric question, which can be readily expressed by an indicative sentence: “It is a joy to learn something and then put it into practice at the right time.” Let us take the interrogative mood-indicator ‘zhu’ as another example. The particle ‘zhu’ appears 14 times at the end of a sentence in the Analects. It turns out that they are not always being used to ask questions or make requests: Sometimes the sentence is used to ask a question (see 6.6, 7.35, 9.13, 13.1, and 13.15); sometimes it is just a rhetorical question, which is equivalent to an assertion (6.30, 12.11, and 14.42). How do we make sense of these cases? Should we conclude that classical Chinese is uniquely different from all the other languages because the mood-force correlation thesis does not apply to it? I believe this is where Davidson comes in. These examples from the Analects sharply highlight the issue that is at the heart of the DummettDavidson debate, in which Davidson argues against Dummett’s moodforce correlation thesis. For Davidson, communication is possible only because there is no correlation between the grammatical features of a language and what people can do with it. Hence, if Davidson is right, there is nothing unique about the lack of such a correlation in classical Chinese.

3. One of Davidson’s early arguments against the mood-force correlation thesis is based on the existence of counterexamples. After having cited a passage from Dummett, in which Dummett gives his version of the correlation thesis,18 Davidson comments: [W]hat bothers me is the implied claim that assertion and the indicative mood can be this closely identified. For there are many utterances of indicative sentences that are not assertions, for example indicative sentences uttered in play, pretense, joke, and fiction; and of course assertions may be made by uttering sentences in other moods. (Utterances

grammatical mood of the original Chinese sentence. But in order to emphasize that the force of the utterance is actually a rhetorical question, a better translation might be: “To learn something and then to put it into practice at the right time: isn’t this a joy?” 18 See Dummett, Michael (1973), Frege: Philosophy of Language, London: Duchworth, pp. 315, 316.

reading the

ANALECTS

with davidson

259

of ‘Did you notice that Joan is wearing her purple hat again?’ or ‘Notice that Joan is wearing her purple hat again’ may on occasion simply be assertions that Joan is wearing her purple hat again.) And similarly for other moods; we can ask a question with an imperative or indicative (‘Tell me who won the third race’, ‘I’d like to know your telephone number’), or issue a command with an indicative (‘In this house we remove our shoes before entering’).19

There is a passage from the Analects that is similar to Davidson’s last example (I use Simon Leys’s translation here): 13.18 The Governor of She declared to Confucius: “Among my people, there is a man of unbending integrity: When his father stole a sheep, he denounced him.” Confucius said: “Among my people, men of integrity do things differently: a father covers up for his son, a son covers up for his father—and there is integrity in what they do.”

This is another counterexample to the mood-force correlation thesis, because the indicative sentences here are used to issue a normative instruction, just as in Davidson’s example, ‘In this house we remove our shoes before entering’. Davidson is aware that it is not enough to refute Dummett’s moodforce correlation thesis simply by giving a list of counter-examples; he has to respond to Dummett’s conventionalist version of the thesis, which is supposedly capable of dismissing these counterexamples. I now turn to Davidson’s arguments against Dummett’s conventionalist theory of force. In “Mood and Performance”, Davidson mentions that Dummett can explain away the counterexamples by saying that they are all deviant, abnormal or non-serious cases. Dummett claims that it is “normal”, “natural” or “serious” that indicative sentences are always used to make assertions, imperative sentences are always used to issue commands, and interrogative sentences are always used to ask questions. Davidson argues that Dummett’s solution doesn’t work: It is easy to see that an appeal to what is ‘serious’ or ‘normal’ does not go beyond an appeal to intuition. It is no clue to the seriousness of a command that it is uttered in the imperative rather than the indicative; similarly, a serious question may be posed in the imperative rather than the interrogative mood. And if ‘normal’ means usual, or statistically more frequent, it is dubious indeed that most indicatives

19

Davidson, 1984a, p. 110.

260

chapter ten are uttered as assertions. There are too many stories, rote repetitions, illustrations, suppositions, parodies, charades, chants, and conspicuously unmeant compliments. And in any case the analysis of mood cannot plausibly rest on the results of this sort of statistical survey.20

According to Davidson, a way for Dummett to rescue his thesis is to drop the concepts of normal or serious cases, and switch to a conventionalist version of the thesis. Instead of saying that an assertion is an indicative sentence uttered in the normal case, Dummett could say that an assertion is an indicative sentence uttered under conditions specified by convention. Dummett has a specific proposal about the convention for assertions, which is that “assertion consists in the (deliberate) utterance of a sentence which, by its form and context, is recognized as being used according to a certain general convention.”21 But for Davidson, this is just the definition of assertion, not the convention of assertion: This [proposal of Dummett’s] also seems to me to be wrong, though in a somewhat different way. What is understood is that the speaker, if he has asserted something, has represented himself as believing it— as uttering a sentence he believes true, then. But this is not a convention, it is merely part of the analysis of what assertion is. To assert is, among other things, to represent oneself as believing what one asserts.22

Therefore, for Davidson, the real issue is: Can there be a convention that can always tell us whether a speaker believes in what she utters? To this question, Davidson’s answer is no. His argument goes like this. Let us assume, for the sake of argument, that we do have an assertion-indicator such as Frege’s assertion sign. That is to say, we have a sign that is not just the formal equivalent of the indicative mood, but also a conventional sign of the force of assertion. Suppose that we always use this strengthened mood whenever we make an assertion. Davidson then argues,

20

Davidson, 1984a, p. 111. Dummett, 1973, p. 311. 22 Davidson, Donald (1984b), “Communication and Convention”, in Inquiries into Truth and Interpretation, Oxford: Oxford University Press, p. 270. For Dummett’s rebuttal, please see Dummett, Michael (1993), “Mood, Force, and Convention”, in The Seas of Language, New York: Oxford University Press, pp. 203–23. But I cannot deal with Dummett’s rebuttal here. 21

reading the

ANALECTS

with davidson

261

It is easy to see that merely speaking sentence in the strengthened mood cannot be counted on to result in an assertion; every joker, storyteller, and actor will immediately take advantage of the strengthened mood to simulate assertion. There is no point, then, in the strengthened mood; the available indicative does as well as language can do in the service of assertion. But since the indicative is not so strong that its mere employment constitutes assertion, what must be added to produce assertion cannot be merely a matter of linguistic convention.23

In another essay, “Communication and Convention”, Davidson formulates the argument as follows: It is clear that there cannot be a conventional sign that shows that one is saying what one believes; for every liar would use it. Convention cannot connect what may always be secret—the intention to say what is true—with what must be public—making an assertion. There is no convention of sincerity. If literal meaning is conventional, then the difference in the grammatical moods—declarative, imperative, interrogative, opative—is conventional. These differences are in the open and intended to be recognized; syntax alone usually does the job. What this shows is that grammatical mood and illocutionary force, no matter how closely related, cannot be related simply by convention.24

Note that Davidson’s conclusion that there cannot be conventional indicators for assertion applies to all languages or linguistic practices. It is not a unique feature of the Chinese language that it does not have assertion-indicators or force-indicators, because no language does.

23

Davidson, 1984a, p. 113. Davidson, 1984b, p. 270. Here we should be very careful not to take Davidson as saying that the illocutionary force is a purely private, interior, and mental act. Elsewhere he does make it clear that this is not what he means: “The argument [for the autonomy of linguistic meaning] has a simple form: mood is not a conventional sign of assertion or command because nothing is, or could be, a conventional sign of assertion or command. The reason for this, it should be stressed, is not that the illocutionary force of a speech act is a purely mental, interior, or intentional aspect of the act” (1984a, p. 114; emphasis added). Nevertheless, Davidson’s point certainly has to do with the fact that speech act has a mental, interior, or intentional aspect. Right after the passage cited above, Davidson adds, “Of course assertion or command must be intentional, as must meaning in the narrow sense. But it is part of the intention that the act should be interpreted as assertive or commanding, and therefore part of the intention that something publicly apparent should invite the appropriate interpretation” (p. 114). 24

262

chapter ten 4.

Davidson’s point that there is no linguistic convention of sincerity (or insincerity) can be illustrated through a very interesting example from the Analects: 17.4 The Master went to Wucheng [where Ziyou was the governor],25 where he heard the sound of string instruments and hymns. He was amused and said with a smile (wan er er xiao ): “Why use an ox-cleaver to kill a chicken?” Ziyou replied: “Master, in the past I have heard you say: ‘Gentlemen who cultivate the Way love people; ordinary people who cultivate the Way are easy to govern.’” The Master said: “My friends! What Ziyou said is true. My earlier remark was a joke.”

Let me list Confucius’s two utterances, as well as the statement he makes in the past, as follows: (1) Why use an ox-cleaver to kill a chicken? (2) Gentlemen who cultivate the Way love people; ordinary people who cultivate the Way are easy to govern. (3) My friends! What Ziyou said is true. My earlier remark was a joke.

There can be a variety of interpretations of the forces of Confucius’s utterances. Let us start with (1). Many readers would agree with Ziyou’s taking (1) as an assertion, and understand (1) as saying that Ziyou should not have bothered to cultivate ordinary people with music and rituals. This is analogous to killing a chicken with an oxcleaver; Confucius wants to say that Ziyou only needs to cultivate the gentlemen. But this is inconsistent with Confucius’s own teaching, which Ziyou cites in his response to the Master. Can there be any conventions to determine that (1) must be an assertion? As we can see in the passage, the editors of the Analects actually report that Confucius says it ‘with a smile (wan er er xiao)’. Does this indicate that it is not an assertion, but an ironic remark? Can we then say that it is a ‘convention of insincerity’ that a speaker is making an ironic remark rather than an assertion when the speaker utters the sentence with a smile? It is indeed the case that people do sometimes tell a joke with a smile, and yet they can also make an assertion with a smile, and tell a joke with a straight face. One 25 In 6.14, it is mentioned that Ziyou is the governor of Wucheng. I use Simon Leys’s translation here.

reading the

ANALECTS

with davidson

263

may claim that the convention of insincerity should include more factors in order to be a real convention; for example, it is possible that Ziyou does not detect the force of (1) because he doesn’t pay attention to the tone of Confucius’s utterance. Davidson does not deny that these factors (such as tone and gesture) play key roles in the making and detecting of assertions, but he argues that they can’t conclusively determine whether the speaker is sincere: It must also be conceded that interpreters and speakers of a language are generally able to tell when an assertion has been made, and that this ability is an essential part of their linguistic competence. Furthermore, knowledge of linguistic and other conventions plays a key role in the making and detecting of assertions. Costume, stance, tone, office, role, and gesture have, or may have, conventional aspects, and all these elements can make a crucial contribution to the force of an utterance. We may easily allow all this without agreeing that merely by following a convention, indicative or imperative utterances become assertions or commands.26

Therefore, our conclusion has to be that there is no convention telling us that Confucius’s utterance (1) must be an assertion or that it must be a joke. That is to say, we cannot say that Ziyou must have got it wrong in his reaction to (1) when he takes it as an assertion, and is puzzled by the fact that it is not consistent with (2). In fact, it is an entirely plausible scenario that Ziyou has got the force of (1) right. That is to say, Confucius originally does make an assertion when he utters (1). It is only after Ziyou points out that (1) is not consistent with (2), which is Confucius’s own belief in the importance of using rituals and music to cultivate ordinary people, that Confucius gives a retrospective articulation of the force of the remark, claiming that (1) is actually intended as an ironic remark and should have not been taken seriously in the first place. Since a joke cannot be inconsistent with an assertion, Confucius can thus explain away the seemingly contradiction between (1) and (2). Now let us turn to the interpretations of Confucius’s utterance (3), which is a remark about the force of (1). There are two possible readings of the force of (3). The first is to assume that it is an assertion. The second is to assume that it is still an ironic remark.

26

Davidson, 1984a, pp. 112–3.

264

chapter ten

If we take (3) ‘My earlier remark was a joke’ as an assertion, we would have to agree that Confucius’s earlier remark, (1), was indeed a joke. How do we make sense of this? Li Zhi (1521–1602), a Ming dynasty scholar, has suggested that Confucius is extremely pleased about what he has seen in Wucheng, where Ziyou is the governor, and that is why he intentionally makes a ‘fan yu ’ (ironic remark).27 In other words, Li Zhi here is alluding to an interesting phenomenon, which is that when people are extremely pleased they often feel that they have to use irony to express it. But Ziyou fails to understand Confucius’s intention, and, as Li Zhi puts it, “gets very serious about it.” Confucius then has to get serious as well, making an assertion regarding the force of (1) to clarify his intention.28 Can we determine conclusively that (3) must be an assertion? As Dummett would suggest, we can get clues from the manner in which (3) is uttered, or from certain linguistic conventions such as the grammatical indicators. The editors report that the Master speaks with a smile when he utters (1), but they do not say anything about the manner in which the Master utters (3). It’s simply ‘The Master said’. However, as we have argued earlier, this kind of description does not necessarily mean that Confucius is making an assertion. Now let us look at the grammatical indicators, in this case, the ending particles. In the original Chinese version of (3), we find two sentences with ending particles: there is the particle ‘ye ’ at the end of the sentence ‘What Ziyou said is true’, and there is the particle (‘er ’) at the end of the sentence ‘My earlier remark was a joke’. Traditional Chinese scholars agree that ‘ye’ and ‘er’ are two typical indicative particles ( jue-ci ).29 But this does not mean that these two utterances in (3) must be assertions. As Davidson has argued, we cannot say that the indicative sentences are always used

27 Li, Zhi (1975), Si-Shu-Ping (Comments on the Four Books), Shanghai: Shanghai renmin chubanshe, p. 146. 28 Li, 1975, p. 146. 29 For example, we find the following observation from the preface to a Yuan Dynasty monograph on particles: “‘hu ’, ‘yu ’, ‘ye ’, ‘zai ’, ‘fu ’, these are inquisitive particles (yi-ci ); ‘yi ’, ‘er ’, ‘yan ’, ‘ye ’, these are assertive particles ( jue-ci )” (Lu, Yiwei [1988], Zhu-Yu-Ci-Ji-Zhu [Collected Comments on the Particles], edited and commented by Wang Kezhong, Beijing: Zhonghua shuju, p. 183). Scholars believe that Lu Yiwei was the first to write a book-length study of the particles. We do not know much about the author except that the book was written no later than 1324.

reading the

ANALECTS

with davidson

265

to make assertions in ‘normal’ and ‘serious’ situations. Davidson’s point becomes especially obvious in our case: There is absolutely no use to appeal to the notion of the ‘serious’ situations, because whether this very situation is a serious one is exactly what we are trying to determine here.30 Would the appeal to convention help? Dummett suggests that an assertion is an indicative sentence uttered under conditions specified by linguistic conventions. Our case here shows that Davidson is right to claim that Dummett’s suggestion won’t work. If there is any convention that can help us decide whether (3) is an assertion or a joke, it would have to be the non-linguistic ones, such as the conventional image of Confucius as always being deadly serious, or the conventional wisdom that the Analects is a collection of Confucius’s sincere moral instructions and commands. Christopher Harbsmeier has shown convincingly that, in the Analects, we can often find Confucius teasing his students, playing with words, and amusing people by saying things jokingly. Contrary to the conventional image, Confucius is actually “an impulsive, emotional, and informal man, a man with wit and humor, a man capable of subtle irony with an acute sensibility for subtle nuances”.31 In his comments on 17.4, Harbsmeier suggests that Confucius is probably still joking when he makes his last remark: “My earlier remark was a joke.” In other words, Confucius might have been joking all the way through.32

30

I have argued elsewhere that traditional Chinese scholars also believe that the syntactic features of a sentence do not determine the pragmatic uses of the sentence. Here is one example. In his 1687 commentary on Lu Yiwei’s book on particles, the Qing Dynasty scholar Chen Lei says, “[The ending particle] ‘zai’ indicates interjection, or interrogation, or interruption, or assertion, or just the completion of the utterance. We should always look at the total context of the speech and text; we cannot just focus on the word ‘zai’” (Lu, 1988, p. 17). Other Qing scholars who have written on particles, such as Yuan Renlin, Wang Yinzhi, and Liu Qi, have made similar observations. It can be argued that Chinese scholars do not necessarily see these particles as just grammatical mood indicators. In fact, when the Chinese scholars write about particles, they discuss not only their syntactic functions, but also (and even more often) their functions and uses in composition, style, persuasion, rhetoric, and argumentation. 31 Harbsmeier, Christoph (1990), “Confucius Ridens: Humor in the Analects”, Harvard Journal of Asiatic Studies 50, 1990, p. 131. I am grateful to David Keightley for having urged me to read this article. 32 One way to make sense of this possibility is to think about Harold Pinter’s plays. Besides the Analects, Pinter’s plays can be read as another massive set of examples of how conventionally simple utterances, such as ‘I don’t know’, ‘You are right’, ‘Yes’, ‘Well’, or non-utterances (pause, silence) are able to do a wide range of things

266

chapter ten

Relying on Davidson’s arguments, we can see how it is possible to make different judgments about whether certain utterances in the Analects are assertions or jokes.33 Obviously, whether an utterance is a joke or an assertion has great implications when we interpret a text. I believe what Bernard Williams has to say about how to interpret Plato’s Theaetetus applies to the Analects as well: “If we are going to get the most from reading one of Plato’s dialogues, we have to keep in close touch with its tone, sustaining a sense of what is a joke, what is merely provisional, what is being tired out or tried on.”34

5. Let me summarize my arguments in this paper. If we take a pragmatic perspective, we will realize that it is not enough to know the grammatical mood of the sentences, for the mood is the feature of a sentence that remains the same, regardless of the different situations in which it is uttered, whereas the force is the feature of the utterance that varies from situation to situation. To put the point in a nutshell, there is no strict correlation between grammatical moodindicators of a sentence and the forces of the utterances of the sentence, because the speaker can always intend to use the sentence to do things that are not determined by its grammatical or conventional features. That is to say, we must take into account the “total speech situation” in which the speaker makes the utterance.35 In other words, the grammatical or conventional features of linguistic expressions do not determine how they can be used pragmatically, and this applies to all languages. Davidson takes this general in our daily, domestic life. The force of a sentence in a script, as every good director or actor knows, is not determined by the literal meanings of the sentence or any other linguistic conventions, and hence can always be interpreted (and delivered) differently. Directors and actors thus can always have a new interpretation of a play in a new production. 33 Elsewhere I have shown that generations of commentators in China have made different judgments regarding the forces of the utterances in the Analects; please see Xiao, Yang (2006), “The Pragmatic Turn: Articulating Communicative Practice in Early China”, Oriens Extremus. 34 Williams, Bernard (1999), Plato, New York: Routledge, p. viii. 35 The term “total speech situation” is from Austin, J.L. (1975), How to Do Things with Words, Cambridge: Harvard University Press, Second Edition, p. 52 and p. 148. He emphasizes that “[i]t is important to take the speech-situation as a whole” (p. 138).

reading the

ANALECTS

with davidson

267

point as a basic trait of human language and calls it the autonomy of linguistic meaning: What this argument illustrates is a basic trait of language, what may be called the autonomy of linguistic meaning. Once a feature of language has been given conventional expression, it can be used to serve many extra-linguistic ends; symbolic representation necessarily breaks any close tie with extra-linguistic purpose. Applied to the present case, this means that there cannot be a form of speech which, solely by dint of its conventional meaning, can be used only for a given purpose, such as making an assertion or asking a question.36

In another place, Davidson states the thesis of the autonomy of linguistic meaning as follows: “Once a sentence is understood, an utterance of it may be used to serve almost any extra-linguistic purpose. An instrument that could be put to only one use would lack autonomy of meaning; this amounts to saying it should not be counted as a language.”37 People were quite shocked when they first heard Davidson declaring at the end of his 1986 essay “A Nice Derangement of Epitaphs” that “there is no such thing as a language”. This sentence, taken out of its context, does sound shocking. But let us cite the whole passage here: I conclude that there is no such thing as a language, not if a language is anything like what many philosophers and linguists have supposed. There is therefore no such thing to be learned, mastered, or born with. We must give up the idea of a clearly defined shared structure which language-users acquire and then apply to cases. And we should try again to say how convention in any important sense is involved in language; or as I think, we should give up the attempt to illuminate how we communicate by appeal to conventions.38

There is nothing one should feel shocked about if one is familiar with Davidson’s thesis of the autonomy of meaning, first defended in his 1976 essay “Mood and Performance”. One should see clearly

36

Davidson, 1984a, pp. 113–4. Please also see 1984b, p. 274. Davidson, Donald (1984c), “Thought and Talk”, in Inquiries into Truth and Interpretation, Oxford: Oxford University Press, p. 164. 38 Davidson, Donald. (1986), “A Nice Derangement of Epitaphs”, in Philosophical Grounds of Rationality: Intentions, Categories, Ends, edited by Richard Grandy and Richard Warner, Oxford: Clarendon, p. 174. This paper, with comments by Ian Hacking and Michael Dummett, is also included in Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, edited by Ernest LePore, Oxford: Blackwell, 1986. 37

268

chapter ten

that Davidson is simply saying that we should not focus on words and sentences, or the conventions and rules that are supposed to govern them; instead, we should look at people’s communicative practice, and how they actually do things with words and sentences. And no convention can capture our communicative practice. Since classical Chinese can be easily perceived as lacking a “clearly defined shared structure which language-users acquire and then apply to cases”, reading the Analects together with Davidson should make— and I hope, has made—it much easier for us to see that “we should give up the attempt to illuminate how we communicate by appeal to conventions.”

PART FIVE

TRUTH CONCERN AND DAO CONCERN

CHAPTER ELEVEN

FROM DONALD DAVIDSON’S USE OF “CONVENTION T” TO MEANING AND TRUTH IN CHINESE LANGUAGE Chung-ying Cheng

Introductory Remarks In this article I shall advance in the first part the Donald Davidson’s theory of truth interpretation in terms of the Convention T as originally established by Tarski in his formalization of semantics for finite languages.1 Davidson has expanded the Tarski’s formal-semantic notion of truth interpretation and applies to natural languages in order to show how the meaning of a sentence could be given in terms of truth conditions of the sentence. This is an ingenious move. But we wish to ask whether the Davidsonian approach to interpretation of truth in terms of the Tarskian Convention T is sufficient to explain meaning and truth in a living natural language like the Chinese and whether it is itself should be further broadened for both semantic and hermeneutical representation of our understanding reality based on experience. In other words, we shall ask whether an ontological re-interpretation and application of Convention T could serve an important purpose of ontological and cross-ontological understanding that I call onto-hermeneutical understanding.2 1 See Donald Davidson’s book (1984), Inquiries into Truth and Interpretation, Oxford: Oxford University Press (1991 printing), p. 85. Specifically the Essay 4 is titled “Semantics for Natural Languages”, Essay 1, “Theories of Meaning and Learnable Languages” and Essay 5 “In Defense of Convention T” are also directly relevant. A. Tarski proposes the idea of Convention T in his article “The Concept of Truth in Formalized Languages”, published in Polish in 1933, in German in 1936, and in English in 1956 (in Logic, Semantics, Metamathematics, Oxford: Clarendon Press). 2 See my explanation of the notion of onto-hermeneutical interpretation in my article “Inquiring into the Primal Model: Yijing and the Onto-Hermeneutical Tradition”, in Journal of Chinese Philosophy, vol. 30, double issue 3–4, pp. 285–288. In a situation where both understanding and communication are important, one must interpret the situation in light of one’s experience of reality in the situation together with both one’s belief and knowledge of the situation. It is an interpretation that is both hermeneutical because it involves understanding of language, and ontological because it involves reference to a reality.

