Beyond the Analytic-Continental Divide: Pluralist Philosophy in the Twenty-First Century (Routledge Studies in Contemporary Philosophy) [1 ed.] 9781138787360

Table of contents :
Cover
Title
Copyright
Contents
1 Introduction: Contemporary Philosophy as Synthetic Philosophy
PART I Methodologies
2 The Emergence of the Concept of the Analytic Tradition as a Form of Philosophical Self-Consciousness
3 Philosophy as Articulation: Austin and Deleuze on Conceptual Analysis
4 Conceptual Genealogy for Analytic Philosophy
PART II Truth and Meaning
5 Truth and Epoché: The Semantic Conception of Truth in Phenomenology
6 From Difference-Maker to Truthmaker (and Back)
7 Reasons, Epistemic Truth, and History: Foucault’s Criticism of Putnam’s Anti-Realism
8 Metaphor without Meanings: Derrida and Davidson as Complementary
PART III Metaphysics and Ontology
9 Why Is Time Different from Space?
10 Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein: Thinking Language Bounding World
11 The Answer to the Question of Being
PART IV Values, Personhood, and Agency
12 Relativism and Recognition
13 Revolutionary Actions and Events
14 Varieties of Shared Intentionality: Tomasello and Classical Phenomenology
List of Contributors
Index

Citation preview

Beyond the Analytic-Continental Divide

This forward-thinking collection presents new work that looks beyond the division between the analytic and continental philosophical traditions—one that has long caused dissension, mutual distrust, and institutional barriers to the development of common concerns and problems. Rather than rehearsing the causes of the divide, contributors draw upon the problems, methods, and results of both traditions to show what post-divide philosophical work looks like in practice. Ranging from metaphysics and philosophy of mind to political philosophy and ethics, the papers gathered here bring into mutual dialogue a wide range of recent and contemporary thinkers and confront leading problems common to both traditions, including methodology, ontology, meaning, truth, values, and personhood. Collectively, these essays show that it is already possible to foresee a future for philosophical thought and practice no longer determined either as “analytic” or as “continental,” but, instead, as a pluralistic synthesis of what is best in both traditions. The new work assembled here shows how the problems, projects, and ambitions of twentieth-century philosophy are already being taken up and productively transformed to produce new insights, questions, and methods for philosophy today. Jeffrey A. Bell is Professor of Philosophy at Southeastern Louisiana University, USA. He is the author of Deleuze’s Hume (2009) and Philosophy at the Edge of Chaos (2006). Andrew Cutrofello is Professor of Philosophy at Loyola University Chicago, USA. He is the author of Continental Philosophy: A Contemporary Introduction (2005). Paul M. Livingston is Associate Professor of Philosophy at the University of New Mexico, USA. His most recent book is The Politics of Logic: Badiou, Wittgenstein and the Consequences of Formalism (2011).

Routledge Studies in Contemporary Philosophy

For a full list of titles in this series, please visit www.routledge.com

37 Contemporary Feminist Pragmatism Edited by Maurice Hamington and Celia Bardwell-Jones 38 Morality, Self Knowledge, and Human Suffering An Essay on The Loss of Confidence in the World Josep Corbi 39 Contrastivism in Philosophy Edited by Martijn Blaauw 40 Aesthetics After Metaphysics From Mimesis to Metaphor Miguel de Beistegui 41 Foundations of Freedom Welfare-Based Arguments against Paternalism Simon R. Clarke 42 Pittsburgh School of Philosophy Sellars, McDowell, Brandom Chauncey Maher 43 Reference and Structure in the Philosophy of Language A Defense of the Russellian Orthodoxy Arthur Sullivan

44 Civic Virtue and the Sovereignty of Evil Derek Edyvane 45 Philosophy of Language and Webs of Information Heimir Geirsson 46 Disagreement and Skepticism Edited by Diego E. Machuca 47 Philosophy in Schools An Introduction for Philosophers and Teachers Edited by Sara Goering, Nicholas J. Shudak, and Thomas E. Wartenberg 48 A Philosophy of Material Culture Action, Function, and Mind Beth Preston 49 A Philosophy of the Screenplay Ted Nannicelli 50 Race, Philosophy, and Film Edited by Mary K. Bloodsworth-Lugo and Dan Flory 51 Knowledge, Virtue, and Action Essays on Putting Epistemic Virtues to Work Edited by Tim Henning and David P. Schweikard

52 The Ontology of Psychology Questioning Foundations in the Philosophy of Mind Linda A.W. Brakel 53 Pragmatism, Law, and Language Edited by Graham Hubbs and Douglas Lind 54 Contemporary Dualism A Defense Edited by Andrea Lavazza and Howard M. Robinson 55 Reframing the Intercultural Dialogue on Human Rights A Philosophical Approach Jeffrey Flynn 56 How History Matters to Philosophy Reconsidering Philosophy’s Past After Positivism Robert C. Scharff 57 The Affordable Care Act Decision Philosophical and Legal Implications Edited by Fritz Allhoff and Mark Hall 58 Realism, Science, and Pragmatism Edited by Kenneth R. Westphal 59 Evidentialism and Epistemic Justification Kevin McCain 60 Democracy in Contemporary Confucian Philosophy David Elstein

61 Deleuze and Pragmatism Edited by Sean Bowden, Simone Bignall, and Paul Patton 62 Mind, Language and Subjectivity Minimal Content and the Theory of Thought Nicholas Georgalis 63 Believing Against the Evidence Agency and the Ethics of Belief Miriam Schleifer McCormick 64 The Essence of the Self In Defense of the Simple View of Personal Identity Geoffrey Madell 65 Personal Autonomy and Social Oppression Philosophical Perspectives Edited by Marina A.L. Oshana 66 Domination and Global Political Justice Conceptual, Historical, and Institutional Perspectives Edited by Barbara Buckinx, Jonathan Trejo-Mathys, and Timothy Waligore 67 Hate Speech Law A Philosophical Examination Alexander Brown 68 Music and Aesthetic Reality Formalism and the Limits of Description By Nick Zangwill 69 Beyond the Analytic-Continental Divide Pluralist Philosophy in the Twenty-First Century Edited by Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston

This page intentionally left blank

Beyond the Analytic-Continental Divide Pluralist Philosophy in the Twenty-First Century Edited by Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston

First published 2016 by Routledge 711 Third Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2016 Taylor & Francis The right of the editors to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Beyond the analytic-continental divide : pluralist philosophy in the twentyfirst century / edited by Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston. — 1 [edition].    pages cm. — (Routledge studies in contemporary philosophy ; 69)   Includes bibliographical references and index.   1.  Philosophy, Modern—21st century.  2.  Analysis (Philosophy)  3.  Continental philosophy.  I.  Bell, Jeffrey A, editor.   B805.B38 2015   190.9'051—dc23   2015014644 ISBN: 978-1-138-78736-0 (hbk) ISBN: 978-1-315-76662-1 (ebk) Typeset in Sabon by Apex CoVantage, LLC

Contents

  1 Introduction: Contemporary Philosophy as Synthetic Philosophy

1

JEFFREY A. BELL, ANDREW CUTROFELLO, AND PAUL M. LIVINGSTON

PART I Methodologies   2 The Emergence of the Concept of the Analytic Tradition as a Form of Philosophical Self-Consciousness

17

JAMES CONANT

  3 Philosophy as Articulation: Austin and Deleuze on Conceptual Analysis

59

RICHARD ELDRIDGE AND TAMSIN LORRAINE

  4 Conceptual Genealogy for Analytic Philosophy

75

CATARINA DUTILH NOVAES

PART II Truth and Meaning   5 Truth and Epoché: The Semantic Conception of Truth in Phenomenology

111

DAVID WOODRUFF SMITH

  6 From Difference-Maker to Truthmaker (and Back) JEFFREY A. BELL

129

viii  Contents   7 Reasons, Epistemic Truth, and History: Foucault’s Criticism of Putnam’s Anti-Realism

151

LEE BRAVER

  8 Metaphor without Meanings: Derrida and Davidson as Complementary

172

SAMUEL C. WHEELER III

PART III Metaphysics and Ontology   9 Why Is Time Different from Space?

193

JOHN MCCUMBER

10 Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein: Thinking Language Bounding World

222

PAUL M. LIVINGSTON

11 The Answer to the Question of Being

249

GRAHAM PRIEST

PART IV Values, Personhood, and Agency 12 Relativism and Recognition

261

CAROL ROVANE

13 Revolutionary Actions and Events

287

ANDREW CUTROFELLO

14 Varieties of Shared Intentionality: Tomasello and Classical Phenomenology

305

DAN ZAHAVI AND GLENDA SATNE

List of Contributors Index

327 329

1 Introduction Contemporary Philosophy as Synthetic Philosophy Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston

The thirteen new essays collected here, written from diverse perspectives, address many more or less interrelated topics of contemporary philosophical discussion and inquiry. What they most centrally share, though, is a common orientation toward the contemporary reality of what might be called synthetic philosophy: philosophical work that situates itself beyond the problematic “divide” between the “analytic” and “continental” traditions, which still largely shapes and constrains philosophical work in the English-speaking world.1,2 As students of academic philosophy learn, typically relatively early in their careers, most of the work that goes on in many departments—at least most of the time—remains recognizably “analytic,” in drawing on the work of philosophers such as Gottlob Frege, Bertrand Russell, and W.V.O. Quine. This is work that—characteristically but not exclusively—privileges logic, practices or responds to projects of linguistic or conceptual analysis, and often (though again not invariably) models philosophical inquiry itself as methodologically analogous to, or continuous with, empirical-scientific inquiry into the natural world. As beginning students also quickly discover, analytic philosophy in this sense is routinely contrasted, in the pedagogical practices as well as sociological divisions of academic philosophers, with what is called “continental” philosophy. This is—as a quick characterization might run—philosophy that draws on the work of thinkers such as Georg Wilhelm Friedrich Hegel, Edmund Husserl, Martin Heidegger, and Jacques Derrida, and develops diverse projects and methods (such as dialectics, phenomenology, hermeneutics, existentialism, Marxism, critical theory, and deconstruction). These projects and methods are, at least at first glance, generally more or less distinct from those characteristic of the “analytic” tradition. The starkness and apparent exhaustiveness of the division as it exists and is often maintained in academic philosophy today can present students on their way to specialization with a difficult and problematic choice: either to work in areas and pursue topics and questions to which they are not naturally drawn, or to risk marginalization or dismissal by pursuing the further development of the extant work that they see as most promising.

2  Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston At the same time, students whose abilities or inclinations make them more comfortable with “analytic” methods and concerns often undergo the frustrating experience of recognizing the relevance of texts and thinkers situated outside the analytic tradition, but lacking—due to the one-sidedness of their training—the resources seriously to engage with them. More broadly, though it is of relatively recent origin (coming to characterize academic philosophy generally only as recently as the 1960s), the division between the traditions has caused dissension, mutual distrust, and institutional barriers to the development of common concerns and problems among working philosophers and so has significantly limited, in many cases, the range and fruitfulness of philosophical discussions and debates. The essays that comprise this volume attempt to move beyond this problematic situation by construing philosophy as a cosmopolitan endeavor that should not be constrained by any a priori presumption of metaphilosophical division. To this end, each of the essays draws centrally on problems, methods, and results of both traditions to produce new work across a variety of areas. The contemporary possibility of such work and its concrete reality today, as indicated and exemplified by the essays, also points toward a future of philosophical thought in the twenty-first century, no longer determined as “analytic” or “continental” but, instead, as a pluralistic synthesis of much of what is best in the diverse legacy of twentieth-century approaches. In drawing in this way on the best outcomes of these methodologically diverse twentieth-century traditions, the new work assembled here thus shows how the characteristic problems, projects, and ambitions of twentieth-century philosophy are already being taken up and productively transformed today. Collectively, they thereby point toward a future of philosophical thought and research that brings the best results and most significant outcomes of both traditions to bear in synthetic fashion on problems with which both have been deeply concerned. In so doing, they also indicate some of the possible forms of a philosophical future of discussion and inquiry that substantially inherits both traditions but is not limited to either one, and so point toward, at least in broad outline, some of the distinctive concerns and broad contours that might characterize much good philosophical work in general over the next several decades.3 The past twenty-five years have witnessed a large number of analyses of the causes or origins of the analytic-continental divide, or of its persistence in contemporary academic philosophy.4 Going back even further and continuing through the present, there have also been many attempts at what may be understood as “bridge-building”: projects that seek to make figures, themes, or topics identified with one or the other of the traditions available and intelligible to those trained, and at home, in the other.5 We do not dispute the usefulness and importance of either of these two kinds of projects, but the approach of this volume is different from either of them. Rather than attempting simply to comprehend the divide or place into dialogue two traditions still conceived as essentially distinct, the essays in this volume situate

Introduction  3 themselves beyond the divide in that they operate in a context wherein the divide is no longer assumed to determine or constrain philosophical thought and inquiry. Here, the right metaphor is no longer, we think, that of building bridges or routes of access between two distinct and (as it is often assumed) mutually incommensurable traditions. Rather, a better picture would be that of the confluence of two streams that, having a common origin and trajectory for much of their course, have recently diverged and run roughly in parallel for a while (although sometimes at significant distance from each other), before converging once again.6 In thus situating themselves in the dynamic confluence of the two traditions and realizing this confluence as a reality of contemporary thought, the essays of this volume also move beyond work that is simply “comparative” in the sense of identifying static similarities and differences of doctrine, theory, or result across the two twentieth-century traditions. Instead, in developing the implications of problems that are posed by the joint legacy of the two traditions as sites for contemporary philosophical research and thought, the essays in this volume pose and re-open longstanding philosophical questions and provide positive directives for future work on them in a pluralist context beyond the limitations of the analytic-continental divide, a context that is already—as the work collected here attests—recognizably “ours” today. Doing so also involves reflecting the reality of the rapidly changing situation of the institutions and forms of academic philosophy today, as new technologies and media increasingly shape and modulate philosophical discussion, both within and without academia itself. This volume has its proximal origin in an online symposium that took place in 2013 on the philosophy blog NewAPPS, devoted to the discussion of a paper by one of the editors (Livingston) and including extensive critical commentary by another editor (Bell) as well as one of the current contributors (Novaes). The discussion of the paper—on connections between Derrida’s deconstructive notion of the “undecidable” and the classic formalization of that notion in the context of the axiomatic theory of arithmetic by Kurt Gödel—drew in comments by a wide variety of discussants, ranging from graduate students to professional analytic as well as continental philosophers, and thereby bearing witness to the actual possibility (though also some of the limitations and inherent obstacles) of pluralist philosophical discussion on the web. In the spirit of this discussion and its contemporary relevance—but also in the belief that the more traditional media of essay and book still provide the essential conditions for the development of philosophical problems and questions—the essays collected here all draw together currents of recent and contemporary philosophy usually held apart and illustrate some of the genuine emerging possibilities for synthetic philosophy today. In addition to drawing together themes, topics, and concrete problems of recent philosophy on both sides of the divide, the issues developed in the various essays that comprise this volume already point toward at least some

4  Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston of the likely contours of philosophical discussions to come. In the remainder of this introduction, we adumbrate briefly some of these contours, as they are already evident in various and overlapping ways in the texts collected here, and as they point toward problems and questions likely to be discussed much more, in a pluralist context, over the next several decades. These contours, as we shall see, pass through a wide variety of topical areas and also crosscut the fourfold division of (rough) topic areas we have adopted for the organization of the essays in the volume itself. Their eclectic traversal of the various topic areas into which academic philosophical activity is traditionally divided indexes, we think, not only the possibility but the contemporary reality, evident in the approaches and arguments of the essays here included, of new kinds of philosophical discussion and practice beyond the divide.7 1.  SENSE BEYOND LANGUAGE: AFTER THE LINGUISTIC TURN In 1967, Richard Rorty used the phrase “The Linguistic Turn” (adopted from Gustav Bergmann) as the title for an anthology of papers by a variety of representatives of (what was then just becoming defined as) analytic philosophy.8 The anthology comprised essays by, and about, representatives of logical positivism and the “ordinary language” school of philosophical analysis, as well the newer project of applying insights from scientific linguistics to philosophical and linguistic analysis. As Rorty suggested in the volume’s introduction, these partisans of the turn to language might be identified as those holding the view that “philosophical problems . . . may be solved (or dissolved) either by reforming language, or by understanding more about the language we presently use,” and this view was (at the time of Rorty’s writing) taken by many of those proponents as “the most important philosophical discovery of our time.”9 Beyond giving unity and shape to the “analytic” tradition for many—though by no means all—of its practitioners, this kind of commitment to the revolutionary implications of language and the analysis of its structure for traditional philosophical problems has also played a decisive role for many (though again not all) of the diverse projects grouped as “continental” philosophy in the twentieth century. For structuralism, philosophical hermeneutics, phenomenology, critical theory, and many of the varieties of “post-structuralism” (to name just a few), reflection on language and its structure has provided both a key methodological resource and an essential thematic dimension, determining or articulating projects and problems, as well as the prospects for their solution, resolution, or transformation. Today, however, many philosophers on both sides of the divide have come to see their projects as located, in important ways, beyond or after the linguistic turn. This does not necessarily mean that they philosophize as if the turn had never occurred, or that they repudiate or challenge the important and varied results (for instance those about linguistic structure,

Introduction  5 meaning, reference and sense) that have emerged from it. But it does mean that, while philosophers continue to employ methods and presuppose results drawn from the linguistic turn, there is a greater sense today of the genuine depth of the questions that are inherently involved in the use of language as a philosophical resource or the analysis of its structure as a means of resolving philosophical problems.10 For instance, one is today likely to see analyses of referential and other forms of meaning undertaken—on the “analytic” side—as analyses of the structure of mental or cognitive, rather than strictly linguistic, content.11 Likewise, one is today more likely to see projects of analysis and critical reflection on the “continental” side taking up in a renewed way the broader questions of the relationship of language to being and the world, rather than simply assuming this application, as in the forms of linguistic idealism or constructivism that have characterized some such projects in the past.12 One way to receive the linguistic turn in this partially transformed context is to see its methods and results as pointing toward the possibility of a broader conception of the foundations of meaning, concepts, and their analysis. This kind of conception draws on the implications and the results of specifically linguistic analysis but generalizes them beyond language itself or considers their broader bearing on questions of the basis of meaning or sense, without prejudice to its linguistic foundation or determination. In the current volume, David Woodruff Smith’s essay, “Truth and Epoché: The Semantic Conception of Truth in Phenomenology,” might be considered exemplary in this regard. In the essay, Smith shows how the “semantic” conception of truth developed by Alfred Tarski on the basis of his formal analysis of the definition and structure of truth-predicates for specific formal languages can be generalized to provide a phenomenological analysis of the phenomenon of truth as it occurs in experiencing consciousness, without prejudice to its linguistic or non-linguistic status. Here, then, and for projects of this sort more broadly, it is not taken for granted either that the analysis of linguistic structure must be rigorously separated from the analysis of phenomenological experience, or that one kind of analysis must serve as a foundation for the other. Instead, the focus is on developing the forms and structures that underlie linguistic meaning specifically alongside homologous or shared ones that plausibly underlie the phenomenon of meaning or sense in broader or less language-specific terms. Also along these lines, Richard Eldridge and Tamsin Lorraine’s essay, “Philosophy as Articulation: Austin and Deleuze on Conceptual Analysis,” suggests that the ordinary-language analyses of J.L. Austin might be related to the more temporally fluid and less language-specific considerations of sense proposed by Gilles Deleuze, for instance in his important 1968 analysis of the logical and metaphysical preconditions for structural meaning, The Logic of Sense.13 Jeffrey A. Bell’s essay, “From Difference-Maker to Truthmaker (and Back),” considers how problems recently discussed by analytic philosophers about the form and nature of

6  Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston “truthmakers”—whatever makes the propositions expressed in (linguistic) sentences true—might be resolved by appealing to Deleuze’s idea of a more basic and non-linguistic (as Bell suggests) underlying metaphysics of differentiation or “difference-making.” Samuel C. Wheeler III’s essay, “Metaphor without Meanings: Derrida and Davidson as Complementary,” considers how the different versions of the analysis of language’s structure pursued by Donald Davidson and Jacques Derrida nevertheless lead the two philosophers to complementary views about the relationship of metaphor to metaphysics. And Paul M. Livingston’s essay, “Wittgenstein Reads Heidegger, Heidegger reads Wittgenstein: Thinking Language Bounding World,” considers the relationship between two of the twentieth-century philosophers who are often considered most central to the linguistic turn, in its analytic and continental variants, respectively. The brief commentaries of these two philosophers on each other, he argues, point to a broader set of problems about the relationship of sense to the structure and totality of the world as such. These problems, he argues, are not simply “about” language as a historical phenomenon or human artifact, but just as much about how the metaphysics of sense is related to the structure of finitude and the infinite, a question that remains very much open today as one of the (various) legacies of the linguistic turn. 2.  REALISM(S) AND THEIR OTHERS If, as has often been noted, the analytic tradition has at least one of its origins in Russell and Moore’s realist rebellion against Hegel and the British forms of post-Hegelian idealism that then represented the dominant philosophical approach at Cambridge, the subsequent analytic tradition has also developed rigorous methods and formalizations for discussing the implications of realism in various domains, as well as the various kinds of positions opposed to it.14 There are, for example, many different rigorously formulated positions of realism, whether “naïve,” “metaphysical,” “scientific,” or “internal” (to name a few), and various positions of “anti-realism,” ranging from verificationism and intuitionism to diverse forms of pragmatism, “coherence” theories of truth and knowledge, linguistic idealism, and other positions captured formally by what Michael Dummett described broadly as “anti-realism” of one type or another. On the “continental” side, twentieth-century projects have often—though by no means invariably—conceived of language, truth, meaning, and value as decisively constrained or determined by the forms and temporality of human life. This has lent some of these projects a distinctively constructivist or “anti-realist” flavor, even where they do not simply replicate classical forms of idealism or subjectivism.15 Recent years have witnessed, however, a renewed discussion of realism and the positions opposed to it, alongside a renewed and sharpened critique of many of the forms of subjectivism and

Introduction  7 constructivism that characterized many twentieth-century continental positions and interpretations of continental philosophers.16 Many of the essays included here address the broad issue of the possibility and stakes of realism and anti-realism today. For example, Lee Braver’s “Reasons, Epistemic Truth, and History: Foucault’s Critique of Putnam’s Anti-Realism” considers the relationship between Putnam’s “internal realism” (actually a variety of anti-realism, according to Braver) and Foucault’s project of historical genealogy. While the two positions have common roots in Kant’s rejection of realism in favor of transcendental idealism, Braver argues, Foucault’s position goes farther than Putnam’s in recognizing the meta-ethical and normative problems posed by realist attitudes and positions insofar as they tend to naturalize or obscure asymmetric power relations, and thereby more clearly affirms the actual political and ethical stakes of anti-realist projects. Other articles, by contrast, argue for realism about sense, content, or truth: David Woodruff Smith’s essay, for example, indicates how an attitude of realism about the basis of truth might be combined with the classical methods of Husserlian phenomenology, despite this project’s (oft-noted) “subjectivist” commitments and orientation. Jeffrey A. Bell’s essay, in a partially similar vein, considers how the phenomenon of truth might have realist but nevertheless pre-linguistic foundations in a metaphysics of underlying difference, and Paul M. Livingston’s essay considers the specific problem of realism about the world—understood as the totality of all that exists—as such. Another essay that considers some of the stakes and implications of realism about the world—this time in a more explicitly ethical/political register—is Carol Rovane’s “Relativism and Recognition.” Here, Rovane considers the implications of commitment to the existence of a single world in which all disputes over values and ethics take place and have their consequences, as opposed to pictures on which there are an irreducible multiplicity of worlds, and argues (against standard rejections of the ‘relativism’ apparently involved in them) that pictures of the latter sort provide distinctive and valuable resources for the logical, practical, and metaphysical consideration of ethical issues and debates. 3. THE ‘RETURNS’ OF METAPHYSICS: CRITIQUE OF THE CRITIQUE Another characteristic feature of much twentieth-century philosophy, for both analytic and continental philosophers, has been a renewed and linguistically motivated pursuit of the critique of metaphysics originally inspired, in many ways, by Kant. Both the forms of this critique and the specific projects it has supported have been diverse, ranging from the Vienna Circle’s attempt to “overcome” metaphysical thinking through the logical clarification of language to Heidegger and Derrida’s discussions of the contemporary closure or exhaustion of the tradition of metaphysics begun with

8  Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston the Greeks and the critical challenge posed by Adorno and other critical theorists to “identity thinking.” Recent years, however, have witnessed a variety of “returns” of metaphysical theorizing for both analytic and continental thinkers. On the analytic side, in the wake of developments of the implications of modal logic and philosophical naturalism by writers such as David Lewis and W.V.O. Quine, philosophers have recoiled against the older attempt to eliminate metaphysics in favor of the philosophy of language, and recent methodological discussions have taken up the question of the relationship of linguistic reflection to metaphysical theory in a renewed spirit. Meanwhile, contemporary continental philosophers such as Alain Badiou and Slavoj Žižek, moving beyond the terms of the Kantian critique, have emphasized the contemporary possibilities for a resurgence of metaphysical thought about ontology and the subject, while those who take up Gilles Deleuze’s philosophical thinking of difference (in this volume and elsewhere) expand upon the ideas of a contemporary philosopher who always considered himself a “pure metaphysician.”17 Several of the essays in this volume bear witness to the varied forms of this contemporary return of metaphysics. John McCumber’s essay “Why is Time Different from Space?” takes up a classical problem of metaphysical theory—that of the nature and relationship of time and space—from a perspective informed both by the outcomes of twentieth-century phenomenological analyses and by “analytic” thinking about the problems of tense, change, and becoming. The essays by Livingston and Bell (discussed above) take up the question of the metaphysical structure of the world as underlying the possibilities of sense and truth. Graham Priest’s essay “The Answer to the Question of Being” argues that an answer to Heidegger’s signature question of the meaning and truth of being—in Priest’s formulation, the question of what makes anything be—can be found by drawing on the resources of non-classical (in particular paraconsistent) logic. Other papers in the volume extend and develop the implications of the twentieth-century critique of metaphysics in a contemporary pluralist context. Wheeler’s paper on Derrida and Davidson, for instance, considers some of the ways in which these two philosophers’ respective considerations of metaphor unites them in the critique of the metaphysical assumption that Quine called the “museum myth” of meanings—the assumption that meanings are substantial entities individually correlative to words and open to an intellectual gaze analogous to vision.18 4.  TRUTH AND TIME As has often been noted, one of the (many) ways in which typical practitioners of analytic philosophy may often be distinguished from their “continental” peers is in terms of their differing attitudes toward the temporality and historicity of knowledge and truth. While analytic philosophers, whether

Introduction  9 they conceive their project as an analysis of logically structured concepts or as continuous with empirical, natural-scientific inquiry, tend often to conceive of the topics of their inquiry in static or ahistorical terms, continental philosophers have often emphasized the inherent historicity of knowledge claims and the temporal transformation of claims to truth across historical periods, languages, or changing conditions of power and culture. This deep-cutting methodological distinction may be seen as having roots in, on the one hand, Frege’s conception of propositional senses and logical laws as atemporal idealities, independent in themselves of history, culture, or temporal becoming, and, on the other, the emphasis placed by continental philosophers including Derrida, Deleuze, and (perhaps above all) Heidegger on what they see as the inherent connections among being, truth, and temporal change and becoming. The question or problem indicated here can also, of course, be taken up as an explicit and theoretical one: if (as it seems) it is no longer tenable to construe the conceptual, semantic, normative, or regulative truths in which philosophers are characteristically interested as existing (as Frege suggested) in a timeless “third realm” separate from the realms of physical existence and lived experience, how then should we construe their relationship to time, change, becoming, and history? Of course, as for virtually any claim about “the” difference between analytic and continental philosophy, there are exceptions to the rule that analytic thinkers treat truths as atemporal and continental thinkers treat them as temporally situated.19 Nevertheless, as several of the essays in this volume suggest, the question of how we should understand the relationship of truth to time, especially when considering philosophical modes of reflection that are at least partly directed toward concepts and their nature and basis, seems an appropriate and potentially fruitful one for the development of new methods of philosophical inquiry and practice in a contemporary and synthetic context. James Conant’s essay, “The Emergence of the Concept of the Analytic Tradition as a Form of Philosophical Self-Consciousness,” considers some of the ways that analytic philosophy has been, and is being, defined and maintained today and argues for a continuation of the tradition in a mode of more explicit self-awareness of its own traditional contours and specific history. Along somewhat similar lines, in “Conceptual Genealogy for Analytic Philosophy,” Catarina Dutilh Novaes defines and defends a methodology of historical and conceptual “genealogy” drawn partly from Foucault, arguing in particular that such a methodology is well-suited for conceptual analysis in an “analytic” mode, or for a future continuation of that mode. Again somewhat similarly, Lorraine and Eldridge argue for the compatibility of Deleuze’s explicitly temporal modes of analysis and reflection on the basis of sense with the conceptual-analytic methodology of J. L. Austin’s variety of “ordinary language” philosophy. Finally, both John McCumber’s essay and Andrew Cutrofello’s “Revolutionary Actions and Events” address the question of how we might understand time, change,