272

chapter eleven

In the second part of the article I shall advance five basic principles for understanding meaning and truth in Chinese language in view of Chinese philosophy as a theory of truth. To do this is to show how meaning and truth in Chinese language or more precisely how meaning and truth of sentences made in Chinese language could have a theory of truth that explain their meaning and truth in a framework of Chinese philosophical reflection formulated in the same language. Thus we may regard these five principles as a theory of truth of the language, namely a theory that gives the categories of truth conditions for whatever meaningful sentences can be made in the language. There is no doubt that I am inspired in reformulating the ontology of Chinese language in this manner by Donald Davidson in view of his work on the semantics for natural languages as mentioned above. Davidson of course was himself inspired by Tarski ‘s work on formal semantics that is developed for the purpose of clarifying the meaning structure of sentences in a formal language in terms of the truth conditions of the sentence to be given in the same language (as a meta-language). It is conceivable that the meta-language could be a different kind of language other than the object language, but the sentences of the object language must be assumed to be translatable into the meta-language in order to speak of the truth conditions of the later.3 In this sense the five principles of a theory of truth for Chinese language belong to the meta-language of Chinese language, which would direct how a meaningful sentence in the Chinese language should be translated. The novelty of my approach is that I make the assumption that I have come to a truth theory of those five principles by way of comprehensive observation and comprehensive reflection on the use of the language.

3 Tarski’s Convention T demands that each sentence of object has a translation in the meta-language. Tarski also proposes that natural languages are essentially inter-translatable. Davidson mentions this in his essay “In Defense of Convention T” (Ibid., p. 72). Davidson also argues for rejecting the idea of conceptual scheme in his article “On the Very Idea of a Conceptual Scheme”, Essay 13 in his book Inquiries into Truth and Interpretation. The insight here is that any two systems of language have to face reality as a common base for each system to make its interpretation, of not only the reality that is transcendentally presupposed by any system, but of each other in light of one’s interpretation of the reality. An alternative language therefore receives a putative translation and interpretation in a given language in light of the given language ‘s experience of reality.

from donald davidson’s use of “convention t”

273

An immediate question arises: How about a language that cannot be translated into the given meta-language? The reply is: One has to interpret the alien language in terms of what we know about our own language in light of whatever evidence or reason there is available. This is what Quine has suggested in his radical translation of an unknown word ‘Gavagai’ in a local language in view of the collateral information a linguist may gather in a situation of observation or confrontation.4 After explaining these principles, we may see how Chinese philosophy plays a part in explaining the meaning and references of Chinese language in general. For our purpose we just assume that Chinese philosophy could be said to play the role as a theory of truth for the Chinese language, because it could lay down the truth conditions for sentences said in the language. The truth conditions could be conveyed by the same sentence the name of which is used in the object language. It could be also conveyed by a different sentence in the meta-language that interprets the sentence in the object language due to one’s understanding. One must also bear in mind two matters in Chinese philosophy concerning things in the world. For one thing, the concept of truth is not treated independently of our concepts or talk of the real, because there is no intention to separate truth from reality that is in turn not separable from experience of natural things in the world. For another, the lack of a universal copula in the Chinese language to make reference to an abstract universal being can only be explained by the fact that Chinese philosophy begins with observing change and transformation of all things with reference to no essence of being,5 but instead with reference to only an ultimate source of creativity which is infinite and indeterminate and indeterminable.

4

See W.V. Quine (1960), Word and Object, chapter 1, Cambridge: MIT Press. There is no reason why a universal essence of being must be assumed in the beginning. The use of a universal copula in Ancient Western languages such as the Greek may be a result of convenience in objectifying reality as fixed and unchanging so that reference to things in such a reality can be made. 5

274

chapter eleven Part 1 Davidson’s Use of Tarkian “Convention T”: A Critique

Donald Davidson (1917–2003) has followed the general trend in analytical philosophy and contemporary philosophy of language as shaped by W.V. Quine (1908–1998) in having developed a theory of truth and meaning as shown in his books Essays on Action and Events (Oxford: Clarendon Press, 1980) which provides an logical analysis of adverbs in terms of logic of action and events of which can be extensionally identified, and Inquiries into Truth and Interpretation (Oxford: Clarendon Press, 1984, 1991 printing) which is a collection of 18 essays dealing with theories of meaning and truth for natural languages. The essential contribution of Davidson lies in developing a model of semantics for the truth and meaning for natural languages in light of the formal theory of semantics as developed by Tarski and adapted by Quine. This basic model of semantics for truth and meaning would aim at having a theory for specifying the meaning of any sentence in a language from its componential structures defined in primitive terms and a set of finite rules with finite number of axioms. It may be asked: Why semantic theory? The answer is that we want to specify the meaning and reference of a sentence as its truth conditions in a semantic theory. It can be further asked: Why finite number of axioms? The answer is that we want to make the language learnable from its rules and primitive terms, learnable in the sense that we are able to specify the meaning and reference of the sentence from the theory of the semantics. This theory is required or assumed to be extensional and referential. For this purpose, Davidson uses the Convention T of Tarski to make a specification of the meaning of sentence p, namely the convention that articulates the conditions that a predicate has to satisfy if that predicate is to count as a truth predicate. Convention T so understood would generate a T-sentence for each truth-claiming sentence in a given language of the following form: P is true in L if and only if p

where P is a quotation of p in the form of ‘p’ or a standardized structural description of the sentence p mentioned (named) in the left side of the T-sentence and used in the right side of the T-sentence in the given language. To be more precise, we may see P as the name of p which can be spelt out in a structural description on the

from donald davidson’s use of “convention t”

275

left side in the quotation and which can be used to refer to a reality with which it corresponds. This formulation reflects the constraining condition on the material adequacy for a formally correct definition of the truth predicate ‘true in L’ which entails the T-sentence for each truth-claiming sentence in L.6 Sentences of this form, namely what are called the T-sentences are true and for that matter trivially true. For this reason T-sentences does not make any reference to a world or a fact,7 but this does not mean that we cannot talk about objects, properties and relations. This is because these sentences enable us to specify the meaning of sentence from a truth-functional meta-language, which would describe the world in terms of extensionality and objectivity in objects, properties and relations. These sentences that specify the truth conditions for sentences in a language therefore constitute the theory of truth for that language. Davidson states a theory of truth for a language L (namely the theory satisfying Tarski’s Convention T) as “a sentence T containing a predicate t such that T has as logical consequences all sentences of the form ‘s is true if and only if p’ with ‘s’ replaced by a canonical description of a sentence of L, ‘p’ replaced by that sentence (or its translation), and “is true” replaced, if necessary, by t.”8 Alternatively, he says: “By a theory of truth I mean a set of axioms that entail, for every sentence in the language, a statement of the conditions under which it is true.”9 To be as brief as possible, Convention T just specifies a condition for a given language on how a truth predicate such as ‘is true’ is to be applied to a sentence s in that language. This condition is to be expressed by the T-sentence of the form ‘s is true if and only if p’. It seems clear to me that Convention T captures not only our habit of using a truth predicate but our intuition on what truth is about. Most of all, it also captures our sense of reality and our sense of a reciprocal and mutual definability of meaning and truth in language and meaning and truth in reality. But Convention T is no definition of truth, but a powerful observation 6 See Tarski (1944), “The Semantic Conception of Truth and the Foundations of Semantics”, from Philosophy and Phenomenological Research, volume 4, pp. 341–375. Specifically section 4 in the essay. 7 See his Essay 13, “On the Very Idea of a Conceptual Scheme”, in op. cit., p. 194; and his Essay 15, “Reality without Reference”, in op. cit., p. 220ff. 8 See his Essay 5, “In Defense of Convention T”, op. cit., p. 66. 9 See his Essay 4, “Semantics for Natural Languages”, op. cit., p. 56.

276

chapter eleven

on our intuitions of truth about truth. It can also serve as a normative requirement for formulating semantics of meaning and truth as it was formally for Tarski. For myself it serves as a normative requirement for linking our use of language to our actual experience of reality and vice versa. Davidson’s contribution to the use of the Convention T consists in seeing truth (he uses the capital T) as a universal predicate of our natural language which requires us to reconstruct or re-describe our sentences in our natural language in standard first order predicate logic to meet the implicit demand of the logic for extensionality. This is of course a tenet from W.V. Quine. He appears to have concentrated on the language side of the Convention T whereas I wish to take the reality side very seriously and argue for a natural determination of our sentence in light of experience of reality. Of course, it is not to say that Davidson does not pay attention to experience of reality. He does so in the spirit of Quine by rejecting conceptual relativism and the dualism between scheme and content which gives rise to conceptual relativism. But this is not yet to focus on our experience of reality as manifested in many forms of our language as many forms of life. Davidson has this to say: “One thing that only gradually dawned on me was that while Tarski intended to analyze the concept of truth by appealing (in Convention T) to the concept of meaning (in the guise of sameness of meaning, or translation), I have the reverse in mind. I considered truth to be the central primitive concept, and hoped, by detailing truth’s structure, to get at meaning.”10 For me, what makes truth is reality and I take reality that is presupposed in the mention of p in the Convention T to be a primitive notion and a primitive experience and hope that by interpreting p to get the truth of ‘p’ as described in a semantic theory satisfying the Convention T. I see that the semantic notion of truth and its application as a predicate arise as a rationalistic move of understanding and rational assessment of experience. This semantic notion of truth must presuppose our intuitive sense of truth that gives rise to interpretation in terms of belief and knowledge that are experiential in character. Given this basic characterization of the Davidsonian project, I wish to make the following observations on how this project could

10

See Introduction to his book Inquiries into Truth and Interpretation, p. xiv.

from donald davidson’s use of “convention t”

277

relate to the Chinese language as a language which is rich with ontology and cosmology and humanological experiences of ontology and cosmology to be described in the five principles in Part 2 of the article.11 I shall raise the question whether the Davidsonian project provides a way for us to understanding the meaning of Chinese language and the reference of Chinese philosophy or for that matter, the concept of truth in Chinese language as prescribed by some theory of truth as provided in the Chinese philosophy. First, the present idea of a semantic theory is geared toward extensional analysis, nominalistic reduction and effective recursive computability. It comes from a logical positivistic tradition of philosophy of science in which ethics is regarded as emotive and metaphysics is nonsensical. In order for such semantic theory to conform to these requirements it is necessary to have a paradigm of ontological commitment in logic and physical science where logic is understood as the standard logic of predicates with identity and physical science is understood as the new physics with relativity and quantum mechanics. Thus a semantic theory is one with finite number of semantic primitives (corresponding to the primitive predicates given in a language) with finite rules of deduction or application which operate on wellformed formulae (grammatically corrected sentences) to yield meaningful sentences with interpretation from the finite number of axioms which specify the initial postulates of meaning and truth or ontology of the language. The Tarskian Convention T will provide the truth condition of a sentence in terms of, or in light of, the understanding of the ontology from the given axioms of the language. The model obviously is very restricted but it has the elegance of logical simplicity and analyzability. The question is whether this is the only possible theory and whether we could have a theory, which allows an open set of axioms and an open set of semantic primitives. Of course we may wish to point to the fact that we learn language from a simple beginning in ostentation, perception and imitation. But then does that mean that our

11 I distinguish the philosophical study of human beings in humanology from anthropology in that the humanology deals with human beings in all their aspects, ontological, cosmological, moral and political, and hence is different from anthropology which treats human beings and human groups from merely a social scientific point of view which lacks a dimension of reflective understanding of human morality and human spirituality which are ontologically and cosmologically rooted.

278

chapter eleven

learning process of a language can be captured or mapped into a logically structured semantic theory as described here? It seems clear that our learning of a language need not to be and is actually not a matter of logical deduction or application. The learning process is full of various kinds of moves of the human mind, conjecturing, deduction, induction, imposition, regulation, and creation or invention of being, such as actually described by Quine in his Word and Object (Cambridge: MIT Press, 1960) and In Pursuit of Truth (Cambridge: Harvard University Press, 1992). We may consider Peirce’s notion of abductive inference or hypothesis making as essential to any language learning process. In this sense the semantic theory Davidson is talking about is only a formalization when the language is fully settled and formed a closure. It would not deal with a living natural language, which is evolving and developing. Even for a language that is closed or confined to published texts we still cannot rely on the semantic theory to produce meaning and truth for a sentence. Because meaning is matter of insight into an underlying ontology which is often not formulated or unformulated for one reason or another, whereas truth is matter of disclosing new reality or presenting a new perspective on reality. Both of these are not to be identified or reduced to an empiricist or phenomenalist epistemology as traditionally described. The important point is that given science as built up from a logic of discovery and abduction, why must semantic theory as proposed by Davidson assume the form of deductive application for generation of meaning from a logical atomic structure and an rudimentary empirical science? It is no wonder that such a semantic theory has not yielded any result in analyzing any major natural language except in regard to a small segment of the language as isolated or insulated from other parts of the language and also without regard to the holistic context of the speaker as a human person. It is in this regard that logicians such as Alonzo Church, Rudolf Carnap and others have proposed semantic theories, which attempt to do more than extensional deduction or application. Yet, unfortunately, these semantic theories also have finite specific-purpose uses only. Second, it is obvious that there is no such an idea of semantic theory in Chinese philosophy, but Chinese language as a living natural language no doubt cannot be conceived even theoretically as necessarily complying with the requirements of the semantic theory in order to be semantically analyzed. It is interesting to raise the

from donald davidson’s use of “convention t”

279

question as to whether Chinese language and the language of traditional philosophers have a finite semantic primitives, finite rules of deduction and finite umber of axioms. Given what is said about Chinese language as an open system that requires attention and experience of the speaker to make a meaningful sentence, it is not possible to limit the language to finite numbers of semantic primitives. In fact the semantic primitive would vary with invention of new terms and new concepts and hence the language would become nonlearnable. Perhaps at the point we should distinguish learnable from the non-learnable because the later requires a participation in experience and a creative insight or reflection on the part of the speaker or language user. Not only Chinese language is non-learnable in the sense of having many undefined primitives in metaphysics and ethics that require experience and reflection on experience, Chinese philosophy is also highly non-learnable because of its search for creative synthesis and reconstruction. One may of course point out to the finite number of basic concepts in the Chinese philosophy if one restricted to texts of a certain philosophy or even to all texts of all past philosophers. But still there are objections to such application of the model, because meaning need not to be fully specified by truth conditions as acknowledged by Davidson himself and because understanding a philosophical concept and grasping its meaning requires understanding the whole system which is not formulated explicitly. The philosophical wisdom often comes in the insight into a meaning, which is a gestalt experienced as reality. Third, it is interesting to note that for Chinese philosophy, many of the semantic primitives as we have enumerated under what we shall see under Five Principles, are not intentional concepts, which as we shall see in the following section are actually co-extensive in extension and hence co-applicable to the existing universe and things in the universe. For example, in speaking of the dao of the heaven, the dao of heaven is coextensive with the heaven and even the dao by itself is conceived as generatively related to this world and things and need not be conceived as an independent and transcendent entity. Hence we may ask the question whether a semantic theory of some finite number of semantic primitives including those of the dao can be built so that we come to know an extensional analysis of relevant sentences can be given in such a theory. But the difficulty is that dao and other terms like the dao are not explicitly and fully determined and defined in meaning. Hence the semantic application

280

chapter eleven

or deduction of meaning for a sentence containing the “dao” may not lead to any true understanding of the dao and the sentence involving a reference to the dao. Hence we must allow more relevant principles such as direct experience, experience of the process of development, and integration of total consciousness involved. The norm is that we should not cut our language to suit a formally targeted semantic theory in order to yield meaning: instead we should modify or revise the given semantic theory in order to catch our natural language or our philosophical thinking. Fourth, this is not to say that we cannot benefit from the learning of semantic theory in regard to our Chinese language and our Chinese philosophy. We learn that extensionality and objective reference are important criteria of meaning. We learn that sentential meaning must be structurally related to some fundamental semantic primitives and it is usually necessary that we learn what are our semantic primitives. Both Chinese language and Chinese philosophy need a format and a form that would lead to logical analysis and semantic understanding, even though our philosophical understanding would exceed such understanding. If Chinese language and Chinese philosophy are not subsumable to a semantic theory because of its openness of experience of reality and the richness of onto-cosmological or onto-ethical and onto-humanological reflections,12 we should know where and how such aspects of Chinese language and Chinese philosophy would lie and how important and significant they are. It is important and significant because they are needed for comparison with the scientific image of the human person and for developing a full picture of the human person in which science still should occupy a relevant place. Fifth, the most important point suggested by Davidson in his theory of meaning and truth is the use of Convention T. From a Chinese philosophical point of view, the use of Convention T not only provides a truth condition for a sentence in an objective theory, it also provides a way of interpreting or reinterpreting a given sentence or

12 I introduced these new terms in order to focus on the holistic and ontological aspects of the subject of human beings. For onto-ethics I stress the importance of understanding the reference to reality for decisions to be made for achievement of actions. For humanology, I intend to establish a view and a study of the human person or human beings which is not confined to any discipline but which would synthesize all legitimate approaches to human beings into a philosophical theory or a metaphysical doctrine.

from donald davidson’s use of “convention t”

281

even a set of sentence (as a theory) in a suitable priorly or “transcendentally” presupposed implicit ontology of the language in which the theory is formulated. As a matter of fact, the meta-language in terms of which the truth condition of a sentence is to be given could contain the theory. This is like to treat English (with all its common semantics in a given period) as a theory into which a sentence from another language could be translated by the English language and into the English. Furthermore, it is quite possible that one could redescribe or re-interpret the given sentence into a theory that is formulated in English. This is possible because the truth of a sentence often requires a projection and insight into what the sentence is about ontologically. The basic meaning of truth here comes from a type of trustworthiness as to what is said about what is regarded as reality. Hence to inquire into the truth of a sentence or a theory is to inquire into what a sentence or theory presents as reality which should be trusted. In this sense, we can see truth as presented in language and language is the symbolic vehicle for truth that needs to be communicated to those reading the linguistic symbols. Once the linguistic symbol (the sentence of a theory) with its modality and the semantic roles it plays in the theory communicates what there is truth condition of the given sentence, one can represent or reconstruct the concept of truth in other linguistic symbols which can be said to interpret the given symbol for the hearer or interpreter. Hence to construe the truth of a sentence is to interpret the sentence regarding its ontological content from the point of view of the same language user or another language user. In the latter case we have translation whereas in the former case we have interpretation. Hence we may say that to construe a sentence according to the requirement of Convention T is to perform an onto-hermeneutical interpretation, which consists of understanding the underlying reality and re-presenting the understood ontology or understanding of the ontology in new terms of the language. This onto-hermeneutical interpretation of the use of Convention T has the following steps: 1. A sentence p which is named P is given in L (similarly for sentence q named Q ) 2. P is true in L-A if and only if Q is true (with q in the metalanguage, presenting or conveying an ontology O of the object language) 3. Re-describe or interpret P as Q in light of understanding O -> p, O -> q 4. p if and only if q

282

chapter eleven

It cannot be denied that in order to reach Q from P the interpreter needs to have an insight into an ontology and this ontology could be given or not yet given but needs to be formulated. Interpretation is interpretation in the language of a given ontology that is embodied in the language of the theory or is provided by the understanding of the interpreter. It is possible that understanding P may inspire the interpreter to find a new ontology or it may enable the interpreter to relate to a commonly assumed reality of daily life as suggested by the Yijing and Classcial Confucian philosophers. With this interpretation, the T-sentences become a mechanism or rule for prompting and encouraging ingenious insight into what reality is in order to reach a truth-condition fort the given proposition or sentence. In other words, I see in the use and appeal to T-sentences an introduction or a turn to my onto-hermeneutics for understanding. In the case of translation, what is in risk is the meeting of the two systems of understanding which exhibit in two different languages. Again it is clear from Quine and Davidson that we should construe a native sentence with our language to translate into. But this presupposes our own understanding our language and construal of a native sentence must be guided by evidence from observation and theory from our own tradition or philosophy. Here we see the Tsentence according to the Convention T can be used cross-linguistically from a meta-linguistic point of view, namely a view of a metalanguage in which both object-languages are presumably translated. For we could have the following form of equivalence: P is true in L-A at time t if and only if Q is true in L-B at time t

Which can be said to derived from the following truth equivalences: P is true in L-A at t if and only if p at t Q is true in L-B at t if and only if q at t p at t if and only if q at t

One can see that interpretations of P and Q could come from two different theories, but they could come from the same theory as two different ways of expressing the same truth. What is significant is that we have to understand P and Q relative to the same or different theories and warrant their equivalence in truth values if want the two sentences “P” and “Q” to be mutual translations of each other. It is necessary that we must treat our own language as ontological or forming an ontological meta-language in order to render the native

from donald davidson’s use of “convention t”

283

sentence from its origin into our language. Hence, P and Q need not to be described literally as P and Q , but instead, they could be described in sentences in the meta-language in which the theory is formulated. Will this be a fair representation? The reply is that all depends on how evidential is the native sentence in a given situation. In this connection, Davidson has this to say: “The present enterprise is served by showing how the theory can be supported by relating T-sentences, and nothing else, to the evidence. What is clear is that the evidence, whatever it is, cannot be described in terms that relate it to any particular language, and this suggests that the concept of truth to which we appeal has a generality that the theory cannot hope to explain.”13 It is important for Davidson to point out that the concept of truth has a generality which a the theory cannot explain, because it is the generality of the concept of truth which allow that no particular language will evince the evidence: the assumption is that the evidence must be experienced and to be expressed in any language in which the theory is formulated. The evidence to the theory relation is one of logical and normative support to be derived from our understanding and experience of a situation. It is assumed by the Song Neo-Confucian Lu Xiangshan that all men have the same mind and all minds share the same reason or principle. There is of course no objective proof or guarantee of the truth of this motto, but what is first required for the given sentence is that it is believed and found true in one’s language with regard to your own observant sense experience. The evidence is your own evidence to reach an interpretation that you can call true. Perhaps until one has commanded both patterns and principles of both languages, the onto-hermeneutical interpretation will still have to yield to a truth-functional description, which may not have to pin down the exact reference of the terms in the original sentence in a language. This is precisely the point that Davidson wishes to make when he suggests his T-sentence form of interpretation. Sixth, Davidson is quite aware of the difficulties of giving suitable referential reading to quotations, indirect discourse and belief sentences or other propositional attitudes. These and other matters require close attention for logical clarification. It is clear that both

13 Quoted from his essay “Reality Without Reference”, Essay 15, in his book Inquiries into Truth and Interpretation, ibid., p. 223.

284

chapter eleven

Chinese philosophy and Chinese language are full of belief sentences and quotations and indirect discourses. In this regard, we should not take a restricted approach, but instead let the semantic theory flows with the dao so that a mutual development will be helpful and beneficial to the theory and the language. Seventh, there is the issue of eliminability of singular terms in traditional languages. Quine has shown how such a project could be managed and reduction or eliminated as claimed. I have argued that in Chinese language and in Chinese philosophy singular term is often present without linguistically recognized.14 It is part of the presence of the consciousness of the speaker. The availability of T-sentences makes it possible to see how a singular term has been badly ignored and thus is subject to reaffirmation and reinstitution.

From Anomalous Monism to Normalized Uni-Binary Realism Now I wish to consider Davidson’s thesis of anomalous monism as a case where reality of mind and brain both counts in asserting or holding a statement concerning the mental state in relation to a brain state statement. Davidson has titled the mind-brain co-happening and has shown how interdependence is essential for mind activity, but there is no special determination by brain or by mind that leads to a special mind or brain state to be a part of the property of brain or a part of the property of mind. This suggests that mind-brain connection could be an emergence from brain to mind, with mind supervient over the brain. This to me has a metaphysical significance, because this shows that there is a binary structure linking mind and brain in understanding mind. Whether a brain state has a corresponding mind state is another question. Interpreting mind state or giving the truth condition of a mind statement requires consideration of the binary structure of mind brain, but the meaning of the statement however seems to be determinable by mind statement alone, which may have or may have not its brain correspondence. This shows the non-reducibility of mind statement to brain statement or to their identity in brain language. Quine has remarked on this phenomenon in the following statement: 14 See my article “Elimination of Singular Terms Reconsidered”, in Journal of the Foundations of Language, No. 4, 1968, pp. 282–295.

from donald davidson’s use of “convention t”

285

It is at this point that we must perhaps acquiesce in the psychological dualism of predicates, though clinging to our effortless monism of substance. It is what Davidson has called anomalous monism. Each occurrence of a mental state is still, we insist, an occurrence of a physical state of a body, but the groupings of these occurrence of a physical state of a body, the groupings of these occurrences under mentalistic predicates are largely untranslatable into physiological terms. There is token identity, to give it the jargon, but type diversity.15

Quine recognizes the irreducibility of the mental to the physical, even though he still sees a token identity between the two. From his positivistic stance, he does not wish to give an ontological status to the mental because the mental is not really quantifiable nor verifiable as a substance. He would not speculate whether this irreducibility of the mental could mean that Reality as a whole network of relations must condone the mental as a level or a part in its network. The fact of the mental as an experience of the human person suggests that the mental is on the one hand epi-phenomenal based on the phenomenon of the material, on the other emergent and supervenient to a point it has its autonomy of expression and forming the subject matter of independent meaningful discourses in all human cultural and linguistic traditions. The anomaly of the type duality between the mental and the material actually exhibit a fairly interesting feature of developmental capacity of tolerance and expressibility of computer languages in modern software industry which forms a hierarchy of lesser and greater capacity of memory, calculation and expressibility. One sees in using Microsoft Windows XP, one can do programs written in MS Windows 2000, or MS 1998 or MS 1997. But one can do many things written in Windows XP or 2000 in MS 1998 or 1997 or any earlier versions of Windows. This is because Windows 2000 or Windows XP is considered more finely developed so that it could contain Windows versions of earlier years but not vice versa. Similarly, could we also consider he mental language as a more powerful language in some sense than the simply materialistic language? The mental language has a partial realization in the brain language, but we cannot infer the mind language from the brain language just as we may consider that Windows XP has a partial realization of

15 See Quine (1995), From Stimulus to Science, Cambridge: Harvard University Press, p. 87.

286

chapter eleven

Windows 98 but we cannot infer Windows XP from Windows 98. One might point out that not everything in XP gets expressed in 98, even though the hardware of the calculating unit still can be said to have an electronic counterpart action in the hardware computer. But on this basis it is difficult to infer software program from the hardware qualities. Of course one must recognize that we need a certain up-grade of a computer hardware to run strong software programs. The point I am making here is that the mental must be considered a creative emergence from the material brain and this creativity even though internal (naturally assumed to be internal than external) to the brain in bringing out the mental has also made the mental transcending the material. Once it crosses certain boundaries the mental brings out a meaning and value of the material, which may not explicitly attach to the material. The mental needs not to deny the existence of the material or transforms the material into the mind (even mind is more subtle than matter). There is close link between the mental and the material, mind and brain. There is meaning difference, but there is root or core identity. The question of how this identify and difference are o be reflected in the brain and mind language requires close consideration of the autonomy of the mind discourse: it leads to a non-reductive and creative-emergent ontology and cosmology of mind and matter in which the world of mind and thinking could be recognized and studied on their own even though it is deeply integrated with the world of things in its root. In doing so we have to perhaps take Whitehead’s suggestion of creativity as an ultimate principle seriously.16 With this understanding we may introduce the notion of onto-cosmic states that would correspond to the mental states that exhibit themselves in mental language and mental discourse. A mental state that has a corresponding brain state also has a corresponding ontocosmic state of the world that is disclosed by the mental state. The onto-cosmic state of course is part of the whole onto-cosmology of the world in which material things and brain states also are included. There is fundamental unity between the mind and matter that could be described as unity of the li and qi or yin and yang of a life force.

16 See Alfred Whitehead’s book Process and Reality where creativity is taken to be primitive notion.

from donald davidson’s use of “convention t”

287

The onto- part of the onto-cosmic state indicates the ontological root of the mental reality, which is also the root for the material reality. The cosmic- part of the onto-cosmic state on the other hand represents the creative emergent phenomenon of the state of ontocosmology, which comprehends both the brain state and the mental state. In this fashion we may speak of the mental state not simply as a psychological or subjectivist entity but as an aspect and dimension of the world reality realized by the mental activities of the human person. With this understanding we may be also said to have normalized the anomalous monism of Davidson into a normalized monism of creative advance, or to put in a Chinese philosophical jargon, a normalized one-body two-dimensional (uni-binary) realism ( yi-ti-er-yuan-shi-zai-lun). The consequence of this understanding is that we could take the world of mental words as constituting a relatively independent discourse for which an independent semantics of truth and meaning can be defined and developed. The primary model for truth and meaning semantics in the original formulation of Davidson could then be applied to this discourse in such a way that we could treat all mental words have an extension which is coextensive with the world but which has a meaning which exceeds the meaning of the extension in a material context.17 This semantics helps us to introduce learnable logic and syntax into the onto-cosmological talk in Chinese philosophy, which however will not place a reduction or restriction on the Chinese discourse of onto-cosmology and ontohumannology or onto-epistemology and onto-ethics. In such learnable contexts it is believed that deeper meaning of the mental terms will acquire an onto-cosmic meaning while the onto-cosmic terms will acquire a mental meaning. Although such meanings are not learnable because they are not finitely realized in semantic primitives (what are designated as semantic primitives) and are not quantifiable in open contexts of metaphysical or onto-cosmological statements, nevertheless they produce a domain based on which individual minds would exercise their creative insights and allow realization of creative onto-cosmic states to emerge and materialize. In 17

See my article “On Referentiality and Its Conditions”, presented at American Philosophical Association’s Eastern Division Annual Meeting, New York, December, 1969. Abstract in Journal of Philosophy, November, 1969. Full article in Notre Dame Journal of Formal Logic, 15:2, April, 1974, pp. 245–264.

288

chapter eleven

this context we could speak of a dimension of creative meaning and truth for the ontological discourse and experience on the same par.

Part II Chinese Model of Truth Interpretation: Five Basic Principles In the above I have introduced the major theses of Donald Davidson on truth interpretation by utilizing Convention T of Tarski. I have also raised critical questions in regard to their use in illustrating and clarifying concepts of meaning and truth in Chinese language and Chinese Philosophy. We find how limited the use is if not broadened with a fresh understanding of what is ontologically presupposed in a living natural language. I shall now actually make an attempt to construct a theory of truth for Chinese language in terms of five fundamental principles of understanding reality in a framework of Chinese philosophy and show how understanding and explication of these five principles would bring our interpretation of truth of sentences to a new level and a new horizon beyond Davidsonian use of the Convention T. We shall see how the Tarski Convention T could receive a new interpretation in light of the relationship between Chinese language and Chinese philosophy as the truth theory of the Chinese language. We can see that we have re-described, re-interpreted, and re-written the Davidsonian theses in light of our analysis and understanding of Chinese language in relation to Chinese philosophy. There are five fundamental principles which can be formulated to capture the fundamental referential and meaning-generating features of Chinese language in its actual use which in turn lead to the revelation of the basic insights into reality in the framework of Chinese philosophy.18 These five principles could be said to delineate a model of truth interpretation for the Chinese language as a natural language. In formulating these five principles, as we indicated above, we are treating Chinese philosophy as a theory of truth for sentences made in the Chinese language. Before we explain how this could be understood, let us first elucidate these principles and then take up 18 These principles are formulated on the basis of my research work and publications in the last twenty years. See specifically my book (1991), New Dimensions of Confucian/Neo-Confucian Philosophy, State of University of New York Press.

from donald davidson’s use of “convention t”

289

the question as to how they can function as a theory or part of a theory of truth for the Chinese language. 1>. The Principle of Open Syntax and Grammar: The general grammar (syntax) of Chinese language has left many kinds of formal specifications open (e.g. gender, tense, numbers, subject or subject-matter, and thus more like a logical formula) so that one can semantically formulate and pragmatically understand a sentence (with a formal structure as understood in the logic of predicate with identity) in light of concrete situations and direct or indirect experiences of the language speaker. This scheme also makes direct presentation of a given situation in the Chinese language possible as the morphology of the dictions and words in the language tends to draw perceptual and especially visual experiences of the speaker as topical resources of focal understanding. The important point of openness of the syntax is that it will impose as minimal structure or form on experience as possible. The experience to be normally conveyed is to do with daily life activities, which are geared toward fulfilling the internal functions of nature and needs of human individual and communities. This daily experience of a community life reveals a common sensible world of things, people, events and affairs which allow a deepening and a widening in terms of processes of natural nature and reflections on the meanings of human feelings and thoughts together referred as heart-mind. The openness of syntax and grammar is further evidenced in the restraining from using a universal being-term (namely the copula or verb to be): there is no underlying desire to see things as if it is made of one mode such as having substances as the Greek has done. 2>. The Principle of Natural Creativity: There is an underlying assumption that any meaningful sentence must find a proper place in the total and ultimate system of understanding the world: whatever terms or ideas or language which are used must point to such a level of reality or lead to its understanding. In this sense Chinese language contains a naturalism or natural realism, which presents the world in its natural reality and natural totality as we have experienced. Indeed, human experiences are also projected into such a natural reality to gain meaning: any experience must conform to this natural reality or lead to this natural reality to qualify as reality-revealing experience or to be regarded or explained as subjective prejudices

290

chapter eleven

or private aberrations. We may even regard the meaning of a sentence as necessarily coming from reference to a natural world of things and is based on our total sharable experiences in the community. In brief, language is founded on and constantly funded from such a community of sharable experiences. In this sense Chinese language has two aspects that can be understood in terms of the naturalism of Daoism and the humanist-historical consciousness of Confucianism. Because it is in both that we can see a basic realism which is naturalistic and which is also the basis for a development of human world of history and morality. It is in this sense the experience and vision embodied in the development of Chinese language resonate with the experience and vision of the dao and Humanity espoused by Daoism and Confucianism. It is also a far cry from the deep experience and deep understanding of the multiple dimensions of reality, the changeless, the changing, the transforming, the contingent and the creative. Hence the language is as much a testimony of our human sense of reality as a manifestation of reality in terms of its poetic and indexical or iconic properties to the human eyes. 3>. The Principle of Internal Reality Externalized: Just as our total experience may point to an external—objective level of natural reality, the believing or positing of which may explain, evaluate and guide our experiences past and present, such total experiences may at the same time point to an internal-subjective level of human existential structure, the believing or positing of which may be required for giving meaning to human self-understanding and self-knowledge and giving guidance to human moral life and action. Reality is not only disclosed or considered from external observation, but also disclosed or considered from internal reflection. Philosophical position such as Daoism thrives on deeper and more comprehensive observation which gives rise to a feeling of non-attainability and profundity and a vision of the human self to be dissolved in such deep experience of reality. On the other hand, philosophical position such as Confucianism devotes itself to the cultivation of a social-moral world as embodied in human interaction with nature and inter-human relations, resulting in a conscientious growth of humanity. In both a conceptual framework is advanced so that reality must be seen and evaluated in such a conceptual framework of the dao (the Way) or the ren (the Humanity). But the success and relevance of the conceptual framework which is concretely evolved and creativity presented still depend

from donald davidson’s use of “convention t”

291

on how much more stimulating and inspiring or fulfilling experience one could provoke and arouse so that a value of harmony, peace, and internal tranquility and freedom from harm and freedom to creativity can be achieved. In this sense conceptual framework is in essence a system of dispositions and abilities in pursuing action and change. 4>. The Principle of Historical Reference (Historicity) and Life-World: In terms of content and semantics, a sentence in Chinese language has embodied or presupposed a meaning background which may reflect some historically accepted pristine insights into reality and life taken as a generality shared by a community of common people. These insights are oftentimes articulated in well-understood idioms and metaphors which can be used to give meaning to present experience of one’s life and understanding of the life-world. The balance of history and present experience is to be achieved by our understanding of how history illuminates experience, and how experience on the other hand illuminates history in yielding insights into the nature of events and processes of nature and the intentions, beliefs and actions of the human person in reference to time, space and other people. 5>. The Principle of Comprehensive Totality and Ultimate Origin: In regard to understanding and interpretation of the meaning or new meaning of a sentence, considerations of both the linguistic and non-linguistic contexts of the sentence are crucial and important because they are clues to how to specify dimensions of meaning of the sentence. Besides, a consideration of a level of reality can become the qualifying and constraining condition for understanding the meaning of a sentence: the key point of such consideration is how total and how ultimate is the realization of one’s understanding on such a level. The gist of such considerations regarding the ultimate and the totality is both a conscious deepening and widening of reference to our experienced nature and its changes. It leads to a presentation and formulation of how things arise and how the processes of events disclose an ultimate source from which continuous change becomes possible. The reality is now described as both a matter of experience and a matter of external reference. It is conceived as ontological and cosmological and hence as onto-cosmological. It is in reference to this emerging and open onto-cosmology that human life-world, human history-world and the individual world of commonality can

292

chapter eleven

be said to be a result of creative development of human experience and a natural evolution from environment.

Interpreting Truth as Reality Presented The five principles mentioned above give a broad outline of the model for truth interpretation of Chinese language. We need to recognize their limitations. First of all, it is to be admitted that they are general in nature, and do not determine specifically how a sentence is constructed. They only give an indication as to how sentences constructed in the Chinese language fall under these principles. Secondly, the five principles are suggested or derived from our understanding of the underlying use of the language in contexts that give rise to philosophical thinking in relation to reality. In brief, they are generated for the purpose of explaining how sentences generally make claims on reality and hence how they could be ontologically conceived. Third, although these principles may not be sufficient to explain details of the meaning and reference of a sentence, they are necessary for demarcating the general meaning and general references of the sentences. In light of these clarifications, we may say that the five principles function as norms prescribing how a Chinese sentence is constructed and how it is to be understood in terms of meaning and reference. At this juncture we may ask how these five principles can yield truth conditions for sentences in Chinese and thus provide an interpretation of truth in the meta-language of Chinese philosophy. In order to answer this question, we may conceive all meaningful sentences formulated in Chinese as true statements, true in the sense of presenting and revealing a reality which have a structure and which also changes according to inner restraints and deeper incentives of time. Of course, this is an intuitive understanding of truth as related to or reflective of reality of some sort where both true and real are taken in an undefined and un-definable sense. In this sense it is clear that the truth of the totality of Chinese sentences simply consists in the presentation of reality for our viewing or reviewing. It consists in taking the totality of the sentences as mentioning or pointing to the reality as an object. Or to put in a different way, the truth of the totality of Chinese sentences thus conceived is a view of reality to be understood or to be consciously entertained by

from donald davidson’s use of “convention t”

293

a human person. For this reason we must take these sentences seriously and treat them as a transparently showing reality or as even embodying the reality as an iconic sign. In doing so we eliminate or forsake the request or determination of the truth of the totality, because the so-called truth is dissolved into our sense of reality, which can be said to exhibit a special direct relation with us through our knowing and understanding. This means that our abilities in knowing and understanding naturally enable us to develop and produce the language in a process of development. It is in this sense we need to appeal to our abilities of apperception and understanding for accounting for our devising language as an instrument of knowledge. On the basis of these abilities we are also able to interpret a sentence in terms of our understanding and thus able to convey to other people how the sentence is constructed and used and what it is intended to refer to. In thinking toward an interpretation, we may come across the five principles as explanatory of how any sentence is used and what truth condition it may entail. In this regard the five principles are my hypothesis for understanding the use or meaning and reference of a sentence. We come also to see our five principles as a portion of a larger theory that would interpret all sentences in terms of our experiences and understanding of reality. It is on this basis that we may construct an interpretation that would illuminate the given sentence. Our interpretation could be hypothetical, but they must be consistent with the five principles. Although the five principles merely tell us how we can form acceptable and meaningful sentences, they also invite us or allow us to introduce more specific principles for making specific and relevant explanation of the meaning of the sentence, namely the specific and relevant truth conditions of the sentence. This is because to understand the truth of a sentence in reference to a description of the real we need to reflect on the real and understand the real. We could therefore appeal to a principle or observation from Chinese philosophy so that we could transform truth into reality or let us see truth as reality. This transformation actually is very simple: Just regard the totality of sentences as presenting truth or reality. The successful interpretation in Chinese philosophy of what is presented or used can be regarded as the theory that is relied upon to be able to explain or interpret all actual and potential sentences. It is clear that such a theory has to interpret not only the meaning a sentence in terms of the reality the sentence presents

294

chapter eleven

as differences of roles of language terms in forming the sentence and thus in structuring the experiences. It has also to explain what a sentence suggests and refers to relative to our understanding and belief. Hence Chinese language stands to Chinese philosophy just as our consciousness of reality stands to an interpretation or explanation of consciousness. It stands as experiences stand to explanation of experiences. It stands as reality stands to explanation of reality. The language becomes the language once we become aware of what it is about and seeks what it means, seeks, shows or intends to represent. We come to see language as representing the reality once we come to see its truth as reality presented. It is also equally valid to say that reality becomes language once it is formulated symbolically or becomes conscious of itself. Interpretation is an act of restoring truth to reality. Using language is then an act of separating truth from reality. If we make sense in making the distinction between what the language says and how it is interpreted, we make a distinction between using language and interpreting language. But if we see using language as mentioning a reality, we need also make sense of the reality. Hence we can speak of interpreting the reality as mentioned by the language. With this said, we see how Davidson comes to introduce the Convention T as a principle of eliminating truth predicate so that we may face truth as suggesting a reality open to our interpretation. In this sense language may be regarded as a tool to bring truth to bear on reality. Davidson has seen how powerful and how significant is interpretation needed for separating language from reality. On the other hand, it is also important to point out that we may also see reality as language-productive relative to our understanding and belief concerning reality. In the following we shall explore how each principle actually leads to the application of some basic insights of Chinese philosophy and therefore contribute to a formation of an enriched context of Chinese philosophy as a theory of truth which interprets truth of language into reality of experience.

Integrative Layers and Underlying Unity It may be said that the five basic principles are formulated as a result of my overarching understanding of the use of Chinese language in both Classical and Modern contexts of life and from both

from donald davidson’s use of “convention t”

295

philosophical reflection and experiential observation. Doubtless, they reflect my intuitions as a native speaker, but at the same time they also reflect a deeper perceptions of time, space, being, non-being, mind and nature as verified on the basis of my study and analysis of Chinese philosophy in its early formation age. I have in mind specifically the philosophical worldview emerging from and presented in the symbolic realism of the creative change in the system of the Zhou-Yi (the Book of Changes). It is clear to me that it is in light of this deeper understanding the practice of making meaningful statements in the Chinese language and in Chinese philosophy can be said to presuppose these five principles as its macro-truth conditions which are absorbed in the formulation of Chinese philosophy at large as a theory of truth for the Chinese language. Chinese language without Chinese philosophy is deprived of meaning and reference. Chinese philosophy without Chinese language will lose its natural form of expression and conscious activity of articulation. In light of the distinction made between a language and a theory of truth for the truth of sentences formulated in the language, it is revealing to see the Chinese language functions as a language for which Chinese philosophy would function as a theory of truth for the language. Here it is important for us to recognize the distinction between macro-truth conditions and micro-truth conditions of the use of a language form in which truth conditions can be strategically attributed and empirically distributed. It is on the basis of the primary reference to the experiences of the Zhou-Yi that the essential insights and conceptions of Confucianism and Daoism can be said to arise.19 Conventionally, a declarative statement would have a truth-value truth if its projected reference in an object or in a state of affairs can be independently ascertained by experiential identification. The projected reference under whatever interpretation or description or re-description would be its micro-truth condition or conditions. But the sentence is made possible by the language to which it may be said to belong. The existence of the language as a holistic and virtual repertoire and mechanism for producing all the legitimate sentences

19

See my article (2000), “Confucian Onto-Hermeneutics: Morality and Ontology”, in Journal of Chinese Philosophy, Vol. 27, No. 1, pp. 33–68. See also my article (2004), “Dimensions of the Dao and Onto-Ethics in Light of the DDJ”, Journal of Chinese Philosophy, Vol. 31, No. 2, pp. 143–182.

296

chapter eleven

under its syntactic, semantic and pragmatic rules of usage is provided by the macro-truth-conditions of the language: it is that way or form of thinking which has captured the essence of an indeterminate reality. If conceptually formulated, it becomes the explicit system of understanding of a reality (by way of reference of language) based on a people’s collective experience and communicative reason. In trying to describe the hidden Chinese understanding of reality in the use of the Chinese language by the Chinese people or/and in a system of philosophical onto-cosmological reflection of the Chinese thinkers, I am quite aware that more detailed description of the functions of each principle needed be specified. Apart from this, we must also beware that there are other principles governing formation and re-formation of meaningful sentences in the Chinese language as a system of virtual ontology-epistemology and onto-phenomenological way of thinking. What we have given is a necessarily minimal outline or a broad framework of reality for identifying what a sentence and its meaning and its reference would be. Another observation needs to be made clear at this point: even though the primary or primal framework of meaning and truth in the Chinese language as outlined has a peculiarity of its own, it could be compared and even integrated with other framework of the same in another language in so far as its macro-truth-conditions are similarly given so that it could be understood as meaningful and truthful. Why is this possible? The answer leads to the following onto-phenomenological reflections. In the first place, given the Chinese language as C and a second language as E, either E can be understood in terms of C because of the sharing of common and basic experiences in the life-world and historicity, or E has to be translated into C by reference to a deeper reference system D which can be constructed by incorporating both C and E. In virtue of this deepening or receding reference system both C and E could be interpreted, re-described and hence rewritten. As C has the ability to recede to a deeper reality and remains ontologically open and semantically generative, D could be considered as a more differentially structured version of C into which E could be incorporated in light of the difference in E. In this way the integration of C and E could be made not by reducing E into C but by accentuating and taking advantage of the difference of E for a differentially structuring of C. With other languages the

from donald davidson’s use of “convention t”

297

similar procedure of differentiation and integration can be adopted to generate more and more comprehensive and yet more and more differentiated system. This process of integration by differentiation and differentiation by integration can be illustrated in the following diagram: C”’ C” C’ C, E, F, G, . . . . . . . . . . . . . .