10  Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston and becoming themselves in a context shaped by the joint reception of contemporary analytic and continental discussions. 5. NEW (PROBLEMATIC) FOUNDATIONS FOR ACTION AND PRACTICE If it is thus already possible today to envision a transformed philosophical discussion that inherits both the analytic and continental traditions in their twentieth-century developments, this discussion will inevitably suggest new directions for critical and applied thought about the contemporary problems of social, political, and ethical thought and action. Here, although there is no single political program or axiological orientation to be discerned, it is thus already possible to see some of the ways in which an explicitly pluralist philosophical discussion might offer to contribute to the understanding and resolution of the problems of collective practice that (arguably) matter most today in the contemporary context of rapid globalization, multiculturalism, the dominance of capitalist forms of economic production and activity, and technologically mediated forms of life. Several of the essays collected here explicitly take up these problems, considering how they might be addressed within and by the pluralist philosophical projects that are already being pursued today. Andrew Cutrofello’s essay, “Revolutionary Actions and Events,” for instance, draws on the work of Hannah Arendt, Alain Badiou, and David Lewis, among others, to consider the structure of politically transformative events, those capable of bringing about fundamental structural changes in existing political communities and regimes. Carol Rovane’s essay (discussed above) takes up the problem of how to understand the possible rationality of genuine disagreement on values between distinct cultures or traditions. Dan Zahavi and Glenda Satne’s essay, “Varieties of Shared Intentionality: Tomasello and Classical Phenomenology,” draws on both contemporary psychological research and classical phenomenology to consider the foundations of the collective intentionality at the basis of shared communities, values, projects, and practices. The essays by Livingston and Braver, finally, offer suggestions for the contemporary continuation of critical theory in its dialogue with, and struggle against, prevalent contemporary forms of economic, political, and social dominance and power. NOTES   1 We do not attempt here to characterize in general terms the (large amount of) high-quality recent and contemporary non-Anglophone philosophical work that has not yet been translated into English. All of the essays in the current volume are written originally in English, but this should not be taken as implying any endorsement of the general predominance often accorded in

Introduction  11 the institutions and practices of philosophy worldwide to English-speaking philosophy today. To the contrary, we regret this predominance and hope for a philosophical future that is more pluralistic, not only thematically and methodologically, but also linguistically.   2 One of the editors—Livingston—first heard the term “synthetic philosophy” used in roughly this sense (and as a partial contrast to “analytic philosophy”) by Brent Kalar in the spring of 2008. Compare also the title and argument of Thomson (2012).   3 In thus suggesting that the future in philosophy may turn in part on the development of “synthetic” philosophy in the sense of philosophy that draws on both of the two twentieth-century traditions considered here, we do not intend to make any claim of exclusiveness of this inheritance. For instance, philosophy in the twenty-first century appears likely to draw just as deeply, and importantly, on various non-Western traditions, including the diverse traditions of Indian, Chinese, and other varieties of Asian philosophy (among others). We do not mean to exclude these traditions from the broader discussion envisioned here, but only to indicate the (necessarily but unfortunately) limited scope of the present volume with respect to them.  4 See, e.g., Critchley (2001), Cutrofello (2005), Dummett (1993), Føllesdal (1997), Friedman (2000), Glock (2008), Hacker (1997, 1998), Hylton (1990), McCumber (2001, 2011), Ross (1998), Steinbock (2014), West (2010), and the essays collected in Biletzki and Matar (1998), Glock (1997), Prado (2003) and Reynolds et al. (2010).  5 See, e.g., Bell (1998), Brandom (2002), Dreyfus (1991); McDowell (1996), Redding (2007), Rorty (1991), Tugendhat (1982).   6 For a related image, see Dummett (1993, p. 26). In particular, Dummett compares the two movements of phenomenology and “analytical philosophy,” at their beginning, to “the Rhine and the Danube, which rise quite close to one another and for a time pursue roughly parallel courses, only to diverge in utterly different directions and flow into different seas.” (Though we agree with this description of the beginning of the divide, we think that the present moment in philosophy witnesses rather the possibility of the two streams once more running together).   7 For another listing of contemporary aspects of philosophical discussion and research that develops some partially parallel themes, see Leiter (2004).  8 Rorty (1967). For the attribution of the phrase to Bergmann, see Rorty’s “Wittgenstein, Heidegger, and the Reification of Language” in Rorty (1991). For another discussion of the extent to which contemporary philosophy may be said to be beyond the linguistic turn, also taking a cue from Rorty’s title and introduction, see Williamson (2004).   9 Rorty (1967, p. 3). 10 For one consideration of some of these problems as they occur in the appeals to language made by twentieth-century philosophers on both sides of the divide, see Livingston (2008). 11 An early and influential example of this kind of analysis is Evans (1982). For more recent examples, see e.g., McDowell (1996) and Chalmers (2002). For a recent defense of the analysis of intentional content (as opposed to its linguistic structure) that also establishes some connections to the phenomenological tradition, see Crane (2014). 12 In Being and Event (Badiou 1988), for example, Alain Badiou develops an elaborate consideration, based partly on set theory and model theory, of the structure and limits of the language characteristic of a particular historical situation, and the possibilities of its transformation by means of what he terms an “event” (see Andrew Cutrofello’s essay in this volume for some discussion). 13 Deleuze (1968).

12  Jeffrey A. Bell, Andrew Cutrofello, and Paul M. Livingston 14 For the story of Moore and Russell’s rebellion against Hegel and idealism, see Hylton (1990) and Glock (2008), among others. 15 For an analysis of the projects of Heidegger, Foucault, and Derrida (among others) as anti-realist in this sense, see Braver (2007). 16 See, e.g., Badiou (1988, 2006), Meillassoux (2008), Hägglund (2008), and many of the essays collected in Bryant, Srnicek and Harman (2011). 17 Villani (1999); cited in Smith and Protevi (2013). 18 Quine (1968), pp. 186–88. 19 In particular, Husserl and Badiou might both be considered exceptions to the general “rule” that continental philosophers see truths as irreducibly temporally situated, insofar as both (in very different ways) adopt conceptions of (at least some) truths according to which they are “ideal” in the sense of being eternal or atemporal. And on the “analytic” side, philosophers such as Kuhn, Rorty, and Brandom have treated truths as having an important and irreducible diachronic and historical dimension.

WORKS CITED Badiou, A. 1988. Being and Event. New York: Continuum, 2005. Badiou, A. 2006. Logics of Worlds: Being and Event, 2. London: Continuum, 2009. Bell, J. A. 1998. The Problem of Difference: Phenomenology and Poststructuralism. Toronto: University of Toronto Press. Biletzki, A. and Anat Matar, eds. 1998. The Story of Analytic Philosophy: Plot and Heroes. London: Routledge. Brandom, R. 2002. Tales of the Mighty Dead: Historical Essays in the Metaphysics of Intentionality. Cambridge, MA: Harvard University Press. Braver, L. 2007. A Thing of This World: A History of Continental Anti-Realism. Evanston, IL: Northwestern University Press. Bryant, L., N. Srnicek, and G. Harman, eds. 2011. The Speculative Turn: Continental Materialism and Realism. Melbourne: re.press. Chalmers, D. 2002. The Components of Content. In David J. Chalmers, ed., Philosophy of Mind: Classical and Contemporary Readings. Oxford: Oxford University Press, pp. 608–633. Crane, T. 2014. Aspects of Psychologism. Cambridge, MA: Harvard U. Press. Critchley, S. 2001. Continental Philosophy: A Very Short Introduction. Oxford: Oxford University Press. Cutrofello, A. 2005. Continental Philosophy: A Contemporary Introduction. New York: Routledge. Deleuze, G. 1968. The Logic of Sense. New York: Columbia University Press, 1990. Dreyfus, H. 1991. Being-in-the-World: A Commentary on Heidegger’s Being and Time, Division I. Cambridge, MA: MIT Press. Dummett, M. 1993. The Origins of Analytical Philosophy. Reprint Edition. Cambridge, MA: Harvard University Press, 1996. Evans, G. 1982. The Varieties of Reference. Ed. J. McDowell. Oxford: Oxford University Press. Føllesdal, D. 1997. Analytic Philosophy: What Is It and Why Should One Engage in Tt? In H. J. Glock, ed., The Rise of Analytic Philosophy. Oxford: Blackwell, 1–16. Friedman, M. 2000. A Parting of the Ways: Carnap, Cassirer, and Heidegger. La Salle, IL: Open Court. Glock, H.-J., ed. 1997. The Rise of Analytic Philosophy. Oxford: Blackwell. Glock, H.-J. 2008. What Is Analytic Philosophy? New York: Cambridge University Press.

Introduction  13 Hacker, P. M. S. 1997. The Rise of Twentieth Century Analytic Philosophy. In H. J. Glock, ed., The Rise of Analytic Philosophy. Oxford: Blackwell, 51-76. Hacker, P. M. S. 1998. Analytic Philosophy: What, Whence, and Whither? In A. Biletzki and A. Matar, ed., The Story of Analytic Philosophy: Plot and Heroes. New York: Routledge, 3-34. Hägglund, M. 2008. Radical Atheism: Derrida and the Time of Life. Stanford, CA: Stanford University Press. Hylton, P. 1990. Russell, Idealism, and the Emergence of Analytic Philosophy. Oxford University Press. Leiter, B. 2004. Introduction. In B. Leiter, ed., The Future for Philosophy. Oxford University Press, 1–23. Livingston, P. 2008. Philosophy and the Vision of Language. New York: Routledge. McCumber, J. 2001. Time in the Ditch: American Philosophy and the McCarthy Era. Evanston, IL: Northwestern University Press. McCumber, J. 2011. Time and Philosophy: A History of Continental Thought. New York: Acumen.: McGill-Queens University Press. McDowell, J. 1996. Mind and World. Cambridge, MA: Harvard University Press. Meillassoux, Q. 2008. After Finitude: An Essay on the Necessity of Contingency. New York: Continuum. Prado, C. G., ed. 2003. A House Divided: Comparing Analytic and Continental Philosophy. Amherst, NY: Humanity Books. Quine, W. V. O. 1968. Ontological Relativity. The Journal of Philosophy 65 (7): 185–212. Redding, P. 2007. Analytic Philosophy and the Return of Hegelian Thought. Cambridge: Cambridge University Press. Reynolds, Jack, James Chase, James Williams, and Edwin Mares, ed. 2010. Postanalytic and Metacontinental: Crossing Philosophical Divides. New York: Continuum. Rorty, R., ed. 1967. The Linguistic Turn: Essays in Philosophical Method. Chicago: University of Chicago Press. Reprinted edition, with two retrospective essays. 1992. Rorty, R. 1991. Essays on Heidegger and Others (Philosophical Papers, volume 2). Cambridge: Cambridge University Press. Ross, J. 1998. Analytical philosophy as a matter of style. In A. Biletzki and A. Matar, eds., The Story of Analytic Philosophy: Plot and Heroes. New York: Routledge, 56-70 Smith, Daniel and John Protevi. 2013. Gilles Deleuze. In Edward N. Zalta, ed., The Stanford Encyclopedia of Philosophy (Spring 2013 Edition).http://plato.stan ford.edu/archives/spr2013/entries/deleuze. Steinbock, A. J. 2014. “SPEP Co-Director’s Address: SPEP and the Continental Divide.” The Journal of Speculative Philosophy 28(3): 256–272. Thomson, I. 2012. “In the Future Philosophy Will be Neither Continental nor Analytic but Synthetic: Toward a Promiscuous Miscegenation of (all) Philosophical Traditions and Styles.” The Southern Journal of Philosophy 50(2): 191–205. Tugendhat, E. 1982. Traditional and Analytical Philosophy: Lectures on the Philosophy of Language. Cambridge: Cambridge University Press. Villani, Arnaud, 1999. La guêpe et l’orchidée: Essai sur Gilles Deleuze. Paris: Belin. West, D. 2010. Continental Philosophy: An Introduction. Malden, MA: Polity. Williamson, T. 2004. Past the linguistic turn? In Brian Leiter, ed., The Future for Philosophy. Oxford University Press, 106–128.

This page intentionally left blank

Part I

Methodologies

This page intentionally left blank

2 The Emergence of the Concept of the Analytic Tradition as a Form of Philosophical Self-Consciousness James Conant 1. ON THE VERY IDEA OF AN ANALYTIC TRADITION IN PHILOSOPHY It is not uncommon for people to speak of something called “analytic philosophy.” We will worry in a moment about what sort of philosophy this label is supposed to single out. Let us begin just by noting one of the most straightforward ways in which people will sometimes try to pick out analytic philosophy from other sorts of philosophy—perhaps the most superficial way of all of trying to do this—namely by trying to say something about where on the planet analytic philosophy has and has not flourished. For example, we are apt to be told that analytic philosophy has been comparatively dominant for much of the twentieth century in the United Kingdom and, since the middle of that same century in the United States, as well as in other English-speaking countries, and even in many of the Scandinavian nations. Conversely, we are equally likely to be told that various forms of non-analytic philosophy—sometimes termed “continental philosophy”—at least until very recently, have been no less dominant in France, in most parts of Germany, and in many of the remaining parts of Continental Europe. Anyone who makes statements such as these must be relying upon some principle other than a merely geographical one for distinguishing the two sorts of philosophy at issue here. Whatever that further principle is, in recent years many have taken offense to the very idea of such a distinction. Through such ways of speaking, we are told, countless diverse strands of the Western philosophical tradition since Kant are shoehorned into a pair of intellectually uninformative rubrics—one of them turning on the supposedly crucial “analytic” aspect of the philosophical work in question, the other on the arguably even more awkward geographical rubric of the supposedly “continental” feature of the philosophy in question. Surely there is much to sympathize with in such a complaint. A host of difficulties necessarily attend any effort to deploy a pair of classificatory terms that are as orthogonal to one another as are the categories “analytic” and “continental.” On the surface, this would appear

18  James Conant to be no more promising a principle for classifying forms of philosophy into two fundamentally different kinds than would be the suggestion that we should go about classifying human beings into those that are vegetarian and those that are Romanian. There are other reasons to be suspicious of the classificatory categories at work here. The employment of the term “continental philosophy” arguably evolved historically in order for there to be some single thing to which analytic philosophy as a whole could be opposed—while leaving unclarified (and, without implicit reference to analytic philosophy, unclarifiable) wherein the unity of the contrasting category of (“continental”) philosophy is supposed to lie.1 As a matter of practice, this meant that the unity of “continental” philosophy was in fact construed by most analytic philosophers through recourse to a via negativa: analytic philosophers specified for themselves what continental philosophy was, in effect, by thinking of it as an enormous garbage bin into which any outwardly apparently non-analytic form of post-Kantian philosophy was to be dumped. As with the contents of any garbage can, so too with this one: after a great many items came to be tossed into the can, it was no longer possible to discern what united them all, without reference to something not to be found among the contents of the can—namely, an appreciation of the aims and interests of those doing the tossing. Whatever original appearance of neatness and appropriateness may have attached to the respective geographical locations of the two traditions has largely dissipated over the past several decades. There are many philosophy departments in the United States and the United Kingdom now specializing in so-called “continental” philosophy. Some of the leading academic positions in the Francophone world are held today by French analytic philosophers, and the German Society for Analytic Philosophy now attracts a level of attendance at its conferences of which it can and does proudly boast. These and other developments have rendered it increasingly difficult to specify these two traditions via anything as superficial as a principle of geographical location.2 For these and other reasons, it has come to seem clear to some that the very idea of a distinctively analytic tradition in philosophy should be abandoned. This is not the conclusion that I shall wish to draw from the above considerations. Nevertheless, it is a matter of some delicacy and difficulty to say with any accuracy what is distinctive about the tradition in question—at least if one wishes to do so in a manner that is both historically informed and philosophically non-partisan.3 With these caveats in mind, I propose to address the following question: what would it be to characterize what is distinctive about the analytic mode of philosophy from within, rather than from without? Specifically, what are analytic philosophy’s own answers to the following two questions: (1) What is analytic philosophy? and (2) what is an analytic philosopher? We shall call these our guiding questions. If we wish to make progress with our guiding questions, we need to look at very

The Emergence of the Concept of the Analytic Tradition  19 different sorts of statements from analytic philosophers regarding what philosophy is—ones that are better suited to bringing out what is distinctively analytic about the sort of philosophy in question. But there now is a further difficulty we encounter here. Philosophers concerned to issue anything resembling such a proclamation often also take the occasion to try to prescribe what analytic philosophy should be, as they conceive it, as a matter of philosophical doctrine. Consider the following three representative cases: (1) At the outset of the tradition, the early Moore and Russell each ­understood what was distinctive in their new philosophical point of departure to lie not just in their new method of analysis, but equally in the power of that method to demonstrate that the philosophical doctrines of holism and idealism, which they opposed, were false and that philosophical doctrines of atomism and realism, which they sought to champion, were true. (2) In a second major phase of the tradition, Carnap thought that the analytic philosophy of his day partly reached its maturity not only by practicing a new method of philosophical elucidation, but by showing, through its application, that a comparatively resolute form of empiricism (according to which most truths were a posteriori) was true, while also revealing that the empiricist must make one crucial concession to the rationalist (namely, that the truths of logic and mathematics were a priori, after all). (3) In its most recent phase, many a representative of the analytic tradition today thinks that anyone who has appropriately internalized the canons of rigor that analytic philosophy seeks to uphold ought to concede (even if it requires one to issue a great many philosophical promissory notes that we at the moment have no idea exactly how to cash) that some form of scientific naturalism is the only intellectually respectable position open to a serious analytic philosopher. Despite the predominance of these very specific doctrinal commitments at each of these three phases, at that very same moment in the history of the analytic tradition there were other philosophers (whom we today regard as belonging no less to that tradition) who were determined to contest these very doctrines. Moreover, even if one restricts oneself to the three phases of the tradition mentioned above, it is worth noticing how dramatically different the supposed doctrinal upshot of a commitment to (something nominally referred to at all three of these phases in the tradition as) “the analytic method in philosophy” was taken to be. What this shows is that doctrinal categories such as realism, empiricism, scientific naturalism, etc., are simply inadequate terms for specifying what unifies and distinguishes the analytic tradition as a whole. The nature of analytic philosophy simply cannot be captured in terms of a shared assent to anything having the form of a philosophical “-ism.”4 What it is to be an analytic philosopher has more

20  James Conant to do with a certain conception of how one ought to do philosophy than it does with what one ought to conclude on the basis of so doing it—with the character of the activity of philosophizing rather than with the body of doctrine in which it issues. Attempts by analytic philosophers to articulate their conceptions of what is distinctive about analytic philosophy generally also express a particular attitude toward the traditional questions of philosophy and which of them are worth answering—thereby usually also furnishing clues regarding how this particular analytic philosopher would go about answering some of these questions and why she would reject others (or, in the case of certain figures, why she is determined to unmask them all as based on either confusion or nonsense). So in the cases of a great many authors, it would be artificial to attempt to separate what they deem to be matters of method from matters of doctrine. But even if these two aspects of what they consider analytic philosophy to be are often inextricably intertwined, it remains possible to select representative quotations that help to highlight what various of them at their particular junctures in the tradition take to be distinctive about the manner in which they think philosophy ought to be practiced. Such statements tend to be produced under the felt pressure to ward off some alternative conception of how philosophy ought to be practiced. Thus they are akin to bulletins from the front—statements written in the heat of a contest to determine what analytic philosophy is to be, in which each contestant attempts to carry the day for what she is seeking to initiate, or inherit, or redirect, in undertaking to do analytic philosophy in her own distinctive way. Here are some characteristic statements of this sort: A proposition is composed not of words, nor yet of thoughts, but of concepts. . . . They are incapable of change; and the relation into which they enter with the knowing subject implies no action or reaction. . . . It seems necessary, then, to regard the world as formed of concepts. These are the only objects of knowledge. . . . From our description of a judgment, there must, then, disappear all reference either to our mind or to the world. . . . The nature of the judgment is more ultimate than either, and less ultimate only than the nature of its constituents—the nature of the concept or logical idea. (G. E. Moore)5 Modern analytical empiricism [. . .] differs from that of Locke, Berkeley, and Hume by its incorporation of mathematics and its development of a powerful logical technique. It is thus able, in regard to certain problems, to achieve definite answers, which have the quality of science rather than of philosophy. It has the advantage, as compared with the philosophies of the system-builders, of being able to tackle its problems one at a time, instead of having to invent at one stroke a block theory of the whole universe. Its methods, in this respect, resemble those of science. I have no doubt that, in so far as philosophical knowledge is possible, it is by such methods that it must be sought. (Bertrand Russell)6

The Emergence of the Concept of the Analytic Tradition  21 Philosophy is not one of the natural sciences. (The word “philosophy” must mean something which stands above or below, but not beside the natural sciences.) The object of philosophy is the logical clarification of thoughts. Philosophy is not a theory but an activity. A philosophical work consists essentially of elucidations. The result of philosophy is not a number of “philosophical propositions”, but to make propositions clear. Philosophy should make clear and delimit sharply the thoughts which otherwise are, as it were, opaque and blurred. (Early Wittgenstein)7 In most cases future philosophers will have to be scientists because it will be necessary for them to have a certain subject matter on which to work—and they will find cases of confused or vague meaning particularly in the foundations of the sciences. . . . I am convinced that our view of the nature of philosophy will be generally adopted in the future; and the consequence will be that it will no longer be attempted to teach philosophy as a system. We shall teach the special sciences and their history in the true philosophical spirit of searching for clarity and, by doing this, we shall develop the philosophical minds of future generations. (Schlick)8 Logic is the method for doing philosophy. [. . .] There is no such thing as philosophy in the shape of a theory, i.e. a system of distinct propositions separate from science. Doing philosophy means nothing else but this: to clarify the concepts and propositions of science through logical analysis. (Carnap)9 It was true to say that our considerations could not be scientific ones. It was not of any possible interest to us to find out empirically ‘that, contrary to our preconceived ideas, it is possible to think such-andsuch’—whatever that may mean. . . . And we may not advance any kind of theory. There must not be anything hypothetical in our considerations. We must do away with all explanation, and description alone must take its place. And this description gets its light, that is to say its purpose, from the philosophical problems. These are, of course, not empirical problems; they are solved, rather, by looking into the workings of our language, and that in such a way as to make us recognize those workings: in despite of an urge to misunderstand them. The problems are solved, not by giving new information, but by arranging what we have always known. Philosophy is a battle against the bewitchment of our intelligence by means of language. (Later Wittgenstein)10 This book offers what may with reservations be described as a theory of the mind. But it does not give new information about minds. We possess already a wealth of information about minds, information which is neither derived from, nor upset by, the arguments of philosophers. The philosophical arguments which constitute this book are intended not to increase what we know about minds, but to rectify the logical geography of the knowledge which we already possess. (Gilbert Ryle)11

22  James Conant In view of the prevalence of the slogan ‘ordinary language’, and of such names as ‘linguistic’ or ‘analytic’ philosophy or ‘the analysis of language’, one thing needs specially emphasizing to counter misunderstandings. When we examine what we should say when, what words we should use in what situations, we are looking again not merely at words (or ‘meanings’, whatever they might be) but also at the realities we use the words to talk about: we are using a sharpened awareness of words to sharpen our perception of, though not as the final arbiter of, the phenomena. For this reason I think it might be better to use for this way of doing philosophy, some less misleading name than those given above—for instance, ‘linguistic phenomenology’, only that is rather a mouthful. (J. L. Austin)12 Metaphysics has been often revisionary, and less often descriptive. Descriptive metaphysics is content to describe the actual structure of our thought about the world. . . . How should it differ from what is called philosophical, or logical, or conceptual analysis? It does not differ in kind of intention, but only in scope and generality. Aiming to lay bare the most general features of our conceptual structure, it can take far less for granted than a more limited and partial conceptual inquiry. Hence, also, a certain difference in method. Up to a point, the reliance upon a close examination of the actual use of words is the best, and indeed the only sure, way in philosophy. But the discriminations we can make, and the connections we can establish, in this way, are not general enough and not far-reaching enough to meet the full metaphysical demand for understanding. For when we ask how we use this or that expression, our answers, however revealing at a certain level, are apt to assume, and not to expose, those general elements of structure which the metaphysician wants revealed. The structure he seeks does not readily display itself on the surface of language, but lies submerged. (P. F. Strawson)13 I have . . . an unswerving belief in external things—people, nerve endings, sticks, stones. . . . I believe also, if less firmly, in atoms and electrons and in classes. How is all this robust realism to be reconciled with the barren scene that I have just been depicting? The answer is naturalism: the recognition that it is within science itself, and not in some prior philosophy, that reality is to be identified and described. (W. V. Quine)14 The first of these statements is from Moore’s aforementioned essay, “The Nature of Judgment.” Though it expresses the essential conviction that Moore took to distinguish his philosophical project from that of his predecessors and it seeks to show how that conviction is to be secured through a different approach to philosophical problems, it still retains much of the flavor of traditional philosophy. The second quotation is one of the later of Russell’s many attempts to define and call for the continuation of the movement that he considered Moore, Frege, and himself (among others) to

The Emergence of the Concept of the Analytic Tradition  23 be initiating. At the center of this conception of it is the idea that philosophy should take the mathematical and natural sciences as a model. Russell’s original conception of the way in which this was to proceed took it for granted that, even while doing so, philosophy would remain in some respects importantly distinct from science. The third quotation is from Wittgenstein’s Tractatus—the first of many influential works within the analytic tradition concerned to take up a position toward the prior chapters of that tradition akin to that which Kant had assumed with regard to the early modern tradition. Like Kant, early Wittgenstein sought to show how the entire collective enterprise had taken a wrong turn and hence why a new beginning is now required. Here, as with Kant, an insistence on the sharp difference between the character of the questions treated by the philosopher and those treated by the natural sciences comes again to play a central role. The fourth and fifth quotations—from Moritz Schlick and Rudolf Carnap respectively—are representative of a number of the thinkers in the succeeding generation who struggled to incorporate the critical edge of Wittgenstein’s apparently devastating, tradition-exploding ideas into an undertaking retaining the overall outline of Russell’s programmatic, tradition-inaugurating ambitions—thereby introducing considerable internal tension into their conception of philosophy. They are eager to retain Russell’s idea that philosophy has a special kinship with the natural sciences, and yet also want to agree with Wittgenstein that the propositions of philosophy must be sharply distinguished from those of the natural sciences—and indeed from empirical propositions in general. In the second half of the above collection of quotations, we have to do with reactions to the first generations of analytic philosophy. The task at that point had become one of redefining an ongoing tradition. In the sixth quotation, we have a characteristic reflection on the nature of philosophy by the later Wittgenstein. Here we encounter a thinker who is in equal measure concerned, first, to repudiate the inheritance of his early work by Schlick and Carnap (and thus to insist that philosophy must be conducted in a spirit completely alien to that of the natural sciences); second, to move beyond that work itself, on the grounds that it failed to live up to its aspiration to fully break with the preceding tradition (and thus to repudiate that work in a very different way from that of its most outspoken critics); and third, to retain and extend the philosophical aspirations at the heart of that early work (and thus to emphasize that the later work can be understood only against the background of the earlier). In the final four quotations, we encounter four alternative proposals for alternative new beginnings for analytic philosophy—each of which seeks to offer a new form of positive program for philosophy, and each of which draws inspiration from sources that come from outside analytic philosophy. The first three of these quotations (from Ryle, Austin, and Strawson) share with the later Wittgenstein the idea that philosophy ought to do justice to the most fundamental aspects of our everyday understanding of

24  James Conant ourselves, the world, and the medium-sized dry goods (as ordinary objects were nicely referred to) and other persons we find there. Ryle harks back to Frege and Wittgenstein; Austin especially to Moore. Both Ryle and Austin were trained as scholars of ancient philosophy and both strove, albeit in very different ways, to reincorporate into the analytic tradition certain neglected insights they find in Plato and Aristotle. Austin, Ryle, and Strawson all belonged to a generation of analytic philosophers who had begun to become suspicious of an overreliance on technical tools in philosophy. Paul Grice later offered a diagnosis of the waning enthusiasm among certain members of the postwar generation for inventing new forms of logical notation (along with an increasing interest in ordinary language) in the following terms: I have little doubt that a contribution towards a gradual shift of style was also made by a growing apprehension that philosophy is all too often being squeezed out of operation by technology; to borrow words from Ramsey, that apparatus which began life as a system of devices to combat woolliness has now become an instrument of scholasticism.15 Others, however, in that same postwar generation strongly dissented from this burgeoning consensus and deplored the depreciation of formal logic that had become so fashionable in the work of many of their contemporaries. Perhaps the most influential of these dissenters was W. V. O. Quine. Carnap and others had viewed the tools of logic as providing the instruments for an articulation of the aim of philosophy that would allow it to continue to differentiate itself from natural science. Quine championed the position that philosophy and science should be regarded as two aspects of a single enterprise—intertwined partners caught up in a single form of pursuit: the pursuit of truth. While his contemporary Strawson looked back especially to Kant for philosophical stratagems designed to vindicate our ordinary conceptual scheme, Quine looked sideways to developments in empirical psychology and physics to overturn that scheme. Whereas Strawson sought to recover the Kantian project of reconciliation in philosophy (making room again for the idea that we can, without contradiction, conceive of ourselves as both material bodies and free agents), Quine sought to revivify the Humean project of debunking the ordinary man’s view of the world (breathing new life into something akin to the very sorts of naturalism that Frege and the early Wittgenstein sought to exorcise from philosophy). Indeed, much of postwar twentieth-century analytic philosophy can be seen as involving a contest between this standard analytic conception of what it is to be a Humean (debunking our ordinary view of the world in the light of a properly naturalized understanding of what it can contain) and the opposed standard analytic conception of what it is to be a Kantian (seeking to respect both the natural-scientific understanding and our everyday understanding of the world).

The Emergence of the Concept of the Analytic Tradition  25 Each of the ten statements on the list above was made by an important analytic philosopher at a very particular point in his career. Each gives expression to a conception of analytic philosophy that proved to be influential for a time, thereby affecting the trajectory of the analytic tradition as a whole. When taken together, they bring out certain characteristic features of the tradition. No one of them, however, suffices as a summary of what “analytic philosophy” as such is. None of them articulates a conception of philosophy—let alone of what makes a particular sort of philosophy analytic—that they would have expected the other nine authors and most of their contemporaries to endorse, at least not without considerable qualification. Each is speaking for himself, attempting to give voice to his own distinctive conception, at the time of writing—a conception that, in many of their cases, underwent one or more dramatic shifts over the course of their careers. Thus each statement offers a characterization of what philosophy is that is simply far too distinctive, too idiosyncratic, and too historically indexed to a particular moment in the unfolding of the analytic tradition to serve as a blanket characterization of what makes analytic philosophy as a whole something that might plausibly be considered a unitary intellectual movement. The statements above appear in roughly chronological order. With respect to at least the first five of them, at the time of their writing, there was not yet anything they were aware of (and to which they could have thought of themselves as contributing) that could have been designated as “the analytic tradition.” The very possibility of discerning anything of the sort requires the attainment of some historical distance from its beginnings, hence of a standpoint from which one is able to compare and contrast an extended stretch of this development in philosophy with other philosophical traditions, such as those to which the early analytic philosophers were reacting—such as German Neo-Kantianism, Austrian Realism, British Idealism, and American Pragmatism. The history of the relations between analytic philosophy and these neighboring traditions is no less tangled than that of analytic philosophy itself. Analytic philosophy arose partly as a reaction to these other forms of philosophy—and, yet, as so often in the history of philosophy, it bears deep traces of the very traditions it sought to resist and replace. No less significantly, some later practitioners in the analytic tradition sought to reincorporate insights from those same traditions—insights they thought their analytic predecessors had either unduly neglected or over-hastily rejected. This subsequently gave rise to the emergence of developments in the analytic tradition that would have astonished many of its earlier figures—bearing such labels as Analytic Kantianism, Analytic Hegelianism, and Analytic Pragmatism. Russell and Moore understood the tradition they were seeking to inaugurate in philosophy to be a revolt against Idealism—and against Kant, Hegel, and the British Idealists in particular. A half-century later, Sellars and Strawson saw themselves as trying to recover insights from Kantian

26  James Conant Idealism and to reincorporate them within the analytic tradition; while some contemporary analytic philosophers, such as John McDowell and Robert Brandom, in the generation thereafter, have become no less concerned to recover and revive what they regard as philosophically valuable and vital in Hegel. While Russell and Moore sought to distinguish themselves sharply from American Pragmatists like William James and John Dewey, and while the pragmatists of the next generation (such as C. I. Lewis) often tended to distinguish themselves sharply from their analytic contemporaries (such as Hans Reichenbach and Carnap), many recent analytic philosophers (notably Hilary Putnam and Richard Rorty) see no essential tension between the best insights of analytic philosophy and Pragmatism, and seek to develop philosophical syntheses of elements drawn from each. But these are all relatively recent developments, of which an early analytic philosopher could not have had any inkling. It is a historical truism to say that the early analytic philosophers themselves did not yet have (because they could not have had) a historical consciousness of their own work as forming the first chapter in a more extended intellectual adventure—an adventure that would eventually become what we now call “the analytic tradition.” Thus they could hardly have reflected on that tradition, or compared and contrasted it with other equally sustained philosophical traditions. Such a perspective on the significance of their own accomplishment would have required of the early analytic philosophers that they be able to step outside of their own moment in time and travel to a moment in the philosophical future from which they could assess how their own chapter in the history of philosophy formed the first chapter of something on the order of a tradition. This important point, upon a moment’s reflection, is an obvious one, but it is often overlooked or at least underappreciated: The founding fathers of analytic philosophy did not take themselves, and could not have taken themselves, to be founding what we today think of as the analytic tradition. In thinking of them as analytic philosophers, we think of them in a way that they could not have thought of themselves. This is not to say that the authors of our first five quotations—G. E. Moore, Bertrand Russell, the early Ludwig Wittgenstein, Moritz Schlick, Rudolf Carnap—had no interest in opening up a new chapter in the history of philosophy. They each wanted to change philosophy in a way that would affect how everyone after them who might venture to practice it would (or at least should) do so. Each sought to transform philosophy, and each ended up figuring in a cumulative development that, as a whole, did transform philosophy—or at any rate philosophy as practiced in the Anglophone mainstream of the discipline—into something new. So, in one sense, each can be said to have contributed materially to a revolution in philosophy. But, in another sense, none of them can simply be said to have authored that revolution, not only because the transformation in question was a collective effort to which they all contributed, but also because none of them would have regarded most of what followed as realizing their original conception

The Emergence of the Concept of the Analytic Tradition  27 of what they sought to initiate. The so-called “founders of analytic philosophy”, in this sense, did not know what they were actually founding. They could not have foreseen—nor, at least in some instances, would they even have been willing to endorse—many subsequent developments within that tradition. In some cases, they would have had difficulty even comprehending how what followed could be supposed to represent an inheritance and continuation of their own endeavors. The experience of being unable to comprehend a later stage in the development of the tradition is something that a number of major figures experienced in a fashion that was devastatingly immediate and personal. The sense of having been betrayed by a member of the next generation, or even of one’s own, sometimes played out very painfully between pairs of individuals who were initially as intellectually close to one another as one can imagine—sometimes standing to one another in the relation of teacher to student, or that of master to disciple, or that of comrades in arms. Indeed, a pattern of this sort, of initial intellectual proximity giving way to subsequently unbridgeable tradition-splintering differences, came to be a recurring feature of the analytic tradition. Some of the most dramatic cases included such pairs as Russell and Wittgenstein, Carnap and Quine, Quine and Donald Davidson, Austin and Paul Grice, and Putnam and X (where, to mention only some of the important instances, X at different points equaled Reichenbach, Quine, Fodor, and Rorty). Ideological statements about the nature of analytic philosophy often seek to give the impression that philosophical personality plays a much lesser role in the practice of this sort of philosophy and that the respective temperaments of its practitioners are fully subordinated to shared philosophical methods and canons of argument. In point of fact, however, some of the most riveting moments in its history have involved dramatic clashes of personality—between initially apparently likeminded individuals, starting out with common sets of aims and concerns and ending up gravely at odds with each other, at least philosophically and sometimes even personally. Some might contend that this historical fact is merely a bit of trivia having no real significance for an understanding of the nature of the analytic tradition in philosophy. Others might want to claim that it is an important fact that reveals something fundamental about the nature of analytic philosophy. We need not decide this question here. Its relevance to our present purpose is simply that the early analytic philosophers themselves were in no position to have a view about the matter, for in order to be able to assess the possible significance of any such factual feature of a tradition, the larger shape of that tradition must already be in view. None of the first five philosophers cited above were in any position to assess how their own contributions would bear on and be further subsumed by the efforts of the later five authors quoted above—let alone to be able to apprehend the cumulative efforts of these ten philosophers, each of whose work makes up only a single stone in the overall mosaic of the analytic tradition.

28  James Conant That overall image at least begins to come into view with the addition of the contributions of the generation of mid-century thinkers who are the authors of our five later quotations. They were writing at a time when something like our contemporary idea of this tradition was beginning to emerge. There was at that point already a substantial tradition of some sort well underway—one whose physiognomy could be discerned in any number of overlapping yet divergent ways. Nevertheless, as in the case of the first five quotations, we do not take these five later quotations, either individually or collectively, to stand as or add up to a characterization of “analytic philosophy as such” (whatever that might mean). Rather, we see them each and all to be expressions of particular conceptions of philosophy, each of which stands in some sort of significant continuity with some of the statements presented in our first five quotations. That is a difference; for unlike the first set of authors, these authors all see themselves as continuing something inaugurated by at least some (though not necessarily all) of our five earlier authors. Yet each such instance of significant continuity is folded within a discontinuity, which each author takes to be potentially transformative for the tradition that they all seek to continue. None of the ten quotations can serve as an adequate or satisfactory answer to our guiding questions. Yet this collection of partially overlapping, crisscrossing, occasionally sharply diverging statements, taken together, does begin to reveal some of the shape of the tradition. It certainly displays some of the most characteristic features of the tradition’s images of itself—self-images that have vied with one another over the course of its history. The sheer diversity of these statements, however, may still (rightly) leave the reader feeling that we have yet to arrive at anything like a useful (let alone succinct) answer to either of our guiding questions. Where else might one look for such answers? 2. SOME ASPECTS OF THE IDEOLOGY OF ANALYTIC PHILOSOPHY There are countless philosophy textbooks and encyclopedias in which members of the discipline deemed qualified to do so undertake to tell their reader just what analytic philosophy is. These accounts are often written by the sort of author one might call an ideologue of analytic philosophy. The sorts of answers such texts provide are not without interest. They, too, in a rather different way, tell us something about what analytic philosophy has been, or is now, or at least is now declared to be and to have always been. In the account of an ideologue of analytic philosophy, we typically are confronted with the views of an author who is reasonably well acquainted with analytic philosophy and has an interest in practicing it (according to a certain understanding of what that means), but who on this occasion is standing back from this practice and delivering a programmatic statement

The Emergence of the Concept of the Analytic Tradition  29 regarding its nature. The sort of programmatic statement here in question purports to provide a valid general description not simply of one individual’s own ongoing practice of philosophy, but of something much broader, namely, the conception of philosophy that animates an entire tradition. Such programmatic statements of what analytic philosophy is tend to take the form of persuasive definitions. On the one hand, they purport to define what analytic philosophy is; on the other, they seek to recommend the very way of doing philosophy that they purport to describe. In recent years, a great many cases in point have appeared in print. If one seeks to take one’s bearings from this array of ideological pronouncements made by analytic philosophers on behalf of analytic philosophy regarding what analytic philosophy really is, the first thing that ought to strike one is how profoundly their pronouncements differ from one another. One recurring theme is the idea that analytic philosophy bears a close relation to the natural sciences, or should take the natural sciences as its model, or perhaps even should come to regard itself as just one more—albeit unusually self-reflective and abstract—branch of natural science. This aspect of the ideology of analytic philosophy has had a number of very concrete institutional effects that have, in turn, occasioned various sorts of heated disagreement among analytic philosophers about the shape their own discipline should strive to assume. As analytic philosophers have self-consciously adopted methods of publication and other forms of institutionalization that characterize the sciences, it has inevitably developed in the ways those forms of publication and institutionalization promote. In particular, it has become increasingly stratified into ever multiplying subdisciplines of philosophical research (such as the philosophy of language, the philosophy of mind, the philosophy of perception, and the philosophy of action). Each of these, in turn, has spawned its own increasingly esoteric sub-literature of puzzle cases and niche controversies, published in professional journals addressed to an increasingly exclusive form of professional readership. This has led to a situation in which, if someone wishes to claim professional competence in a given “area” of philosophy, she has to wonder whether this requires that she forfeit the better part of her philosophical life to staying on top of the growing body of journal literature devoted to that “area.” On a certain conception of what it is to have a suitably developed professional conscience, such a narrowing of one’s philosophical focus has come to appear to be a compulsory feature of what it now means to be a serious analytic philosopher. This has significant consequences. The sheer amount of time and energy required thus “to professionalize oneself” in one of these sub-disciplines increasingly precludes the possibility of a single philosopher contributing significantly to several areas of philosophy at once. Yet the capacity to do just this was frequently held up, throughout much of the history of the tradition, as one sign of a genuinely philosophical mind. These forms of institutionalization and professionalization within analytic philosophy have provoked a variety of counter-reactions. They have

30  James Conant led some major figures during their later years—perhaps most notably, Hilary Putnam—to push back against this development and to press the argument that the very possibility of doing path-breaking work in philosophy requires (what the Germans call) an Übersicht—a synoptic overview—of the whole of the subject.16 Others have argued that this tendency towards an ever-increasing speciation of sub-disciplines obscures from view something essential to the very nature of the philosophical enterprise itself: its underlying unity. Paul Grice, in the last phase of his career, became a forceful advocate for this point of view: [I]t is my firm conviction that despite its real or apparent division into departments, philosophy is one subject, a single discipline. By this I do not merely mean that between different areas of philosophy there are cross-references, as when, for example, one encounters in ethics the problem whether such and such principles fall within the epistemological classification of a priori knowledge. I mean (or hope I mean) something a good deal stronger than this, something more like the thesis that it is not possible to reach full understanding of, or high level proficiency in, any one department without a corresponding understanding and proficiency in the others; to the extent that when I  visit an unfamiliar university and (as occasionally happens) I am introduced to, ‘Mr Puddle, our man in Political Philosophy’ (or in ‘Nineteenth-century continental philosophy’ or ‘Aesthetics’, as the case may be), I am immediately confident that either Mr Puddle is being under-described and in consequence maligned, or else Mr Puddle is not really good at his stuff. Philosophy, like virtue, is entire. Or, one might even dare to say, there is only one problem in philosophy, namely all of them.17 We find ourselves increasingly in a situation today in which some analytic philosophers view Grice’s remarks above as getting at something essential to the very nature of the philosophical enterprise (something that is in danger of being lost through its present form of institutionalization and professionalization), while others look upon them as vestigial traces of a vanishing philosophical era (which we can leave behind without great intellectual cost to anything in philosophy about which we should care). This disagreement itself constitutes a significant crossroads at the heart of contemporary analytic philosophy, at which the very soul of analytic philosophy—what it is, what it wants to be, and what it shall become—is itself at stake. Some ideologues of analytic philosophy who have wished to be able to sum up what analytic philosophy is in a slogan, while also doing justice to the internal diversity of the tradition, have sought to do so by speaking of a characteristically analytic style of philosophy—a single consistent style that supposedly cuts across the many differences in analytic philosophers’ conceptions of philosophical method. Here, again, the attempt to capture the entire breadth of analytic philosophy in a single formula—in this case one

The Emergence of the Concept of the Analytic Tradition  31 pertaining to its style—runs into problems not unlike those we have encountered above. This way of trying to get at the essence of analytic philosophy is no less beset than the others by the twin dangers of total vacuity or inaccurate partiality. A full unpacking of what any such a conception of the supposed style of analytic philosophy comes to would require an in-depth exposition of what the philosophers who extol the virtues of analytic philosophy in the above ways are concerned to shy away from or shun—of what it is of which they are afraid. This is a topic upon which some figures in the analytic tradition have been moved to reflect (perhaps most notably, in their somewhat different ways, the later Wittgenstein, Iris Murdoch, Bernard Williams, Stanley Cavell, and John McDowell). Those who have so much as broached this topic thoughtfully and judiciously (rather than abruptly or polemically), however, have been a distinct minority. 3.  RITUALS OF EXCOMMUNICATION The roots of the distinction between two kinds of philosophy—of which the rigorous and respectable sort is of the analytic variety and the other sort is one which bears the philosophical shortcomings characteristic of certain French or German thinkers—like so much else in the analytic tradition, arguably can be traced back to Russell. His 1912 essay on “The Philosophy of Bergson” represents possibly the earliest attempt within the tradition to represent the contemporaneous intellectual landscape as presenting us with a fundamental choice between the intellectual virtues and vices of these two philosophical types. This, in turn, gives rise to an implicit conception of the character of those who belong within the community of genuinely serious philosophers and those who should be excommunicated from any such community. In that essay, we find passages such as the following: There are in Bergson’s works many allusions to mathematics and science, and to a careless reader these allusions may seem to strengthen his philosophy greatly. As regards science, especially biology and physiology, I am not competent to criticize his interpretations. But as regards mathematics, he has deliberately preferred traditional errors in interpretation to the more modern views which have prevailed among mathematicians for the last half century. In this matter, he has followed the example of most philosophers. In the eighteenth and early nineteenth centuries, the infinitesimal calculus, though well developed as a method, was supported, as regards its foundations, by many fallacies and much confused thinking. Hegel and his followers seized upon these fallacies and confusions, to support them in their attempt to prove all mathematics self-contradictory. Thence the Hegelian account of these matters passed into the current thought of philosophers, where it has remained long after the mathematicians have removed all the difficulties upon

32  James Conant which the philosophers rely. And so long as the main object of philosophers is to show that nothing can be learned by patience and detailed thinking, but that we ought rather to worship the prejudices of the ignorant under the title of “reason” if we are Hegelians, or of “intuition” if we are Bergsonians, so long philosophers will take care to remain ignorant of what mathematicians have done to remove the errors by which Hegel profited.18 Characteristically for Russell, much is made to turn here on a proper understanding of developments in mathematics and science—an understanding that he finds sorely lacking in Bergson. For our present purpose, however, the more interesting theme present here is the idea of there being a fundamental opposition—one that structures the terms of a choice forced upon the contemporary reader: a choice between Russell and Bergson, between mathematical thought and unreflective intuition, between logic and mysticism, between reason and unreason. This opposition eventually assumes an increasingly strident shape in Russell’s thinking. What Russell is especially concerned to ward off in his later writings is that any undue element of anthropocentrism be permitted to slip into our metaphysical view of the nature of the universe and man’s place in it. We must be on guard against any form of philosophy that fails to appreciate how very puny we really are, when viewed from the perspective of the larger scheme of things. Misguided forms of philosophy attempt to provide us with an image of the very nature of reality in which the aims and purposes that we happen to have at the present moment would appear to have a proper place in the universe such as it is: “In this way they interfere with that receptivity to fact which is the essence of the scientific attitude towards the world” (Russell 1917). This sentence of Russell’s can still serve helpfully to sum up a certain inchoate distinction between appropriately hard-headed philosophers and comparatively soft-headed philosophers—a distinction which has played a role both in shaping analytic philosophy’s understanding of its other and in shaping certain controversies that have in recent years increasingly come to dominate the internal discourse of analytic philosophy itself. Many analytic philosophers today might well be willing to admit that they are in no position to specify the conditions that philosophical work must satisfy in order to count as “analytic” or genuinely “hard-headed,” while also being passionately concerned to retain their right to enter the charge that the work of some particular author be deemed unworthy of an analytic philosopher—as being insufficiently rigorous or overly soft in some respect. On what basis is this sort of judgment made? Those who make it are likely to insist that they simply can tell a work of analytic philosophy when they see one. Conversely, they can just tell when someone is no longer producing analytic philosophy, even if the work in question is authored by someone who was previously considered (and still considers herself) to be

The Emergence of the Concept of the Analytic Tradition  33 an analytic philosopher. That is, they can tell when a certain tipping point has been reached: when too many of the virtues of such philosophy have fallen away, or when too many of the vices characteristic of the writings of French or German “Continental” luminaries obtrude themselves, or when there is a bit of both. This can lead to impassioned denunciations—episodes in which one analytic philosopher accuses another putative member of the guild of having betrayed a communally shared conception of the philosophical calling. Yet even when such intramural denunciations are made (and they are no longer as infrequent as they once were), questions naturally arise about whom the denouncer is speaking for and how the legitimacy of the charge is to be adjudicated. Consider the following remarks by Crispin Wright, made in the context of the closing remarks of a review of John McDowell’s Mind and World: If analytical philosophy demands self-consciousness about unexplained or only partially explained terms of art, formality and explicitness in setting out of argument, and the clearest possible sign-posting and formulation of assumptions, targets, and goals, etc., then this is not a work of analytical philosophy. . . . At its worst, indeed, McDowell’s prose puts barriers of jargon, convolution and metaphor before the reader hardly less formidable than those characteristically erected by his German luminaries. . . . .[T]he stylistic extravagance of McDowell’s book—more extreme than in any of his other writings to date—will unquestionably color the influence it will exert . . . [T]he fear must be that the book will encourage too many of the susceptible to swim out of their depth in seas of rhetorical metaphysics. Wittgenstein complained that, “The seed I am most likely to sow is a certain jargon.” One feels that, if so, he had only himself to blame. McDowell is a strong swimmer, but his stroke is not to be imitated.19 Crispin Wright is one of the leading analytic philosophers of the present day. John McDowell’s Mind and World is arguably one of the single most influential works of analytic philosophy of the past quarter of a century. Or, perhaps we should say, in order not to beg a question here: It is arguably one of the single most influential works of the past quarter of a century written by someone who, at least for most of his career, was deemed, by at least most of his contemporaries, to be a practicing analytic philosopher. Perhaps, with the publication of this book, John McDowell suddenly ceased to be an analytic philosopher; perhaps Crispin Wright was the first to publicize the fact of McDowell’s exodus from the community. Yet the relevant passages in Wright’s text read less like a report of an astonishing discovery (news flash: McDowell has emigrated to a different philosophical continent!) and more like a plea for an edict of excommunication (proposed motion: respect due to a member of our community no longer to be accorded to McDowell!).

34  James Conant But on what grounds is such a charge entered and before which tribunal? And how is its validity to be determined? Wright need not have had a clear view of how such questions are to be answered in order to feel that he is, nonetheless, in the right—and about something important. He is evidently writing, even here in this part of this review, as someone who is not without considerable admiration for McDowell’s abilities as a philosopher. Yet he is also writing from a sense that some line has been crossed in McDowell’s latest work, so that this product of philosophy, by this erstwhile analytic philosopher, is one which has gone too far. It is important to make clear that, once a work of philosophy has reached the point where it looks and sounds like this, then (as Wright bluntly puts it) “this is not a work of analytical philosophy.” Notice that the fundamental ground of the criticism, at least in the above passage, appears in the first instance not to lie in a charge directed against either the character of the doctrines McDowell upholds or the method of philosophy that he practices. The charge is quite explicitly directed at the style of the work. Apparently a work that courts such a style may no longer be counted as analytic philosophy. Some of the vices of style are linked by Wright to features that analytic philosophers in the past have often regarded as characteristic of “continental” efforts at philosophizing (fuzziness of thought, liberal employment of metaphors, extravagance of expression). Other aspects of the vicissitudes of McDowell’s style are linked by Wright to more time-honored complaints—familiar already to Socrates—leveled against forms of philosophy that are feared because of their potential to win a following (to corrupt the youth, inspire imitation, and lead the next generation astray). This can readily lead to a situation in which two sets of readers, equally familiar with the philosophical temperament of the reviewer, are drawn to opposite conclusions: One set, upon reading such a review, concludes that the work is one with which they need not bother further (given that it permits itself such forms of stylistic license), whereas the other concludes, against the reviewer’s own intentions, that the work might well be of philosophical interest (just because the danger it appears to pose to this reviewer is of this sort). It is a noteworthy feature of analytic philosophy in its most recent Anglophone phase that increasing numbers of philosophers who regard themselves as members of “the analytic tradition” have in this way often become more concerned to differentiate themselves from certain others who also so regard themselves than they are to differentiate themselves from any current species of non-analytic philosopher. Just as in the aftermath of the Russian revolutions both Stalin and Trotsky were far more able to tolerate a temporary truce with Churchill or Roosevelt than either was to tolerate one between themselves, so, too, there are now subcommunities of analytic philosophers who find it far easier to enter into non-aggression pacts with those who are simply outsiders to their internecine quarrels than they are to make peace with those within their community whom they view

The Emergence of the Concept of the Analytic Tradition  35 as having placed themselves beyond the pale of respectability through the character of their thought or writing. The remarks in the preceding paragraphs about certain features of the most recent phase of Anglo-American analytic philosophy are far less true of the current dispensation of analytic philosophy on the European Continent. On that side of the English Channel, where the position of analytic philosophy as a dominant tradition of philosophizing has been far less secure, one still encounters frequent attempts (undertaken by figures on either side of the mutually contested terrain) to draw bright red battle-lines between the analytic and non-analytic ways of doing things. A visiting Anglo-American analytic philosopher suddenly finding herself amid a diverse gathering of European philosophers may be left with the impression that she has stepped into a time machine; the dominant ideological struggles (along with other aspects of how the respective German, French, or Italian dispensations of analytic philosophy currently define themselves) may strike her as a surreal recapitulation of a whole series of episodes from the past of her own tradition, only now all compressed into a single episode. One particular recent development within the Anglophone analytic tradition has therefore been greeted with particular dismay in such combatively minded Continental European analytic circles: a minority of influential figures within the Anglo-American analytic community have become increasingly vocal in their expressions of annoyance at efforts (by both proponents and critics of the analytic tradition) to make too much of the idea that there is a philosophically significant contrast to be drawn between analytic and other kinds of philosophy. Bernard Williams is an example of a major figure in the analytic tradition whose later writings manifest a leaning in this direction—and thus also a concern to deny that the differences in question reflect anything philosophically deep. In a characteristic passage, he writes: The contrast between ‘analytic’ philosophy and ‘continental’ philosophy is not at all an opposition of content, of interest, or even of style. Indeed, there are some differences, some of which are important, between typical examples of philosophical writing to which these terms could be applied, but these differences do not rest upon any significant basic principles. It could even be said that these terms mark a difference without a distinction.20 The terms “analytic” and “continental” mark a difference without a distinction, for Williams, if the purpose to which they are to be put is to provide a philosophical account of how the very essence of the analytic way of doing philosophy must of necessity differ from that of any other way of doing philosophy if it is to retain its integrity qua analytic philosophy. If, however, the point of using this terminology is merely to mark a difference between the sorts of writing more typically found in one tradition than in another, then he is perfectly willing to grant that the terms in question may helpfully