There may not be an absolute ultimate point to which the given integration could reach. But that is because an infinite space allows infinite receding and deepening construction of reference in so far we have various systems which show the differences from a given level of integration. It is important to keep in mind that with the deepening system the original differential systems be whatsoever they are could be interpreted, re-described and rewritten in the language of the participating differential systems. In the case of Chinese language system of the Zhou-Yi, not only Confucian philosophy could be re-interpreted, re-described and re-written, Daoist philosophy could be similarly but differently re-interpreted, re-descried and re-written. This will no doubt give rise to the result of a more differentially structured system of the Zhou-Yi onto-cosmology or onto-phenomenology. With this language the Greek philosophy of atomism and dualism could be also re-interpreted, re-described and rewritten, as well as the Indian philosophy of Brahmanism and Advaita Vedenta. By the same token, with reference to the Chinese language at large, modern Western philosophy of whatever kind, and contemporary postmodern philosophy of whatever kind (Nietzsche, Heidegger and Whitehead etc.) can be equally re-interpreted, re-described and rewritten. Of course Chinese language is not the same thing as Chinese philosophy or a system of Chinese philosophy such as the Zhou-Yi system, hence the interpretive/re-interpretive effort or activity in redescribing and re-writing needs not be so explicit as to warrant being called interpretation or re-interpretation. Instead, it is usually called translation. But then we could no doubt regard translation as a procedure theoretically composed of interpretation, re-description and rewriting in a different language.

298

chapter eleven

It is to be noted that the more translated materials, the more influential on the translating language and thus the more natural for the translating language to be subject to the influence of the translation because it has appropriated the translated materials as part of its system. This phenomenon has certainly occurred in the arising of the new literature, new vocabulary, new ways of expression and even new ways of grammatical organization in modern and contemporary Chinese after 1919 New Culture Movement in China. This phenomenon can serve as a corroboration of the effectiveness of integration of two systems of language or two ways of thinking. Similar examples no doubt can be drawn from other traditions as well. In this connection, it is not an understatement to say that all languages must share something in common on some levels of its understanding of the world. This is so because we live in the same world and we live as human beings as well. The ability to recognize the same world on some basic level and to recognize the same species of the humankind is inherent in the establishment of language, as language is a tool invented by the human kind to deal with and cope with the world and with itself. This point is made to simply point to the fact that there cannot be empirically incommensurable systems of understanding between two systems if these systems are not to be applied to the world and to the human beings. The differences comes on levels of interpretation which become the sources or resources of meaning and truth conditions for the given language system of a people. To go back to our five principles of the above, it is important to point out that each of the principles illustrates a type of balance, equilibrium, and harmony where balance is achievement of forces or elements without necessarily requesting any elimination of mutual conflict and struggle, equilibrium is achievement of balance with conflicts reduced or eliminated, leading to a state of peace and rest, and harmony is to establish positive reciprocal support and interested interaction among the forces and elements which will lead to creativeness and production of novelty in reality. Hence, in the first principle, we shall have the balance, equilibrium and harmony of syntax and pragmatics in Chinese language in such a way that the meaning of a sentence is determined by both the explicit rules of grammar and the implicit customs or cases of use without imposing unnecessary prior restrictions on reality or form.

from donald davidson’s use of “convention t”

299

The idea is to achieve utmost simplicity and conciseness of form with utmost scope and comprehension of content. This of course gives incentive to seeking the simplest and most direct experience and understanding of reality and life to be formulated in simple and yet profound pristine insights of reality and life. This is highly illustrated by the formation of the philosoph(ies) of tian (heaven), dao (the way), xing (human nature), ming (human destiny), li (principle), qi (vital force), xin (heart-mind), wu-ji (the no-ultimate), tai-ji (the ultimate), wu (nothingness), kong (emptiness) and other onto-cosmological or onto-phenomenological categories of reference variously present in the systems of Yi-Jing (Zhou-Yi ), Yi-Zhuan, Daoism, Confucianism, Neo-Daoism, Chinese Buddhism of Tiantai / Huayan / Chan and Neo-Confucianism in Chinese philosophy. In a sense Chinese philosophy as understood in these important traditions is a philosophical extension of the formation and structure of Chinese language in reference to its structure and use; similarly, the use and grammar or syntax of Chinese language can be equally explained philosophically as a linguistic realization and concretion of Chinese philosophy of yi (transformative change) / dao (total ordering) / li (intrinsic ordering) / qi (creative exploration). In the second and third principles we see also the mind and a way of thinking at work: the world is considered as comprehensively and ultimately real and natural in the sense that the world reality is independent of my mind and consciousness and yet can be learned and represented in my consciousness through our perceptual experience and reflective understanding. To say this however is not to say that the world is not related to my mind and consciousness in a profound and fundamental way: the world which generates the world also generate my mind, and one can realize and has to realize that the mind and the world meet in the deep level of reality which is the dao or tian or tai-ji as those terms are used. This is an insight to be gained from comprehensive observation and reflective understanding on many levels: the onto-phenomenological level, the onto-cosmological level, the historical-political level and the humancultural level. This point shows the difference between the Greek way of thinking or the modern European way of thinking which separates reality from appearance, the subjective from the objective, the human from the natural, and the Chinese way of thinking which does not make such a dichotomy. We see such a dichotomy in

300

chapter eleven

Parmenides, Plato, Augustine, Aquinas, Descartes, Kant and even modern Science which is reductionist in methodology.20 On the other hand, we see a unison and unity of the subjective and objective in understanding an onto-cosmology that is seen to give rise to the difference and provides a way of fruitfully reconciling the difference in the philosophical systems of Daoism and Confucianism. Specifically, we see the principle of unity of opposites at work: the tai-ji generates the opposites of the yin and the yang which gives rise to the world and human person.21 The human person which arises from Heaven and Earth in unison mediates heaven and earth in fulfilling the potential creativity of the tai-ji or the dao so that he can act right and live well and harmonize things in the world. There is a natural conformity between the objective and subjective because our experiences would naturally form a host of concepts which applies and which can be also corrected from constant learning from experiences. Besides, there is practice in the widest sense of action that can be relied as lessons to provide validity and substance to our ideas and knowledge. In this sense the so-called Principle of Charity in the philosophy of understanding and language is not to be understood as just a logical or strategic principle geared toward understanding, but a principle of reality, which we can verify and confirm in our own experience by way of observation and reflection. It is in terms of such experience by way of observation and reflection that certain fundamental concepts of reality arise and certain insights into the nature of reality and human self ensue that are axiomatic and defining. After these concepts are formed, knowledge of reality and self becomes not only possible but also justified. Here we have spoken of an onto-epistemology based not on the atomic perception as such but on the macro—understanding of the larger universe with its macro-truth conditions. The onto-hermeneutical circle principle is also at work here: we learn from the larger universe 20 Dualism is pervasive in views on man and God, mind and body, this world and the other world, the phenomenon and noumenon, in the tradition of Western philosophy. It implies a methodology of qualitative differentiation and separation in metaphysics and epistemology. In modern science, we see more reductionism than dualism of the traditional sort. But in demarcating the physical world from our subjective experience of the world, modern science has implicitly affirmed the tradition of metaphysical dualism. 21 This view of onto-cosmo-humano-gony is well presented in the Neo-Confucian philosopher Zhou Dunyi’s work titled Discourse on the Diagram of the Great Ultimate.

from donald davidson’s use of “convention t”

301

in order to understand the micro-universe and we learn from the underlying unity to understand the emerging diversity and vice versa. We learn both so that we can found and justify the post-established harmony among different things and especially between our mind and the world, or between meaning and truth in our language. The interesting question to ask is how these fundamental ontocosmological concepts arise which shows or suggests underlying unity and integration of diversity: The answer is that they arise by comprehensive observation which leads to understanding of pattern and order in things under the large scope of space and long process of time in our observation of things in the world. Eventually one sees things in clusters and in relations and in process of eventuation and happening. One also sees the inception, the emergence, the formation and the transformation, the ending and disappearance of things. One sees being in nonbeing or void and void in being. Then with reflection in our minds which wish to have order and pattern further organized and explained, this leads to the insight into the fundamental concepts of origins and development process which are presented in the terms of tian, dao, tai-ji, li and qi and etc. Each of these onto-cosmological terms acquires a definite scope of reference and can be said to denote a reality, which is seen as a whole, a dynamic force, a process and a creative advance and transformation. There is an overlapping of reference of these terms as they focus on different aspects of reality or levels or aspects of the same reality. Hence these concepts also form a developing and logical structure of their own. Hence one sees how a creative power in tian gives rise to things in change that leads to the notion of the dao. Yet the creativity of the tian warrants the reality of all things for harmony in the dao. Tai-ji suggests origin and leads to unity of opposition of the yin and yang as two modes of movement and presentation. From a material point of view, reality is also seen as a matter of qi, the vital force that can take forms of many systems of material organization. The order and organization which illuminates our minds and which explains the structure and happenings of the things are then seen as principles or li that assumes the ultimate role of explanation and justification. In this use of these terms, one sees how a philosophy of reality emerges that relates reality to appearance, things to events, reasons to causes and origins to systems and their applications (ben to ti and ti to yong). In terms of this understanding, one sees that Chinese philosophy is fundamentally experiential and

302

chapter eleven

fundamentally realistic in the sense of natural realism, namely our experience of the larger nature and natural events defines what reality is. The origin and further development of this reality must be related to this presented nature for understanding. Understanding means finding relevant central concepts of unity, which unifies difference and diversity of experiences and things in a holistic system of interrelationship and productive ordering. Apart from these fundamental concepts which deal with underlying reality and appearance in the natural world and which we describe as a system of onto-cosmology, there is also the development of the system of concepts for the understanding of the human person: These are the concepts which Confucianism has paid particular attention and which actually arise from self-reflection and self-cultivation of human persons which Confucian scholars have taken to be most important from the time of Confucius. These concepts deals with the human inner life and structure of the human self and therefore function not only as a way of understanding the human self but as a basis and source from morality and ethical thinking on norms and ways of right action. Here we can mention such concepts as ming (mandate, endownment), xing (nature), xin (mind, heart), qing (emotions), yu (desires), cai (abilities), yi (intending), zhi (will, willing), si (thinking), gan (responsive feeling), li-jie (understanding), wu (total awakening), jue (consciousness, awakening), zhi (knowing), xin (believing) etc in an effort to grasp the nature and reality of human person. Again these concepts are nodes formed from human experiences in both observing the behavior of others and reflection and introspection of oneself. They are not abstract concepts for deductive explanation but concretely linked to our experiences and induction from experiences. Hence they are onto-psychologically real as we understand what reality is. The subject person is a real entity with real structure of mind and nature, which we need to cultivate in order to maintain and develop our humanity and human status in the world as a creative power. Confucianism is serious on the principle of self-cultivation of the human person into a moral and ontomoral entity and this of course presupposes the self-reflection that leads to the self-knowledge of mind and human nature, which functions as source for moral virtues and moral action. Hence based on these concepts which point to an internal structure of humanity we come to the basic concepts of morality and ethics such as ren, yi, li, zhi, xin, zhong, xiao, cheng, ming, zhong (central equilibrium) and he (fitting harmony).

from donald davidson’s use of “convention t”

303

In fact, these concepts may be said to arise also from human experience in interaction of oneself with others in a community. They reflect a culture and a community in which one person grows. But they also point to an origin and a world of reality. They form a ben-ti or origin-body. Hence it is in observing and reflecting on the human feeling of empathy-sympathy that Mencius comes to speak of the existence of human nature (xing). Similarly, the distinction of mind into reason and feeling in Xun Zi shows the wisdom of selfreflection: Reason embodies a norm whereas feeling offers a value. The language of moral understanding, like the language of human psychology, is formed from experience of the reality in the human person. The need for morality in politics and government also leads from reality of heaven to the mandate of heaven (tian-ming). In order to link morality to onto-cosmology, one also comes to see how the mandate of heaven creates a norm for political behavior but provides a reason for the existence of the human nature or the nature of being human: It is the mandate of heaven which gives rise to human nature as the Zhong-Yong says. The implication of this statement is such that human nature, human feelings and human actions being subjective as they are, are also equally ontologically real and relevant. They are ways of being and modes of becoming that can lead to formation and transformation of the whole world. Here we see the onto-cosmologization of moral cultivation of the human person as the project of the Confucian thinking. The fourth principle represents a balance, equilibrium and harmony between historical experiences with present-at-hand confrontation of a natural reality. It is a balance struggling for fusion and harmony between culture and nature, the timed reality and the timing reality. History has produced cultural forms and language forms that codetermine the content of a semantic meaning of a sentence or the potential reservoir for sentence formation. The history also means that philosophical insights into reality become historically verified or fulfilled so that its validity and acceptability must take historicity seriously. There is no meaning of the Chinese words, terms and phrases which may not accrue and relate to a historical model or case as originating point or a paradigm of understanding. This shows how certain pristine insights of onto-cosmological or moral- psychological understanding have become instituted in common discourse and daily use as we see from the following examples: li-yi-fen-shu (from Zhu Xi), shu-tu-tong-gui (from Yi-Zhuan), sui-shi (from Yi-Zhuan), pi-ji-tai-lai (from Zhou-Yi), wu-wei-er-wu-bu-wei (from Dao-De-Jing), li-shang-wang-lai (from

304

chapter eleven

Li-Ji ), tui-ji-ji-ren (from Confucisus), ting-tian-you-ming (from popular Confucianism), cheng-ren-qu-yi (from Mencius). This is not to say that one may not question or controvert these idiomatic statements in common use, but in doing one has to cope with both the balance of philosophical wisdom from history and one’s own insights into reality and life in such a way that one must achieve a new insight and establish a new paradigm or give an enriched interpretation of those idioms for one’s understanding and add this to the repertoire of historical uses of the language. Given this enriched content of these two principles, we are provided another set of fundamental and pristine insights into reality and its levels or processes of development. They form the basis for generating the language of moral terms and moral discourses in Chinese language that no doubt becomes enriched because of this. Again it must be remembered that the moral language is still fundamentally founded and justified in a methodology of experience by way of observation and reflection. Consequently, a methodology of unification of universes into a unified and harmonized discourse of understanding exhibits the following four levels: the ultimate and the total which is also the unifying integration, the background and source of difference and diversity; the level of external world of things and processes; the level of internal nature and mind of humanity; and the level of moral norms and moral action.

Understanding the Ultimate Way as Theory of Truth for Language Finally, the fifth principle represents a balance, equilibrium and harmony between the partial and the total, the transitory and the perennial, the surface and the depth or the ultimate as the foundation (matrix) and fountain (source or origin). In a sense this principle shows how language needs to be transcended in a given form in order to be enriched, and to be put aside in order to be re-introduced or re-established in a new form. This is the principle to see language as part of the larger whole even though it can struggle to represent or point to the larger whole. For the question of totality one should understand how hermeneutically Chinese words have to be understood in a context of sentence and a discourse of use and Chinese sentential meaning have to be understood in even larger

from donald davidson’s use of “convention t”

305

context of discourse and use such as a text or a host of texts. This is the hermeneutical circle in action. But the meaning of a word, a sentence or a text is not only determined by the hermeneutical circle of part and whole of a text, it is also determined by the experience and understanding of reality as revealed in previous use of language or previous text available historically. The reference to such a reality is essential for determining the meaning of a word or term or sentence because it is believed that reality invites our experience and our experience suggests or point to such a reality which would then give rise to the meaning of words in use or the new forms of expression for understanding. Hence the hermeneutical-circle eventually becomes the onto-hermeneutical circle as I have explained, and the hermeneutical understanding becomes the onto-hermeneutical understanding. There is the constant effort or drive or will toward seeking and seeing what reality with its origin ultimately is and how it leads to or support our experiences and understanding of reality. This of course presupposes a searching mind at work, an intelligence that is capable of understanding reality and interpreting reality according to our desire or will to explore origin and development. This desire or will is realized in the notion of the ben-ti or Origin-Body that refers to a reality as embodying its origin or source and base. On the basis of this understanding, ben-ti is to be also understood as the basis and origin for the understanding of reality with its source and base. The ultimate and the total in this principle actually speak of the origin and the system in the ben-ti notion. The ultimate speaks to the ben idea and the total speaks to the ti concept,22 both of which gives rise to a context of understanding and a level of reality for the reference and meaning of understanding. We may in effect use the formation of the dao (the Way) in DaoDe-Jing as an illustration of the principle at work: the concept of the dao is explored to approximate the ultimate and to give rise to the total of reality by Lao Zi via his pristine insights into reality as he understands. Why choosing the word ‘dao’ which he says does not represent the Constant Way (chang-dao), which is the ultimate and the total? This is because the dao has been used in such a way it

22 See my entry of ti and benti in Antonio Cua ed. (2001), Encyclopedia of Chinese Philosophy, edited by Antonio Cua, New York: Routledge Publishing.

306

chapter eleven

can generalize on ways of approach to reality and yet needs to be further expanded and deepened beyond its present reach in experience or use in the language. This is also because the dao has been revealed in well-experienced and well-explained idioms such as the ‘way of heaven’ and the ‘way of earth’ and the ‘way of the human person’. As heaven has itself acquired a dimension of the ultimate and totality (as the source of life and reality), the way of heaven suggests both ultimateness and totality of the way of the heaven and hence the way as heaven and hence the way as realized and revealed in things and processes in the world. Hence the dao becomes both concrete and abstract, both particular and general, and even comes to present a dimension of unlimitedness and infinity of the reality. On this basis one can see how the term dao points to something which is not confined to its ordinary uses and references and which needs and can be explored by new experiences and new insights into reality. It is clear that the statements of the dao in the Dao-DeJing result from such an exploration and expansion. The first sentence of the Dao-De-Jing speaks of the dao as unspeakable: the dao that can be spoken of is not the constant dao. Yet the unspeakable constant dao is called forth from spoken dao: the speaking both defines and un-define the dao so that one can liberate the dao from its historical confines to receive new meanings under axiomatic and pristine insights as the ultimate and the total, the flowing and the given. As we can see from reading the Dao-De-Jing we have the dao described in the following 8 insightful proposition:23 a) the dao cannot be spoken but speaking requires the presence of the dao and reveals the dao as the background and the origin (#1); b) the dao is the void of being as things are beings to be experienced in terms of actions, limits, desires, knowledge (#4, #32); c) the dao is the origin or mother of all beings, it leads from starting with heaven and earth to all things (#25); d) the dao is not only creative of all things of the world but also circulate within them which implies the dao as the creative process of things as events and happenings (#25) and yet in doing so remains void without taking possession or ownership (#40, hence unlike God or even Heaven who exercise a will or dominion over all things and human beings); 23 I use for reference the Chinese text Lao-Zi-Zhu-Shi-Ji-Ping-Jie (Commentaries and Comments on Lao-Zi ), by Chen Guying, Beijing: Zhonghua Publishing, 1984.

from donald davidson’s use of “convention t”

307

e) the dao even though being invisible and void is still open to our understanding for we can still open our eyes and our mind to see and observe how things originate and develop and decline and therefore comes to see the dao as something like an image, a spirit or a message (#21); f ) the dao acts in terms of not acting so that everything acts and acts in naturality or natural spontaneity so that everything flows in unison or in connection with everything else and the total nature (#25, #34, #37); g) the dao therefore acts in the direction of opposition and goes back to origin after its origination of things which lead to its own laws of rise and decline for things in the world. h) the dao is the model for our self-cultivation toward well being and moral action; following the dao in communal life is also the golden key toward a good government and peaceful world (#51, #63, #74, #77).

We may regard these propositions that formulate an open concept of the dao and present a philosophy of the dao in action as axioms for the rise of a new language, namely the language of the dao. It is obvious and evident to the students of Chinese philosophy, history and literature that this language of the dao has conditioned and enriched the development of Chinese language up to this very day: they are the axioms for a theory of truth for the language from which they arise and from which the truth conditions for a special class of sentences to do with the dao are generated. Given the above description of Chinese language and Chinese philosophy and their inner relationships of substance and function, theory and practice, semantic meaning and ontological truth, experience and interpretation, the Davidsonian Convention T, even itself a larger expansion of the Tarskian Convention T, can perform a useful job of distinguishing these aspects of relationships so that we come to have a clear logical distinction between language and theory and between meaning and truth which leads to their mutual defining and mutual determination. Although we could learn how to formally formulate the truth conditions of a sentence asserted to be true in the language, the truth of the statement or the sentence does not illuminate what the truth is as it is related to the natural world of experience and reality which are embodied in the development of a natural and onto-cosmologically well-cultivated language such as Chinese or any other world language. The analysis and explication of the basic principles of Chinese language and their enriched

chapter eleven

308

philosophical consciousness in Chinese philosophy serve to demonstrate the logical usefulness of the Davidsonian Convention T and yet the ontological limitedness and uselessness for illuminating truth in a fundamental sense of truth.

Concluding Remarks In the above I have dealt with the theory of truth based on Convention T and related topics from Davidson in reference to Chinese language as a natural language and in reference to Chinese philosophy as a non-idealistic, non-transcendental onto-cosmology: 1. 2. 3. 4. 5. 6.

Convention T and semantic theory of meaning Truth conditions or theory of truth for a language Philosophy of the mental events: anomalous monism Radical interpretation: belief, knowledge and reality Five fundamental principles for Chinese language Holism and indeterminism and Chinese onto-cosmic realism

In conclusion, we have critiqued Davidson’s theory of meaning and truth in light of a close consideration of Chinese language and Chinese philosophy. We come to see how Chinese language and Chinese philosophy each provides a revised thick model for re-interpreting the conditions of truth and meaning theory in Davidson and how his major theses should be modified and yet could be re-interpreted relative to our experience of reality to be described in the meta-language and acquire a new meaning in a philosophical reflection on Chinese language in relation to truth. Yet it is in light of this philosophical reflection his theory of interpretation in the spirit of nonreduction, integration, and robust realism becomes vindicated in Chinese language and Chinese philosophy.

CHAPTER TWELVE

TRUTH PURSUIT AND DAO PURSUIT: FROM DAVIDSON’S APPROACH TO CLASSICAL DAOIST APPROACH IN VIEW OF THE THESIS OF TRUTH AS STRATEGIC NORMATIVE GOAL Bo Mou

1. Introduction In this essay, I focus on two related issues. First, I explore the due point of the thesis of truth as (strategic) normative goal (the TNG thesis below) through analyzing Donald Davidson’s approach to the TNG thesis and drawing a number of morals from his approach via three distinctions. Second, based on the preceding understanding of the point of the TNG thesis, I explore one crucial dimension of the relation between the truth concern, which has been considered as a dominant concern in the Western philosophical tradition, and the dao concern, which has been considered as a trade-mark fundamental concern in the Chinese philosophical tradition, that is, the relation between truth pursuit and dao pursuit regarding their roles as normative goals respectively in Western and Chinese reflective practice. As far as the first issue is concerned, I argue that the due point of the TNG thesis concerning truth pursuit is that truth (nature) as understanding and capturing the way things are is an explanatory norm to regulate and explain one strategic goal of philosophical inquiries. As far as the second issue is concerned, I argue that the dao pursuit of the classical Daoism is essentially the truth pursuit in one crucial

* This essay is the first part of the complete contents of a longer paper “Truth Centrality and Dao Pursuit: Davidson’s Thesis of Truth Centrality and Philosophical Daoism”, which also includes another part on Davidson’s Slingshot approach to the thesis of truth as explanatory basis (the TEB thesis) and a Daoist thick-object account of how correspondence with reality is possible. The latter part is the content of my presentation paper at the 1st ISCWP international conference “Davidson’s Philosophy and Chinese Philosophy” ( June 8 & 9, 2004; Beijing). Due to space limitation, this part is not included here.