36  James Conant be employed to indicate characteristic differences in forms of philosophical prose. What he is most concerned to deny is that the differences thereby indicated are in any way a function of a philosophically significant opposition between two fundamentally different kinds of philosophy. 4. ANALYTIC PHILOSOPHY’S RELATION TO THE HISTORY OF PHILOSOPHY Analytic philosophers have differed markedly amongst themselves in their attitudes with respect to the history of philosophy. Some major figures have wanted to understand what is essential to analytic philosophy as requiring a sharp break with the entire past of philosophy. For them, there is no longer any need or reason for philosophers to occupy themselves with the writings of figures belonging to the prehistory of analytic philosophy. Others have spoken (only slightly less immodestly) of a form of philosophical inheritance of the past in which our understanding of the very nature of the activity undergoes radical transformation. Their view is that we may continue to be concerned with the philosophical writings of the past, but in a sufficiently novel manner that we will, in effect, introduce (as later Wittgenstein put it) “a kink” in the history of philosophy. Yet others have seen their own philosophical projects as directly inheriting those of the great figures in the past. Our relation to them need not require any specifically historical form of understanding of the past; it should involve nothing more than direct philosophical engagement with the writings of these “mighty dead.” Analytic philosophers today, for a variety of reasons, are increasingly inclined to regard the very idea of a department of philosophy lacking capable historians of philosophy as existing in a condition of impoverishment. One set of concerns that play a role here comes from those practitioners of analytic philosophy (and there have always been some) who look upon their philosophical questions as stemming directly from those of a broader philosophical tradition. They have good reason to regard those who seek to acquire a broader and deeper understanding of that tradition as colleagues engaged in a form of inquiry continuous with their own. This sort of connection to the community, however, can still leave the historian of philosophy (rightly) feeling that the reigning conception of the distribution of labor presupposes a historically parochial perspective on the philosophical bearing of the past on the present. For even among analytic philosophers who have in this way been open to the philosophical importance of cultivating such forms of familiarity with ancient, medieval, or early modern texts, there sometimes still lingers a tendency to regard the proper purview of the professional historian of philosophy as coming to an end at that moment in the history of the subject when the analytic tradition begins. On this way of looking at things, philosophers such as Frege, Russell, Carnap, Wittgenstein, and Quine are to be regarded as forming a part

The Emergence of the Concept of the Analytic Tradition  37 of “our” analytic-philosophical present in a way that no merely “historical” figure could. One is thus thought to be doing a special sort of philosophical violence to such authors if one treats them as fit subjects of “historical” inquiry. Conversely, it has not been uncommon for those trained as professional “historians of philosophy” to view the manner in which even their comparatively sympathetic analytic colleagues take up the ideas of the great figures of the past as evincing a peculiarly ahistorical relation to the history of philosophy. This quarrel, which has been conducted throughout the history of the analytic tradition, between card-carrying analytic philosophers and their historically scrupulous professional colleagues, has involved a number of different aspects. There is something to be said for and against each of the parties in this quarrel. Speaking first in defense of the analytic philosopher, it should be noted that it is by no means evident that these tensions are to be traced solely to an unusual degree of hostility on the part of analytic philosophers towards the philosophical past. They may be a function of very different ways of engaging with the past—among which the attitude of the typical analytic philosopher towards prior tradition may in fact represent the more time-honored alternative, far more closely resembling the ongoing philosophical activity of past historical epochs than is generally conceded by the contemporary working historian of philosophy. On this account of the matter, the source of tension enters into the practice of philosophy not through what is strange in the analytic philosopher’s attitude towards the history of philosophy, but rather through what is in fact historically quite parochial in the attitude of the contemporary historian towards the history of philosophy, namely, an insistence on the cultivation and maintenance of a certain form of historical self-consciousness. The form of self-consciousness at issue here was first introduced into the history of philosophy, now itself understood as a form of philosophy, comparatively recently—arguably beginning with Hegel—in any case not much over two centuries ago. Its arrival on the analytic scene is a far more recent—and hence all the more unsettling—event. The irony underlying this line of defense is that the source of the conflict is thus attributed to a respect in which analytic philosophy is actually more traditional in its approach to philosophical problems (precisely in its not requiring the cultivation of historical self-consciousness in order to get down to philosophical business). Or, at any rate, it is far more traditional in its mode of philosophizing than the contemporary historian of philosophy hostile to analytic philosophy has usually been prepared to acknowledge. There has been, and still is, a strongly cultivated tendency within analytic philosophy to approach the writings of the great figures of the past, as nearly as possible, as if they were attempting to make direct contributions to current debates and to treat “the mighty dead” not just as philosophical equals but as philosophical contemporaries. Grice famously remarked that we “should treat great but dead philosophers as we treat great and

38  James Conant living philosophers, as having something to say to us.” Such an approach to the history of philosophy hardly constitutes an unprecedented form of philosophical engagement with the past. In commenting on how best to understand Plato’s concept of an Idea, Kant sums up a longstanding method of engaging with the great figures of philosophy’s past—which he takes to permeate the writings of his great predecessors, such as Aristotle (in his relation to Plato), Aquinas (in his relation to Aristotle), and Leibniz (in his relation to all three). Here is how Kant puts it: I shall not engage here in any literary enquiry into the meaning of the expression. I need only remark that it is by no means unusual, upon comparing the thoughts which an author has expressed in regard to his subject, whether in ordinary conversation or in writing, to find that we understand him better than he understood himself. As he has not sufficiently determined his concept, he has sometimes spoken, or even thought, in opposition to his own intention.21 The interpretive ideal here is to understand a philosophical author better than he understood himself. On the modern historian’s conception of what it is to grasp a philosophical author’s intention, the first order of business is to overcome hindrances introduced by intervening episodes in the history of philosophy—episodes that necessarily obstruct our view of the original intention. On the traditional understanding of the interpretive ideal—to which Kant here gives eloquent expression—the intervening history of philosophy is an indispensable aid in fully determining the author’s concept. (For that might well require forms of philosophical proficiency unknown to the original author.) There is much in the contemporary analytic philosopher’s way of inheriting this traditional ideal that might be irritating to the working historian of philosophy. Nevertheless, the fact remains that the prevalence among contemporary analytic philosophers of a version of this mode of engagement with past philosophy cannot be attributed solely to an unprecedented benightedness in analytic philosophy’s relation to the philosophical past. Williams puts the point well in the following passage, regarding the relation that almost all philosophy has had to at least certain portions of its past—most notably, to Plato and Aristotle: The involvement of Greek philosophy in the Western philosophical tradition is not measured merely by the fact that ancient philosophy originated so many fields of enquiry which continue to the present day. It emerges also in the fact that in each age philosophers have looked back to ancient philosophy—overwhelmingly, of course, to Plato and Aristotle—in order to give authority to their own work, or to contrast it, or by reinterpretation of the classical philosophers to come to understand them, and themselves, in different ways. The Greek philosophers

The Emergence of the Concept of the Analytic Tradition  39 have been not just the fathers, but the companions, of Western philosophy. Different motives for this concern have predominated in different ages. . . . But from whatever motive, these relations to the Greek past are a particularly important expression of that involvement in its own history which is characteristic of philosophy and not of the sciences. . . . [W]e might say that the classical philosophers Plato and Aristotle are classics in the sense that it has been impossible, at least up to now, for philosophy not to want to make some living sense of these writers and relate its positions to theirs, if only by showing why they have to be rejected.22 Nevertheless, it is one thing to view certain philosophers as having the status of living classics in this sense (so that it is impossible for the practicing philosopher not to want to make some living sense of their writings). It is another and much more problematic matter to insist that the terms in which that task of making sense is to be achieved are fully specifiable prior to such a philosophically sustained encounter with the past. Williams’s target here is the attitude towards the history of philosophy famously summarized in Ryle’s frequent injunction to treat something written by Plato as though it had just come out in the most recent issue of Mind.23 The advocate for the professional historian of philosophy might well be able to argue that the attitude towards the past expressed in that injunction is at best naïve and at worse historically obtuse. One reason this quarrel is no longer quite as heated as it once was is because there has recently been a surprising amount of fruitful intellectual interchange between the original parties to the dispute. Some of the recent attempts on the part of scholars trained within the analytic tradition to read major figures of the philosophical past—and, in particular, to read them as far more sympathetic to some particular contemporary analytic project than one might have supposed possible—have occasioned fascinating and influential monographs. They have given rise to further historical scholarship on these figures, which, in turn, has been shaped by these monographs—which, in turn, have led these analytic philosophers to rethink aspects of their original readings of these figures. Indeed, this tradition of analytic historical writing is arguably as old as the analytic tradition. One might even argue that it was initiated by Russell himself, in his lively book, The Philosophy of Leibniz,24 and that the first fruitful instance of an interchange of the aforementioned sort (between analytic philosophers and historians of philosophy) was the one that occurred, immediately after the publication of this work, between Russell and the distinguished French historian of philosophy and Leibniz specialist Louis Couturat. Many of these more recent analytically minded historical monographs have sought to contest the roles in which prior tradition (including the prior analytic tradition) has sought to cast various central personae dramatis in the stories that philosophers have kept repeating to themselves, their

40  James Conant students, and each other. They have challenged the standard analytic conception of what it is to be a Humean, for example, as well as the standard analytic conception of what it is to be a Kantian. Whereas the Humean was once understood by the analytic philosopher to be the figure who debunked our ordinary view of the world in the light of a properly naturalized account of what it might contain, some analytic historians now cast Hume as seeking, by the end of The Treatise on Human Nature, to vindicate much of what we are pre-theoretically disposed to believe in our everyday commerce with the world. Whereas the Kantian was once understood by the analytic philosopher to be the figure who sought to maintain a lasting peace between our everyday and scientific images of the world, some analytic historians now cast Kant as the precursor of modern cognitive science, seeking to show how a properly reconceived form of philosophical psychology furnishes us with a fully naturalized account of the human mind. This has had an effect not only on how the history of philosophy is done, but also on how analytic philosophy understands what it is doing. For many of the major figures in the history of philosophy have become unmoored from the fixed positions once assigned to them in analytic philosophy’s own narrative about how the previous history of philosophy is supposed to have led up to its philosophical present. These analytically informed revisionist readings and re-readings of the history of philosophy have played a part in the analytic tradition’s gaining an increasingly historically informed perspective on its own place within the broader sweep of the history of philosophy. It is now more widely acknowledged than it once was that the analytic tradition is in fact one philosophical tradition among others—rather than a development that culminates and so stands above and beyond the history of philosophy. Contemporary analytic philosophers have begun to recognize that their tradition has nourished stereotypes about its differently minded (non-analytic) neighbors that were as uninformed as they were dismissive, regarding them as, for example, sloppy and overwrought. Their disparaged counterparts have been only too ready to return the disfavor, with equally uncomprehending and dismissive slurs (of which “fussy” and “boring” have been among the more polite). Encouraged by individual efforts at perestroika stemming from each side, there are signs of a gradual thaw in this philosophical cold war. These stereotypes have increasingly come to be regarded as equally prejudicial and uncomprehending on both sides. There has, non-coincidentally, come to be a surge of historical scholarship investigating the ways in which, throughout the history of the analytic tradition, there have been important junctures at which analytic philosophers sought to engage in fruitful dialogue with interlocutors outside their tradition. (To name only three notable examples that have attracted recent scholarly attention on both sides of the Atlantic: Frege’s influential correspondence with Husserl, Ryle’s sympathetic early review of Heidegger’s Being and Time, and Rawls’s late dialogue with Habermas.) The tendency to view such episodes as merely momentary

The Emergence of the Concept of the Analytic Tradition  41 thaws in the cold war has now given way to an interest in the various ways in which the two traditions may have repeatedly cross-fertilized one another in the past—and (even more importantly) how they may continue to do so, for as long as their philosophical identities remain sufficiently distinct to permit such forms of intellectual commerce to be mutually enriching. As the analytic tradition entered the last quarter of the twentieth century and moved into the twenty-first, its resistance to the idea that it represents only one continent in the larger world of philosophy (rather than a movement with a rightful claim to dominate the whole of that world) began to fade. This has helped to transform not only its attitude toward its neighbors as a matter of contemporary philosophical practice, but also its attitude toward itself. A correlative shift has taken place within analytic philosophy in recent decades in the way in which the relation between its philosophical past and its present is conceived. This shift has taken place along a number of dimensions. One aspect of it is the present frequency with which analytic philosophers now seek to enrich their own tradition, and contribute to its further evolution, by working self-consciously to incorporate this or that philosophical line of thought or intellectual strategy drawn from another tradition—in some cases, a contemporaneous one, in others an early modern one, or one that goes back as far as Plato and Aristotle. The beginnings of this development were already occurring in the 1950s (in the work of figures such as Sellars, Strawson, Elizabeth Anscombe, and Peter Geach). By the 1980s, it had become a commonplace to speak of movements and strands within analytic philosophy, such as those of analytic Aristotelianism, Thomism, Pragmatism, Kantianism, Hegelianism, and even analytic Marxism. As we have noted, Bertrand Russell, writing in the middle of the twentieth century, was happy to combine the terms “analytic” and “empiricism” into the novel compound “analytic empiricism,” using the first term to designate what was new in the form of philosophy he was championing and the second term to identify an older strand in the broader philosophical tradition that he sought to inherit, transform, and carry forward. If one had told him that soon there would be philosophers who purported to belong to a tradition that was built in part upon his own early work, but who would describe their philosophical outlook using compound expressions such as those just mentioned, he would have been mystified—and, in some cases, dismayed. For Aristotelianism, Thomism, and the rest were among the very movements in philosophy he was vigorously fighting to displace in favor of his own conception of philosophy as logical analysis. The early Russell would have had difficulty comprehending how the term “analytic” in these different compounds could have anything to do with what he had originally meant by it, and so could amount to anything more than a mere homonym in relation to his own use of it. He could not have foreseen the development of a tradition that would both draw inspiration from the analytic philosophers of his generation and also seek to reincorporate so much

42  James Conant that he himself was determined to eliminate. More generally, from that early vantage point, it would have been impossible for anyone to make out how a tradition might develop out of the work of Russell and the other early analytic philosophers that would be robust and capacious enough to be able to retain its distinctive identity, while reincorporating so many aspects of the previous strands of philosophy that the founders had sought to vanquish. Of the labels mentioned above, the most baffling to Russell himself would have been “analytic Hegelianism.” If there was anything that Russell and Moore in the early years of the twentieth century had been against, and that the first phase of the tradition had succeeded in freezing out as a philosophically respectable option, then it was Hegelianism. This aspect of the thaw was certainly very gradual in coming. Figures (such as Wilfrid Sellars) sympathetic to German Idealism in the generation of analytic philosophers who came of age after World War II, even when outspoken in their enthusiasm for Kant, tended to remain circumspect and guarded in their expressions of admiration for Hegel. It is a mark of how far the situation has evolved since then that, in an omnibus review of five recent works of Anglophone Hegel scholarship, all published in the year 2012, one finds a leading Hegel scholar, Robert Pippin (himself once a student of Sellars’s), reflecting on a robust ongoing tradition of analytic Hegelianism.25 These two forms of interest in the philosophical past—first, the longstanding interest on the part of analytic philosophers in the classic authors of the philosophical tradition (such as Plato, Aristotle, and Kant), and second, the far more recent resurgence of interest in figures previously excluded from the canon (such as Hegel and Marx)—have been further nourished by, as well as themselves, in turn, contributing to, the cultivation of yet a third kind of interest in analytical philosophy’s relation to the past, more specifically a new kind of interest in its own past. 5. THE HISTORY OF ANALYTIC PHILOSOPHY AS A NEW FORM OF PHILOSOPHY This new kind of interest fully matured with the gradual emergence of something called “the history of analytic philosophy”—where the phrase in question refers to an area of philosophical research in its own right within the ongoing pursuit of contemporary analytic philosophy. The aforementioned quarrel between analytic philosophers and professional historians of philosophy—epitomized in Gilbert Ryle’s notorious remark about how one ought to go about approaching a text by Plato—is presently further altering its shape, partly owing to pressures exerted on it by this new form of professional subspecialty within analytic philosophy. As this field has gradually developed, so, too, has a new form of philosophical self-consciousness on the part of many analytic philosophers with respect to the nature and extent of that which is historically local in their own philosophical tradition. It

The Emergence of the Concept of the Analytic Tradition  43 has given rise to the possibility—for practitioners and students of analytic philosophy alike—of encountering aspects of analytic philosophy’s own history as something remote and even alien, so that a confrontation with that history can itself become an occasion for philosophical reflection.26 The writings of the historian of analytic philosophy provide yet another perspective upon our topic than those of its practitioners and ideologues—one that may likewise serve as a resource in seeking answers to our two guiding questions. In this case, it would be rather more difficult to compile a comparably perspicuous list of statements representative of the various outlooks harbored by practitioners of this newly emerging discipline. For our purposes, it will suffice to remark briefly upon some of what an examination of exemplary instances of their work would bring to light that is relevant to our task here. One thing it would quickly reveal is that a good historian of analytic philosophy is not merely a historian of ideas. She is also a philosopher—and necessarily so, for several reasons. First, the task of grasping the philosophical power of a way of thinking that is occluded by our present preconceptions is always a philosophical as well as a historical one. Second, many a historian of analytic philosophy is moved in part by philosophical motives—sometimes seeking to make something in the analytic past that has become alien to many today an available resource for understanding what analytic philosophy might or ought to be in the future, and sometimes simply desiring to recover some bit of lost treasure from an earlier stratum of the tradition. When practiced with an eye to changing the present of philosophy, the discipline of the history of analytic philosophy can become saddled with difficulties that do not as obstinately beset scholarship on the history of other philosophical traditions—at least not in the same way and to quite the same degree. Correlatively, the pronouncements of the historian of analytic philosophy can meet with visceral forms of resistance from contemporaries in the discipline who are deeply invested in certain entrenched narratives of how the tradition unfolded. A convincing unmasking of these narratives requires the attainment of a form of self-understanding that is in equal parts historical and philosophical. The claim by a historian of analytic philosophy that the early Russell’s or Frege’s conception of “logic” or “analysis” is quite different from the manner in which these terms have come to be construed by contemporary analytic philosophers, for example, may be received by some with bitterness and resentment. This form of historical claim can seem to threaten certain essential aspects of a contemporary analytic philosopher’s sense of her own philosophical identity. Analytic philosophy, throughout much of its history, has been extraordinarily resistant to the very idea that it so much as has a history (in the relevant sense of what it means to say that a tradition “has a history”). Of course, no one denies that some authors lived before others and influenced

44  James Conant successors who in turn lived and worked at some later point in time. In this trivial sense of what it means to “have a history,” analytic philosophers are happy to regard what they do as participating in an ongoing enterprise that has a history. Indeed, they tend to be deeply committed to a certain tidy account of what that history must have been—who the founding fathers were, what the defining statements of the tradition were, which pieces of writing count as paradigms of philosophical analysis, and the like. This potted account of the history of the tradition—now enshrined in numerous introductory textbooks and encyclopedia articles—often plays a constitutive role in various analytic practitioners’ respective understandings of the very enterprise that they themselves seek to continue in doing (what they themselves still want to call) “analytic philosophy.” What analytic philosophers tend to resist is the far more unsettling idea that the tools of the historian’s trade are relevant for getting at the truth about those very philosophical episodes—those that play a tradition-defining role in this internally propagated narrative. What is unsettling is the idea that those tools might turn out to be essential for achieving a faithful understanding of what prior generations of analytic philosophers actually meant when they employed terms that continue to circulate widely throughout the writings of analytic philosophers today—terms such as “logical constant,” “syntax,” “semantics,” “proposition,” “concept,” “meaning,” “reference,” “language,” “judgment,” “inference,” “justification,” and the like. Analytic philosophy has tended to want to imagine that it does not have a history in just this sense; it has wanted to believe that its philosophical past is fully transparent to its philosophical present. For example, contemporary analytic philosophers have been prone to assume that they can just pick up an early classic of the tradition (such as Frege’s essay “On Sense and Reference” or Russell’s essay “On Denoting”) and fully unpack its intended upshot simply by drawing on their (present-day) understanding of the terminology used, without needing first to examine how their assumptions about how philosophy ought to be done relate to those of these earlier authors. They likewise tend to assume that such a text may simply be placed into the hands of their students to be read and understood by them, without any prior effort on their part to properly orient the students in relation to a way of thinking that may well be philosophically foreign to them. The assumption that analytic philosophy’s past must be transparent to its present goes together with the supposition that there is no special need for analytic philosophers, when reading a text from an earlier moment in their own tradition, to seek out the expertise of the historian of analytic philosophy. There is no sense that forms of historical sensitivity might be cultivated that would enable them to attain a perspective on what is going on in that text, which, in turn, could open up a further perspective on their own practice of philosophy—vastly expanding their sense of the philosophical distance that separates analytic philosophy’s present from its past.

The Emergence of the Concept of the Analytic Tradition  45 The problem here is not only that a certain obliviousness tends to prevail among analytic philosophers regarding what such a historically informed perspective might disclose. It is that there often is a positive repression of the possibility of such disclosure—for the very reason that many a potentially liberating insight is repressed—namely because it threatens to unsettle our fantasies regarding who we are and what we can do. When the analytic philosopher confronts the historian of analytic philosophy’s way of reading one of the tradition’s cherished classics, an enormous chasm can suddenly appear to open up between what the text has always officially been held to have said and what it now actually seems to be concerned to say. The text suddenly takes on the double-aspect of a duck-rabbit figure—with two mutually occluding aspects. It can come into view under its familiar and reassuring aspect, induced by the reading encouraged by a certain tidy canonical account of how analytic philosophy’s past is related to its present. Or an entirely new aspect can dawn, once the reader’s entire experience of the figure/ground relation in the text (between its sentences and the thoughts they express) is framed in an entirely new way. One of the methods historians of analytic philosophy have effectively employed to trigger such gestalt-shifts is through revealing the extent of the gulf that looms between what the key terms in the text once meant (when understood against the background of the no longer easily visible philosophical assumptions of the original author) and what they—or, in important instances, the English words used to translate them—are generally understood to mean today (when read against the background of the often invisible and thus generally unexamined philosophical assumptions of the present). In order to be able to survey the overall topography of such a gulf, one must also possess the particular philosophical capacities required to survey the fine structure of both backgrounds: both its historically proximate edge (such as the unexamined philosophical assumptions against which contemporary analytic philosophers approach their problems and those of their interlocutors) and its historically distal one (such as the shifts in meaning that a shared philosophical terminology may undergo over the course of several intellectual generations). Such forms of recognition of what is intellectually consequential in “what goes without saying” are forms of philosophical achievement: they serve to render visible much that otherwise remains invisible in contemporary analytic philosophical discussion, precisely because the discussants themselves are wont to regard those ways of thinking that come most naturally to them as those that are philosophically least problematic and most self-evident. The vocation of the historian of analytic philosophy can appear to both the contemporary analytic philosopher and the contemporary historian of philosophy to fall between two stools. It can seem, on the one hand, to be too committed to and involved in historical scholarship to count as genuinely analytic philosophy, and yet also to be too narrowly preoccupied by the methods, concerns, and aims peculiar to the analytic tradition to count

46  James Conant as serious history of philosophy. What the good historian of analytic philosophy can do, however, is to demonstrate that this pursuit is an integrated form of inquiry that requires the cultivation of the virtues and competences of both a scrupulous historical-philosophical scholar and a sophisticated participant in contemporary analytic philosophical practice. Good historians of analytic philosophy can show where and how the assumptions and concerns of contemporary analytic philosophers are not those of their analytic forebears only if they have attained a fully integrated mastery of these two forms of philosophical competence. Such a twofold fluency is essential, if they are to be able to reveal how methods and aims (and, along with them, the meanings of many a familiar piece of philosophical terminology) have shifted over the course of the history of the analytic tradition, and to identify and illuminate cases in which forms of philosophical statement employed by contemporary analytic philosophers belong to frameworks of thought very different from those that conferred meaning on the apparent linguistic twins of those statements in the writings of their analytic predecessors. As noted above, there are some respects in which the difficulties faced by a historian of analytic philosophy resemble those that beset a historian of science more than those typically encountered by the philosophically minded scholar of other chapters of the history of philosophy. Correlatively, the forms of resistance the historian of analytic philosophy faces can resemble those encountered by, say, the historian of twentieth-century physics. Contemporary physicists often find themselves disturbed by the accounts of major revolutions in the history of physics—especially some involving comparatively recent episodes (such as those that led to the theories of relativity and quantum mechanics)—advanced by historians of science. The practicing physicist, like the practicing analytic philosopher, is wedded to a narrative in which the achievements of figures such as Albert Einstein and Niels Bohr are presented in a very particular way, namely, as responses to challenges and difficulties that are describable in terms equally intelligible to both the past and the present practitioner of the subject. It is this transparency of past physics to the present that a sensitive historian’s account often threatens to undo—thus apparently depriving contemporary physicists of their working understanding of the place their own contributions assume in a single ongoing enterprise. The historian of science seeks an entirely different order of intelligibility in the past than that which is conferred on it by an official textbook-level narrative of how the innovations of the mighty dead led to our contemporary understanding of the topic under consideration. What the historian of science wants to understand is not the reason that, with the hindsight of later development, now seems to a contemporary physicist to be the obvious basis for adopting our contemporary understanding of what Einstein’s, or Bohr’s, original conclusion must have been. The historian rather is concerned to uncover and sort out the tangle of now forgotten, but back then nagging (and—but only for those who had eyes to see—deeply significant)

The Emergence of the Concept of the Analytic Tradition  47 puzzles and anomalies that moved an Einstein or a Bohr, at that particular moment in the history of physics, to draw what could only seem to his contemporaries to be an altogether surprising (and not at all easily intelligible) conclusion. From the point of view of the historian, this requires doing full justice to every nuance of the many large and small differences between our present and (sometimes even only slightly) earlier ways of thinking about physical reality—nuances that are all simultaneously erased from view in Whiggish textbook accounts of the history of science. From the point of view of the physicist, such a historically nuanced account—with its seemingly myopic preoccupation with theoretically and experimentally secondary considerations—is heedless of the substantial extent to which our contemporary understanding of physical reality is able—indeed, must be able—to encompass and comprehend the point of view of an Einstein or a Bohr. In the cases of both contemporary physics and contemporary analytic philosophy, an investment in a similarly deeply entrenched and internally institutionalized narrative plays a parallel role in the quarrel with the historian. It is no accident that, in each case, this same narrative plays a pedagogical role in initiating students into the subject. And these are narratives that the conscientious historians of both subjects may well feel they must at least question and complicate (if not altogether subvert), if the actual contours of the relation between the present and the past—recent as well as more distant—are to come into view. These internally propagated disciplinary narratives of how past achievements led up to the present—in both theoretical physics and analytic philosophy—tend to represent the terminology, methods, and aims of the past as essentially homogeneous in intellectual form and content with those of the contemporary practitioner. They present the original problems, concerns, and aims of the founders as versions of current ones in the disciplines. To observe that the historical soundness and adequacy of such narratives cannot simply be taken for granted is not to deny that they have any legitimacy or usefulness. Indeed, they may have an essential role to play in helping to articulate and promulgate a certain widely shared (albeit often largely inchoate) understanding of the ongoing practice. We encountered various versions of such understandings in the statements presented above as representative of putatively authoritative stances taken by (those whom we referred to as) ideologues of analytic philosophy. And we suggested that collections of statements of that sort, when appropriately arranged and displayed, can serve to bring out significant features of analytic philosophy’s own multifarious self-image. What we have seen now, however, is that the history of analytic philosophy, if it is to perform its office as a serious branch of the discipline of history, must call into question and be prepared to contest such disciplinary self-images and related proclamations. But, unlike the case of the history of science, this is not its only office—nor is it anything like the primary reason why analytic philosophers are generally moved to