310

chapter twelve

dimension as captured by the point of the TNG thesis: both render the pursuit of understanding and capturing the world (the way things, as due objects of studies, are) normative in regulating a strategic goal of philosophical inquiries. I further show how the classical Daoism as presented in the Daoist classical texts, the Dao-De-Jing and the Zhuang-Zi, can make its substantial contribution to our understanding of the truth concern in philosophically interesting ways. My strategy in the following discussion is this. First, in the remaining portion of this part, I will introduce what is at issue concerning the TNG thesis through specifying some apparent way of delivering it in the context of a more comprehensive thesis concerning the central role played by truth and Davidson’s apparent attitude towards it. Second, in the second part, I will give an analysis of the nature and point of the TNG thesis through examining Davidson’s approach and clarifying some conflations via three distinctions. Third, in the third part, based on the preceding discussion, I will give a comparative analysis of the relation between the truth concern in Western tradition and the dao concern in the classical Daoism of Chinese philosophy through examining how the dao-pursuing approach of the Dao-De-Jing illustrates and enriches the TNG thesis and how some ideas in the classical Daoism as delivered in the Dao-De-Jing and the Zhuang-Zi can contribute to our understanding of the philosophical issue of truth on some fronts. The reflective concern with truth1 in philosophical inquiries has its various aspects and dimensions and has revealed itself in various ways in different philosophical traditions. One important dimension of the philosophical concern with truth, as emphasized in the Western tradition, consists in the reflective concern with its indispensable, central explanatory role in philosophical inquiries and in our folk life. This point can be highlighted by a comprehensive thesis of truth centrality concerning its explanatory role, the TCER thesis, which can be formulated as follows:

1 Given the distinction between use and mention, the term ‘truth’, whenever appearing without quotation marks besides it in this essay, is used, instead of being mentioned, to indicate either the non-linguistic thing talked about by using, say, ‘truth’ in English or our folk/reflective notion of it, both of which might be indicated by different linguistic expressions or phrases in different languages.

truth pursuit and (TCER)

DAO

pursuit

311

The Thesis of Truth Centrality Concerning Explanatory Role: Truth (or the concept of truth) plays its indispensable, central explanatory role in philosophical inquiries.

How is such an indispensable explanatory role played by truth to be understood? Traditionally, in the Western philosophical tradition, such an indispensable and central explanatory role has been considered to serve an explanatory norm to regulate and explain one (central) goal of philosophical inquiries. This view can be highlighted by a thesis of truth as (strategic) normative goal, the TNG thesis, which is sometimes formulated as something like this: (TNG)

Truth is an explanatory norm to regulate and explain one goal of philosophical inquiries.2

Although, as we will see, the TNG thesis as presented in terms of seemingly ambiguous (TNG) has yet to be clarified, many think that there is some reasonable point of the TNG thesis which is considered to be prescriptively adequate. Nevertheless, some think that the indispensable, central explanatory role played by truth does not consist (merely) in its role to regulate an normative goal in philosophical inquiries but in its serving as an explanatory basis for some other important things both in philosophical inquiries and in our life, as emphatically and systematically advocated by Donald Davidson. This view is highlighted through a thesis of truth as explanatory basis, the TEB thesis, which can be formulated as follows: (TEB)

(The concept of ) Truth is an explanatory basis to explain other important things in philosophical inquiries and in our life (such as propositional thought and understanding).

The two theses, the TNG thesis and the TEB thesis, which seem to highlight distinct dimensions of indispensable, central explanatory role played by truth if any, might as well be considered as two subtheses of the TCER thesis.

2 Or, more simply, ‘truth pursuing is a goal of philosophy’. Various expressions of the TNG thesis can be found in the extensive literature of philosophy. Take as example one recent expression in Earle Conee and Theodeore Sider (2005), Riddles of Existence: A Guide Tour of Metaphysics (Oxford University Press) when they characterize philosophy: philosophers “criticize each others’ ideas ruthlessly in pursuit of truth. . . . Philosophy is an intellectual quest, with rigorous rules designed to help us figure out what is really true” (pp. 2–3).

312

chapter twelve

The TNG thesis and the TEB thesis, and thus the TCER thesis, can be viewed both as descriptive theses and as prescriptive theses, though the two identities cannot be absolutely separated. The TNG thesis, as a descriptive thesis, has been considered as true by many especially concerning the philosophical practice in the Western tradition but challenged by some scholars in view of, say, the situation of the Chinese philosophical tradition. Whether or not the TNG thesis is descriptively true concerning the past philosophical practice even within the Western tradition, its core idea has been considered by many as prescriptively adequate for the reason to be addressed below, although the TNG thesis as a prescriptive thesis has been challenged by some philosophers who reject the ‘correspondence’ understanding of truth and, seemingly odd enough, even by Davidson in a way to be discussed. In contrast to the TNG thesis, the TEB thesis seems to be less widely recognized. It is Davidson who has systematically advocated the TEB thesis. The advocates of the TEB thesis argue that this thesis is both descriptively true and prescriptively adequate. For the purpose of this essay, I focus on discussing the prescriptively adequate point of the TNG thesis through examining Davidson’s approach to it and the dao-pusuing approach as presented in the Dao-De-Jing, because how to understand Davidson’s relevant points is an effective and illuminating way to explore some involved theoretic issues and clarify some conflations, and because how to understand the nature and point of the TNG thesis is closely related to how to evaluate the relation between the truth concern and the dao concern. It is known that Davidson has his fundamental concern with truth and emphasizes the central explanatory role played by the concept of truth. That is not merely because the concept of truth plays its key role in his own theoretic work (say, in his theory of meaning and his theory of interpretation), but also, or primarily, because the concept of truth simply plays “its key role in our understanding of the world and of the minds of agents”;3 “Without a grasp of the concept of truth, not only language, but thought itself, is impossi3

Davidson (1997), “Truth Rehabilitated”, in Davidson (2005a), Truth, Language, and History, Oxford: Clarendon Press, p. 3. [This paper and another paper “The Centrality of Truth” are slightly different versions of the same essay; the later appears in J. Peregrin ed. (1999), Truth and its Nature (if any), Dordrecht: Kluwer, pp. 105–115.]

truth pursuit and

DAO

pursuit

313

ble”; “without the idea of truth we would not be thinking creatures, nor would we understand what it is for someone else to be a thinking creature.”4 That is, according to Davidson, the concept of truth occupies a central place in philosophical inquiries because of its fundamental explanatory role. But what seems to be interesting or even puzzling is this. Davidson appears to reject the TNG thesis while advocating the TEB thesis. Davidson maintains, “I do not think it adds anything to say that truth is [should be] a goal, of science or anything else. We do not [should not] aim at truth but at honest justification. Truth is not, in my opinion, a norm.”5 It appears that, according to Davidson, the truth-centrality in the preceding sense amounts to claiming neither that truth is a goal to be pursued nor that truth is a norm to have its regulative function.6 Now, how to understand Davidson’s remarks here in view of his relevant views as a whole, and how to understand the nature and point of the TNG thesis? I will examine these issues in the next part for the sake of exploring the relation between the truth concern and the dao concern of the classical Daoism.

2. Davidson’s Approach and the Point of the TNG Thesis In this part, I analyze the point of the TNG thesis through examining Davidson’s approach and clarifying certain involved conflations via three distinctions. These three distinctions are not all new. The first distinction, the one between truth nature and truth criterion, has been widely recognized, though this distinction is sometimes 4

Davidson (1997), op. cit., p. 16. Davidson (1999): “Reply to Pascal Engel”, in L.E. Hahn ed.: The Philosophy of Donald Davidson, Chicago, Ill.: Open Court, p. 461. 6 Davidson advocates and celebrates the TEB thesis but in his own way. He opposes explaining truth in terms of correspondence to facts for two reasons: first, there is no need: people have already had their good enough pre-theoretic understanding of truth without the need of further theoretic elaboration; second, there is no possibility: “. . . if we try to provide a serious semantics for reference to facts, we discover that they melt into one; there is no telling them apart” (1997, p. 5); the argument for this claim is often called ‘The Slingshot”. In this way, Davidson associates his version of the TEB thesis with the Slingshot to dismiss the possibility of explaining truth in terms of correspondence with facts. The evaluation of Davidson’s attitude towards the TNG thesis bears on how to evaluate his approach to the TEB thesis. Due to space limitation, I will not discuss this here but in another essay. 5

314

chapter twelve

either somehow neglected or purposely ignored. The basic idea involved in the second distinction, the one between pursuing truth as a strategic goal and pursuing truths as a tactic goal on the way towards a certain strategic goal, should not be new, though it has yet to be formulated into an explicit distinction in view of the issue under examination. To my knowledge, the third one, the distinction between the semantic-ascent version and the paraphrase-explanatoryreduction version of the TNG thesis, has not been suggested before and might appear trivial at the first blush. This distinction, however, is arguably significant, and it is crucial to my major argument concerning the due relation between the truth pursuit and the dao concern as well as how to understand Davidson’s point. My strategy is this. (1) I first analyze Davidson’s (at least apparently) negative attitude towards the TNG thesis which has been shown through his criticism of some usual expressions of the TNG thesis like (TNG); in so doing, I address the distinction between truth nature and truth criterion, which is needed to understand Davidson and the point of the TNG thesis. (2) I then examine Davidson’s clearly positive position concerning the normative role of truth nature, given that truth nature consists in understanding and capturing the way things are in two senses to be explained, and formulate the distinction between truth pursuit as a strategic normative goal and truth pursuit as a tactic normative goal, which is also needed to adequately understand the point of the TNG thesis. (3) Finally, I draw some moral concerning the point of the TNG thesis from Davidson’s approach, which consists of its positive and negative dimensions, and make the distinction between the semantic-ascent version and the paraphrase-explanatory-reduction version of the TNG thesis, which is significant both for delivering the point of the TNG thesis and for adequately understanding the due relation between the truth pursuit and the dao pursuit in regard to their normative roles in regulating a goal of reflective inquiries. 2.1. Truth Nature, Truth Means, and Justificatory Norm Davidson apparently rejects such sayings that indiscriminately call truth a goal. He emphasizes, “. . . what we will never know for certain is which of the things we believe are true. Since it is neither visible as a target, nor recognizable when achieved, there is no point

truth pursuit and

DAO

pursuit

315

in calling truth a goal”;7 and, as cited before, he says, “I do not think it adds anything to say that truth is [should be] a goal, of science or anything else. We do not [should not] aim at truth but at honest justification. Truth is not, in my opinion, a norm.”8 Now let me give a close examination of Davidson’s point in the context of his whole view on the issue. Two questions immediately emerge to understand Davidson here: First, what does he mean by ‘norm’ and ‘goal’? Second, what does he mean by ‘truth’ here? To figure out Davidson’s point, the two questions are actually entangled together. To fully answer the two questions in view of the point of the TNG thesis, one needs to spell out the issues to be addressed in the next two sections. Nevertheless, let me give a preliminary exploration of the issue addressed by the two questions. From the preceding citations, it is not difficult to see that, in this context, Davidson talks about ‘norm’ and ‘goal’ in their epistemic or justificatory sense while talks about ‘truth’ in its non-epistemic sense. On the one hand, in the preceding citations, Davidson takes a goal as some visible target that is explicitly recognizable when achieved; such a goal is intrinsically related to the means by which to achieve the goal; to this extent, such a goal is regulated and intrinsically specified by its justificatory means as a norm. In this sense, such a goal might as well be called ‘tactic goal’ in contrast to a strategic goal to be explained in the next section. On the other hand, Davidson categorically rejects any epistemic notions of truth that somehow render truth intrinsically related to a certain justificatory means. Rather, it seems to Davidson, truth (nature) is essentially nonepistemic and thus objective instead of subjective,9 substantial instead of deflationary,10 in the following sense. It should be captured based on our pre-theoretic understanding of truth,11 and its non-epistemic

7

Davidson (1997), op. cit., p. 6. Davidson (1999), op. cit., p. 461. 9 Davidson (1997), op. cit., p. 16. 10 For a detailed discussion of deflationary approach in contrast to substantive approach to the issue of truth, see my article “A Metaphilosophical Analysis of the Core Idea of Deflationism,” Metaphilosophy, vol. 31, No. 3 (2000), pp. 262–286. 11 Davidson (2005b), Truth and Predication, Cambridge, MA: Harvard University Press, p. 24. [The chapters 1, 2 and 3 of the book, pp. 1–75, are the contents of Davidson’s paper “The Structure and Content of Truth”, The Journal of Philosophy vol. 87, 1990, pp. 279–328.] 8

chapter twelve

316

and substantial character can be captured by such a reflectively innocent saying to the effect that sentences are true or false because of “the way things are”.12 Therefore, the non-epistemic truth (nature) cannot be the same as an epistemic means as a justificatory norm to regulate the goal as specified above. Actually, Davidson has another consideration to reject truth as a goal. If the preceding consideration is metaphysical one in nature concerning the (metaphyhsical) difference between truth nature and truth criterion (truth means), the current consideration is based on epistemological reason, though it also presupposes the metaphysical distinction. Given the foregoing non-epistemic and substantial understanding of truth (nature), it seems to Davidson, truths-seeking should not be a goal that is intrinsically related to, and regulated by, epistemic means and justificatory norm. What Davidson is concerned with here is the validity of any epistemic means and practical criterion, or ‘truth’ as truth means if any, by which one can tell which of our beliefs are true. Davidson is quite pessimistic in this regard for the sake of epistemological reason. For this consideration, Davidson further rejects ‘truth’ as truth means or as justificatory norm: there no such truth criterion or truth means that is epistemologically available and reliable to tell for certain which of the things we believe are true. So there is no wonder that Davidson denies that there is any truth means as justificatory norm by which one can tell for certain which of our beliefs are true and thus that truths-seeking via such truth means can make it as a justificatory goal. In this way, for the preceding two considerations, Davidson renders it mistaken to characterize truth (both truth nature and truth means) as a justificatory norm: truth nature cannot be such a norm because of their metaphysical difference; truth means fails to be such a norm because of the epistemological difficulty. It is noted that, if Davidson’s first consideration does make sense given his specific understanding of ‘norm’ and ‘goal’, his second consideration is controversial because there might be some ways to overcome or dissolve the addressed epistemological difficulty. What kind of positive moral, so far, can one draw concerning an adequate understanding of the TNG thesis from Davidson’s seemingly negative attitude towards some typical expressions of the TNG

12

Davidson, 2005b, p. 126.

truth pursuit and

DAO

pursuit

317

thesis, like (TNG) above? It is the distinction between truth nature and truth criterion which Davidson implicitly resorts to in his first consideration above, whether or not one would agree with Davidson about his second consideration in claiming that truth is not a goal. The distinction between truth nature and truth criterion is this: the former is examined by asking what constitutes truth, what truth consists in or what it is for a statement (or belief, etc.) to be true, while, given a certain understanding of truth nature, the latter is examined by asking what is a criterion, standard or means by which one can identify, judge and distinguish truths or true statements from false ones. Our pre-theoretic understanding of truth and its non-epistemic reflective account is about the truth nature instead of truth criterion. This distinction is not new but has been either sometimes carelessly neglected or occasionally intentionally ignored. The latter way is typically taken by those who subscribe to certain epistemic conceptions of truth in a broad sense, such as the coherence one, the anti-realist one and the pragmatic one. One point shared by various epistemic notions of truth is that truth (nature) is not independent of, but intrinsically related to, what one believes or can know; consequently, it seems to those who maintain certain epistemic notions of truth that there is no essential distinction between truth nature and truth criterion as justificatory means by which to reach things possessing truth nature. As explained above, Davidson thinks that there is a fundamental mistake in an epistemic account of truth: what is true does not amount to what is believed to be true. This distinction would bear on a due understanding of the TNG thesis in this way: if what is expected to regulate a normative goal of philosophical inquiries is truth (metaphysical) nature or what constitutes truth instead of truth (epistemic) criterion or truth as means, that is, if truth as criterion or means is not intrinsically or necessarily involved in regulating a normative goal as addressed in the TNG thesis, any epistemological difficulty with the viability of such truth criterion or means would not constitute a rejection or challenge to the TNG thesis per se. In this sense, Davidson’s second consideration for rejection of truth (truths-seeking) as a goal can be only available, not to mention viable, when his first consideration is working, that is, his talks about ‘norm’ and ‘goal’ in an epistemic and justificatory sense as explained above. Nevertheless, we have a number of questions concerning his first consideration: if non-epistemic truth (nature) is not a justificatory norm and goal in the preceding

318

chapter twelve

sense, and given that truth nature consists in capturing the way things are, how does Davidson look at the normative role of such truth nature to regulate one goal of philosophical inquiries? What kind of further moral can we draw concerning the point of the TNG thesis from Davidson’s treatment? Now let me move onto the issue in the next section. 2.2. Truth Pursuit as Strategic Goal and Truths Pursuit as Tactic Goal I start with examining a straightforward question: given that ‘truth’ is understood as truth nature and that such truth nature consists in understanding and capturing the way things are, does Davidson really reject truth (nature) as a normative goal when ‘goal’ is understood in some non-epistemic but still philosophically interesting way? If my understanding of Davidson is correct, I think (or Davidson would respond) that the answer would be no in view of the following points of his works. First, as far as the metaphysical study that seeks exploring “the most general aspects of reality” is concerned, Davidson certainly does not abandon but advocates such a metaphysical study in the traditional sense through what he calls ‘the method of truth’ as highlighted in the following citation: We suppose that much of what we take to be common is true, but we cannot . . . assume we know where the truth lies. We cannot interpret on the basis of known truths, not because we know none, but because we do not always know which they are. . . . Successful communication proves the existence of a shared, and largely true, view of the world. But what led us to demand the common view was the recognition that sentences held true . . . determine the meanings of the words they contain. Thus the common view shapes the shared language. This is why it is plausible to hold that by studying the most general aspects of language we will be studying the most general aspects of reality . . . What we must attend to in language, if we want to bring into relief general features of the world, is what it is in general for a sentence in the language to be true. The suggestion is that if the truth conditions of sentences are placed in the context of a comprehensive theory, the linguistic structure that emerges will reflect large features of reality.13 13 Davidson (1977), “The Method of Truth in Metaphysics”, in Davidson (2001), Inquiries into Truth and Interpretation (second edition), Oxford: Clarendon Press, pp. 200–1.

truth pursuit and

DAO

pursuit

319

It is clear that Davidson takes it as a strategic or ultimate goal of such a metaphysical study to understand and capture “the most general aspects of reality” which Davidson considers as the due object of study in metaphysics. Although we cannot take pursuing truths as a goal on the way towards this strategic goal because “we cannot assume we know where the truth lies”, we can obtain a shared, largely true, view of the world through the method of truth. That is, if my understanding of Davidson is correct, although it is ‘tactically’ impossible (due to some alleged epistemological difficulty) to tell for certain which of our beliefs are true on our way towards understanding and capturing the way things are, it is ‘strategically’ achievable to understand and capture the way things are, which as what truth (nature) consists in constitutes a strategic normative goal in regard to metaphysical study. Second, as far as Davidson’s truth-condition approach to meaning is concerned, when the concept of truth is taken as the explanatory basis for some other important concepts, Davidson in fact assigns the truth condition a central status as explanatory norm to regulate the meaning of sentences that deliver thoughts or beliefs involved in any reflective examination (including philosophical inquiries). As I see it, this is one specific way to implement the point of the TGN thesis. For understanding and capturing the truth condition (say, the truth bearer, the sentence ‘Snow is white,’ capturing the truth condition, snow being white), which presupposes our pre-theoretic understanding of truth (nature), has been rendered foundational to any goal (whether it is to understand the world or to successfully steer by or to achieve intellectual satisfaction . . .) of any reflective examination (including philosophical inquiries), given that a goal is intrinsically related to, and eventually based on, our understanding the truth condition of any propositional thought necessarily involved in such a reflective examination. Third, more generally speaking, as far as all those objects of study that Davidson explores in his own philosophical inquiries are concerned, Davidson, like almost all serious reflective explorers, intends and thus implicitly sets it up as a normative goal to understand and capture the way those objects of study are; it is clear that those reflective explorers do not intend to obtain distorted or misrepresented accounts of the objects of their studies. In this way, with the minimalmetaphysical-commitment character of the pre-theoretic understanding of truth which is likewise presupposed in these cases, Davidson’s

320

chapter twelve

own exploration of those objects of study, implicitly, set it up as a regulative norm to capture the way those objects of study are. In the foregoing three ways, Davidson does not reject but rather, explicitly or implicitly, insists on taking it as one strategic goal of his own philosophical practice to understand and capture the way things are. Davidson himself never abandons but maintains this strategic goal in terms of his method of truth in metaphysics, his characteristic truth-condition approach to meaning, and his way of emphasizing “the key role” of the concept of truth “in our understanding of the world and of the minds of agents.”14 Davidson’s approach is thus not merely completely compatible with, but actually maintains, taking truth nature (capturing the way things are), instead of truth means, seriously as a strategic normative goal in the sense as specified in the foregoing discussion. Now what kind of further moral can one draw concerning the point of the TNG thesis from Davidson’s clearly positive position concerning the normative role of truth nature, given that truth nature consists in understanding and capturing the way things are in the preceding sense? As I see it, it is the distinction between truth pursuit as a strategic goal, on the one hand, and truths pursuit as a tactic goal on the way towards a certain strategic goal, on the other hand. The basic idea involved in this distinction, as indicated before, should not be new, though it has yet to be formulated into an explicit distinction in view of the issue under examination. Let me start with a platitude: at least conceptually speaking, a purpose is the same neither as a means by which to fulfill the purpose nor as a justificatory norm by which to regulate the means. As discussed in the last section, truth nature is not truth means and thus not justificatory norm. The goal that is intrinsically related to, or regulated by, a justificatory norm might as well be called a ‘tactic goal’: a tactic goal per se is a means of achieving a certain desired result or realizing a certain purpose. Let me call a ‘non-tactic goal’ such a desired result or purpose which one can achieve by means of some tactic goal(s) and which per se is not intrinsically related to a justificatory norm. Among non-tactic goals, a goal can be characterized in terms of ‘strategic goal’ insofar as it constitutes a general objective or ultimate purpose

14

Davidson (1997), op. cit., p. 3.

truth pursuit and

DAO

pursuit

321

in a certain discipline. In philosophical inquiries, one can identify its strategic goal as understanding and capturing the way things are in two direct senses. First, in a general sense, a reflectively serious philosopher, like serious practitioners in other serious academic disciplines, intends to understand and capture the way the due objects of her reflective studies are, instead of obtaining distorted or misrepresented accounts of these object of studies. (Even those who oppose the ‘correspondence’ concept of truth and attempt to redefine truth, like pragmatists, implicitly presuppose such a ‘truth-telling’ commitment, whether or not they are willing to admit it.) Second, in a specific sense, philosophy via one of its core areas, metaphysics, is intended to understand and capture “the most general aspects of reality” (in Davidson’s terms). In the foregoing two senses, it is reasonable, or prescriptively adequate, to say that truth nature, or understanding and capturing the way things are, is an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries. One note is due about the distinction between truth pursuit as a strategic goal and truths pursuit as a tactic goal on the way towards a certain strategic goal. The former is not necessarily related to the latter in the following sense. Given that one talks about truth as truth nature based on our pre-theoretic understanding of truth, one can take as a strategic goal truth pursuit (that is, understanding and capture the way things are in the foregoing two senses) in one’s philosophical inquiries while without taking truths pursuit as a tactic goal on the way towards the strategic goal. One does not look forward to telling for certain which of our beliefs are true either, negatively, because such a tactical goal is considered as impossible due to some alleged epistemological difficulty or, positively, because one intends to take some characteristic non-truths-pursuing way towards one’s strategic goal of understanding and capturing the way things are. This kind of strategy is certainly not a mere possibility but has been already implemented and illustrated in certain actual approaches taken by some philosophers in both Western and Chinese traditions. Davidson’s own approach as spelled out above is a good illustration of such a strategy. It might be arguably right to say that a Chinese case of illustrating such a kind of strategy is that of some Daoist thinkers of Jin dynasty, such as Wang Bi (226–249), who take a ‘forgetting-speech-once-achieving-meaning’ approach to capturing

chapter twelve

322

the dao. With this distinction, one can see the primary concern of the TNG thesis in a clearer way: the TNG thesis is primarily concerned with pursuing truth as a strategic goal rather than with pursuing truths as a tactic goal; the latter is at most considered as secondary in the sense that one can take a truths-pursuing way as a means towards one’s strategic goal of truth pursuit (i.e., understanding and capturing the way things are) without any conceptual inconsistency and (arguably) without conflict in reflective practice given that the alleged epistemological difficulty can be somehow overcome. It is also noted that, insofar as such a tactic goal can thus be viewed as a means to achieve a strategic goal, one might as well call a tactic truths-pursuing goal a ‘truths means’ concern, in contrast to a ‘truth means’ concern addressed in the previous discussion of the distinction between truth nature and truth criterion (means). At this point, with the foregoing two distinctions in place, one can see that the TNG thesis is primarily concerned neither with truth means nor with truths means; one thus can reformulate the TNG thesis from a not-so-clear and ambiguous formulation (TNG) into this one: (TNG*)

Truth (nature), instead of truth means, is an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries, whether or not truths pursuing is taken as tactic goal (as truths means) on the way towards the central strategic goal.