48  James Conant become serious historians of their own tradition (while usually also seeking to remain analytic philosophers). Its most important function is arguably to enhance, deepen, and further orient analytic philosophy’s own ongoing philosophical understanding of itself—upon which its developing practice depends. What this suggests is that the history of analytic philosophy’s most important function is not one that it shares—or even could share—with the history of science. For there is an absolutely crucial difference between the history of a science (like theoretical physics) and that of analytic philosophy—a difference that is clearly visible in the very different ways in which these two disciplines are generally practiced. Competent historians of physics are not out to make (and do not see themselves as seeking to make) contributions to contemporary physics. They think of themselves as historians rather than as physicists. As we have already noted, however, good historians of analytic philosophy tend to be (and to want to be) practitioners of the very discipline of which they also undertake to be historians. In non-analytic philosophical circles, there is seldom any presumption that these two forms of identity and inquiry (that of the historian and that of the philosopher) must exclude each other (even though they may be recognized to stand in a certain productive tension with each other, requiring careful negotiation), whereas among analytic philosophers—as among physicists—there has, until recently, often been such a presumption. Thus the capacity to fuse—or otherwise juggle—these two forms of identity within the space of a single philosophical life has until recently remained a comparatively rare achievement in the analytic tradition, exemplified in the work of only a handful of figures. This is gradually beginning to change: leading figures in the tradition now find themselves not only increasingly pushed but also naturally inclined to articulate their own respective plausible readings of just what it was that Carnap, Ryle, or Anscombe might originally have wanted to mean in this or that frequently quoted (but previously seldom carefully read) remark, article, or book. While historians of physics have no stake in the outcome of ongoing disputes in the current cutting-edge of theoretical physics, historians of analytic philosophy generally do have such a stake in contemporary philosophical disputes—not necessarily in their minutiae, but in larger questions prompted by their ongoing conduct. These often involve questions such as whether a current controversy is a repetition (albeit in a different guise) of a prior one, or whether (its novelty notwithstanding) the present game is worth the candle, or whether (in dominating the contemporary horizon) it obscures from view promising avenues for philosophical reflection. Motives such as these often enter into the historian of analytic philosophy’s particular way of reframing and narrating a carefully selected episode from the history of the tradition. The misunderstood or neglected aspects of the philosophical past thereby displayed are therefore almost never selected with the mere scholarly aim of setting the historical record straight, but

The Emergence of the Concept of the Analytic Tradition  49 with a further specifically philosophical aim. That motive may be just to display the sheer provinciality of the philosophical present, or the power of a philosophical idea of the past, or the unwitting faithfulness with which the present recapitulates the past due to ignorance of it—but, most likely, it will be a combination of all three, along with yet others. The historian of analytic philosophy’s motives are thus generally mixed, springing no less from a desire to show how certain—often unduly neglected—aspects of the past can still speak to us (thereby holding open the promise of enriching the subject as we know it) than from a desire to show how certain—often wrong-headedly celebrated—aspects of the past cannot any longer speak to us (thereby allowing us to appreciate what is philosophically and historically specific to our present moment of philosophy). Conscientious historians of analytic philosophy tend to be cautious in approaching the central terms analytic philosophers have continually employed over the course of the tradition to characterize their topics, problems, or methods (including terms that have occurred frequently in the preceding pages—such as “formal,” “logic,” and “analysis”). They probe carefully to ascertain the extent to which these terms have or have not retained their original significance across their successive occasions of use. They are uniquely positioned—and needed—to bring out the character and structure of certain sorts of moments that occurred at significant junctures in the development of the tradition—moments in which two major figures inspired by the same philosophical texts and teachers, apparently pursuing a common project within a shared framework of philosophical endeavor, employing the same terms and declaring allegiance to the same intellectual paradigms, nonetheless utterly fail to engage one another philosophically—talking by or otherwise misunderstanding one another—despite the appearance of their pursuing a single common project and working within a single shared framework of philosophical endeavor. The good historian of analytic philosophy may bring out how two philosophers who appear to agree on fundamentals are only apparently in agreement with one another, as well as how two philosophers who appear to disagree actually do not—either because they really agree when they take themselves not to or because they are philosophically so far apart that their positions are not even sufficiently aligned to permit of disagreement in the first place. Finally, the good historian of analytic philosophy can reveal how two figures in the history of philosophy—perhaps only one of whom is an analytic philosopher—may actually have far more in common with one another than either one of them would have been willing to allow or could have been in a position to comprehend. This requires showing how the underlying projects of these two philosophers, belonging to different movements of thought (outwardly characterized by utterly different intellectual styles and temperaments), are inwardly bound together by profound affinities. In more extreme cases, the good historian of analytic philosophy may

50  James Conant even be concerned to reveal how an entire episode of analytic philosophy is to be seen as significantly related to some prior moment in philosophy’s past. Gilbert Ryle has figured in some of our remarks above as typifying a certain sort of somewhat dismissive analytic attitude towards the philosophical past. Yet he, too, in his own way, was deeply concerned to make contributions to this genre of analytically informed history of philosophy. For example, we may see Ryle’s work on the relation between Plato and logical atomism as animated by such a concern. This work was concerned to reveal how six figures in the history of philosophy, four of whom (Moore, Frege, Russell, and early Wittgenstein) were analytic philosophers and two of whom (Plato and Meinong) were not, may have had more in common with one another than at least several of them could possibly have been in any position to comprehend. Ryle argues that Moore’s 1899 essay on “The Nature of Judgment” provides an account of the distinction between propositions (in a non-linguistic sense of that term) and concepts (taken to be the elements of propositions) that doesn’t work and that runs into essentially the problem that is expressed in the Theaetetus (as how we can think that which is not). He also argues that what we find in Meinong’s treatment of “objectives” and Russell’s treatment of facts (especially in his Lectures on Logical Atomism) are unsuccessful attempts to resolve this problem by continuing to frame it in the manner in which Moore takes it up. He contrasts this with the approach to the problem we find first in Frege and then, to an even clearer degree, in the Tractatus—one which in various respects resembles that proposed by Socrates in the dialogue to a surprising degree. Ryle then summarizes what he has shown thus far as follows: I now urge that it is pretty clear that the issue that Socrates was discussing is the same as or at least overlaps with the issue that was being discussed fifty to thirty years ago by, among others, Meinong, Moore, Russell and Wittgenstein; and that Socrates at least adumbrated certain ideas very much like those which were rendered necessary by some of the inherent defects of the theories of objects or concepts originally put forward by Meinong and Moore.27 To defend a claim of this form requires making clear the affinities between the philosophical preoccupations of those who are traditionally counted as analytic philosophers and those who are not. To the extent that one judges Ryle’s attempt to do this to be successful, one must also concede that it permits one to see the form of a widely shared philosophical problem more clearly than one had before. Once affinities of this sort between the analytic and the non-analytic past are brought sharply into view, this may enable us to discern more clearly not only the historical landscape, but also the philosophical landscape. For it can enable a clear apprehension of the very form of a philosophical problem for the first time—allowing us to separate the real form of the problem from the superficial guises through which

The Emergence of the Concept of the Analytic Tradition  51 it simultaneously manifests itself in the work of apparently very different thinkers. In this and other ways, the work of the good historian of analytic philosophy—utterly unlike the work of the historian of science in its relation to contemporary science—can, indeed, contribute to the achievement of new and surprising modes of philosophical progress. The analogy between the history of physics and the history of analytic philosophy breaks down in a further—equally instructive—way. The analogy, as framed above, encourages one to think that the primary misrepresentation of the past the historian must seek to undo is one that arises from an institutionalized tendency to represent the problems, concerns, and aims of earlier heroes in the discipline as if they were immediately recognizable as versions of contemporary ones. Yet some analytic philosophers have been drawn to re-narrate episodes from the history of their tradition out of a desire to correct a roughly opposite form of misrepresentation of the past. Consider the following three texts—each of which marked at the time of its publication a significant transformation in the tradition’s self-image: (1) G. E. M. Anscombe’s 1959 book An Introduction to Wittgenstein’s Tractatus, (2) James Griffin’s 1964 book Wittgenstein’s Logical Atomism, and (3) Peter Hylton’s comparatively recent 1990 book Russell, Idealism and the Emergence of Analytic Philosophy.28 The internally propagated narratives that these three authors aim to subvert, in their three remarkably different books, stand in a different sort of relation to each of the then dominant self-understandings of analytic philosophy. Anscombe and Griffin aimed in different ways to subvert an account of early Wittgenstein’s place in the history of analytic philosophy advanced in numerous contemporaneous narratives—perhaps none more influential at the time than J. O. Urmson’s 1956 book Philosophical Analysis: Its Development between the Two World Wars.29 Like many others writing at that moment, Urmson wanted to bring out the radical nature of a recent shift in analytic philosophy (and hence to minimize its discernible continuity with even the quite recent past). On Urmson’s telling of the story, the author of the Tractatus fully shared his teacher Russell’s stridently empiricist, atomist, and largely Humean philosophical orientation. Urmson (along with his likeminded philosophical cohort) did not regard the common thread he discerned throughout the early analytic tradition (epitomized by Russell and early Wittgenstein) to be anything like a version of his own generation’s central philosophical concerns and aims. That was his whole point: we have broken with even the very recent past of analytic philosophy. He spoke for a generation eager to see themselves as beneficiaries of a mid-century philosophical revolution that had effectively separated and liberated them from the concerns and aims of their earlier analytic forebears. The larger historiographical point here is that the shape of the target of the historian of analytic philosophy is sometimes roughly the opposite of that of the historian of science—Urmson’s account was to be exposed as insufficiently, rather than excessively, Whiggish—as overstating, rather

52  James Conant than understating, the seismic character of the revolution. His work sought to disseminate an account of how very sharp the break with the past really was—to make palpable the extent to which it had issued in forms of philosophical practice unimaginable to the original founders of the analytic tradition. Anscombe and Griffin, in their respective books, were challenging an account of this form (though each took issue with very different aspects of the then prevailing narrative). Their work thus offers an example a very different kind of philosophical motive a historian of analytic philosophy may have for seeking to overturn an institutionalized narrative. The concern may be to show the depth of philosophical discontinuity where seamless continuity is the prevailing assumption of the day, but it may equally well be the reverse. One of the many interlacing philosophical motives of Hylton’s superb book on Russell bears a certain resemblance to an aim of Anscombe’s and Griffin’s, namely, to shake us free of an account of the thought of a central figure in the analytic tradition in which that figure is cast as a representative of a tradition of philosophy involved in a fairly direct inheritance of the central assumptions of British empiricism. Whereas Anscombe and Griffin were primarily concerned to overturn a Humean reading of early Wittgenstein, Hylton is concerned to complicate a reading of Russell’s philosophy as developing too simply and too directly out of a set of philosophical commitments he supposedly shares with a figure such as Hume. Though such exercises in the history of analytic philosophy are driven by philosophical motives, this does not necessarily mean that they come with explicit methodological reflections on the forms of philosophical insight that the history of analytic philosophy can afford. Neither Anscombe nor Griffin in their aforementioned books offers anything like a self-conscious statement of what the historian of analytic philosophy can or should do. In this respect, they are typical of even the most philosophically original contributors to the history of analytic philosophy prior to the 1980s. In this respect, Hylton’s book is more representative of the best recent work, displaying forms of self-reflexivity seldom found in earlier work in the genre. In a particularly eloquent passage from the Introduction to the book, we find the following set of remarks: Philosophy cannot, as the natural sciences perhaps can, absorb what is correct in its past and conclusively refute what is incorrect, for the difference is unsettled. There is as little finality in our views as to what is correct in the philosophies of Plato or Hume or Kant or Russell as there is in our views on the most contemporary issue. . . . Philosophy thus always has the hope of learning neglected lessons from its past. It also, and perhaps more characteristically, is always in a state of potential rivalry with its past, defining itself against its past, and threatened by it. It is for this reason that the history of philosophy often has an evaluative and judgmental tone—precisely not the tone of one who has a

The Emergence of the Concept of the Analytic Tradition  53 secure understanding of the matters at issue, but the tone of one whose understanding is threatened. The deliberately ahistorical character of much history of philosophy seems to me not accidental, but a product of this insecure relationship between philosophy and its past. We approach the past ahistorically in order to refute it—as if the past of philosophy will not stay in the past, but constantly threatens to come back to life. Our uncertainty over the history of philosophy—whether it is history, whether it is philosophy, whether it can be both—seems to correspond to the uneasiness of the relation between philosophy and its past, and to our unease about the status of the subject as a whole.30 This is as thoughtful and penetrating a set of opening remarks regarding the relation between philosophy and its past as one is likely to find at the outset of any work on a topic in the history of philosophy. Or, to put the point the other way around, it is thoroughly unrepresentative of what one finds, throughout most of the history of analytic philosophy, in writing devoted to furnishing a historical overview of some major period or figure or movement within analytic philosophy. The topic of Hylton’s own book (the development of Russell’s philosophy, his early revolt against British Idealism, and his ongoing responses to the resulting internal tensions in his thought) is a classic topic in the genre of the history of analytic philosophy—as classic as you can get. Most of the work done on this topic—and, indeed, in this whole genre—is written in just the tone Hylton mentions above, and for the reason he gives. The unacknowledged unease in the tradition’s relation to the past at issue here is due in no small part precisely to a desire to have that relation be an easy one—one of total continuity or sharp discontinuity—as long as it allows analytic philosophers to look back upon their tradition simply with a view to absorbing what is correct in it and conclusively refuting what is false in it. The ensuing unease in the relation to the subject as a whole is therefore nourished by a desire to rid our relation to the past of the very dimensions of complexity and ambivalence that form constitutive aspects of philosophy’s ongoing encounter with its past. Indeed, analytic philosophy’s ambition to free itself from certain forms of preoccupation with history—an ambition characteristic of so many of the founding projects of the tradition—is part of what has given rise (at this much later stage in the history of the tradition) to the present felt need for a particular sort of philosophically sensitive work in the history of analytic philosophy—a specifically historical-philosophical form of exercise in remembering, repeating, and working through—able to undo specific forms of philosophical repression induced by the original founding ambition. Hylton’s official topic in the above quotation is the relation between philosophy as such and the entirety of its past—not our present and narrower topic, namely, analytic philosophy’s relation to that tiny chapter in the history of philosophy that is its own past. Yet what Hylton shows in his book

54  James Conant bears directly on our topic. For the book brings out any number of ways in which analytic philosophers have become invested in narrating the past of their own tradition in ways that repress to an extraordinary degree any consciousness of the forms of difficulty that (as the passage above suggests) necessarily attend philosophy’s relation to its past. Viewed from this angle, one of analytic philosophy’s most characteristic features would appear to be one that failed to show up on our collection of statements above, namely, the tradition’s sustained investment in trying to rid itself of the awareness that it (like any other form of philosophy) is subject to the vicissitudes of philosophy’s relation to its history. And it is no accident that it failed to appear there—not only because it is a more subtle sort of feature than any of those we have discussed. It is a sort of feature that can come properly into view only once the tradition’s prior retrospective relation to itself is viewed through the lens of the sort of philosophically sensitive work in the history of analytic philosophy discussed in the previous paragraphs. This, in turn, suggests that—as the history of analytic philosophy practiced as a form of analytic philosophy itself comes to be an increasingly significant and respectable subspecialty within the discipline—this characteristic feature of the tradition (like so many others) must gradually mutate, eventually coming to be an ever less definitive mark of the tradition as a whole. That this particular subspecialty has come to be conducted with an eye to transforming the shape of ongoing contemporary philosophical debate is non-accidentally related to the way in which it has also gradually come to be regarded as itself constituting a self-standing form of philosophically inquiry in its own right. This is a genuine and significant development within the analytic tradition. It involves the emergence of a philosophically self-conscious form of historical inquiry in the history of analytic philosophy conducted by analytic philosophers writing primarily for an audience of analytic philosophers. It is a form of historical inquiry not subordinated to any particular philosophical agenda, but open to the whole range of forms of understanding that may be afforded when analytic philosophy’s present is confronted with the sort of philosophically informed selective focus on aspects of its past discussed above. Hylton’s book, for example, seeks simultaneously to make some aspects of Russell’s philosophy seem far stranger than they had been taken to be, while revealing other aspects to have far more bearing on the fundamental difficulties that plague the present moment of analytic philosophy than might previously have seemed possible. Russell’s thinking is thus shown to be, in some ways, far more surprising than the tradition’s self-image had been willing to allow, while other supposed historical platitudes about the tradition are shown to cover up the most interesting ways in which his concerns are far more philosophically akin with our own than had previously been recognized. Good historians of analytic philosophy will by no means simply converge upon some single alternative to the currently institutionalized account of the

The Emergence of the Concept of the Analytic Tradition  55 history of analytic philosophy. Here, as elsewhere in the practice of history, uncovering the historical past involves appreciating the revelatory powers of different forms of account. But for all of their differences, they would not be good historians if they were to take as their point of departure any particular definition of what analytic philosophy or the analytic style as such is—unless their purpose in doing so is to call it into question or employ it to illustrate how misleading such blanket statements prove to be. They will seek instead to characterize the historical episode of thought at issue precisely as part of an ongoing and internally evolving tradition—with all of the internal complexity and disagreement that is apt to characterize any interesting historical tradition of thought, be it literary, mathematical, or philosophical. Thus the historian of analytic philosophy is far more likely than the ideologue of analytic philosophy to see the history of analytic philosophy as consisting in a series of successively mutating conceptions of philosophy, rather than as the grand unfolding of a unitary something called “analytic philosophy” that can be aptly summed up in the form of a definition or summary statement of its aims, commitments, or style. But if that is true, then what is the history of analytic philosophy a history of? What unifies the diverse, evolving, and contested enterprise that the historian of analytic philosophy seeks to display? To answer this question as well as it can be answered, we must have recourse to the concept of a tradition. The unity and identity of a tradition is not explicable in terms of a collection of features each of its members fortuitously happens to instantiate. It is explicable only through a form of understanding that seeks to grasp a specific sort of historical development—one in which each moment is linked to the others in a significant way. Reflection on the significance of each such moment possesses the power to illuminate that of any other—but only when they are collectively considered in the light of their partially overlapping and mutually intertwining relations with one another. The concept of a tradition shows its worth when, through concerted attempts to engage in such reflection, we actually do find our appreciation of each of the elements in a series of historical episodes coming to be deepened in this mutually illuminating way. When such acts of reflection bear fruit in this manner, what they uncover is revealed to be not merely a “series of historical episodes,” but, rather, the successive moments of the internal unfolding of a tradition. The unity of analytic philosophy here at issue is to be sought not at the level of the doctrines, or the conception of philosophy, or the style of the writing of its practitioners, but rather in the manner in which it forms a distinctive tradition of thought. Once the concept of the analytic tradition comes to function in this sort of way in shaping the self-understanding of practicing philosophers seeking to inherit and develop the tradition in question, it becomes—not merely the concept of a certain philosophical ideology or sensibility, but rather—the concept of a form of philosophical self-consciousness.31

56  James Conant NOTES  1 For an illuminating discussion of this point, see Simon Glendinning’s provocative editor’s introduction to The Edinburgh Encyclopedia of Continental Philosophy (Glendenning 1999).   2 This is not to say that a merely geographical principle for distinguishing between the traditions of philosophy here at issue was ever of much merit. (Michael Dummett has pointed out its limitations as a way of distinguishing the early history of the two traditions in question; see Dummett (1993). It is only to say that the time has come when it is now a genuinely comical way of distinguishing them.   3 It should be noted that, if my remarks below do go any way in illuminating how the term “the analytic tradition in philosophy” might pick out something with a certain sort of unity, this does not thereby show wherein the unity might lie in something genuinely deserving of the name of a Continental tradition in philosophy.   4 This point is forcefully argued by Peter van Inwagen; see van Inwagen (2006).   5 Moore (1993).   6 Russell (1945), p. 834.   7 Wittgenstein (1922), §§4.111–2.   8 Schlick (1927), p. 223.   9 Carnap (1959), §§1 & 9. 10 Wittgenstein (1953), §§109. 11 Ryle (2002), p. 1. 12 Austin (1961), p. 182. 13 Strawson (1964), p. 9. 14 Quine (1981), p. 21 15 Grice (1986), p. 61. 16 One way to summarize the point here at issue is to say that Putnam holds, along with Kant, that the Schulbegriff of philosophy must be brought into equipoise with its Weltbegriff; for further discussion, see my editor’s introduction to Hilary Putnam, Realism with a Human Face (Conant 1990), pp. xxiv–xxxii. 17 Grice (1986), p. 64. 18 Russell (1912), p. 345. 19 Wright (1996), p. 252. 20 These remarks form part of a text that Williams specially wrote for a francophone audience—namely, the Preface to L’éthique et les limites de la philosophie (Williams 1990), the French translation of his Ethics and the Limits of Philosophy. Related remarks, developing his thoughts on this subject at greater length, can be found in English in his “What Might Philosophy Become?” included in his collection Philosophy as a Humanistic Discipline (Williams 2008). 21 Critique of Pure Reason, B 371–372. 22 Williams (2006a), p. 9. 23 In his “An Essay on Collingwood,” Williams (2006b) discusses the conception of the history of philosophy he takes to be implicit in Ryle’s injunction; see especially p. 344. 24 For a forerunner to the later fashionable distinction between two essentially opposed ways of doing the history of philosophy—merely historically and genuinely philosophically—see the opening pages of Russell’s Preface to the First Edition of The Philosophy of Leibniz (Russell 1900). For further discussion, see Ayers (1978) (see especially pp. 42–46).

The Emergence of the Concept of the Analytic Tradition  57 25 Pippin remarks: “Given the origins of analytic philosophy in anti-Hegelianism, perhaps the most surprising new Hegel is the Anglophone Hegel . . . show[ing] that Hegel could make a living contribution to contemporary debates in philosophy, in the way that Anglophone philosophy has long done for philosophers like Plato, Aristotle, and Kant.” (Review of Andrew Shanks, Hegel and Religious Faith; Peter Hodgson, Shapes of Freedom: Hegel’s Philosophy of History in Theological Perspective; Sally Sedgwick, Hegel’s Critique of Kant; Terry Pinkard, Hegel’s Naturalism; Christopher Yeomans, Freedom and Reflection, in Times Literary Supplement, Spring 2013). 26 I am indebted to the discussion of this topic in Michael Kremer’s “What is the Good of Philosophical History?” (Kremer 2013), as well as to Kremer’s review of Scott Soames’s Philosophical Analysis in the Twentieth Century. 27 Ryle (1990), p. 42. This article was published posthumously. 28 Anscombe (1959), Griffin (1964); Hylton (1990) 29 Urmson (1956). 30 Hylton (1990), pp. 6–7. 31 The present essay was extracted from a longer manuscript and overlaps at diverse points with material originally written for various sections of The Norton Anthology of Philosophy, Volume V. After Kant: The Analytic Tradition (Conant and Elliott 2016). I am indebted to Cora Diamond, Jay Elliott, and Richard Schacht for comments and suggestions on the original manuscript, as well as to the assistance of Jeffrey Bell, Andrew Cutrofello, and Paul Livingston in helping me to trim the material down to its present comparatively reasonable length.

WORKS CITED Anscombe, G. E. M. 1959. An Introduction to Wittgenstein’s Tractatus. London: Hutchinson University Library. Austin, J. L. 1961. Philosophical Papers. Ed. J. O. Urmson. Oxford: Oxford University Press. Ayers, M. 1978. Analytical Philosophy and the History of Philosophy. In Philosophy and Its Past. Sussex: Harvester Press, 42–66. Carnap, R. 1959. The Old and the New Logic. In A. J. Ayer, ed., Logical Postivism. New York: The Free Press, 133–146. Conant, J. 1990. Introduction. In Hilary Putnam, ed., Realism with a Human Face. Cambridge, MA: Harvard University Press, xv– lxxiv. Conant, J. and J. Elliott, eds. 2016. The Norton Anthology of Philosophy, Volume V. After Kant: The Analytic Tradition. New York: W. W. Norton & Co. Dummett, M. 1993. The Origins of Analytic Philosophy. Cambridge, MA: Harvard University Press. Glendinning, S. 1999. What Is Continental Philosophy? In S. Glendinning, ed., The Edinburgh Encyclopedia of Continental Philosophy. Edinburgh: Edinburgh University Press, 3–20. Grice, P. 1986. Reply to Richards. In Richard Grandy and Richard Warner, eds., Philosophical Grounds of Rationality. Oxford: Clarendon Press, 45–108. Griffin, J. 1964. Wittgenstein’s Logical Atomism. Oxford: Clarendon Press. Hodgson, P. 2012. Shapes of Freedom: Hegel’s Philosophy of History in Theological Perspective. Oxford: Oxford University Press. Hylton, P. 1990. Russell, Idealism and the Emergence of Analytic Philosophy. Oxford: Clarendon Press.

58  James Conant Kremer, M. 2005. Review of Scott Soames’s Philosophical Analysis in the Twentieth Century. (published in Notre Dame Philosophical Reviews, available online at http://ndpr.nd.edu/news/24868-book-1-philosophical-analysis-in-thetwentieth-century-vol-1-the-dawn-of-analysis-book-2-philosophical-analysis-inthe-twentieth-century-vol-2-the-age-of-meaning.) Kremer, M. 2013. What Is the Good of Philosophical History? In E. Reck, ed., The Historical Turn in Analytic Philosophy. Palgrave MacMillan, 294–325. Moore, G.E. 1899. “The Nature of Judgment,” Mind 8 (30):176-193. Moore, G. E. 1993. The Nature of Judgment. In Thomas Baldwin, ed., G. E. Moore: Selected Writings. London: Routledge, 1–19. Pinkard, T. 2012. Hegel’s Naturalism. Oxford: Oxford U. Press. Pippin, R. 2013. Review of Andrew Shanks, Hegel and Religious Faith. XXXX: Times Literary Supplement. Quine, W. V. O. 1981. Things and Their Place in Theories. In Theories and Things. Cambridge, MA: Havard University Press, 1–23. Russell, B. 1900. The Philosophy of Leibniz. London: George Allen & Unwin. Russell, B. 1912. The Philosophy of Bergson. The Monist 22: 321–347. Russell, B. 1917. “Scientific Method in Philosophy,” in Mysticism and Logic: and Other Essays,” London: Longmans, Green, and Co. 107. Russell, B. 1945. A History of Western Philosophy. New York: Simon & Schuster. Ryle, G. 1990. Logical Atomism in Plato’s Theaetetus. Phronesis 35(1): 21–46. Ryle, G. 2002. The Concept of Mind. Chicago: University of Chicago Press. Schlick, M. 1927. The Future of Philosophy. In Henk L. Mulder and B.F.B. van de Velde-Schlick, ed., Philosophical Papers, Volume II: 1925–1936. Dordrecht: Reidel, 171–175. Sedgwick, S. 2012. Hegel’s Critique of Kant. Oxford: Oxford University Press. Strawson, P. F. 1964. Individuals: An Essay in Descriptive Metaphysics. New York: Methuen. Urmson, J. O. 1956. Philosophical Analysis: Its Development between the Two World Wars. Oxford: Oxford University Press. van Inwagen, P. 2006. What Is Naturalism? What Is Analytical Philosophy? In A. Corradini, S. Galvan, E. J. Lowe, eds., Analytic Philosophy Without Naturalism. London: Routledge, 74–78. Williams, B. 1990. Preface to L’éthique et les limites de la philosophie. Paris: Gallimard. Williams, B. 2006a. The Legacy of Greek Philosophy. In M. F. Burnyeat, ed., The Sense of the Past. Princeton, NJ: Princeton University Press, 3–48. Williams, B. 2006b. An Essay on Collingwood. In M. F. Burnyeat, ed., The Sense of the Past. Princeton, NJ: Princeton University Press, 341–360. Williams, B. 2008. What Might Philosophy Become? In A.W. Moore,, ed., Philosophy as a Humanistic Discipline. Princeton, NJ: Princeton University Press, 200–214. Wittgenstein, L. 1922. Tractatus Logico-Philosophicus. Trans. C K. Ogden. London: Routledge & Kegan Paul. Wittgenstein, L. 1953. Philosophical Investigations. Trans. G. E. M. Anscombe. Oxford: Blackwell. Wright, C. 1996. Human Nature? European Journal of Philosophy, 4:235–254.