At this point, Davidson’s approach raises one further question concerning how to understand the point of the TNG thesis that has yet to be explored. If Davidson takes such straightforward sayings as ‘understanding and capturing the way things are’ to indicate a due strategic goal of, say, metaphysical study, and if this is one way of delivering the point of the TNG thesis, what kind of further moral can one draw from this? In other words, is the difference between this kind of straightforward way and the ‘truth’-mentioned way of expressing the TNG thesis just trivial or somehow philosophically interesting? In the next section, I examine this issue for the sake of fully understanding the point of the TNG thesis and the due relation between the truth pursuit and the dao pursuit of philosophical Daoism. For this purpose, we need to address one more distinction.

truth pursuit and

DAO

pursuit

323

2.3. The Semantic-Ascent Version and the Paraphrase-Explanatory-Reduction Version of the TNG Thesis I think that the difference between the two ways of delivering the TNG thesis as addressed at the end of the last section indicates something philosophically significant. What is involved here is the distinction between the semantic-ascent way and paraphrase-explanatory-reduction way of addressing truth (nature) and thus the distinction between the semantic-ascent version and the paraphrase-explanatory-reduction version of the TNG thesis.15 To my knowledge, this distinction has not been suggested before, though it appears to be trivial at the first blush. I do not know how Davidson would look at the distinction; but, it seems to me, this distinction is important to understanding the point of his approach, the point of the TNG thesis, and the due relation between the truth pursuit and the dao pursuit. If my understanding of Davidson’s relevant ideas is correct, I believe that Davidson might agree to the point I will make concerning this distinction, especially in view of the situation of Chinese philosophy to be examined in the next part. Let me begin with one of Quine’s points concerning the role of the truth predicate, which is relevant to the point to be made here. On the one hand, Quine regards Tarski’s (T), whose instances is like ‘The sentence “Snow is white” is true if and only if snow is white’, as capturing the (partial) core idea of our pre-theoretic ‘correspondence’ understanding of non-linguistic truth. He emphasizes: . . . truth should hinge on reality, not language; . . . . No sentence is true but reality makes it so. The sentence ‘Snow is white’ is true, as Tarski has taught us, if and only if real snow is really white.16

On the other hand, Quine also considers (T) as explicitly expressing the ‘cancellatory force’ of the linguistic truth predicate the use of which cancels the quotation mark or serves as “a device of disquotation”. He says:

15 Note that the term ‘truth’ used in this passage is used, rather than mentioned, to talk about that nature or property as captured in our pre-theoretic understanding of non-linguistic truth, which might be referred to via various linguistic expressions in different linguistic communities. 16 W.V. Quine (1970), Philosophy of Logic, Cambridge: Harvard University Press, p. 10.

chapter twelve

324

. . . we already saw how to express generalizations of the desired sort without appeal to propositions, by just going up a step and attributing truth to sentences. This ascent to a linguistic plane of reference is only a momentary retreat from the world, for the utility of the truth predicate is precisely the cancellation of linguistic reference. The truth predicate is a reminder that, despite a technical ascent to talk of sentences, our eye is on the world. This cancellatory force of the truth predicate is explicit in Tarski’s paradigm: ‘Snow is white ‘is true if and only if snow is white. Quotation marks make all the difference between talking about words and talking about snow. The quotation is a name of a sentence that contains a name, namely ‘snow’, of snow. By calling the sentence true, we call snow white. The truth predicate is a device of disquotation. We may affirm the single sentence by just uttering it, unaided by quotation or by the truth predicate; but if we want to affirm some infinite lot of sentences that we can demarcate only by talking about the sentences, then the truth predicate has its use. We need it to restore the effect of objective reference when for the sake of some generalization we have resorted to semantic ascent.17

Although Quine here talks about the role of using the truth predicate, the point is the same concerning any relevant truth talk: when one uses such terms as ‘(be) true’, ‘truth’ or other cognate terms in Western phonetic languages to express our pre-theoretic understanding of non-linguistic truth or its reflective counterpart(s) (thus rendering a truth talk relevant), such a truth talk, either using the truth predicate to talk about sentences or using its noun counterpart ‘truth’ to talk about non-linguistic truth (nature), is essentially a kind of ‘semantic ascent’ way of talking about (the truth bearer’s) capturing or ‘corresponding to’ the way things are in the world, in which non-linguistic truth consists. In Quinean terms, one might as well say that the truth talk, when ‘truth’ or its cognates are used to express our pretheoretic understanding of truth, is a reminder that, despite a technical ascent to talk on truth, “our eye is on the world”, i.e., (the truth bearer, whatever it is) capturing or pursuing the world. In this way, given that the TNG thesis is primarily concerned with non-linguistic truth (nature) as an explanatory norm to regulate a strategic goal, there are actually two ways to formulate the thesis. One way is its typical or by-default way in the Western tradition to formulate thesis in its ‘semantic ascent’ version, i.e., (TNG*) above, in which the term ‘truth’ in English or its counterparts in other Western phonetic languages (say, ‘Wahrheit’ in German and 17

Quine, op. cit., p. 12.

truth pursuit and

DAO

pursuit

325

‘vétité ’ in French) are used to talk about truth instead of directly talking about what truth (nature) consists in. (TNG*)

The semantic-ascent version of the thesis of truth centrality as normative goal: Truth (nature) is an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries, whether or not truths pursuing is taken as tactical goal on the way towards the foregoing central strategic goal.

The other way is its paraphrase-explanatory-reduction way in which truth nature (i.e., understanding and capturing the way things are), according to a standard or conventional paraphrase of the term ‘truth’ in the linguistic community which uses the term, is directly and straightforwardly given as a strategic normative goal: (TNG**)

The paraphrase-explanatory-reduction version of the thesis of truth centrality as normative goal: Understanding and capturing the way those things in reality that are due objects of studies of philosophical inquiries are is an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries, whether or not truths pursuing is taken as tactical goal on the way towards the foregoing central strategic goal.

Several notes are due at this point. First, though we usually talk about ‘truth’ with a by-default assumption that the truth bearer is the mental one like thought and belief or the linguistic one like sentence and statement, the logical subject of understanding (i.e., the agent who carries out understanding) as indicated in (TNG**) is the human agent whose understanding the way things are results in the mental and linguistic truth bearer which captures the way things are. It might as well say that the human agent is the primary truth bearer while the mental or linguistic truth bearer the secondary in the following sense and to the following extent. It is the human agent who generates her thought and beliefs and produces her linguistic expressions to express her thoughts and beliefs; it is the human agent who can hold a holistic understanding that would provide various relevant background supports for the truth of her beliefs or sentences as the mental or linguistic truth bearers. Nevertheless, (TNG**) still allows the mental or linguistic truth bearer which can be the subject of capturing as indicated in (TNG**). In this way, (TNG**) is rendered more inclusive and more penetrating in regard to the identity and nature of truth bearers than (TNG*) is.

326

chapter twelve

Second, such a saying in (TNG**) as ‘understanding and capture the way things are’ presupposes or implies not merely the objectivity of truth which, as Davidson puts it, distinguished what is true from what we believe to be true, but also a kind of ‘corresponding’ (instead of distorting or misrepresenting) relation of the truth bearer (the logical subject of understanding or capturing, whatever it is) to the way things are. One might well say that the latter meaning is simply implied in such phrase as ‘understanding’ and ‘capturing’ in this context. There is no wonder why the term ‘correspondence’ is both often used to indicate our pre-theoretic understanding and taken as a trade-mark label of any theoretic accounts that are intended to elaborate our pre-theoretic ‘correspondence’ understanding of truth. It is in the preceding pre-theoretic, innocent sense in which I use the term ‘correspondence’ throughout this section to characterize the relation between the reflective agent’s understanding, or the result of such understanding, and the way things are, though it is a further issue how to theoretically elaborate such a correspondence relation. Third, it is important to note that the metaphysical commitment of our pre-theoretic understanding of truth per se as resorted to in the TNG thesis is minimal in the following sense. It does not commit to any ad hoc metaphysical account of what counts as reality, given that reality or the way things are would render truth objective to this extent: what is true in regard to a thing in reality is so because of ‘the way it is’; it is different from what one believes to be true in regard to that thing. For example, an Aristotelian realist (or Daoist naturalistic) pre-theoretic ‘correspondence’ understanding of truth is actually a combination of our pre-theoretic understanding of truth and a Aristotelian realist (or a Daoist naturalistic) ontological understanding or explanation of what counts as, say, snow’s being white. It can be compatible with various kinds of reality: physical reality, social reality, or any object of study that renders truth objective in the above sense. In this way, in (TNG**), ‘things’ in the phrase ‘the way things are’ means various kinds of ‘real’ things that are considered to be due objects of philosophical studies and that would render truth about them objective in the preceding sense. Fourth, as already suggested above, philosophical inquiry surely is not supposed to indiscriminately explore the way all things are in the world but only those things that are adequately considered to be due objects of philosophical studies. Indeed, once upon time when almost all the scientific disciplines on certain subjects had yet to

truth pursuit and

DAO

pursuit

327

become mature enough, philosophy did assume such a role as a global intellectual stepmother, and even right now philosophy still takes it as one of its roles to assume the intellectual stepmother for some newly developed disciplines (such as the role played by the philosophy of language and mind in the development of cognitive science). But this does not imply that philosophy is supposed to explore the way all things are or all the final truths. Davidson points out that it is “the confused idea that philosophy was the place to look for the final and most basic truths on which all other truths, whether of science, morality, or common sense, must rest.”18 Whether or not, or no matter to what extent, Davidson is right on the issue, the TNG thesis per se does not necessarily imply the idea which Davidson dismisses. To highlight this, (TNG**) explicitly indicate that what constitutes an explanatory norm to regulate one central goal is “understanding or capturing the way those things that are due objects of philosophical studies are”. This would provide enough open space to accommodate various interpretations of what counts as due objects of philosophical studies. Fifth, it is also important to note that there are two kinds of reduction in philosophical inquiries. One is to explain a term or notion by virtue of its supposed paraphrased meaning that is based on its pre-theoretic understanding. The other is to explain a notion or a term away by replacing it with another notion that is different in nature. (One example of the latter is some thorough materialist attempt in philosophy of mind to reduce mind entirely to physical body.) It is clear that (TNG**) belongs to the former; for the involved reduction is to reduce the semantic ascent expression back to its paraphrased basis. The logical order here is that the semantic-ascent version per se presupposes an already understood notion of truth, i.e., our pre-theoretic understanding of truth; the paraphrase-explanatory reduction version is simply to reduce its semantic ascent version, or bring it back, to the original point of understanding in a way that gives a more or less reflective generalization of our pre-theoretic understanding of truth which is already presupposed in the semanticascent version. It is noted that, in my previous examination of Davidson’s point in this connection, I actually resort to the paraphrase-explanatoryreduction version of the TNG thesis to deliver the point of Davidson’s 18

Davidson (1997), op. cit., p. 3.

328

chapter twelve

approach. One might object: doesn’t Davidson reject any reductive treatment of truth? In other words, doesn’t Davidson take truth simply as irreducible to other, more basic concepts? Yes and No. Yes, Davidson rejects any reduction of truth to something else that would be substantially different from our pre-theoretic understanding of truth, and thus, as far as the issue of truth nature is concerned, he gives the pragmatic and coherence accounts a summary dismissal. However, on the other hand, Davidson objects to neither characterizing truth as correspondence with reality nor capturing the way things are; he does not oppose saying that sentences are true or false because of the way things are, especially when such paraphraseexplanatory-reduction expressions are used to deliver our pre-theoretic understanding of truth and its reflective counterpart. Davidson not merely renders it innocent but also having some merit: it highlights the objectivity of truth to the effect that something is not true simply because it is believed, even if believed by everyone.”19 Surely, all those concepts like ‘correspondence with’, ‘capturing’ and ‘the way things are’ are more basic concepts to the extent that they paraphrase what the concept of truth is supposed to mean. The trouble with the [so-far-available] correspondence theories, it seems to Davidson, is that they fail to “say, in an instructive way, which fact or slice of reality it is that makes a particular sentence true”;20 for they allegedly suffer from the slingshot argument which aims to show that all true sentences correspond to the same fact. But, whether or not he is right in evaluating some of the current correspondence theories, Davidson certainly does not intend to close the door for any prospective correspondence theory that can provide a viable account “in an instructive way”: to say that no one has succeeded in doing this in the past does not amount to saying that it is impossible to do this in interesting theoretic terms. Actually, with the foregoing distinction between the paraphrase-explanatory reduction and the revisionist reduction, Davidson’s seemingly different attitudes turn out to be compatible. Therefore, it is at least inaccurate to say that Davidson indiscriminately objects to any reduction treatment of truth. All the preceding discussions (conceptual analysis and clarification of the nature and point of the TNG thesis, explanation of needed

19 20

Davidson (1997), op. cit., p. 5; also see Davidson, 2005b, p. 126. Davidson (1997), op. cit., p. 5.

truth pursuit and

DAO

pursuit

329

conceptual resources, and interpretative work on Davidson’s relevant ideas) have also served the purposes of understanding the due relation between the truth pursuit and the dao pursuit of Chinese philosophical tradition in view of the TNG thesis. Now let me move onto this topic in the next part.

3. Truth Pursuit and Dao Pursuit of Classical Daoism As explained above, through an explanatory reduction of the truth property to what the term ‘truth’ is used to really talked about along the line of our pre-theoretic understanding of truth, the paraphraseexplanatory-reduction version of the TNG thesis hits the point in regard to exactly what counts as an explanatory norm to regulate one central strategic goal of philosophical inquiries. In so doing, the explanatory-reduction version of the TNG thesis has another significant role in capturing a due cross-cultural understanding of the nature and scope of the truth concern in different philosophical traditions. Now let us see how the paraphrase-explanatory-reduction version of the TNG thesis, as one positive point drawn from examining Davidson’s approach, would help us identify and characterize the truth concern in Chinese philosophical tradition in view of the classical Daoism. It seems that the truth concern, generally speaking, and the truth pursuit, specifically speaking, is a dominant concern and pursuit in Western tradition while the dao concern, generally speaking, and the dao pursuit, specifically speaking, is a dominant concern and pursuit in Chinese tradition. What is the relation between the truth concern and the dao concern? Are they dramatically and totally different reflective concerns in philosophy? It seems to some authors21 that

21

A.C. Graham said, the crucial question for those traditional Chinese philosophers “is not the Western philosopher’s ‘What is the truth?’ but ‘Where is the Way?’, . . .”. See A.C. Graham (1989), Disputers of the Tao, La Salle, Ill: Open Court, p. 3. To elaborate Graham’s point, David Hall and Roger Ames continued, “The Western ‘What’ question is usually expressed in something like this manner: ‘What kinds of things are there?’ ‘What is the world made of ?’ or simply, ‘What is this?’ Such questions have resulted in a catalog of facts and principles that assist one in taking an inventory of the world about us. The Chinese ‘Where’ question, on the other hand, led to a search for the right path, the appropriate models of conduct to lead one along the path, the ‘way’ that life is to be lived, and where to stand. . . . In the West, truth is a knowledge of what is real and what represents that reality. For the Chinese, knowledge is not abstract, but concrete; it is not representational,

330

chapter twelve

the traditional Chinese philosophy, especially the pre-Han classical Chinese philosophy,22 is not concerned with truth. This conclusion has been drawn based on some seemingly plausible observations or claims as follows. (1) It seems to be the dao, instead of truth, that assumes the primary explanatory norm to regulate one central goal of philosophical inquiries in the traditional Chinese philosophy. (2) In the traditional Chinese philosophy, there appears neither conscious investigation of a general definitional issue of ‘What is truth?’ in meta-discourse nor conscious ‘semantic ascent’ examination of the function, and its philosophical relevance, of the truth predicate. (3) It seems hard to find a unified Chinese character in the pre-Han classical period that would serve as an exact counterpart of ‘truth’/‘true’ in, say, English. (4) Some scholars argue that the dominant portion of the classical Chinese philosophy is non-sentential philosophy in contrast to what is called ‘Western sentential philosophy’ and thus not essentially related to those concepts that are intrinsically connected with sentential philosophy like proposition (or semantic content), truth and belief.23 (5) The significant part and the primary concern of the traditional Chinese philosophy have been considered to be its moral concern and its ethical accounts; and the moral concern is not with how to understand impersonal material world but with the ethical constitution in the human society. In this way, it is not the by-default account of truth (the correspondence account) but a pragmatic account of truth (if any) that plays the role.24 Based on one or more of these, some scholars of Chinese philosophy have concluded that there is

but performative and participatory; it is not discursive, but is, as a knowledge of the way, a kind of know-how.” See Hall and Ames (1998), Thinking from the Han: Self, Truth, and Transcendence in Chinese and Western Culture, Albany: State University of New York Press, pp. 103–4. 22 The phrase ‘traditional Chinese philosophy’ is used to cover various movements of philosophical thought in China from the Zhou dynasty (roughly eleventh century to 256 BC) through the early Qing dynasty (1644–mid 19th century). 23 Cf., Chad Hansen (1985), “Chinese Language, Chinese Philosophy, and ‘Truth’”, Journal of Asian Studies, vol. XLIV, no. 3, pp. 491–519; and (2003), “The Metaphysics of Dao”, in Bo Mou ed.: Comparative Approaches to Chinese Philosophy, Aldershot, UK: Ashgate, pp. 205–224. 24 Cf., David Hall (1997), “The way and the truth”, in Eliot Deutsch and Ron Bontekoe eds., A Companion to World Philosophies, Oxford: Blackwell, pp. 214–224; and (2001), “The Import of Analysis in Classical China—A Pragmatic Appraisal”, in Bo Mou ed.: Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions, Chacago: Open Court, pp. 153–167.

truth pursuit and

DAO

pursuit

331

no significant concern with truth (as correspondence with reality) or with semantic truth in the classical Chinese philosophy. Let me call this claim ‘the thesis of no-truth-concern in Chinese philosophy’, with ‘the NTCP thesis’ for its brevity. Indeed, as suggested in the first observation above, it is agreed that a significant concern, or even the central concern, in the classical Chinese philosophy is with dao, whatever dao would mean in distinctive movements of thought in the traditional Chinese philosophy. The NTCP thesis thus presupposes (or, in some cases, implies) a sister claim, the NTCP thesis*, to the effect that the truth-concern and the dao-concern, generally speaking, and the truth pursuit and the dao pursuit, specifically speaking, are essentially different or even opposing reflective concerns that render the two major philosophical traditions significantly or even totally different in orientation and agenda. It is noted that these views have been voiced prominently and loudly especially in West and thus have left on many who are not familiar with Chinese philosophy the impression to the effect that there is no truth concern in the traditional Chinese philosophy and that the truth concern in Western tradition and the dao concern in Chinese tradition are dramatically different from each other. I disagree to both of the theses above. My view consists of three related points. First, the dao pursuit of the classical Daoism is essentially the truth pursuit in the way as captured by the point of the TNG thesis, although the truth pursuit manifests itself in distinctive ways in the Chinese and Western philosophical traditions. Second, the NTCP thesis is thus untenable; that is, the truth concern via truth pursuit is indeed one significant concern in the classical Daoism. Third, one aspect of the dao concern, i.e., the truth-pursuing agent aspect of the truth concern, as suggested in Zhuang Zi’s approach, has made a significant contribution to the common truth concern in philosophy. The connection of the three points is this: The second point is based on the first point; the first point is presupposed in the third point; the first and second points are further strengthened and enriched by the third point. My strategy in arguing for the three points is this. First, it is known that the classical Chinese philosophy is not a single one philosophical school but consists of a variety of distinctive movements of thought; it is neither necessary due to the purpose here nor practical due to the space limitation to examine all of various versions of the dao concern in different movements of thought throughout the

332

chapter twelve

classical Chinese philosophy. Rather, I will focus on the relation between the truth pursuit and the dao pursuit in view of the classical Daoism as prominently presented by Lao Zi in the Dao-De-Jing and Zhuang Zi in the Zhuang-Zi. For any general claims concerning the traditional Chinese philosophy not merely cannot ignore the case of Daoism, but rather need to bring it in focus especially on the issue of the dao-concern. In so doing, I do not pretend to exhaust all the orientations and styles of the dao-concern in the traditional Chinese philosophy (such as that in Confucianism). Second, both for the sake of strengthening the case for the first point and for the sake of spelling out some distinct contribution by the classical Daoism to the common truth concern in philosophy, I focus on Zhuang Zi’s account of true man and true knowledge which is related to the truthpursuing agent aspect of the truth concern. While doing this, I will, directly or indirectly, also give responses to some of the preceding seemingly plausible supporting observations for the NTCP thesis.25 3.1. Truth Pursuit as Dao Pursuit in the Dao-De-Jing It is known that one central strategic goal of the classical Daoism is to understand and capture the daoa or pursue the dao. Now what is the dao? The dao is not something mysterious beyond the human understanding. The dao, as characterized in the Dao-De-Jing, is primarily the metaphysical dao; Lao Zi’s characterization of the metaphysical dao reveals one basic Daoist metaphysical insight regarding the structure of the world. The metaphysical dao as unifying force that runs through the whole universe. The dao as root is fundamental;26 the dao as origin is universal in the sense that it is the origin of all things;27 the dao is the one in the above two senses; the dao as power is inherent in nature (in each thing of the universe) rather than transcendent beyond and above nature;28 the dao as source is never exhausted;29 the dao as whole is nature (in the above senses

25 Due to the space limitation, in somewhere else instead of here, I discuss the validity of the third observation concerning the truth predicate and explain how the relevant linguistic phenomena concerning the Chinese truth predicate bear on the way in which the truth concern in the classical Chinese philosophy manifests itself. 26 Cf., the Dao-De-Jing, Chs. 1, 6, 21, 25, 34, 42. 27 Cf., the Dao-De-Jing, Chs. 1, 25, 34, 40, 42. 28 Cf., the Dao-De-Jing, Ch. 42. 29 Cf., the Dao-De-Jing, Chs. 4, 6.

truth pursuit and

DAO

pursuit

333

combined); the dao as the way of nature is the way of yin-yang complementary interaction to reach harmonious balance;30 the dao as the way of existence in time is eternal;31 the dao as the way of existence mode evolves itself and keep changing dynamically;32 the dao as the way of development is spontaneous and natural (because the dao is nature).33 In this way, the metaphysical dao is not something like the platonic Form beyond and above, but consists in, particular things in the universe; all particular things in nature, wan-wub (ten-thousand things),34 are manifestations of the metaphysical dao with individualizedparticularized daos within that render them having power. The relation between the metaphysical dao and its manifestations in wan-wu, or dec in the broad sense of the term as used in the Dao-De-Jing, is essentially yin-yangd complementary: the dao and wan-wu are interdependent, interpenetrating, interactive and correlative, although the force and existence of the metaphysical dao cannot be simply reduced to the sum of (the forces and existences of ) wan-wu. Epistemologically speaking, the metaphysical dao can thus be (partially) captured in our thought and language through capturing wan-wu. The metaphysical dao is thus not something mysterious, which neither metaphysically exists beyond and above nature nor epistemologically goes beyond human understanding. In our pre-theoretic terms, the dao is the universe as nature and its way instead of something mysterious beyond nature; pursing dao is to understand and capture the way things are in nature. Daoism takes pursuing, modeling on, and performing the dao as the fundamental mission of the human being in their reflective inquiry. As Lao Zi emphatically points out: “The human being models him/herself upon earth; earth models itself upon heaven; heaven models itself upon the dao; the dao models itself upon what is natural.”35 In this way, the dao pursuit is the most fundamental dao concern of the classical Daoism. If my view about the nature and function of the TNG thesis as given in the previous part and my account of the nature and mission of the dao concern of philosophical Daoism

30 31 32 33 34 35

Cf., the Dao-De-Jing, Chs. 2, 42, 77. Cf., the Dao-De-Jing, Chs. 4, 6. Cf., the Dao-De-Jing, Ch. 1. Cf., the Dao-De-Jing, Chs. 25, 34. Cf., the Dao-De-Jing, Chs. 4, 5, 8, 16, 34, 37, 42, 62, 64. The Dao-De-Jing, Ch. 25.

chapter twelve

334

are right, the dao-pursuing mission of the classical Daoism in the above sense is essentially a kind of truth-pursuit mission which can be delivered in terms of a Daoist way of presenting the TNG thesis as follows: (TNG***) The Daoist reflective way of presenting the explanatoryreduction version of the thesis of truth centrality as strategic normative goal: Understanding and capturing dao (‘dao’ designates the way things are) is an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries; or, simply: The dao pursuit is an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries.