3 Philosophy as Articulation Austin and Deleuze on Conceptual Analysis Richard Eldridge and Tamsin Lorraine

I Whether doing thought experiments from their armchairs, consulting intuitions, or investigating possible worlds, contemporary philosophers often describe what they do as conceptual analysis. This seems reasonable enough, since philosophy is often concerned with highly general cognitive claims and since the use of experimental and archival data is not as central as it is in the natural and social sciences. The distinctive contribution of the philosopher beyond the special sciences seems to involve an effort to make plausible or compelling a way of looking at things that can be expressed in a highly general, conceptual claim: the essence of matter is extension; the right thing to do is whatever will produce the most net good consequences; preferences are transitive; art is embodied meaning, and so on. Claims like these are debated among philosophers, tested by reference to cases (often imaginary ones), and revised, refined, and tested for coherence with other claims, both general conceptual ones and more obviously empirical ones. If one wants a name for these activities, conceptual analysis seems about as good a candidate as there could be. But are we clear either about what concepts are, about how they come about, or about what, exactly, the activity of analyzing them is? In 1903, near the dawn of analytic philosophy, G. E. Moore first formulated a version of what has come to be known as the paradox of analysis.1 Consider a putative result of conceptual analysis expressed in a claim of the form “Concept-word or phrase F is (essentially, necessarily) concept-word or phrase G.” [ (x) (Fx ≡ Gx)]. Any object correctly characterized by conceptword or phrase F is (essentially, necessarily) an object correctly characterized by concept-word or phrase G. Alternatively, F and G are not synonymous; in this latter case, the conceptual claim that all Fs are Gs seems, at best, accidentally true or true as things happen to be, not necessarily true. As a claim about necessary relations among concepts (or about relations between essences) it seems false. Hence it seems difficult to formulate the results of conceptual analysis in a way that is both informative and true.

60  Richard Eldridge and Tamsin Lorraine One way to block this result is to appeal to Frege’s distinction between sense and reference.2 Consider the expressions “the sum of 19 and 8” and “the cube of 3.” These two expressions mean different things or have different sense.3 (A child who had learned to add but not to multiply might fully understand the first but fail to understand the second.) Yet they have the same referent—the number 27—and necessarily so. Perhaps, then, the philosophical analysis of concepts might be comparable to the activity of mathematical proof in yielding results (necessary truths about concepts) involving same referent, different sense. By carefully inspecting, as it were, the senses of two different concept words (an activity analogous to mathematical proof in its abstraction from particular data), the philosopher as analyst of concepts might arrive at results that are all at once true, informative, and necessary. Yet while this is an attractive suggestion, it also faces considerable difficulties. First, in mathematics there are clear, shared standards for counting a sequence of mathematical statements as a proof. Mathematicians know how to check for failures of valid argumentation, involving unconsidered possibilities and the like. It is not clear that there are similar clear, shared standards for successful conceptual analysis. Second, as Frege frankly holds, thinking of mathematical truth as he does involves a commitment to senses (and numbers) as Platonic entities, standing eternally in relations to one another independently of human practice. This may seem, on the face of it, plausible enough for numbers and shapes, say. Hasn’t it always been true that 3 plus 8 = 11? But one may also wonder: were there (always already) those relations among numbers as abstract objects, over and above the fact that a certain pile of eleven acorns is exactly eight acorns bigger than a pile with only three, and so on for similar cases?4 We adopt “3 + 8 = 11” as a convenient norm of representation that applies to a very wide range of objective facts. But is there a matter of eternal mathematical fact over and above the many empirical facts (and norms of adding) that are thus representable? Third, even if we accept a Platonist construal of mathematical objects, it seems yet more strained to suppose that, say, in the Precambrian Era, four billion years ago before there was any life on earth, the sense (of the concept-word) squirrel was already necessarily both contained in that of mammal and waiting around to refer to the furry nut-gatherers who haunt our temperate parks and gardens. And then it seems yet more strained to hold that something like this is true of the interesting concepts that stand in need of philosophical analysis precisely insofar as they are essentially related to various human practices: justice, courage, art, decency, belief, and so on.5 These difficulties—lack of standards comparable to those in mathematics for determining necessary truths, alternatives to mathematical Platonism, and complex, non-natural kind concepts as the natural foci of philosophical analysis—likewise trouble more empiricist approaches to the analysis of

Philosophy as Articulation  61 concepts as essentially psychological entities in the mind or brain. While this latter approach is free of worries about the nature and existence of abstract objects, it also faces difficulties in construing full-blooded necessity with respect to interesting concepts, as well as worries about how to identify concepts ‘in’ either the mind or brain, as entities that underlie and determine the correct uses of concept-words. How could standards of correctness of use and necessary relations of implication be established by merely empirically existing entities with roles governed, at best, by empirical laws?6 Perhaps, then, rather than thinking of concepts as some sorts of fixed entities (Platonistic or psychological) whose natures and relations are open to some kind of ‘inspection,’ it would be apt and useful to consider concepts as essentially the meanings of concept-words as those words have complex, sometimes changing, and sometimes contested uses in practice. This would amount to thinking of the nature of concepts as determined by practice (which might well have its own forms of reasonableness, groundings, and responsiveness to reality in some cases), rather than vice versa, and thinking of concepts in this way would yield a significantly different picture of conceptual analysis as itself essentially practice-directed. What might conceptual analysis then look like? II In this view of conceptual analysis, the increasingly blurry line between analyzing words as they have been used in practice and articulating new uses of those words in the midst of living suggests that a notion of conceptual articulation may be more appropriate. Plausibly, philosophers sensitive to the difficulties we have just surveyed do not simply analyze how words are used; they also put those words into use in ways they hope will affect our perspectives and further thinking. In fact, the idea that conceptual analysis is essentially practice-directed, verging upon a practice of conceptual articulation and even conceptual creation, was significantly developed in the middle third of the twentieth century, primarily by Austin, Wittgenstein, and Ryle, before the developments of Quinean naturalism, Chomskyan psychologism, and Kripkean essentialism, and then again by Deleuze in the latter part of the twentieth century. Within the Anglo-American tradition, however, the relevant understandings of conceptual analysis as essentially practice-directed, together with the reasons for them and the sorts of results such a practice might yield, have been largely forgotten or ignored, largely under the pressures of naturalism, psychologism, and essentialism. Likewise, under the pressures of historicism, it is often not recognized within the Continental European tradition that the notion of philosophy as an activity of conceptual creation put forward by Deleuze is in some ways akin to conceptual analysis in the styles of Austin and Wittgenstein. Hence, comparing and elaborating different developments

62  Richard Eldridge and Tamsin Lorraine of conceptual analysis as essentially practice-directed not only can uncover surprising affinities and points of significant internal debate within otherwise suppressed traditions, but also can potentially reinvigorate conceptual analysis itself and bring two otherwise distinct lines of development into more intimate and productive relation to one another. III Among developers of conceptual analysis within the Anglo-American tradition, Austin has been particularly ignored, neglected not only in the wake of the developments of essentialism and psychologism, but also in virtue of falling under the shadow of Wittgenstein. Yet Austin’s arguments, while in some respects parallel to Wittgenstein’s, are frequently compact, antignomic, and forward-looking toward results in ways that Wittgenstein’s are not. Austin’s earliest published essay, his 1939 “Are There A Priori Concepts?” inaugurates both his attacks on concepts-as-entities and his practice-oriented understandings of concepts and conceptual analysis. Pointedly, Austin criticizes “the nonsense into which we are led through the facile use of the word ‘concept,’ ” in particular the nonsense of treating a concept “as an article of property, a pretty straightforward piece of goods, which comes into my ‘possession,’ if at all in some definite enough manner and at some definite enough moment,” so that “whether I do possess it or not is, apparently, ascertained simply by making an inventory of the ‘furniture’ of my mind.”7 The reasons why this is nonsense are, first, that we possess no direct intellectual, intuitive, or introspective access to concepts as entities (Platonic or psychological). The only thing that can show whether anyone ‘possesses’ a concept is whether that person consistently and reliably uses a concept-word within a roughly identifiable and bounded practice of words-in-uses. Or as Austin puts it, “It seems clear, then, that to ask ‘whether we possess a certain concept’ is the same as to ask whether a certain word—or rather, sentences in which it occurs—has any meaning,” which question is, Austin adds, “likely to be ambiguous” and at least in some cases not to admit of a simple and straightforward yes/no answer.8 Second, the sentences in question, the uses of which may establish mastery of a concept word and thus possession of a concept, are themselves sentences that are available and intelligible within an ordinary, roughly and indefinitely bounded common linguistic practice. “‘Does he, or do they, understand this word?’ . . . means, speaking roughly, [‘does he, or do they] use [it] as we, or as most Englishmen, or as some other assignable persons use [it?].’ ”9 Hence, trafficking in concepts, in the only sense we can give to this notion, involves participating in a roughly and indefinitely bounded practical life with other language users. When we turn our attentions to actual uses of words, then what we find is a range of phenomena that do not fall neatly under the model of

Philosophy as Articulation  63 concept-as-entity together with an extension. First, there is the conceptual priority of sentence meaning over word meaning, in the sense that one must accomplish a range of entire linguistic acts—calling attention to, exclaiming, asserting, laughing at, wondering whether, and so on—before one is properly creditable with mastery of concept-words.10 Second, the fact that all sentences are in some rough sense “about things” (in the broadest and most indeterminate possible sense of “thing”) does not imply that every individual word denotes a thing.11 Prepositions, adverbs, logical connectives, and so on are best conceived of not as names at all (and so not names of functions), but rather as words that have systematic roles in contributing to the meanings of the complete sentences in which they occur. Even for words that are plausibly conceived of as (general) names—that is, for one-place predicate expressions—it is a mistake to suppose that there is for each one a single referent or extension, sharply bounded in the same way from case to case. There is, Austin observes, “no reason whatever to accept” the principle “unum nomen unum nominatum. . . . Why, if ‘one identical’ word is used, must there be ‘one identical object’ which it denotes? Why should it not be the whole function of a word to denote many things? [Footnote: Many similar things, on a plausible view: but other views might be held].”12 In fact, Austin goes on, in “The Meaning of a Word” (1940), to analyze seven different varieties of “obvious cases where the reasons for ‘calling different sorts of things by the same name’ are not to be dismissed lightly as ‘similarity.’ ”13 These seven varieties of cases of uses, in which correct applications of concept words are not bounded by any single, obvious, simple, and univocal rule, are:14 1. paronymous uses: e.g., “healthy body” (a “nuclear sense”) vs. “healthy food” (productive of a healthy body) vs. “healthy glow” (resulting from a healthy body); the body, food, and glow are not in any single, obvious way alike. 2. analogous uses: e.g., “foot of a mountain,” “foot of a page.” 3. nontransitive uses: e.g., Bach’s St. Matthew Passion and Mozart’s Requiem are both profound, and so are Mozart’s Requiem and Beethoven’s Sonata #8 (“Pathetique”), but the Bach and the Beethoven are not in any obvious way like one another. 4. uses of terms with multiple, independent criteria: e.g., “fascism,” “cynicism.” 5. determinates of determinables: e.g., “ultramarine blue” and “indigo blue” are both blue, but otherwise quite different; more radically, “pleasure in solving quadratic equations” and “pleasure in drinking Belgian beer” are both pleasures but otherwise unlike. 6. quality/object ambiguities: “love” (“A’s love for B,” “my love”), “youth” (“early in a developmental history” vs. “the youths”). 7. activity-related terms: “cricket bat,” “cricket umpire,” “cricket pitch” (two senses), “cricket sweater.”

64  Richard Eldridge and Tamsin Lorraine Given such varieties of uses of single terms, where no clear, single rule for application is evident, it is reasonable to conclude, as Austin does, that “An actual language has few, if any explicit conventions, no sharp limits to the spheres of operations of rules, no rigid separation of what is syntactical and what semantical.”15 About any particular use of any particular word in a particular circumstance, it is possible to ask reasonably for clarification. Or, as Austin puts it, we may ask, “What-is-the-meaning-of (the word) ‘rat’?” either in this case (rodent vs. informer) or with a specific, more general range of cases or field of comparisons in view. But it is nonsense to ask, “What is the-meaning-of-the-word-‘rat’?” as though the expression containing dashes were the name of a fixed, univocal, and normatively dispositive substantive lying somehow behind our motley of uses.16 The consequences of this view about the multiplicity and context- and comparison- sensitivity of criteria for the correct usage of many terms are immediate and powerful. Conceptual analysis cannot be any sort of inspection (intellectual, intuitive, etc.) of a fixed ‘meaning-body’ behind or beyond usage. We should instead pay attention to the complex criteria and commitments by which our usages are normatively governed, in multiple ways, within our courses of practical responsiveness to the objects and phenomena of our world. Clarification of cloudy or uncertain uses of concept-words is likely to be piecemeal and field-of-comparison specific. “If we rush up for a demand with a definition in the simple manner of Plato or many other philosophers, if we use the rigid dichotomy ‘same meaning, different meaning,’ or ‘What x means,’ as distinguished from ‘the things which are x,’ we shall simply make hashes of things.”17 Despite, however, his arguments and warnings against a certain picture of conceptual analysis, Austin nonetheless himself clearly practices a form of it. He devotes himself to “examining what we should say when,”18 including both actual and “imagined cases,”19 with the aim, for example, of undoing significant confusions about freedom and responsibility.20 At the same time, in practicing his form of conceptual analysis, Austin concedes that sometimes “people’s usages do vary, and we do talk loosely, and we do say different things apparently indifferently,”21 and that “it cannot be expected that all examples will appeal equally to all hearers.”22 Moreover, ordinary usage, even when it is relatively clear and shared, is not sacrosanct. “It equally will not do, having discovered the facts about ‘ordinary usage’ to rest content with that, as though there were nothing more to be discussed and discovered. There may be plenty which might happen and does happen which would need new and better language to describe it.”23 Yet Austin is practicing a form of conceptual analysis, not doing empirical linguistics in the sense of simply tabulating (sometimes loose, sometimes divergent) usages. What exactly then is Austin doing, when he is displaying cases in which we would clearly call one thing an accident and another a mistake, one thing a case of succumbing to temptation and another quite different thing a case of losing control of oneself?24

Philosophy as Articulation  65 This is not an easy question to answer. But one helpful suggestion, made by Stanley Bates and Ted Cohen, following up on work by Stanley Cavell, is that Austin, along with anyone who enters a claim about what we say when, is speaking “with a universal voice,” in Kant’s sense of this phrase in The Critique of the Power of Judgment.25 According to Kant, when someone calls something beautiful, then that person does not postulate or predict the agreement of others, but rather demands and ascribes it.26 Such judgments are, Kant later adds, arguable by reference to reasons but not disputable by means of proofs or other decisive evidence.27 One might say that the judgment that something is beautiful expresses a sense of having experienced something in a certain way, as a certain kind of achievement of distinctively pleasurable intelligibility, coupled with a sense that it must be so experienced, that others must experience it likewise (if they pay attention to it in the right way). Adapted, then, to the judgment of the ordinary language philosopher about what we say when, the thought is that making such a claim is both enabling others to hear and say likewise and demanding that they do so.28 Such a claim can always fail, just as a critic’s claims directed at the enabling of aesthetic experience can fail or be repudiated. (Perhaps the critic has paid attention in the wrong way, or perhaps the work is simply inaccessible to some others.) But such failures do not impugn the reasonableness of the procedure, which sometimes leads to success in the form, in the case of claims about what we say when, of felt satisfaction and rightness in what one, along with others, clearly and confidently says and means. This picture of the possible achievement arrived at via claims about what we say when thus implies that, prior to such achievements, one, along with others, may have been judging and speaking in a kind of incoherence, in a fugue state of half-meaning or not fully meaning what one had said or thought, as though one were a living victim of cliché and inattentiveness. Hence the claim of the ordinary language philosopher as analyzer of concepts is directed at furthering a kind of awakening to one’s own judgments, thoughts, and experiences, directed toward a kind of heightened, more fluent and apt responsiveness to the things of one’s world. Both awakening and heightened fluency and responsiveness can be shared with others with whom one shares a language and world, and, in following the ordinary language philosopher’s claim with their own ears and minds, its hearers or readers may arrive at such shared fluency and responsiveness for themselves. Or, of course, they may not. But what is at stake in the ordinary language philosopher’s entering of a claim about what we say when is the achievement of a kind of heightened life as a responsive subject in relation to things, under conditions in which, always, that life is liable to become sterile and unthinking. Since such threats are permanent, philosophy is centrally less a body of theory than it is centrally an ongoing critical activity in the service of life.

66  Richard Eldridge and Tamsin Lorraine IV In A Thousand Plateaus, Deleuze, with Guattari, approvingly states with respect to Austin’s pragmatic approach to language that “the meaning and syntax of language can no longer be defined independently of the speech acts they presuppose.”29 Deleuze and Guattari’s theory of language will ultimately challenge traditional notions of the speaking subject by considering speech acts as effects of what they call “collective assemblages of enunciation”—ways of meaning-making with implicit rules that change over time—rather than of the particular subjects making the speech acts.30 Furthermore, they characterize philosophy as a practice that, along with creating concepts, creates what they call “conceptual personae” in defiance of the personal identities of embodied philosophers.31 Nevertheless, they, like Austin, consider philosophy to be a critical activity that can help us to achieve a heightened response to life.32 According to Deleuze, thinking is part of life; a philosophy that would freeze thinking into propositions that it proclaims to be timelessly true is a philosophy that becomes increasingly out of step with the need to make sense of the life in which we are immersed. Just as life cannot be reduced to the forms we perceive at a given moment, so the wisdom philosophy seeks cannot be reduced to propositions that fixate moments of thinking. Life is always more than what is manifest to our conscious awareness—it is also the intensities that are moving what is manifest into new forms, the imperceptible forces insisting in the most fleeting moment that are even now moving whatever is toward something else. To counter the notion of a philosophical concept as a word or term with a fixed meaning that can be cashed out in a set of propositions, Deleuze posits it as an event.33 Like any word, a concept may be attributable to specific states of affairs, but the sense of the concept is always in excess of any given state of affairs; not only are there always other states of affairs to which it may also apply, but the meanings of the concept will reverberate differently in keeping with shifts in the internal relations of its components and its relations to other concepts. Deleuze characterizes concepts in a way that not only attempts to shake us out of the ruts of conventional movements of thought, but also conceives them as integrally related to and yet in excess of the empirical movements of thinking that actualize them. Deleuze’s conception of the time-image that appears in modern cinema presents some aspects of what he thinks philosophical thinking can do. While the perspective of the camera can be taken up by the spectator as a gaze with which the spectator can identify, cinema is also capable of going beyond any one gaze and, by virtue of deliberately playing with the “irrational cuts” made possible by film, evoking a multiplicity of perspectives that cannot be assimilated into one rational whole. Cinema is an art form that can access what Deleuze calls the virtual—what we might here describe as the transcendental field of virtual relations conditioning what actually

Philosophy as Articulation  67 appears—by going beyond any one perspective in order to intimate a whole of multiple perspectives that can no longer be contained within one totalizable whole.34 While the cinematic time-image intimates the intensities that haunt the specific moment in space and time rendered by the cinematic shot, philosophy deliberately confronts the transcendental field of infinite meaning. A philosopher creates a novel perspective from a plane of immanence—a pre-philosophical plane constituting a problem to which thought responds by breaking out of the constraints of common sense perception and opinion.35 This perspective travels from the specific lived experiences of the philosopher back up to a realm of meaning unfettered by the common and by the good sense of a philosopher and her personal identity. The philosopher’s concept, by virtue of its access to a stratigraphic time in which all words are connected to all other words, taps resonances, echoes, and intensities that can incite novel trajectories of meaning out of the transcendental field of sense that conditions specific acts of meaning. Working with sense—paying attention to the components of meaning that make up a particular concept, playing with those components until the concept attains a kind of fullness—a self-referential quality that stabilizes it out of a sea of possible meanings—allows one to move away from a particular state of affairs in order to make generalizations that apply to more than that one state of affairs without losing touch with the actual situation out of which it emerged. A philosophical concept attempts to get at meanings that can be applied to more than one situation, meanings haunted by implicit trajectories that could unfold in more than one way. It is thus an event rather than a representation of a state of affairs, a constellation of meanings that actualize some rather than others of the virtual relations of a transcendental field of sense. It entails tendencies toward further movements of thought that may or may not actually unfold, and it is a configuration of meaning that can be ascribed to multiple states of affairs. A philosopher in doing philosophy breaks from her personal identity and creates perspectives on a plane of immanence drawn from her pragmatic situation—the situation she lives as an embodied individual immersed in the life of a specific social field. The meaning of words plays out against a transcendental field of sense where the meaning of particular sentences stabilizes in the context of the referents and speakers of pragmatic situations. That meaning, however, could always have played out otherwise with different inflections, thus actualizing other nuances in connection with the discursive and non-discursive practices informing the social field from which those meanings emerge. A plane of immanence emerges with the creation of concepts that condense components of thought and link up to other concepts. The self-referential connections that form within and between concepts constitute new forms of meaning that allow one to leave behind the constraints of a conventional perspective on life and yet to stabilize meaning out of the chaotic possibilities of sense.

68  Richard Eldridge and Tamsin Lorraine If language can never completely capture truth in relation to life it is because truth unfolds in time. And time entails the continual unfolding of the new, as well as the repetition of patterns familiar to us due to our experience with the past. If we consider the truth of this moment here and now—this moment of life that I am living—we know that we can never do it justice. Any articulation of this moment must fail because this moment is always unfolding into the next and, even if we can detect patterns that we can recognize as equally true of the past as of the present, those patterns can never capture the truth of this moment. According to Henri Bergson, who is an important influence in Deleuze’s work, in order for conscious perception to be that of a living organism able to act in ways that will ensure its survival, it must suppress the greater part of the complicated enmeshment of material life in order to discern what is of practical interest from the perspective of a particular organism with particular needs. In Matter and Memory, Bergson considers the relationship of material objects to our perceptions of them; he conceives a material object apart from our perception of that object as an image among a universe of images, all of which are material points without perspective that implicate and impact one another. He contrasts “the image which I call a material object” to the represented image that is a conscious perception of that material object: “That which distinguishes it as a present image, as an objective reality, from a represented image is the necessity which obliges it to act through every one of its points upon all the points of all other images, to transmit the whole of what it receives, to oppose to every action an equal and contrary reaction, to be, in short, merely a road by which pass, in every direction, the modifications propagated throughout the immensity of the universe.”36 As Alia Al-Saji puts it, such “an unperceived and unperceiving point virtually implies the rest of the dynamic and interpenetrating universe in its complexity and richness, with its infinite and incompossible relations. Its vision is a non-selective and indifferent kind, which registers everything but discerns nothing.”37 A conscious perspective requires “a process that limits and diminishes the virtual whole. It is in this way that representation and consciousness come about.”38 The futility of attempting to capture life with a static image is apparent when one stares at a photograph of a loved one who has passed away. Words likewise inevitably fail to capture the living presence of one who is no longer with us. What “truth” about a loved one refuses distillation into a form we can grasp even when she or he is gone? It is toward art that we may look to capture something of the truth of concrete forms of life. Deleuze’s books on Francis Bacon, Marcel Proust, and cinema develop an intriguing understanding of this kind of aesthetic truth that is an illuminating counterpart to Deleuze’s conception of philosophy as a thought form that entails creating concepts. If we consider Bergson’s comparison of a material image to the represented image of conscious perception, we can see that anything that we can

Philosophy as Articulation  69 call a perspective is in a sense haunted by interpenetrating influences that always exceed those that emerge in relation to the needs of a particular organism. The force of these influences constitutes a kind of ontological unconscious that is the virtual reality inflecting any actualized present. In his book on the painter Francis Bacon, Francis Bacon: The Logic of Sensation, Deleuze distinguishes between sensation and perception; sensation refers to a kind of experience that is imperceptible, since it refers to forces impinging on unreflective awareness (and also entailing our own impingement on what’s around us) that affect how we experience the world but that lie just beyond the edges of what we can pin down in perceptions stable enough to describe.39 According to Deleuze, art composes monuments of sensation that intimate a visceral becoming-other that haunts our conscious awareness without becoming overtly manifest.40 In The Logic of Sensation, Deleuze describes how Bacon talks about capturing a likeness in a portrait that goes beyond appearance and evokes the intensity of the real; instead of doing portraits that rest with the form of a particular human being as she is present in appearance at a given point in time, Bacon hints at the forces making up that human being—forces that are always in the process of unfolding.41 Thus, Bacon moves beyond the conventions of perception and through his paintings evokes sensation—an experience of visceral affect inarticulable in words and eluding familiar forms of perception and instead evoking a sense of discomfort and unfamiliarity that allows us to experience something about those human beings and the human condition that we had not before. By opening up an aesthetic articulation of the world of the novel through an exploration into the depth of events, Proust similarly investigates the virtual past of those events in a way that shows us the intensity of time.42 His series of novels evokes life in terms of a memory that exceeds any one perspective or a multiplicity of perspectives that can be correlated into one homogeneous whole and instead posits what Deleuze calls “fragments” that cannot be put together and instead are put alongside one another.43 These fragments evoke a past that may not have become explicit to consciousness, but that insists itself in the present in terms of what would have to change in the present, actualized situation in order for it to move over thresholds into a different situation. These fragments cannot be assimilated into one whole because they compose different fragments of duration that could unfold in ways that are incompatible with other durations. Because how each fragment unfolds with respect to other fragments exerts its own effects, there is no way to organize them into a linear chain of cause and effect. Instead, each fragment could set off a whole series of unfoldings, each of which would interact with other series in unpredictable ways. To stay with the appearance of what has already actually unfolded is thus for Deleuze to overlook a crucial aspect of the real. Philosophy, unlike art, does not look at the becoming of specific things—of a specific human body or the portrait of an individual. But according to

70  Richard Eldridge and Tamsin Lorraine Deleuze, it, like art, provides access to a chaotic world that defies containment of the world within categories that we have applied to past events. Philosophy—like art and science—is a distinctive form of thought that creatively grapples with life’s novelty. While art composes monuments of sensation—percepts (perceptions that evoke becoming-other) and affects (emotions that evoke becoming-other)—and science invents functions on a plane of reference, philosophy creates concepts.44 It is a paradox that, even when philosophy is conceived as a set of timeless truths waiting to be discovered, it is never satisfied with its answers; philosophy is an open-ended process in which one must inevitably question the truths one has arrived at and continue the process of thinking. Deleuze’s conception of philosophy as a thought form that creates concepts entails the notion of a concept as an event rather than a fixed truth. A concept as an event of thought is a virtual multiplicity whose internal links among components of meaning and external links to other concepts make up a plane of immanence that evokes the restless movement of thinking as a generative process. Our contemporary situation—one of rapid change and unprecedented problems of almost unthinkable scale—is perhaps one that demands this Deleuzian conception. Only a philosophy that includes a temporal as well as spatial dimension can speak to a thinking that arises from and keeps pace with the accelerating speed of life’s movement. V Austin, by articulating a notion of concepts as practice-directed, turns our attention from concepts as reified entities to which our conceptual activity refers (more or less well) to embodied practices in and through which concepts evolve. Philosophical activity as conceptual analysis becomes a practice-directed by immanent rules in which we attempt to become ever more aware and more precise about how we articulate meanings we share. Such activity demands attention to those rules in the living contexts in which they are applied, rather than accepting past applications as automatically transferable to present circumstances. Austin’s particular style and point in cultivating this active attention are a function of his sense that we are sometimes captivated by impossible images of absolute control in thinking and judging that are associated with metaphysical philosophy. We often enough fail to think clearly and have instead rushed into a theoretical stance that counterfeits our interests by running against the grain of ordinary life that is often meaningful enough. As Austin puts it in “A Plea for Excuses,” “ordinary language is not the last word: in principle it can everywhere be supplemented and improved upon and superseded. Only remember, it is the first word,” in virtue of embodying “the inherited experience and acumen of many generations.”45 Deleuze, with his characterization of concepts as events, indicates something of the past of meaning-making as it informs its present evolution.

Philosophy as Articulation  71 Philosophical activity, by attending to shifts in meaning according to a principle of consistency, settles on some ways in which meaning in particular times and places can unfold, given its past, in keeping with the problems life poses.46 Conceptual analysis, in Deleuze’s view, becomes not just a philosophical practice that attends to meaning in precise ways, but a practice that deliberately unfolds some meanings rather than others from the chaotic field of sense conditioning specific practices of meaning. Deleuze’s particular style and point in this practice of unfolding are largely aimed at uncovering new possibilities of practice against the grain of a common life that is all too likely to be ossified or constrained by implicit rules that refuse life’s experimentation with the new. For both Austin and Deleuze, however, philosophical activity is a form of practice in the articulation of concepts that can break with ritualized ways of thinking by attending to the virtual echoes that inevitably emerge as we articulate our responses to life. If Austin enacts a contemporary sense that human life is both chaotically fragmented and over-intellectualized, Deleuze enacts a contemporary sense of being frozen in one-sidedness and subject to the sedimented patterns of the already said and done. Both senses strike us as reasonable perceptions of threats posed to human flourishing in current circumstances. Given the nature and difficulty of these partly opposed, partly complementary threats, it is clear why, in the view of both Austin and Deleuze, the practice of philosophy as conceptual articulation not only is inevitably open-ended, but also plays a crucial role in the ongoing attunements of humanity to life. NOTES  1 See Moore (1903), p. 442. Moore’s stalking horse example is the Idealist analysis of the concept yellow that is expressed in the claim (Moore calls it a formula) “yellow is the sensation of yellow.”   2 Frege (1892). Following the opening discussion of the problem of cognitively informative identity statements, the sense/reference distinction is introduced on p. 200.  3 The sense of an expression is, roughly, its linguistic meaning plus any further information that is relevant and available in virtue of its context of use. Since the context of mathematics is (usually) completely general, contextual specificities can normally be discounted there and sense can be identified with linguistic meaning.   4 Thoughts like this might suggest a structuralist conception of mathematics: a common pattern is instanced in various groups of acorns, pebbles, kernels of corn, etc., and mathematics might be about such patterns, rather than about independently existing, eternal abstract objects.   5 Similar points haunt a Kripkean approach, in the spirit of Frege, to empirically knowable necessary truths. When does empirical inquiry yield strong, more than pragmatic necessity with regard to natural kinds? And what about the important, complex non-natural kind-concepts that are the traditional concerns of philosophical analysis?   6 Frege’s criticisms of psychologism are to the point here, even if his own Platonist alternative is not free of problems of its own.

72  Richard Eldridge and Tamsin Lorraine   7 Austin (1939), p. 41.  8 Austin (1939), p. 43. Note the plural “sentences;” multiple compliant and coherent uses, not a single use only, are necessary before we will credit anyone with mastery of a concept-word as opposed to the issuing of a happenstantial noise.   9 Austin (1939), p. 42. 10 See Austin (1939), p. 40 as well as Austin (1940), p. 56. 11 Austin (1939), p. 40. 12 Austin (1939), p. 38. 13 Austin (1940), p. 71. 14 The following seven points summarize Austin’s discussion in Austin (1940), pp. 71–74. 15 Austin (1940), p. 67. 16 Austin (1940), p. 55. 17 Austin (1940), p. 74. 18 Austin (1957), p. 181. 19 Austin (1966), p. 274. 20 See the discussions of the importance of Aristotle in having recognized that questions of who is responsible for what (and who may be excused for what) are more natural than and prior to (artificial) ‘theoretical’ investigations of the freedom of the will in Austin (1957), p. 180, and Austin (1966), p. 273. 21 Austin (1957), p. 183. 22 Austin (1940), p. 66. 23 Austin (1940), p. 69. Compare Austin (1957), p. 185. 24 These distinctions appear in Austin (1957), pp. 185n. and 198n. 25 Bates and Cohen (1972), p. 22. Kant introduces the idea of speaking with a universal voice in §8 of the Critique of the Power of Judgment in order to capture the unique kind of claim that is made in calling anything beautiful. Kant (2000), V: 216, p. 101. 26 The postulate/demand and predict/ascribe distinctions are laid out in Kant (2000), §§7, 8, V: 212–13, 216; pp. 98. 101. 27 Kant (2000), §56, V: 338, p. 214. 28 Stanley Cavell develops the parallel between the judgments of the critic of the arts and the claims of the ordinary language philosopher in “Aesthetic Problems of Modern Philosophy,” in Cavell (1969), pp. 88–96. 29 Deleuze and Guattari (1987), p. 77. 30 “It is the illocutionary that constitutes the nondiscursive or implicit presuppositions. And the illocutionary is in turn explained by collective assemblages of enunciation, by juridical acts or equivalents of juridical acts, which, far from depending on subjectification proceedings or assignations of subjects in language, in fact determine their distribution,” (Deleuze and Guattari (1987), p. 78). 31 “The face and body of philosophers shelter these personae who often give them a strange appearance, especially in the glance, as if someone else was looking through their eyes.” Deleuze and Guattari (1994), p. 73. For more on conceptual personae see chapter 3 of Deleuze and Guattari (1994), pp. 61–83. 32 Since the following discussion is framed through Deleuze’s work rather than through that of Guattari, the references to A Thousand Plateaus and Deleuze and Guattari’s book, What is Philosophy? (Deleuze and Guattari 1994) will henceforth be referred to as Deleuze. 33 “Concepts are events” (Deleuze and Guattari 1994), p. 36. Deleuze and Guattari also here characterize the concept as “speaking” or “knowing” the event (pp. 21, 33) and philosophy as drawing concepts from states of affairs “inasmuch as it extracts the event from them” (p. 52). “The concept is neither denotation of states of affairs nor signification of the lived; it is the event as pure sense that immediately runs through its components” (p. 144). The

Philosophy as Articulation  73 concept of the event is developed by Deleuze, and Deleuze with Guattari, in multiple places. For some relevant examples see Deleuze and Guattari (1994), pp. 156–178 and Deleuze (1990), pp. 12–22, 52–57, and 148–153. 34 Deleuze develops the concept of the time-image with respect to Henri Bergson’s notions of the actual and the virtual in his book, Cinema 2: The Time-Image (Deleuze 1989). See p. 74 and pp.105–11 for Deleuze’s characterization of the images related to the mystery of Kane’s dying word, “Rosebud,” in Orson Welle’s film Citizen Kane as an example of an early depiction of the direct time-image in modern cinema. 35 See ch. 2, “The Plane of Immanence” in Deleuze and Guattari (1994), pp. 35–60. 36 Bergson (1991), p. 34. 37 Al-Saji (2004), p. 220. 38 Al-Saji (2004), p. 220. 39 See Bogue (2003), pp. 116–121, for more on Deleuze’s distinction between sensation and perception and its sources in the work of Henri Maldiney and Erwin Straus. 40 “We attain to the percept and the affect only as to autonomous and sufficient beings that no longer owe anything to those who experience or have experienced them: Combray [the town evoked by Proust’s madeleine] like it never was, is or will be lived; Combray as cathedral or monument” (Deleuze and Guattari 1994, p. 168). 41 “But in the end, it is a movement ‘in-place,’ a spasm, which reveals a completely different problem characteristic of Bacon: the action of invisible forces on the body” (Deleuze 2003), p. 36. 42 The connection made here between the concept of sensation that Deleuze develops in The Logic of Sensation and the concept of the fragment he develops in Proust and Signs is indebted to Miguel de Beistegui’s lucid commentary in chapter 6 of his book, Immanence: Deleuze and Philosophy (de Beistegui 2010), pp. 160–191. 43 “Even when the past is given back to us in essences . . . what is given us is neither a totality nor an eternity, but ‘a bit of time in the pure state’, that is, a fragment (Proust, À la Recherche du temps perdu, Bibliothèque de la Pléiade, III, p. 705)” (Deleuze, 2000, p. 122). “By setting fragments into fragments, Proust finds the means of making us contemplate them all, but without reference to a unity from which they might derive or which itself would derive from them” (Deleuze (2000), p. 123). 44 Deleuze and Guattari (1994), p. 197. 45 Austin (1957), p. 185. 46 “[T]he concept in philosophy expresses an event that gives consistency to the virtual on a plane of immanence in an ordered form” Deleuze and Guattari (1994), p. 133.

WORKS CITED Al-Saji, A. 2004. The Memory of Another Past: Bergson, Deleuze and a New Theory of Time. Continental Philosophy Review 37(2): 203–239. Austin, J. L. 1939. Are There A Priori Concepts? In Philosophical Papers, Oxford: Oxford University Press, 32–54. Austin, J. L. 1940. The Meaning of a Word. In Philosophical Papers, Oxford: Oxford University Press, 55–75. Austin, J. L. 1957. A Plea for Excuses. In Philosophical Papers, Oxford: Oxford University Press, 175–204.

74  Richard Eldridge and Tamsin Lorraine Austin, J. L. 1966. Three Ways of Spilling Ink. In Philosophical Papers, Oxford: Oxford University Press, 272–287. Austin, J. L. 1970. Philosophical Papers. 2nd. ed. Eds. J. O. Urmson and G. J. Warnock. Oxford: Oxford University Press. Bates, S. and T. Cohen. 1972. More on What We Say. Metaphilosophy 3(1) (January): 1–24. Bergson, H. 1991. Matter and Memory. Zone Books. Bogue, R. 2003. Deleuze: On Music, Painting, and the Arts. London: Routledge. Cavell, S. 1969. Aesthetic Problems of Modern Philosophy. In S. Cavell, ed., Must We Mean What We Say? New York: Charles Scriber’s Sons, 73–96. De Beistegui, M. 2010. Immanence: Deleuze and Philosophy. Edinburgh, Scotland: Edinburgh University Press. Deleuze, G. 1989. Cinema 2: The Time-Image. Minneapolis: University of Minnesota Press. Deleuze, G. 1990. Logic of Sense. New York: Columbia University Press. Deleuze, G. 2000. Proust and Signs, The Complete Text. Minneapolis: University of Minnesota Press. Deleuze, G. 2003. Francis Bacon: The Logic of Sensation. Minneapolis: University of Minnesota Press. Deleuze, G. and F. Guattari. 1987. A Thousand Plateaus Minneapolis: University of Minnesota Press. Deleuze, G. and F. Guattari. 1994. What Is Philosophy? New York: Columbia University Press. Frege, G. 1892. Über Sinn und Bedeutung. Zeitschrift für Philosophie und Philosophische Kritik 100 (1892); reprinted as “On Sense and Nominatum,” trans. Herbert Feigl, in The Philosophy of Language, 4th ed., ed. A. P. Martinich (Oxford University Press, 2001), 199–211. Kant, I. 2000. Critique of the Power of Judgment. Trans. Paul Guyer and Eric Matthews. Cambridge: Cambridge University Press. Moore, G. E. 1903. The Refutation of Idealism. Mind 12(48): 433–453. Proust, M. À la Recherche du temps perdu, Bibliothèque de la Pléiade, III.

4 Conceptual Genealogy for Analytic Philosophy Catarina Dutilh Novaes

1. INTRODUCTION The significance attributed to the history of philosophy for the systematic investigation of philosophical issues divides the analytic and the continental philosophical traditions. Typically, the continental philosopher sees the historical development of a given philosophical issue or concept as a substantial and perhaps even indispensable element for the analysis, whereas the analytic philosopher tends to treat issues and concepts as if they were ahistorical entities, thus not requiring such a historical contextualization to be properly grasped.1 Influential ‘continental’ authors such as Nietzsche and Foucault have placed historical analysis at the epicenter of their respective philosophical methodologies, in particular with the concepts of ‘genealogy’ and ‘archaeology.’ More recently, ‘analytic’ authors such as Ian Hacking, Edward Craig, and Bernard Williams, among others, have pursued similar lines of investigation. For the most part, however, and despite some notable exceptions (such as Crane 2015), analytic philosophers remain quite hostile to the idea that the systematic analysis of a given concept or issue has something to benefit from becoming historically informed. In this paper, I discuss in detail a philosophical methodology that I call ‘conceptual genealogy.’ This methodology underpins much of my work in the history and philosophy of logic to date (for example, my work on the concept of logical form), which however falls squarely within the ‘analytic tradition.’2 I argue that analytic philosophy in general has much to gain from incorporating the historicist component of genealogical investigations. Analytic philosophers too must take seriously the idea that philosophical concepts may be historical products, rather than atemporal natural kinds or essences, and that they bring along with them traces of their historical development as well as of broader cultural contexts. Indeed, one of the key aspects of typical genealogical approaches (as is clear, in particular, in Nietzsche) is an emphasis on the contingent nature of (philosophical) concepts and phenomena as products of long and winding historical developments. Moreover, conceptual genealogy produces narratives whose protagonists are

76  Catarina Dutilh Novaes concepts, issues, and arguments, not authors; it operates predominantly on philosophical texts, but textual authorship is not the main focus. It focuses on how philosophical concepts are reinterpreted and transformed through their historical developments, while maintaining traces of their previous instantiations. Conceptual genealogy is thus a longue durée approach to the history of philosophy. Before we proceed, however, I must say something about the range and scope of the considerations to follow. I focus almost exclusively on distinctively theoretical, philosophical concepts, such as, for example, the concepts of logical form, truth, knowledge, etc. These concepts may be related to extra-theoretical concepts and phenomena in interesting ways, but here the focus is on the historical development of such concepts as registered in philosophical texts. Thus, projects of evolutionary genealogy, or ‘state of nature’ projects (i.e., quasi-mythological accounts of the ‘original,’ pre-societal stages of human existence, e.g., Rousseau’s account of the origins of inequality), for example, do not count as ‘conceptual genealogy’ in the sense countenanced here. Social and historical developments outside philosophy also fall outside the scope of the present investigation3 though the historian of philosophy does well to pay attention to the material and social backgrounds for the emergence of philosophical theories and concepts.4 For example, Foucault’s (1975, 1976) genealogical investigations of social phenomena such as sexuality and punishment obviously included numerous elements outside the limited scope of the philosophical corpus, and this is not the kind of (sociological) genealogy we will focus on here (though some elements of Foucault’s framework will be in the background). In other words, the genealogical projects to be discussed here pertain to the development of certain key concepts specifically within philosophy, and predominantly as registered in philosophical texts. There is still the question of how these philosophical concepts are related to non-philosophical concepts (and the line between what counts as a philosophical vs. a non-philosophical concept is bound to be rather blurry and contentious), as well as to non-conceptual phenomena, i.e., ‘reality’ as such. But for now, it is sufficient to clarify that the scope of the present investigation is somewhat narrower than some readers may expect, given the various other genealogical projects they may be familiar with. 2. ANALYTIC PHILOSOPHY AND THE AHISTORICAL CONCEPTION OF PHILOSOPHY Williams (2002) and Craig (2007) draw a useful distinction between genealogies that seek to expose the reprehensible origins of something and thereby decrease its value, and genealogies that seek to glorify their objects by exposing their ‘noble’ origins. The former are described as ‘subversive,’

Conceptual Genealogy for Analytic Philosophy  77 ‘shameful,’ or ‘debunking,’ while the latter may be described as ‘vindicatory.’ (I will have much more to say on this distinction later on; I will also discuss a third category.)5 Nietzsche’s famous genealogical analysis of morality is the archetypal subversive genealogy and has given rise to a formidable tradition of deconstruction of concepts, values, views, beliefs, etc. by the exposure of their pudenda origo, their shameful origins (Srinivasan 2011). But naturally, the pull of (Nietzschean) genealogy was not felt to the same extent elsewhere. In effect, so-called analytic philosophers remained by and large resistant to genealogical enterprises (both subversive and vindicatory), so much so that one’s stance towards genealogical projects can be seen as one of the main differences between so-called continental and so-called analytic philosophers.6 What is more, some analytic philosophers seemed to take what they saw as the shortcomings of genealogical projects to be a sign that any kind of historical contextualization of concepts and beliefs in the context of philosophical analysis would be misguided. Thus, analytic philosophy embraced what could be described as a largely ahistorical (or even anti-historical) conception of philosophy, whereas historical analysis remains of crucial importance for continental philosophers (Mulligan et al. 2006).7 Naturally, these are general trends rather than absolute rules, and a number of analytic philosophers have engaged in genealogical projects or, more generally, ensured that their philosophical analyses be historically informed.8 (I take genealogy to be one but not the only way to engage in philosophically relevant historical analysis.) Hacking (for example, in (Hacking 1995), among many other works) is perhaps the most prominent example, along with Craig (1990) and Williams (2002). (Importantly, Williams and Hacking are overtly influenced by so-called continental authors such as Nietzsche and Foucault, respectively). Still, mainstream analytic philosophers tend to believe that the (philosophical) history of a given concept or belief is not likely to be relevant for a philosophical understanding of the concept, or likely to improve our knowledge of the phenomena in reality that the concept in question is a concept of.9 Here is an apt description of this phenomenon:10 Analytic philosophy has largely rejected historical modes of understanding. [. . .] This neglect, however, is not accidental: it is the result of the general repudiation of the historical mode of understanding within analytic philosophy. In particular, analytic philosophy seems to think of itself as taking place within a single timeless moment. (Hylton 1990, p. viii) A number of explanations may be given to account for (or even justify) the analytic philosopher’s rejection of historically informed philosophical analysis in general and genealogical approaches more specifically. Firstly, the analytic philosopher may think that what is philosophically relevant is not

78  Catarina Dutilh Novaes the context of discovery of a concept or belief, but rather its potential justification (to resort to the old but still useful Popperian distinction). In this vein, the origins and historical development of a concept or belief are irrelevant for establishing its (presumably) objective and philosophically relevant properties: What is the extension of the concept? What is the truth-value of the belief? Is it justified? For instance, if the belief is true (and even better, also justified), then the fact of having shameful origins will not make any difference; conversely, if the belief is false, then having noble origins will not change its falsity (Srinivasan ms.) (see section 4.3 for a discussion of the genetic fallacy). Moreover, a tacit (and sometimes explicit) commitment that seems to underpin much analytic philosophy is the idea that the concepts the philosopher studies (or the corresponding non-conceptual phenomena in reality) are natural kinds, defined by immutable, atemporal essences. (See e.g., Kornblith 2011 for the claim that the concept of knowledge is a natural kind, and a critique of Kornblith by Kusch 2013.) Indeed, Frege, Russell, and Moore (arguably, the founders of analytic philosophy as we know it) were all conceptual realists, i.e., they took concepts to have human-independent reality. Now, if the concepts studied by philosophers correspond to (or are themselves) natural kinds, or more generally have a human-independent and atemporal kind of existence, then the different ways in which philosophers conceived of them through time are irrelevant. What matters is to formulate them correctly, for example, by formulating the right necessary and sufficient conditions for something to count as X.11 Assuming that there are such conditions (objectively speaking), if philosophers of previous generations associated different conditions with X, then they were simply wrong, and there is not much of philosophical significance to be gained from their mistakes (except perhaps for avoiding making the same mistakes). Another way to make the same point is to say that what philosophers do is to set out to discover preexisting, possibly immutable, essences. Now, if this is what they do indeed, then there is not much point in historically contextualizing the quest for ahistorical essences. My goal at this point is not to offer a definitive answer to the (largely sociological, but also philosophically interesting) question of why analytic philosophers tend not to be big fans of genealogical projects in general. For now, it is sufficient to notice that this seems indeed to be the case (as also noted by a number of authors before me, such as Hylton 1990 and Crane 2015). Moreover, the tentative explanations just offered also suggest a certain conception of philosophical concepts and of the nature of the philosophical enterprise—discovering truths, be they about concepts or about non-conceptual reality—that must be further discussed and, to some extent, questioned, if genealogical projects are to be relevant at all for the investigation of systematic philosophical questions. Notice also that these considerations may apply more broadly to historical analysis as a whole, not only to the specific kind of historical analysis that is a (conceptual) genealogy.12

Conceptual Genealogy for Analytic Philosophy  79 In the remainder of the paper, I defend the view that a suitable formulation of the idea of conceptual genealogy does represent a fruitful methodological approach also for the analytic philosopher. The idea is not that it should supplant other ‘traditional’ analytic methods, but rather that it may be viewed as a valuable tool in the analytic philosopher’s toolbox, to be combined with the traditional methods.

3.  CONCEPTUAL GENEALOGY

3.1 Nietzschean Genealogy The mundane, commonsensical sense of genealogy is typically related to the idea of vindication, i.e., validation of one’s authority and standing through the narrative of one’s origins. This is particularly conspicuous in historical disputes for political power within traditional monarchic models: a contestant has a claim to the throne if she can prove to be a descendant of the right people, namely previous monarchical power-holders. In such cases, a genealogy is what Geuss (1994, p. 274) describes as ‘tracing a pedigree,’ a practice as old as (Western?) civilization itself.13 The key idea is the idea of transmission of value: a person with noble ancestry inherits this status from her ancestors through their common bloodline. But a genealogy may also have more ‘neutral’ implications: perhaps a person’s ancestors are not particularly distinguished or noble, but she may still wish to know where she ‘comes from.’ In the limit case, a genealogy may also be of the shameful kind, e.g., if what transpires from genealogical analysis is that a person’s ancestors had dubious social standing (e.g., they were convicted thieves). In theory, none of it should matter for a person’s individual worth, and yet in practice we tend to attach a great deal of importance to a person’s ancestry (e.g., the quasi-mythical status of the Mayflower pilgrims and their descendants in the United States). In this sense, a genealogy is a narrative with no gaps: a person’s genealogy is a detailed account of her ancestry, which specifies every relevant parent-offspring step in the chain. (Naturally, it will have to stop at some point, usually at the person who is then viewed as the founder of the dynasty in question—even though this person obviously had parents as well.) Typically, a genealogy may focus on the transmission of a family’s surname through generations, thus indicating continuity (of positive value in particular, e.g., nobility).14 At the same time, a genealogy will always contain an element of change as well, if for no other reason then because parents and offspring are by definition different individuals. In effect, the interplay between continuity and change is one of the fundamental aspects of the concept of genealogy for the present purposes. But naturally, this is not a study of genealogies of people; instead, we are interested in genealogies of (philosophical) concepts, and thus in their

80  Catarina Dutilh Novaes development through time. The idea that thought itself is a historical beast rather than something immutable and atemporal can be traced back (at least) to the German historicist tradition, which emerged in the eighteenth century (Beiser 2011). Nietzsche’s genealogical approach falls squarely within this tradition, even if his own interpretation of historicism in terms of genealogy is arguably quite unique to him. For the present purposes, it will prove instructive to compare Nietzsche’s historicism to that of Hegel,15 who famously said: As far as the individual is concerned, each individual is in any case a child of his time, thus, philosophy, too, is its own time comprehended in thoughts. (Hegel 1820/1991, p. 21) It follows that different times/contexts will give rise to different instantiations of philosophical concepts; thus, philosophical concepts themselves will change over time, following more global changes. However, Hegel’s conception of history in general, and of the history of concepts in particular, is teleological: things could not have taken a different turn, as temporal developments follow an inevitable path. And so, from this perspective, historical analysis tracing the different steps in the evolution of a concept and their mutual relations—a conceptual genealogy—will not have the effect of decreasing the value of the concept in question to us; when there are no other options, there is nothing to compare it to. Despite the already established dominant historicist background in nineteenth-century Germany, Nietzsche is usually seen as the inaugurator of a new approach, namely what can be described as the subversive variant of historicist projects. In his On the Genealogy of Morality (1887), he famously offers a genealogy of Christian morality that is meant to expose its ‘shameful’ origins. Rather than comprising eternal, immutable moral precepts, Christian morality (and the human practices associated with it) is in fact the product of contingent historical developments, more specifically a conjunction of a number of diverse lines of events (Geuss 1994, p. 276)— none of them particularly ‘honorable.’ Specifically, Christian morality arises from the resentment of slaves directed against their masters, having thus distinctively malevolent origins (while currently presenting itself as pure and magnanimous). Besides the idea of a confluence of multiple lines of development, another crucial characteristic of Nietzsche’s genealogy of Christianity for the present purposes is the idea of a superimposition of layers through processes of reinterpretation of previously existing practices, giving rise to new practices that nevertheless retain traces of their previous instantiations. The first significant step in this succession of reinterpretations is the influential conception of Christianity formulated by Saint Paul, which represents, however, a drastic departure from the way of life exemplified by Jesus himself.16

Conceptual Genealogy for Analytic Philosophy  81 Paul’s ‘interpretation’ represents so drastic and crude a misinterpretation of Jesus’ way of life that even at a distance of 2000 years we can see that wherever the Pauline reading gets the upper hand [. . .] it transforms ‘Christianity’ [. . .] into what is the exact reverse of anything Jesus himself would have practiced. (Geuss 1994, p. 280) However, such processes of reinterpretation never manage to quash entirely traces of the original practices: Nietzsche thinks that such attempts to take over/reinterpret an existing set of practices or way of life will not in general be so fully successful that nothing of the original form of life remains, hence the continuing tension in post-Pauline Christianity between forms of acting, feeling, judging which still somehow eventually derive from aboriginal Christianity and Paul’s theological dogmas. (Geuss 1994, p. 281) Central to Nietzsche’s genealogy of Christianity is the idea of constant ‘power struggles’ between different ‘wills,’ attempting to impose their own interpretations and meanings on the practices in question. But even when a particular new meaning manages to impose itself, the old meaning(s) will remain present, albeit in modified form, in the resulting complex. Later on, we will see that this is very much what happens in the historical development of philosophical concepts: they undergo modifications, but the superimposed layers of meaning typically retain traces of their previous stages and instantiations even when acquiring a new meaning. And so, Nietzschean genealogy is characterized by the crucial interplay between continuity and change; indeed, how can we say that a genealogy is a genealogy of X if there is nothing permanent at all in the phenomenon in question through time? As Brian Leiter puts it, how do we fix the object? He says: Genealogy, then, presupposes that its object has a stable or essential17 character—its Brauch—that permits us to individuate it intelligibly over time. What the genealogist denies is that this stable element is to be located in the object’s purpose or value or meaning (its Sinn); it is precisely that feature which is discontinuous from point of origin to present-day embodiment.18 (Leiter 2002/2014, p. 136) Thus, change and continuity are crucial in a genealogy. And so, the components of the Nietzschean conception of genealogy that are particularly relevant for our present purposes are a particular historicist conception of concepts and values; the emphasis on the contingency of the underlying historical developments (i.e., contingentist historicism, different from Hegel’s teleological historicism), usually involving multiple lines of influence; and the superimposition of layers of meaning, resulting in both change (the new meaning) and continuity (traces of the old meanings still present, and

82  Catarina Dutilh Novaes continuity in the phenomenon as such).19 By contrast, Nietzsche’s focus on shameful genealogy is best kept apart for the present purposes; it will be argued later on that conceptual genealogy of philosophical concepts can be either vindicatory or subversive—as well as largely neutral, i.e., expository.

3.2 Canguilhem and the Historicity of Scientific Concepts The thesis of the relevance of historical analysis for philosophical theorizing rests crucially on a historicist conception of philosophical concepts, namely that they are not (or do not correspond to) atemporal essences or natural kinds. However, ‘historicism’ can have different meanings (Beiser 2011, Introduction), so I must spell out in more detail in what sense I defend a historicist conception of philosophical concepts. To this end, let us go back to the anti-historicism of analytic philosophy discussed in section 2. This largely ahistorical conception of philosophy is at least partially related to the close ties between philosophy and the (empirical and formal) sciences defended by key early figures such as Russell and Carnap. Roughly speaking, just as scientists investigate (presumably) immutable, non-historical physical phenomena such as relativity or cell metabolism, so do philosophers investigate (allegedly) immutable concepts: truth, causation, knowledge, logical validity, etc. However, the irony is that scientific concepts themselves have time and again been shown to be everything but immutable and ahistorical (even if the phenomena they describe might be so, in some sense or another). There are vibrant, rich traditions within philosophy of science that operate precisely by means of thorough historical analyses of the emergence and transformation of scientific concepts and theories. For instance, the increasingly influential Integrated History and Philosophy of Science (&HPS) movement defines itself in the following way: The founding insight of the modern discipline of HPS is that history and philosophy have a special affinity and one can effectively advance both simultaneously. What gives HPS its distinctive character is the conviction that the common goal of understanding of science can be pursued by dual, interdependent means.20 Kuhn is of, course, one of the founding fathers of the historically informed approach within philosophy of science, but there are other strands contributing to these developments. In particular, the French tradition of ‘historical epistemology,’ dating back to Bachelard and Canguilhem and including Foucault, has influenced the work of authors such as Hacking and Kusch, among others. Now, if a strong case can be made for the need to adopt a historically informed approach to analyze scientific concepts (and the main support for this claim is the visible success of these programs), why should it be any different for philosophical concepts?