Several explanatory notes are needed. First, with the distinction between use and mention, although the phrase ‘the paraphraseexplanatory-reduction version’ is only mentioned in contrast to ‘the semantic-ascent version’ which is a by-default version in the West tradition but whose counterpart can hardly be found in the literature of the classical Chinese philosophy, the phrase ‘the paraphraseexplanatory-reduction version’ here is used to refer to what it is supposed to refer to, i.e., its content to the effect that understanding and capturing the way things (as due objects of philosophical studies) are is taken as an explanatory norm to regulate and explain one central strategic goal of philosophical inquiries. In this way, what the Daoist way of presenting the paraphrase-explanatory-reduction version presents is such a content, instead of presupposing the presence of (the counterpart of ) the semantic-ascent version in the classical Chinese philosophy. Second, one might object: isn’t the Daoist understanding of dao so different from some typical or representative understanding of reality or the way things are in the West so that the preceding socalled Daoist reflective way, (TNG***), of presenting the TNG thesis actually talks about something else? As explained before, our pretheoretic understanding of truth and the TNG thesis themselves do not intrinsically commit themselves to any ad hoc metaphysical account or elaboration of reality (i.e., on what counts as reality that would renders truth objective in the aforementioned sense, on what is the structure of reality, etc.), and therefore their metaphysical commitment is minimal in this connection. Rather, the TNG thesis is com-

truth pursuit and

DAO

pursuit

335

patible with, and allows for, various reflective ways of presenting it given that these various ways are reflective ways of talking about the way things are—the Daoist way of talking about the dao is one of these ways. Third, related to the point of the preceding note, it is also noted that we might as well replace the Daoist reflective way of talking about the way things are with one Chinese pre-theoretic way of talking about the way things are via such folk phrases as ‘shi-shi-qiu-shi’e. A reflective way of talking about ‘qiu-dao’f and a folk way of talking about ‘shi-shi-qiu-shi’ are both distinctive Chinese ways of delivering the point of the TNG thesis. Fourth, while Davidson actually subscribes to the paraphraseexplanatory-reduction version of the TNG thesis in an implicit and indirect way, the classical Daoism explicitly and directly delivers the point of the thesis through its characteristic dao-pursuing version that captures the crux of the paraphrase-explanatory-reduction version of the thesis. Now one issue emerges: given that the classical Daoism’s dao-pursuing strategic goal captures the point of the TNG thesis through (TNG***) and thus that the dao pursuit in the classical Daoism is the truth pursuit in the way as captured by the point of the TNG thesis, does the classical Daoism make any substantial contribution through (TNG***) to our reflective understanding of the truth concern? Now it is time to explore merits (if any) of each of the two characteristic versions of the TNG thesis. Generally speaking, each of the two versions of the TNG thesis has its own merits (and actual or potential disadvantages in contrast to the other’s merits). The semantic-ascent version talks about truth in an economic and convenient way by using one-word term ‘truth’ with merely one syllable instead of using some multiple-syllable phrases like ‘[a certain truth bearer] in accordance with [a certain] fact’. Another merit of the semantic-ascent version is this: what is (or is supposed to be) shared, or something common, stable, definite, constant, unchanged and thus universal, in all concrete and particular states of (various truth bearers’) corresponding to the ways things are in the world is highlighted and emphasized in terms of one word ‘be true’ or ‘truth’ via such semantic ascent. In the philosophical context, this approach actually reflects a general being-aspect-concerned methodological orientation of the Western tradition: it tends to focus on the being aspect of an object of study, that is, the aspect of the object that is stable and invariable, unchangeable, definite and

336

chapter twelve

constant—i.e., the being aspect; when what is involved is to capture what is stable, constant and invariable among a number of objects of one kind, this orientation thus tends to focus on what are shared, common and thus universal among them. A reflective methodological perspective that is intended to capture such an orientation to look at an object of study or a number of objects that are somehow related might as well be called ‘the being-aspect-concerned perspective’. In this way, one can say that, besides the aforementioned consideration for economy and convenience, the semantic-ascent version of the thesis as strategic normative goal more or less reflects such a being-aspect-concerned orientation or perspective. If the first merit is more or less instrumental in character, the second merit is quite substantial concerning methodological perspective. In contrast, one merit of the paraphrase-explanatory-reduction version is that it delivers the substantial content of truth pursuit in an explicit, straightforward and illuminative way. Another merit is this. One key phrase ‘understanding and capturing the way things are’ not merely delivers the substantial content of truth pursuit but also implies or points to the concrete and particular aspect of truth pursuit: generally speaking, there is some distinct aspect(s) between the way one thing is and the way another thing is; therefore, generally speaking, the way of capturing the way one thing is differs from the way of capturing the way another thing is. One certainly cannot say that this version thus looses sight of, or is inconsistent with, the general and universal aspect of truth pursuit: the phrase ‘understanding and capturing the way things are’ is an abstract and generalization: what is common among many different truth bearers is their capturing the way things are. Notice that the apparent singular term ‘way’ used in the paraphrase-explanatory-deduction version actually covers both ways: one might as well say that it is used both as a collective noun to cover various particular ways things are and as an abstract term to capture the general character or shared dimension of all these particular ‘correspondence’ ways. In this connection, given the specified meaning of ‘(metaphysical) dao’ in the classical Daoism as characterized before, the Daoist dao-pursuing way of presenting the TNG thesis, as a variant of the paraphrase-explanatory-reduction version in regard to content, hits the point of the unification of both ways. If my preceding discussions of the distinction between the two versions of the TNG thesis and of their respective merits concerning methodological orientation are correct, I consider this as one substantial

truth pursuit and

DAO

pursuit

337

contribution by Daoism to our reflective understanding of the truth concern in philosophical inquiries. There is one more merit of the paraphrase-explanatory-reduction version that is already briefly addressed in one note on (TNG***): while the semantic-ascent version by default suggests a non-agent thing as a ‘truth bearer’ like a sentence, statement, belief or proposition, the paraphrase-explanatory-reduction version suggests, or at least is compatible with, the human agent as the primary truth bearer to this extent: it is eventually the human agent who ‘understands’ or ‘captures’ the way things are in the world and thus who generates, possesses and unifies various true beliefs and thoughts that she actually has. It is noted that the dao-pursuing is not limited to a static understanding of the world; it also includes the agent’s dynamic understanding and her putting into effect the understanding via her action that is regulated by wu-weig (going without being against dynamic nature). The point and significance of this merit will be explained in the next section when Zhuang Zi’s relevant point is discussed. In the following, to further understand and illustrate the truth pursuit in the Dao-De-Jing (that is, how the Daoist classic text raises some reflectively interesting issues in this connection and how it responds to these issues in some reflectively interesting ways if any), let me give a case analysis of one passage in the Dao-De-Jing that raises some philosophically interesting issues concerning the truth/dao pursuit. The passage is from Chapter 54 as follows (my translation): Cultivates virtue within oneself as a whole body and it thus becomes genuine (true) [zhen]; Cultivates virtue in one’s family and that thus becomes to overflow; Cultivates virtue in one’s village and it thus becomes long-lasting; Cultivates virtue in one’s state and it thus becomes abundant; Cultivates virtue in the world and it thus becomes universal. Therefore, look at the oneself by virtue of the oneself; Look at the family by virtue of the family; Look at the village by virtue of the village; Look at the state by virtue of the state; Look at the world by virtue of the world. How do I know the world as it is? By virtue of this. —from Chapter 54

As I see it, there are four interesting points concerning the truth pursuit in this short passage. First, Lao Zi here both implicitly makes

338

chapter twelve

his metaphysical point concerning truth nature as well as explicitly makes his epistemological point concerning truth means in line with our pre-theoretic ‘correspondence’ understanding of truth. In the second part of this citation, Lao Zi explicitly raises the issue of how to know the world as it is (he-yi-zhi-tian-xia-zhi-ranh); the criterion or means by virtue which one can know that, according to Lao Zi, is to examine (guani) the object of knowledge (whether it is the human being oneself or family or state or other things in the world) by virtue of the way the object is in the world. (Surely, as we have seen above, Davidson would disagree due to some epistemological difficulties well known in the Western tradition; nevertheless, for a classical Daoist like Lao Zi and Zhuang Zi, one can somehow know the world as it is through, say, joint functions by various knowing organs which are not limited to those inter-subjective ones like our senses and intellectual mind.36) In so doing, with the dual meaning of zhi j (knowing as the process of knowledge and what is known as the result of knowledge), Lao Zi as a matter of fact makes his metaphysical point concerning truth nature which is to be possessed by the result of knowing: the resulting knowledge capturing the world as it is (zhitian-xia-zhi-rank). Note that, in so doing, Lao Zi does not use any one-Chinese-character counterpart, if any, of the one-syllable-word in Western phonetic languages (say, ‘true’ or ‘truth’ in English)—as a semantic-ascent linguistic means of indicating truth nature—to deliver his insight concerning the truth/dao pursuit; that is exactly one point which this essay is intended to make: the truth concern, generally speaking, and the truth pursuit, specifically speaking, are not necessarily related to any ad hoc semantic-ascent linguistic means whose meaning depends on its due paraphrase explanation. Second, it is interesting enough to note that the Chinese character ‘zhen’l, whose current usage in contemporary Chinese language has made it become a by-default one-character Chinese counterpart of English term ‘truth’ or ‘true’, does appear in the first statement in the cited passage: “Cultivates virtue within oneself and it thus becomes genuine [zhen]”; the other occurrence of the character ‘zhen’ in the Dao-De-Jing is in Chapter 21: “[Dao] Deep and far off, there is the essence within; The essence is highly authentic (true) [qi-jingshen-zhenm], and there is evidence within.” Indeed, the logical subject 36 This is a point that is explicitly made by Zhuang Zi in Inner Chapter 2 “QiWu-Lun” [On Equality of Things].

truth pursuit and

DAO

pursuit

339

of zhen, or the zhen bearer, in these two cases is neither the linguistic sentence or statement nor the propositional content of thought or belief; but, at least in the context of the Daoist classic Dao-DeJing, ‘zhen’ is used to indicate something like reaching or capturing (a high level of ) the way things are. In the former case, only when virtue is cultivated within oneself [via wu-wei], the alleged virtue becomes virtue, which, one can say, means shang-den (genuine virtue, a kind of high level of spontaneous virtue)37 and which is thus a kind of the way the genuine human virtue is. In the latter case, the dao as essence and power of wan-wu ‘highly’ reaches, or very much constitutes, the very way the nature is, which is the dao. In this way, although the ancient usage of the term ‘zhen’ here is different from its primary modern usage as a counterpart of ‘true’ and ‘truth’, what it delivers is essentially along the same line as what our pre-theoretic understanding of truth is to deliver, that is, (the truth nature bearer) reaching or capturing the way things are. To this extent, it is not implausible or too odd to translate the two occurrences of ‘zhen’ in the Dao-DeJing into ‘true’, instead of ‘genuine’ in the former case and ‘authentic’ in the latter case. Third, another interesting issue concerning the truth pursuit raised in the passage is that of the truth of human morality. From Lao Zi’s point of view, the human morality in terms of human virtue, as indicated by the term ‘de’ used in the narrow sense in the DaoDe-Jing, is not something like fixed floating entity that can be imposed upon the moral agent from outside but is cultivated “within and through [the moral agent] oneself as a whole body” (xiu-zhi-yu-sheno); in accordance with the broad sense of ‘de’ referring to manifestations of the metaphysical dao, or individualized daos, in particular things, de as human virtue is the manifestation of metaphysical dao in human beings regarding morality, which renders human beings having ‘power’. In this sense, from Lao Zi’s Daoist point of view, it is not only that the truth of a moral judgment, say, ‘John is a moral person with virtue’, has its objective basis that consists in its capturing the way the moral agent is in regard to her cultivated virtue within and through herself as a whole body, but also that the truth of the moral agent’s virtue itself has its objective basis that consists in its following or “modeling itself on” ( fap) (Chapter 25) the dao in the

37

Cf., the Dao-De-Jing, Ch. 38.

340

chapter twelve

way of wu-wei (a Daoist way of presenting the point of capturing the way things are in this context). (It is noted that the latter insight above actually provides a due basis for explaining how it is possible for human virtue to possess truth, an issue is induced by Lao zi’s idea “[human virtue] thus becomes true [qi-de-nai-zhenq]” in his first statement of the cited passage, as mentioned in the preceding second point.) In this way, Lao Zi’s dao-pursuing approach does not exclude but intrinsically includes the moral-truth pursuit in the above sense. Fourth, the foregoing second and third points are actually related to another interesting issue concerning the truth pursuit in philosophy, i.e., the issue of eligible truth bearers, as already more or less addressed in the preceding discussions of the two points. The crux of the issue is this: given that truth nature consists in understanding and capturing the way things are, whether due truth bearers can be only such mental things with conceptual contents as thoughts and beliefs and their linguistic expressions like sentences and statements or can also be other human things like human virtue and the human agent herself as a whole, and how those aforementioned truth-bearer candidates if any are related. Lao Zi indirectly makes his positive point concerning this issue as he favorably talks about both kinds of truth bearers. Let me further address this issue in the next section where I discuss how another important classical Daoist philosopher, Zhuang Zi, explicitly addresses the issue in his account of true man and true knowledge. Before moving onto the subject of the next section on Zhaung Zi’s account, let me give a brief summary of the central point I have endeavored to make in this section. Although the trade-mark version of the TNG thesis is its semantic-ascent version especially in the Western tradition, and although the semantic-ascent version does have its merits, the paraphrase-explanatory-reduction version of the thesis, in my opinion, is more illuminative and hit the point in one crucial aspect. In this crucial aspect, the dao pursuit of the classical Daoism is essentially the truth pursuit in general terms, though it is presented in the classical Daoism in a characteristic way. To further understand and illustrate the truth pursuit in the Daoist classic text Dao-De-Jing, I have then given a case analysis of one passage from Chapter 54 which, as I see it, explicitly or implicitly makes some philosophically interesting points concerning the truth pursuit in philosophy.

truth pursuit and

DAO

pursuit

341

3.2. Zhuang Zi on True Man and True Knowledge: An Account of Truth-Pursuing Agent Dimension of Truth Concern In the preceding section, when I give a general characterization of the dao concern as the truth concern of philosophical Daoism and a case analysis of one passage of the Dao-De-Jing, both discussions end with one point that is somehow related to Zhuang Zi’s accout of truth-pursuing agent, which in my opinion constitutes one significant contribution by Daoism to our understanding of the truth concern. Before looking at Zhuang Zi’s account, I intend to emphasize two points that are related to having a due understanding of his account. First, one needs to understand Zhuang Zi’s general methodological strategy, a kind of objective perspectivism, which consists of two significant and related points.38 (1) Each thing has its various aspects, and one can take a finite perspective [as a working perspective] to look at one aspect: one can look at its this aspect, from a this-aspectconcerned perspective, and sees it as a this, and one can also look at its that aspect, from a that-aspect-concerned perspective, and sees it as a that. Its metaphysical foundation is this: various aspects, the this aspect and the that aspect, ontologically depend on each other; 38 The two points are given in the following passage (my translation) in the Zhuang-Zi: Everything has its that aspect and its this aspect. One cannot see the this aspect of one thing if one looks at the thing from the perspective of the that aspect; one can see the this aspect if one looks at the thing from the perspective of the this aspect. One thus can say that the that and the this come from each other . . . Thus, the sage does not limited to looking merely at the this or that aspect [from the finite point of view] but looks at all the aspects of the thing in the light of Nature. The this is also the that, and the that is also the this. The that has one criterion of right and wrong, while the this has one criterion of right and wrong. Is there really a distinction between the that and the this? . . . When the this aspect and the that aspect cease to be viewed as opposite, it is called ‘the pivot of taking Dao point of view’. One’s capturing the pivot is like one’s standing the center around which all things revolve in endless change: one thus can deal with endless change from the Dao point of view, among whose many manifestations the this and the that are. Therefore it is said that the best way is to look at things in the light [of Nature].—Inner Chapter 2 “Qi-Wu-Lun” [On Equality of Things] In contrast to some interpretations that take this passage as one crucial textual evidence for Zhuang Zi’s alleged radical “anything goes” relativism or a relativistic perspectivism, I think Zhuang Zi’s point here is essentially a kind of objective perspectivism. For, instead of ‘any perspective can go’, Zhuang Zi bases relevance and eligibility of a perspective ( given an object of study) on whether it points to some aspect that is really or objectively possessed by the object of study.

342

chapter twelve

various perspectives, the this-aspect-concerned perspective and the that-aspect-concerned perspective, thus actually complement each other. (2) For the purpose of looking at the connection of various aspects of a thing and/or of having a comprehensive understanding of the thing, Zhuang Zi also encourages us to look at things from a higher point of view which transcends various finite points of view; in this way, those different aspects cease to be viewed as opposite or incompatible but complementary. That is, Zhuang Zi emphasizes a holistic understanding of the world that transcends various local perspectives (at least in one’s background thinking), though it is totally legitimate or even expected for one to take a certain perspective as working perspective depending on one’s purpose. With the understanding of these two strategic methodological points of Zhuang Zi’s objective perspectivism, one can effectively understand Zhuang Zi’s substantial approaches to various issues including the current issue under examination: one might as well say that the latter constitute implementations and illustrations of the former. In the following, to enable the reader to have a close look at Zhuang Zi’s original narrative account of true man and true knowledge, I first make the citations of some relevant passages from Inner Chapter 6 “Da-Zong-Shi” with certain needed paraphrases in bracket parentheses; and then I give an interpretation of Zhuang Zi’s relevant points in the context of Daoism and his whole thought, especially in view of his general methodological strategy to look at various issues as characterized above. Now let us take a close look at how Zhuang Zi makes his point in the text.39 The one who knows what Heaven [Tian] does and what the human does has reached the utmost. The one who knows what Heaven does live with the Heaven. The person who knows what the human does use the knowledge of what one knows to support the knowledge of what one does not know, and one thus completes one’s natural span of life without dying young half way [completely following the dao without failing half way]. This is knowledge at its greatness. However, there is one difficulty. Knowledge must have what it waits for [as its objective basis] and be then applicable, and what it waits for is changeable. How can I know that what I call ‘Heaven’ is not really the human, and what I call ‘the human’ is not really Heaven? [The key

39 Excerpts from Inner Chapter 6 “Da-Zong-Shi” of the Zhuang-Zi, translated by this author.

truth pursuit and

DAO

pursuit

343

to overcome this difficulty is this.] One needs to first become a true man [zhen-renr] and thus has one’s true knowledge [zhen-zhi s ] [that would be sensitive to what is changeable]. What is meant by an ‘true man’? The true man in ancient times did not reject [but were sensitive to] what is little, did not brag about achievements, and did not scheme things [against being natural]. A man like this would not regret it when missing something and would not be complacent when having achievements. A man like this would not feel frightened when climbing the high places, would not feel soaked when entering the water, and would not feel hot when going through fire. [He would not be restricted by apparent limits but transcend them with his vision.] Such is the knowledge by which one can climb all way up on the course of the dao. . . . The true man in ancient times . . . regarded knowledge as product of time and de as what is based on. . . . [For the true man] To regard knowledge as product of time means that he needs to respond to situations and changes as if he could not keep from doing it. To regard de as what is based on is as if the one with two feet needs to walk on one’s way to climb a hill, and the true man makes his diligent efforts to do so. . . . The person who is called a ‘true man’ renders Heaven and the human in accord instead of overcoming each other.

Indeed, given that the English term ‘true’ is used here in line with our pre-theoretic understanding of truth, it does not appear immediately plausible to talk about ‘true man’ or translate ‘zhen-ren’ into ‘true man’. For, after all, we usually consider the bearer of truth to be such mental things as thoughts and beliefs or their linguistic expressions like sentences and statements. (I guess that, with this consideration as a presupposition, some translations avoid translating the term ‘zhen’ in ‘zhen-ren’ here into ‘true’ but some terms else, e.g., ‘authentic’ or ‘genuine’.) However, with the foregoing analysis and clarification of the point of the TNG thesis via its paraphraseexplanatory-reduction version, and if my account there is correct, one thing is certain: given that the truth (nature) as delivered in our pre-theoretic understanding of truth consists in (the truth bearer) understanding or capturing the way things are, it should be neither implausible nor odd to talk about ‘true man’ when ‘true’ is along the line of our pre-theoretic understanding of truth. For it indeed makes sense to say that the subject (or even the primary subject in a certain sense to be explained below) of understanding or capturing the way things are is the human agent, or the thinking creature, instead of some non-thinking thing. But, at this point, two further

344

chapter twelve

questions emerge. First, does Zhuang Zi talk about (mention) ‘zhenren’ in the sense of ‘zhen’ that delivers our pre-theoretic understanding of truth? Second, given that it is plausible or does make sense to interpret Zhuang Zi’s talk about ‘zhen-ren’ into the talk about ‘true man’ in this context, is there any serious reflective need or any theoretic significance to highlight the conception of true man as Zhuang Zi does? Or is this just a kind of insignificant, though innocent, rhetoric saying? As for the first question, it is clear that, in this context, Zhuang Zi relates ‘true knowledge’ with ‘true man’ and uses ‘zhen’ in both case along the same line of our pre-theoretic understanding of truth, that is, capturing (understanding, corresponding to, etc.) the way things are (the way Heaven, the human, etc. are). In this way, I intentionally translate ‘zhen’ in ‘zhen-ren’ into ‘true’ to deliver one point of Zhuang Zi’s account: the bearer of the truth nature, or the subject of understanding and capturing the way things are can be, and is, the human agent in a certain sense, and this is completely compatible with other kinds of the truth nearer (the propositional content of belief or its linguistic expression). The second question above is more reflectively interesting: is there any serious reflective need or any theoretic significance to highlight the conception of true man as Zhuang Zi does? My answer is yes. I think this is exactly where Zhuang Zi’s account of true man and true knowledge would make some significant contribution to our understanding of the truth concern in philosophy. One crucial claim of Zhuang Zi’s account in regard to the relation between true man and true knowledge is this: “One needs to first become a true man [zhen-ren] and thus has one’s true knowledge [zhen-zhi ]”. The point of Zhuang Zi’s claim and its significance if any need to be placed in the textual context and in view of his whole thought. The passages around the claim show that Zhuang Zi addresses some related metaphysical and epistemological issues in that context. First, metaphysically speaking, the object of knowledge is changeable; true knowledge of the object thus needs to be regarded as product of time in accordance with change of the object; but it is the human subject, instead of thought or its linguistic expression as the definite and stable result of previous knowing process, who can be directly sensitive and respond to situation and change. Second, epistemologically speaking, it indicates how to achieve knowledge at its greatness that is comprehensive and holistic (say, both Heaven and the human, various aspects of the object). It is the human subject, instead of piecemeal individual

truth pursuit and

DAO

pursuit

345

beliefs and their linguistic expressions per se, who can autonomously and creatively transcend the limitations of individual beliefs and their linguistic expressions, unify her various individual beliefs into a holistic and comprehensive understanding of the way things are, and thus overcome some epistemological difficulties which it can be hardly overcome by looking at piecemeal individual beliefs and their linguistic expressions. In this sense, to this extent, and for the sake of achieving true knowledge that captures the changing world, one needs to first become a true man who can be sensitive and respond to situation and change (or, more accurately, the changing, dynamic and becoming aspect of the thing as the object of knowledge). In this way, through his conception of true man and his account of the relation between true man and true knowledge, Zhuang Zi actually captures and highlights the pragmatic, becoming, dynamic dimension/aspect/layer of the truth concern involved in philosophy of language, metaphysics and epistemology. One significance of Zhuang Zi’s point is this: (1) from the point of view of philosophy of language, his point calls our attention to the pragmatic dimension of the linguistic truth bearer that involves the speaker’s intention and her situated uses, instead of the semantic dimension alone; (2) from the point of view of metaphysics, his point calls our attention to the becoming aspect of the object of knowledge, instead of the being aspect alone, for the sake of a holistic understanding of all aspects of the way things are; (3) from the point of view of epistemology, his point calls our attention to the dynamic layer, instead of the static-stable layer, of the whole process of capturing the way things are. Zhuang Zi’s view thus enlarges and enriches the reflective concept of correspondence (with reality) as traditionally treated through his conception of true man. One might ask: isn’t it not merely innocent but also more conceptually effective to talk about the propositional content of a belief or its linguistic expression alone as the truth bearer? Does Zhuang Zi indiscriminately render absolutely superior the order of first becoming true man and then achieving true knowledge? Indeed, Zhuang Zi does not directly provide his response to such reflectively interesting questions in the text. Nevertheless, one can base on the point of Zhuang Zi’s general methodological strategy in treating various reflective issues (as given at the outset of this section) and the basic point of Daoist thought about the metaphysical dao (as characterized in the previous section) to provide an adequatel elaboration of the

346

chapter twelve

due implication of his general methodological strategy to the current issue. As emphasized above, the crux of Zhuang Zi’s claim needs to be placed in the textual context and his whole thought which would help us identify for which sake and for what purpose Zhuang Zi takes a certain perspective. From the foregoing discussion, one can see that Zhuang Zi intends to capture the pragmatic dimension of the belief or linguistic truth bearer, the becoming aspect of the object of true knowledge, and the dynamic layer of the process of capturing the way things are. The fact per se that one actually focuses on the becoming aspect and takes a becoming-aspect concerned perspective as one’s working perspective implies neither that one would deny other relevant perspectives as relevant nor that one has an inadequate guiding principle that renders one’s current working perspective absolutely superior while the other relevant perspective irrelevant or absolutely inferior.40 Generally speaking, Zhuang Zi is certainly not so unintelligible that he could fail to realize the being aspects of things: a thing always keeps its own certain identity at any stage of its changing process before, or unless, this thing turns to be something else; changes do not happen in chaos but follow certain ways, and the dao is considered as one fundamental and unifying way throughout the universe. Specifically speaking, in the cited passages where Zhuang Zi gives his account of true man and true knowledge with emphasis on the pragmatic, becoming and dynamic aspect involved in the truth concern, it is arguably correct that Zhuang Zi implicitly presupposes the presence of the semantic, being and relatively-stable dimension/aspect/layer involved in the truth concern. Moreover, both Zhuang Zi’s own reflective practice and his general ‘thing-equality’ methodology are not merely compatible but consistently suggest that, for another sake, one needs to first has one’s true knowledge [zhen-zhi ] and thus become a true man [zhen-ren]: given the wisdom-generating role and character-cultivating role played by one’s true knowledge (including moral knowledge as well as intellectual knowledge), one needs to first achieve or resort to true knowledge for the sake of becoming true man in the sense

40 For a detailed discussion of this methodological point, see my article “An Analysis of the Structure of Philosophical Methodology—in View of Comparative Philosophy”, in Bo Mou ed. (2001), Two Roads to Wisdom?—Chinese and Analytic Philosophical Traditions, Open Court, especially its Part 3, pp. 351–6.

truth pursuit and

DAO

pursuit

347

that, in so doing, one is able to have a (more) comprehensive understanding of the world and cultivate oneself in a right direction and with an adequate guidance. Zhuang Zi’s own teachings per se would help people fulfill this. In sum, one important implication of Zhuang Zi’s ‘things-equality’ (qi-wut) methodological strategy in treating the issue of the truth/dao concern is this: given that the dao-pursuing enterprise has both the static aspect and dynamic aspect, both unchanging and changing aspects, and both being aspect and becoming aspect, they are metaphysically equal in the sense that they depend on each other and are yin-yang complementary, and the becoming-aspect concerned perspective and the being-aspect concerned perspective in our journey of pursuing truth/dao are methodologically equal in the sense that both are relevant, indispensable and yin-yang complementary for a holistic understanding of the issue. In this way, Zhuang Zi’s contribution also lies in his general methodological approach that can be extended or applied to how to look at the relation between various dimensions, and their related projects, of the truth-concern enterprise in philosophy. In sum, Zhuang Zi’s contribution to the truth-concern enterprise in philosophy is dual: one is his substantial contribution to the project that is concerned with the truth-pursuing agent dimension of the truth concern; the other lies in his general methodological contribution to how to look at the relation between various dimensions of the truth concern. * In the preceding discussion, I have endeavored to do the following things. First, through examining Davidson’s approach and drawing on three distinctions, I have argued that, despite some unclear or even misleading expressions of the TNG thesis, its due point is that truth (nature) as understanding and capturing the way things are is an explanatory norm to regulate and explain one strategic goal of philosophical inquiries; especially I have examined the two characteristic versions of the TNG thesis, that is, its semantic-ascent version and its paraphrase-explanatory-reduction version, and explained why the latter delivers the foregoing point of the thesis in a more explicit and effective way. Second, based on the preceding understanding of the point of the TNG thesis, and in view of its paraphrase-explanatoryreduction version, I have argued that the dao pursuit of the classical Daoism is essentially the truth pursuit in the way as captured by the point of the TNG thesis: both render the pursuit of understanding

348

chapter twelve

and capturing the world (the way the things as due objects of studies are) normative in regulating one strategic goal of philosophical inquiries; from this point, I have further shown how the classical Daoism as presented in the Daoist classical text, the Dao-De-Jing and the Zhuang-Zi, can make its substantial contribution to our understanding of the truth concern, as one major cross-cultural common concern of philosophy, in some philosophically interesting ways.

truth pursuit and

DAO

Chinese Glossary a. b. c. d. e. f. g. h. i. j. k. l. m. n. o. p. q. r. s. t.

dao wan-wu de yin-yang shi-shi-qiu-shi qiu-dao wu-wei he-yi-zhi-tian-xia-zhi-ran guan zhi zhi-tian-xia-zhi-ran zhen qi-jing-shen-zhen shang-de xiu-zhi-yu-shen fa qi-de-nai-zhen zhen-ren zhen-zhi qi-wu

pursuit

349

INDEX OF NAMES AND SUBJECTS*

(* The English translations of non-English terms, primarily Chinese ones, are given either following the relevant authors’ paraphrases in certain contexts or merely tentatively for the heuristic sake, especially when the concepts are under reflective scrutinization in this volume. The transliterations of Chinese terms are given their originals when available.) absolutism, 39–40, 45 aesthetics/aesthetic, 109 Allinson, Robert E., 73n, 230, 232–3, 245–6, 238n Ames, Roger, 50n, 65–7, 109, 119, 229n, 330n Analects < > (Lun Yü), 30, 50, 107–8, 110, 243–4, 247–68, 244 analytic method / methodological approach, 5 and passim. philosophy, 1–9 and passim. Angle, Stephen C., 23, 73–100 Anscombe, G.E.M., 193n Aquinas, St Thomas, 300 Aristotle/Aristotelianism, 1, 79, 85–6, 98, 180n atomism, 297 aufheben (sublate), 148 Augustine, 300 Austin, J.L., 248n, 266n Barton, John, 212n Batson, C.D., 113n Benedict, Ruth, 42–3, 45–6 Biagioli, Mario, 97–8 Bloom, Alfred, 253, 255 Bourdieu, Pierre, 97n Brandom, Robert, 23, 74, 76n, 86, 89–98, 198n Bruce, J. Percy, 118 Buddhism, 229–31 Chan / Zen, 63n, 137–52, 158–60, 299 Bunnin, Nicholas, 73 Carnap, Rudolf, 278 Chan, Wing-tsit (Chen, Rong-jie), 50n charity, 24–5, 103–62 Davidson’s principle of charity. See Davidson

Cheng, Chung-ying , 27, 32, 271–308 Chen, Cheryl, 48n Chen, Guying , 306n Chen, Lei, 265n Chinese philosophy, 1–33, 50–1, 55–71, 79–81, 107–16, 117–60, 165, 197, 229–46, 247–9, 251–68, 271–3, 277–308, 309–10, 329–48. See also Confucianism, Daoism, Buddhism, etc. Choi, Incheol, 64n Chong, Kim-chong , 29–30, 229–46 Church, Alonzo, 278 commensurability vs. incommensurability, 37–53, 55–71, 73–98, 185–8, 197–203 comparative philosophy constructive-engagement enterprise in, 1–33 and passim. meta-philosophical examination of, 1–19, 55–71, 73–98. Also see commensurability, conceptual scheme, and relativism compatibility, 240–2 conceptual scheme, 3, 20–3, 28, 37–100, 117–36 Conee, Earle, 311n Confucianism (Ru, Ru-Jia, Ru School), 55, 60, 67–8, 80–1, 98, 106–14, 236, 283, 290, 299–304 Confucius (Kong Zi; Kong, Qiu ), 79, 107–8, 110, 243–6, 302, 304 consistency, 166, 169–70, 240–2 convention, 76, 82–6, 90, 95–6, 189, 248–9, 260–2, 266–8 Cooper, David, 104–5 Cua, Antonio S., 109, 237n, 242n Cutrofello, Andrew, 132

352

index of names and subjects

Dai, Zhen , 244–5 dao , 20, 31–3, 51, 158–9, 279–80, 284, 290, 299, 332–40 Dao-De-Jing < > (the Lao-Zi ), 33, 303, 305–6, 310, 332–40, 348 Daoism , 55, 229–246, 290, 297, 299, 309–10, 313, 329–48 metaphysical dao of, 305–7, 332–6 dao pursuit of 31–33, 309–10, 329–49 Darley, J.M., 113 Davidson, Donald, 1–9, 19–33, 37–53, 55–63, 70–1, 73–98, 103–6, 117–52, 165–88, 189–203, 209–11, 229–33, 245–6, 247–51, 258–68, 271–88, 294, 307–8, 309–32, 347 on conceptual scheme and relativism, principle of charity, 23–5, 28, 48, 77, 91, 93, 103–6, 117–52, 197–8, 199, 300 de (obtainment of dao as power source in particular things or manifestation of dao in particular things; virtue), 333, 339 Derrida, Jacques, 129, 157, 227 Descartes, Rene, 1, 71n, 137, 300 descriptive, 189–96, 253. See also normativity Doris, John, 113n dualism (dichotony), 43–6, 141, 150–1, 300 Dummett, Michael, 30, 248–9, 258–60, 264–5 Dworkin, Ronald, 210

Gadamer, Hans-Georg, 66n Galileo, 98 Genova, A.C., 132 Gibbard, Allan, 113 God, 65–6, 119, 138, 300n gong-fu , 144 Gong-sun Long , 128 Gong-Sun-Long-Zi < >, 128 Graham, Angus, C., 117–36, 153–7, 227n, 252n, 329n Grandy, Richard, 77–8 Grice, H.P., 208, 224n

elenctic method (elenchus), 5 empiricism, 44, 278 epistemology, 71, 155, 278, 300n onto-, 287, 296, 300 equilibrium. See harmony ethics, 25–6, 67–71, 107–11, 114, 165–88. Confucian, 67–71, 107–11, 114. See also Confucianism rational foundation for, 165–88 of cultures, 186–8

incommensurability. See commensurability interpretation, 3, 29, 12–5, 40, 73–98, 112–5, 117–36, 166–70, 197–203, 207–27, 229–68, 292–304 intuition, 64n, 136. prajñà-, see prajñà irrationality, 138–42 Isen, A.M., 113n

fallibilism, 39–40 feminist, 38, 53 Frege, Gottlob, 260 Fung, Yiu-ming 117–162

, 24–5,

Hacking, Ian, 3, 80n, 209–10, 224n Hajek, Alan, 173n Hall, David L., 65–7, 109, 119, 329n, 330n Hanna, Patricia, 46–7 Hansen, Chad, 59n, 60n, 63–4, 128–9, 155, 227n, 253–5, 330n Harbsmeier, Christoph, 155, 265 Harman, Gilbert, 113n harmony, 67, 110–1, 298 Harrison, Bernard, 46–7 heaven, 65–6, 279, 290–300, 303, 306 Heidegger, Martin, 66n, 297 Hempel, Carl G., 168n hermeneutics, 271, 305 onto-, 271, 281, 288–308 Hobbes, Thomas, 226 holism, 202, 308 Hollis, Martin, 77–8, 94 Hopkins, Gerard Manley, 215–9 humanity, human nature, 240–2

Jakobson, Roman, 118, 120 Jenkins, Keith, 227n jing (respect, reverence), 107, 110 Johnson, M., 30, 229–30, 234, 245 Kahneman, Daniel, 175n Kant, Immanuel (Kantianism), 25–6,

index of names and subjects 61, 130–1, 133–4, 165–70, 176–7, 181–3, 300 Kasulis, Thomas P., 56, 66–7 ke-ji (overcoming the self ), 114 Klein, Gary, 115n knowledge. See epistemology. Krausz, Michael, 22, 37–53 Kripke, Saul, 80n Kuhn, Thomas, 37n, 81, 157, 130, 197 Lakoff, G., 30, 229–30, 234, 245 language, Chinese ideographic, 37, 41–7, 52, 57–9, 61n, 73–98 grammatical approach, 247–9, 255 Hopi, 42–7, 52 meaning and truth in Chinese, 271–308 philosophy of, passim. pragmatic approach, 89–98, 247–68 semantic approach via truth, 27–8, 274–8. Also see relevant references in Davidson understanding and interpretation of. See understanding and interpretation Western phonetic (Indo-European), 42–7, 52, 61n, 273n syntax, 254, 261, 287, 289, 298–9, 254, 261 Lao Zi (Lao Tzu; Lao, Dan ), 305, 332–40 Lao-Zi < >. See the Dao-De-Jing Larsen, Richard, 184n Larson, Gerald James, 55–6, 71n Lau, D.C., 50n, 110, 251n Le Gall, S., 118, 155 Legge, James, 110 Levin, P.F., 113n Lewis, David, 85n, 95 Leys, Simon, 251n, 259, 262n li (propriety, rite), 109, 114 li (reason, principle), 118, 286, 299, 301 Li-Ji , 304 Li, Zhi , 264 Lin, Shuen-fu, 237 li-yi-fen-shu (one principle with its many manifestations), 303 liberalism, 111 logic, 63–5, 71, 91n, 117–31. Also see consistency and reasoning

353

Lu, Xiang-shan , 283 Lu, Yiwei, 264n Lukes, Steven, 77–81 Luo, Ji, 257n Lun Yü < >. See Analects McDowell, John, 44 MacIntyre, Alasdair, 41, 50–1, 77–81, 98 Margolis, Joseph, 38 Martinich, Aloysius P., 29, 207–227, 248n–249n Marxist, 38 meaning, 3, 20, 27–31, 73–99, 120, 123, 173n, 207–27, 229–68, 271–308 Mencius (Meng Zi) , 241, 244 Mencius (Meng-Zi ) < >, 304 Metaphor, 3, 29–30, 118, 229–46 metaphysics, 71, 277, 279, 300n. methodology (method), analytic. See analytic. comparative, 1–19 and passim. hermeneutical. See hermeneutics mind, 3, 130–2, 137–50, 284. Also see xin (heart-mind). ming (destiny, fate), 299 Mohism , 128, 236 Mohist Canon (Mo-Jing) < >, 128 monism, anomalous, 284 Moody-Adams, Michele, 105 Mou, Bo , 1–33, 309–49 Möller, Hans-Georg, 233n Nagel, Thomas, 132, 176–7, 181–3 naturalism, 289 Nichols, S., 64n Nietzsche, Friedrich, 297 Nisbett, Richard, 64n Norenzayan, Ara, 64n normativity / normative, 20, 25–6, 86, 96–7, 189–203, 253 objectivity, 3, 44 Ockham’s razor, 190, 199 Oshima, Harold, 237n paradox, 13n, 238 Parkes, Graham, 56 Parmenides, 300 Peirce, C.S., 278 Peng, Kaiping, 64n

354

index of names and subjects

perspectivism, 341 Pettit, Philip, 115n phenomenalist, 278 Plato, 1, 137, 266, 300 Popper, Karl, 40–1 Potter, Karl H., 68n pragmatism / pragmatic, 30, 179, 247–68, 298 prajñà, 140–2 probability conditional-, 91n, 168–75 Putnam, Hilary, 52, 137 qi (vital energies), 118, 155–6, 299 qing (emotion), 304 Qian, Zhongshu , 229, 230 Quine, W.V., 60n, 93, 166, 178–9, 180n, 191, 273–4, 276, 278, 282, 284–5 Ramberg, Bjorn T., 81–6, 95 Rand, Ayn, 111–5 rationality, 3, 20, 24–6, 112, 138–42, 144–50, 165–88 realism, 46–7, 132, 137, 284–7, 289–90, 295, 302, 308 reality, 3, 137, 289–90, 292, 298, 302, 323–8, 332–40. Also see metaphysical dao in dao reasoning, 3, 89–98, 112–5, 165–88, 247. Also see logic. relativism, 20–2, 28, 37–53, 60n, 73, 139, 142, 155, 179, 185–6 relativity / relativization, 170–6 ren (humaneness, humanity), 50, 70–1, 114, 290, 302 Richardson, Henry, 106 Rosemont, Henry, 50n, 60, 66n, 67n Russell, Bertrand, 249n Ryle, Gilbert, 129, 157 samsara, 140 satori, 140, 142, 149–50 self, 234, 239 non-, 139–40 Sellars, Wilfrid, 96 semiology / semiotics, 119, 120 si (thinking), 302 Sider, Theodeore, 311n Simon, H.A., 115n skepticism, 39, 137–9, 144, 187 Skinner, Quentin, 226 Slingerland, Edward, 230, 234–5, 239–46

Stich, S., 64n Strauss, Leo, 212n Strawson, P.F., 133, 248, 249n Stroud, Barry, 132, 134 Structuralism, 129 sunyata, 141 Suzuki, D.T., 63n, 140–50, 158–60 syntax. See language tai-ji (supreme ultimate), 299–301 Tanaka, Koji, 22, 55–71 Tarski, Alfred, 27–8, 32, 271–2, 274–7, 288, 307 ti (physical body), ben(Origin-Body; origins and their manifestations), 303, 305 -yong (origins and their applications), 301 tian (Heaven), 65–6, 299, 301. See also heaven transcendence translation / translatability, 22, 41, 47–53, 57–9, 61n, 73–98, 117–31, 153–7 truth, 3, 20, 27–8, 31–3, 84–6, 165–88, 253, 271–349 Twersky, Amos, 175n utilitarianism, 179, 181, 185 understanding, 3, 37, 41, 47–53 Van Norden, Bryan W., 73n vijnana, 141 virtue, 107–11, 114, 339–40 Wagner, Rudolf, 227n Waley, Arthur, 50n Wang Bi, , 321 Watson, Burton, 238n, 240n Wei-Kuan , 158 Weinberg, J.M., 64n Western philosophy, 1–9 and passim. See also Davidson Wheeler, Samuel C., 25–6, 76n, 91n, 98n, 165–88, Whitehead, Alfred North, 286, 297 Whorf, Benjamin Lee, 42–3, 45–6, 57, 155 Williams, Bernard, 266 Wittgenstein, Ludwig, 80n, 91n, 96, 129, 157, 231–2, 245–6 Wong, David B. , 24, 50–1, 103–16 wu (to have not, nonbeing), 299

index of names and subjects wu (things) wan(ten thousand things), 156, 333 wu-ji (the no-ultimate), 299 wu-wei (nonaction), 337, 340 xiao (filial piety), 107–8 Xiao, Yang , 30, 247–68 xin (heart, mind, heart-mind), 236, 242, 289, 299, 304 xing (predispositions of the heart, nature), 299, 304 Xun Zi , 109, 240–2, 246 Xun-Zi < >, 230, 240, 242, 244 yi (change; transformation), 299 Yi-Jing < > Zhou-Yi < > (the Yi-Jing text in the classical sense), 295, 297, 299, 303

355

Yi-Zhuan < > (the commentaries on the Zhou-Yi ), 299, 303 yin-yang , 119–24, 286, 300–1, 333 Yu, Jiyuan , 73 Zheng, Yujian , 26, 189–203 zhi-ji (knowing), 302 zhi (directions of the heart; will), 302 zhong , 302 Zhong-Yong < > (Doctrine of the Mean, Centrality and Commonality), 303 Zhou, Dunyi (Chou Tun-I), 300n Zhu Xi , 303 Zhuang Zi (Chuang Tzu; Zhuang, Zhou ), 59n, 60n, 229–46, 331–2, 340–8 Zhuang-Zi < >, 33, 229–46, 310, 332, 341–8