Conceptual Genealogy for Analytic Philosophy  83 To clarify further what a genealogy of concepts might look like in philosophical contexts, let us take a brief look at the work of Canguilhem, who has made extensive use of this general idea in his work on the history of medicine and biology—for example, in La formation du concept de réflexe and the classic The Normal and the Pathological (Canguilhem 1978). As is well known, he was a major influence on Foucault, who credits him with promoting a ‘‘philosophy of the concept,’’ as opposed to a ‘‘philosophy of the subject’’ (Méthot 2012, 114).21 Canguilhem distinguishes concepts from theories and describes concepts as theoretically polyvalent (i.e., they can appear, often taking on different meanings, in different theories).22 As described by Méthot (2012, p. 114): [A]n historic-philosophical approach à la Canguilhem consists primarily in tracking scientific concepts over space and time, and across disciplinary boundaries, in order to locate significant shifts regarding meaning, reference, and domains of application. For Canguilhem, a specific instantiation of a scientific concept at a given point in time is closely linked to practices and available technologies;23 in turn, scientific concepts allow for the formulation of new theories and hypotheses, which may provoke further changes in practices and technologies in a feedback loop. However, it would be a mistake to view Canguilhem as a full-blooded ‘social constructivist,’ even if he acknowledges the contribution of social contexts and material conditions for the shaping of scientific concepts. As he notes on the concept of ‘normal’ within medicine and biology: It is life itself and not medical judgment which makes the biological normal a concept of value and not a concept of statistical reality. (Canguilhem 1978, p. 73). The work of Canguilhem in the history and philosophy of medicine and biology provides thus a fruitful point of departure for the method of conceptual genealogy in the context of philosophical analysis, especially given his focus on ‘significant shifts regarding meaning, reference, and domains of application.’ However, one obstacle for the project of genealogical analysis of philosophical concepts is the (still) influential author-centered conception of the history of philosophy. Scholarly work on the history of philosophy typically focuses on authors, in particular, the key figures of the philosophical canon: Plato, Aristotle, Augustine, Abelard, Aquinas, Ockham, Descartes, Leibniz, Hobbes, Locke, Hume, Kant, Nietzsche etc. Now, this is not the place for an extensive discussion of the shortcomings of the author-centered approach to the history of philosophy. For the present purposes, it is sufficient to note that an alternative, concept-centered approach is viable and has, in particular, been developed in detail by A. de

84  Catarina Dutilh Novaes Libera in his recent work (2007, 2008, 2014) presenting an ‘archaeology of the subject.’ In an interview, he describes the enterprise in the following terms: The point for me was to construct a story whose protagonists would not be people, but concepts, problems, rules and arguments.24 And so, there is prima facie no reason why the historian of philosophy or the philosopher should necessarily focus on authors, rather than on concepts or problems, much as the historian of science inspired by Canguilhem and others chooses to do.

3.3 The Historicity of Philosophical Concepts We are now in a better position to describe in more detail what I take to be the five main characteristics of the historicist conception of philosophical concepts that I defend here, borrowing elements from Nietzsche’s conception of genealogy25 and Canguilhem’s concept-centered historical approach.26 These are: 1. Historical change 2. Superimposition of layers of meaning 3. Contingency 4. Multiple lines of influence 5. Connected to (extra- or intra-philosophical) practices and goals 1. Historical change. This is probably the least controversial aspect of the historicist conception of philosophical concepts that I wish to articulate here. Few to none would deny that, as a matter of fact, philosophical concepts do change over time. But the question then becomes how philosophically relevant is it (as opposed to ‘merely historically’ relevant) to track these changes over time? As described in section 2 above, one possible response (and one that is often either explicitly given or presupposed) is that the temporal development of a philosophical concept X is not relevant for the determination of the ‘essence’ of X, i.e., the necessary and sufficient conditions for something to count as X. If past philosophers differed from us in their conceptualization of X, then either they were wrong or we are wrong (most likely they were wrong). We’d learn nothing of philosophical significance about X by examining mistaken past conceptions of X (though some defend the relevance of such ‘mistakes’ in the sense of telling us which mistakes not to repeat; see Crane 2015 for a critique of this argument.) According to the present approach, however, most (if not all) philosophical concepts are by and large theoretical constructs, even if they latch on in important ways to a concept-independent reality (just as Canguilhem’s concept of the ‘normal’ is closely related to the biological phenomenon of life,

Conceptual Genealogy for Analytic Philosophy  85 even if it undergoes modifications through time qua concept). Moreover, as they maintain traces of their previous instantiations throughout their development over time (as will be argued shortly), in particular in the form of presuppositions that then become uncritically accepted, to understand the current instantiation of a particular concept, it becomes crucial to examine how it was construed over time. And thus, the historical development of philosophical concepts is not only a factual observation; it is also a philosophically relevant aspect for the analysis. 2. Superimposition of layers of meaning. The relevance of the philosophical history of a concept for the analysis of its current instantiations relies crucially on the idea of the ‘superimposition of layers of meaning,’ which we encountered when discussing Nietzsche’s conception of genealogy. An alternative conception would amount to, for example, viewing the temporal changes of a concept as corresponding to ‘radical revolutions,’ after which nothing of their previous instantiations stays in place. If this were the case, then the analysis of past instantiations of a philosophical concept with the goal of improving our understanding of its current version(s) would be idle at best, but in fact also potentially distracting and misleading. That the ‘superimposition of layers of meaning’ model offers a more accurate account of the development of philosophical concepts than the alternative ‘radical revolutions’ account is perhaps best argued for by means of concrete examples. So, let me offer a brief example to illustrate the point: the philosophical concept of substance received a number of formulations starting in ancient Greek philosophy, including, in particular, Aristotle’s influential conception. Through the centuries, it received a number of different instantiations, many of which radically different from the Aristotelian conception (e.g., Descartes’ notion of substance), while at the same time retaining some key Aristotelian components such as the idea that there are basic ontological units that are the building blocks for all that exists in reality (Robinson 2014). The main reason why it is of the utmost importance for the philosopher to be aware of the stratified nature of philosophical concepts is the fact that neglecting this dimension leads to the uncritical assimilation of presuppositions and substantive theoretical choices made along the way in the shaping of a concept, which then come to be viewed as truisms. These are often described as one’s absolute ‘intuitions’ about concept X, when in fact they are the products of theoretical choices made (by others) along the way. It is by exposing and investigating these different layers of meaning that the philosopher is able to isolate the theoretical choices that led to the particular shaping of a given philosophical concept over time. Notice also that the ‘superimposition of layers’ model, coupled with the idea of some sort of stable core that fixes the object of the genealogy, sits well with Canguilhem’s idea of the polyvalence and plasticity of a (fruitful) philosophical concept, amenable to receiving a number of theoretical interpretations. Arguably, each of these interpretations will impose a new

86  Catarina Dutilh Novaes meaning on the concept in question, but traces of previous interpretations will remain. 3. Contingency. The process of superimposition of layers of meaning may be understood in at least two radically different ways: either in teleological or in contingent terms. On the teleological conception, each new instantiation of a concept is necessitated by its past instantiations: there is one unique path for the temporal development of a concept (roughly, in the spirit of Hegelian teleology).

X On the contingentist conception, in contrast, each new meaning is one of the many new layers that a given concept is capable of acquiring at any given time and a result of contingent events pertaining to the ‘struggle’ among the different ‘contenders’ (to resort again to Nietzschean ideas). In other words, the historical development of a concept branches to the future:

X

And thus, there will most likely be possible theoretical paths that do not become actualized or instantiated for a given concept.27 Which path does become instantiated (at the expense of other possibilities) is typically a matter of substantive theoretical choices (and possibly of contingent extra-philosophical factors). However, once the choice is made, its substantive content often becomes viewed as a constitutive component of the concept as such, as if it had been there all along, and necessarily so. (In other words, something like a teleological conception is often tacitly endorsed.) But of course, on the contingentist conception, this is a mistake: the historical development of concept X could have taken a rather different turn at some point or another. If this had been the case, then its current instantiations might have become something quite different.28 One of the claims associated with (but not necessitated by)29 the contingentist conception is thus that philosophical concepts are by and large theoretical constructs: they do not track necessary essences that exist in reality (though they do latch on to phenomena in reality in somewhat complicated ways), but rather unfold through time by means of the process of superimpositions of layers of meaning described above. 4. Multiple lines of influence. The contingency aspect just discussed ensures that the developmental path of a concept branches to the future.

Conceptual Genealogy for Analytic Philosophy  87 But what about the past? Well, as mentioned when discussing Nietzsche’s conception of genealogy, the instantiation of a concept or practice at a given point in time is typically the result of a confluence of multiple lines of events and ideas, each having influenced the current status quo in different ways. And thus, the historical development of a concept branches to the past as well, not only to the future:

X

Both Leiter and Geuss note that, in this respect, Nietzschean genealogy differs from genealogy as pedigree, in that the latter features only one non-branching line of ancestry leading to a given person: each person has one unique father. However, this is the case only if the sole lineage that matters in a genealogy is the male lineage. If the female lineages are included, then a person’s genealogical tree is composed of multiple lines of influence, namely the multiple branches and lineages that come together in one and the same person. Viewed from the perspective of multiple lines of influence, the genealogy of a person is not that different from the genealogy of a concept or practice, as described by Nietzsche. More concretely, this means that a philosophical concept will often be shaped by a number of lines of influence. One example is the concept of ‘logical form’; it emerged from the confluence of Aristotelian metaphysical hylomorphism with developments in logic, in particular syllogistic (Dutilh Novaes 2012a). As noted by MacFarlane (2000), Aristotle is both the father of hylomorphism and the father of logic, but he is not the father of logical hylomorphism: he himself never brought the two concepts together. This happened only later, in the tradition of the ancient commentators. (See section 4.2 for further details.) 5. Connected to (extra- or intra-philosophical) practices and goals. The final component of the specific historicist conception of philosophical concepts presented here is the commitment to a view of such concepts not as disembodied constructions, floating above and beyond all human reality and practices. Instead, philosophical concepts are typically embedded in (philosophical as well as non-philosophical) practices and goals and are formulated as responses to specific theoretical (and possibly practical) needs and circumstances of a given intellectual community. The claim is not that philosophical concepts emerge only in response to material conditions and material needs, as the crude Marxist might claim

88  Catarina Dutilh Novaes (though these factors too are more significant than most philosophers tend to recognize). Rather, the idea is that philosophical concepts, theories, and techniques typically emerge or receive new interpretations in response to theoretical or even practical needs, or against the background of preexisting practices. For example, Aristotelian logic is best understood against the background of the practices of ancient dialectic (Kapp 1975; Castelnerac and Marion 2009), which we presume were pervasive in the early Academy (judging from, e.g., Plato’s dialogues). Ignoring the dialectical background gives rise to much confusion concerning otherwise puzzling aspects of Aristotle’s syllogistic such as irreflexivity (Duncombe 2014). Another example might be medieval logical theories. When they began to be investigated more systematically in the mid-twentieth century, scholars tended to look at these theories only from the point of view of modern logic and modern concerns. Thus, Geach (1962) went on to claim that theories of supposition (a very important group of semantic theories in Latin medieval logic) were theories of reference—and, what is more, that they were very deficient theories of reference—without raising the question of whether theories of supposition might be theories of something else altogether. In my own work on theories of supposition, by contrast, the starting point was the question: Why did the medieval authors themselves need something like a theory of supposition? What did they need these theories for? My answer to this question is that a semantic theory providing the tools for textual interpretation was much needed in a tradition in which textual commentary and interpretation played such a crucial role. And so, I claimed that medieval theories of supposition are best understood as theories of sentential meaning, rather than as theories of reference (Dutilh Novaes 2008). An example closer to home is the emergence of modern symbolic logic in the late nineteenth century. It is absolutely imperative to keep in mind the broader context for these developments, namely the projects of axiomatization of portions of mathematics: axiomatizing mathematics was the primary goal and function of the logical systems designed by Frege, Russell, Hilbert, etc. Understanding this background clarifies a number of still pervasive features in the practices of logicians, such as the prominent role of completeness proofs (Awodey and Reck 2002). Thus the theoretical function of a concept, broadly speaking, is typically an important element in how a philosophical concept is shaped and interpreted at a given time. But this of course poses a serious problem for the historian/interpreter, who must resist the temptation to project her own ideas regarding the function(s) of a given concept— which presumably will reflect her own Zeitgeist—into past instantiations of the concept (as Geach seems to have done for supposition/reference). This is a point that Nietzsche was acutely aware of in his genealogical project, and he criticized his predecessors for failing to take it into account: one must avoid projecting current uses and meanings into the past.

Conceptual Genealogy for Analytic Philosophy  89 On this point, Nietzsche follows Darwin, who cautioned against “the mistake of inferring current function or meaning from ancestral function or meaning.” (Leiter 2002/2014, p. 135) In fact, misprojection of meaning and function can take place in both directions: it is a mistake to infer past function or meaning from current ones, but it is also a mistake to infer current meaning and function from past ones. (Notice that this is a potential pitfall for ‘state-of-nature’ genealogical projects such as Craig’s 1990, 2007.) This is so because purpose and function are precisely elements of change and discontinuity in the historical development of a concept—given that theoretical needs, circumstances, and available techniques themselves change over time.30 And thus, the fact that a given concept has had a certain meaning in the past does not mean that the same meaning is the one that we, current philosophers, should attribute to it (see discussion of the genetic fallacy below). However, the point is that traces of these previous meanings (related to past purposes and functions) may still be present in the current embodiment(s) of a concept, which may in fact not necessarily sit well with the current functions and purposes of a concept—and yet are uncritically accepted by current practitioners. Exposing such tensions is indeed one of the goals of conceptual genealogy, just as it is a goal for Nietzschean genealogy.

3.4 Archaeology and Genealogy—Foucault I hope to have argued more or less convincingly by now that, given the specific historicist conception of philosophical concepts I’ve just sketched, genealogy is a particularly suitable method for historically informed philosophical analysis. However, and as mentioned above, I take genealogy to be one among other such historical methods, so there are options. Why is genealogy a better option than the alternatives? In order to address this question, in this section I pitch genealogy against one of its main ‘competitors’ as a method for historical analysis: its close cousin archaeology. Naturally, this confrontation leads me directly to Foucault. As is well known, early in his career, Foucault developed and applied the archaeological method in a number of works, which then received a more explicit methodological reflection in The Archaeology of Knowledge (1969). An “archaeology of knowledge” is an investigation that examines artifacts unearthed in an excavation, but the kind of artifact is not bone, pottery, or metalwork, it is what people said and wrote in the past: their “statements” (in French, énoncé: what has been enunciated or expressed). (Packer 2010, p. 345) The ‘real’ archaeologist digs out material traces of past practices and forms of life, which are then laid out for synchronic analysis (though, of course,

90  Catarina Dutilh Novaes layers of sediments typically convey information about sequences of events). The conceptual archaeologist does something similar with documented discourse, digging deeper towards levels of unconsciousness.31 The premise of the archaeological method is that systems of thought and knowledge (epistemes or discursive formations, in Foucault’s terminology) are governed by rules, beyond those of grammar and logic, that operate beneath the consciousness of individual subjects and define a system of conceptual possibilities that determines the boundaries of thought in a given domain and period. So, for example, History of Madness should, Foucault maintained, be read as an intellectual excavation of the radically different discursive formations that governed talk and thought about madness from the 17th through the 19th centuries. (Gutting 2013, section 4.3) The key word here is ‘excavation’: Foucault took (conceptual) archaeology to allow for the unearthing of what is hidden and unconscious. “It allowed the historian of thought to operate at an unconscious level that displaced the primacy of the subject found in both phenomenology and in traditional historiography” (Gutting 2013, section 4.3). (Recall Foucault’s endorsement of Canguilhem’s rejection of a ‘philosophy of the subject’ in favor of a ‘philosophy of the concept.’) Now, just as the ‘real’ archaeologist does not necessarily seek to establish historical and causal connections between different segments of time, the Foucaultian conceptual archaeologist is not interested in establishing how one way of thinking transitions into another; she mostly looks at different temporal points in isolation. She is, of course, able to notice differences and similarities between different times and ways of thinking, but not much will be said about how such changes come about and how specific modifications to one mode of thinking resulted in a new mode of thinking. Indeed, Foucaultian archaeology is predominantly interested in rupture and discontinuity (Koopman 2008; Packer 2010, chap. 14). In his later work, however, Foucault came to see the essentially synchronic nature of archaeology as a limiting feature of the method, in particular when it comes to stressing the contingency of modes of thinking (Gutting 2013, section 4.3; Koopman 2008). This is when he turned to Nietzschean genealogy (Foucault 1971) in order to fill this lacuna in the archaeological method.32 The shift from archaeology to genealogy (which should, however, be viewed as an addition rather than as a replacement; archaeology remains in the theorist’s toolbox) is a much-debated topic among scholars, and so any brief treatment of it is bound to be superficial. But this seems like an apt description: The much-debated question of why Foucault shifted from archaeology to genealogy can be answered in this way: whereas archaeology offers a static analysis of practices synchronically pulled from the past, genealogy offers a dynamic analysis whereby these practices can be viewed

Conceptual Genealogy for Analytic Philosophy  91 diachronically as historical processes themselves. Genealogy enabled Foucault to explain historical change and continuity. In this way, genealogy was an expansion of archaeology rather than a refutation of it—of course genealogy refutes a few assumptions made by archaeology, but on the whole it refutes these assumptions by reinterpreting the key elements of the earlier approach. (Koopman 2007; emphasis added)33 I submit that the very reasons that led Foucault to supplement (and to some extent revise) his archaeological method with genealogical elements (as suggested in this passage) are also compelling reasons to prefer the genealogical approach over the archaeological one for the kind of historically informed analysis of philosophical concepts that I propose.34 What is needed is a diachronic framework offering the resources to explain not only each particular stage in the history of a philosophical concept, but also the transitions between stages. Nietzschean genealogy, as also developed by Foucault, offers precisely this insofar as it brings to the fore the interplay between continuity and change by means of the key notion of superimposition of layers of meaning. In a slogan, genealogy is about emergence, whereas archaeology is about existence (Foucault 1971; Koopman 2013, 40). Another valuable component of Foucaultian genealogy for the present enterprise (which it shares with Foucaultian archaeology) is its focus on discourse. Recall that the main object of analysis for the theorist engaged in the kind of conceptual genealogy that I propose here are philosophical texts.35 I take it to be of paramount importance that the analysis be firmly grounded in existing documentation—in this case, primarily, but not exclusively, philosophical texts36—rather than being merely speculative. (I worry about the risk of producing philosophical ‘just-so stories.’37 Nietzsche himself already stressed that genealogy is about “that which can be documented, which can actually be confirmed, and has actually existed” (On the Genealogy of Morality, Preface, 7).)38 The focus on texts and documentation naturally still leaves margin for the emergence of selection biases; which texts the theorist will include in her analyses will significantly influence the results, hence the need to be inclusive and to consider a large number of sources (which will of course still typically be a selection). Indeed, a judicious choice of the textual material to work with is one of the main methodological challenges for a conceptual genealogy of philosophical concepts; in particular, it is important to resist the temptation to focus exclusively on a few canonical texts by canonical authors. (In this sense, my proposal differs somewhat from Foucault’s archaeology, which tends to focus on “pronouncements made by figures of authority in positions of power” [Packer 2010, p. 345].) Finally, another commitment that Foucault and Nietzsche seem to share, namely a focus on practices and ‘forms of life’ (to use a Wittgensteinian terminology), represents a useful reminder for the (analytic) philosopher. Analytic philosophers tend to view philosophical theories as ‘disembodied and dis-embedded,’ as if the broader material, social, and cultural contexts

92  Catarina Dutilh Novaes in which they emerge were unimportant (Akehurst 2010). Now, while conceptual genealogy is not the same thing as ‘history of ideas,’ it does recommend that elements outside the purely textual sources be taken into account—including (but not restricted to) facts about circulation and dissemination of texts, such as availability of translations, number of extant copies (in the case of manuscripts), etc. Moreover, the dynamics of how a particular instantiation of a concept becomes influential at the expense of others will often be related to, among other factors, institutional facts pertaining to curriculum and structure of education.39 Such elements (which both the Nietzschean and the Foucaultian would probably be happy to describe as ‘power relationships’) should also be taken into account. Summing up: in the context of the methodological proposal being articulated here, a genealogical approach is to be preferred over its close cousin archaeology for a number of reasons, but most importantly because genealogy is largely diachronic, while archaeology is largely synchronic. Foucault correctly identified this limitation in his earlier archaeological method, and his turn towards Nietzschean genealogy provided the required remedy. 4. APPLICATIONS OF GENEALOGY In the spirit of the functionalist, goal-oriented approach adopted here, a pressing question now becomes: What’s the point of a genealogy? What kind of results do we obtain from performing a genealogical analysis of philosophical concepts? I’ve already mentioned vindication and subversion/ debunking en passant along the way, but now it is time to discuss applications of genealogy in a more systematic way.40

4.1 Genealogy as Vindicatory or as Subversive By now, it should be clear that genealogy is a rather plastic concept, which can be (and has been) instantiated in a number of different ways. Craig offers a helpful description of the range of options: [Genealogies] can be subversive, or vindicatory, of the doctrines or practices whose origins (factual, imaginary, and conjectural) they claim to describe. They may at the same time be explanatory, accounting for the existence of whatever it is that they vindicate or subvert. In theory, at least, they may be merely explanatory, evaluatively neutral (although as I shall shortly argue it is no accident that convincing examples are hard to find). They can remind us of the contingency of our institutions and standards, communicating a sense of how easily they might have been different, and of how different they might have been. Or they can have the opposite tendency, implying a kind of necessity: given a few

Conceptual Genealogy for Analytic Philosophy  93 basic facts about human nature and our conditions of life, this was the only way things could have turned out. (Craig 2007, p. 182) In this section, we are primarily interested in the goals associated with genealogies, in particular the distinction between vindicating and subversive genealogies (the third option raised by Craig, ‘neutral’ genealogies, will be the focus of the next section). However, before discussing these goals more specifically, let us pause for a minute on the kinds of origins that may become the object of genealogical analysis, according to Craig: “factual, imaginary, and conjectural.” We can accordingly distinguish three kinds of genealogical projects. Factual genealogies focus on developments actually having taken place in time and space, as documented in extant sources such as texts, but also other kinds of material evidence. Imaginary genealogies are like foundational myths, which may not be believed to the letter by practitioners (not even as possibilities), but which help them explain and make sense of current practices and beliefs. Conjectural genealogies are different from purely imaginary ones in that things could at least in theory have unfolded as described conjecturally, but these descriptions do not require the kind of evidential documentation involved in factual genealogies. Imaginary genealogies are arguably not particularly prominent within philosophy, but there are a few interesting examples, such as Aristophanes’s myth of the origin of love as described in Plato’s Symposium. Conjectural genealogies, in turn, have enjoyed and continue to enjoy quite some popularity among philosophers. For example, recent uses of evolutionary arguments in ethics—in their majority of the debunking, subversive kind (Kahane 2011)—are typically of the conjectural kind, not necessarily grounded in material documentation or empirical evidence (though there seem to be some exceptions). State-of-nature genealogical enterprises such as the ones by Hobbes, Hume, Rousseau, and more recently by Craig (1990, 2007) and Williams (2002), are overtly conjectural. (Williams in fact describes his genealogy as ‘fictional’ and ‘imaginary,’ but it seems to me to come closer to being a conjectural genealogy.) As such, conjectural genealogies do not seem to offer a fruitful vantage point for the kind of conceptual genealogy of philosophical concepts articulated in this paper; indeed, histories of philosophical concepts based on speculation and conjecture are not going to be very illuminating. In this sense, the philosopher engaged in this enterprise must be more like a ‘proper historian,’ dealing extensively with documented sources.41 And so, the relevant kind of genealogy for our purposes is what can be described as ‘factual genealogy’ (as already suggested elsewhere in this paper). The model here would be that of the French school of historical epistemology as represented by Canguilhem and Foucault, as well as more recent work in the HPS tradition, with emphasis on grounding the analyses in sources and documentation.

94  Catarina Dutilh Novaes And now let us turn specifically to the purposes of engaging in a genealogy. As suggested in the passage by Craig above, on the received view, genealogy involves the idea of passing a judgment of value on a given practice, doctrine or idea by means of a genealogical analysis; genealogies are often used to show that something is good, or else that something is bad. Koopman (2013, p. 62) describes these projects as ‘normatively ambitious’ (and goes on to contrast them with the ‘normatively modest’ genealogical method of Foucault, which he embraces). Craig seems to think that this value component is inherent to any genealogical project, and thus that evaluatively neutral genealogies are in a sense conceptually unviable (or else hopelessly uninteresting—see below). As noted above, vindicatory genealogies are closer in spirit to the commonsensical notion of genealogy, whereby tracing a person’s pedigree serves to legitimate and/or increase her social and political standing. Subversive genealogies, in contrast, turn the commonsensical notion of genealogy upside down; they seek to decrease or question the legitimacy of a given practice or concept by exposing its ‘shameful origins.’ As examples of vindicatory genealogies, Koopman (2013, p. 59) cites the ‘new British genealogists’ Williams, Craig, and Skinner, possibly inspired by the ‘old British genealogist’ Hobbes. Nietzsche is, of course, the quintessential example of a subversive genealogist, but Koopman also mentions a few precursors, such as Darwin42 and Hume. More recently, evolutionary debunking arguments (in ethics (Kahane 2011; Street 2006) as well as elsewhere) can also be viewed as examples of subversive genealogies. (Koopman 2013 focuses specifically on Williams and Nietzsche as representatives of vindicatory and subversive genealogies, respectively.) To further discuss these two kinds of genealogy in general terms, we can continue to follow Craig (2007). He distinguishes genealogies that are intrinsically vindicatory/ subversive from genealogies that are merely accidentally so and then goes on to describe the four categories. He starts with intrinsically subversive genealogies: In the intrinsic type we have an account of the history of certain attitudes, beliefs or practices that their proponent cannot accept without damage to his esteem for, and certitude in, the attitudes, beliefs or practices themselves. For one thing, it may in some cases actually be a part of the belief-system that the belief-system itself had a quite different kind of origin—most religions are like this, perhaps all. (Craig 2007, p. 182) He then goes on to describe how Hume’s account of the origins of monotheistic belief as related to processes that have no apparent connection to truth (some of which are based on motivations that are ‘positively disreputable’) will surely negatively affect the faith of the believer who takes Hume’s story onboard. Similarly for Nietzsche’s account of Christian morality as

Conceptual Genealogy for Analytic Philosophy  95 a self-deceptive expression of ‘hatred, resentment, and bewilderment.’ (His example of an accidental subversive genealogy is Darwinism.) Vindicatory genealogies are described in the following terms: Some genealogies, by contrast, are vindicatory: the story they tell is in one way or another a recommendation of whatever it is they tell us the story of. [. . .] The genealogies—by which I mean the causal stories—of many of our beliefs are intrinsically justificatory in a very strong sense: they give an essential place to the very facts believed in, so if that is how they came about they must be true. Or a genealogy may vindicate a practice, exhibiting it as arising out of the need to find a solution to a problem; and we may then regard it as intrinsically vindicatory if the problem is one that any human society [. . .] will want to solve. [. . .] A genealogy is accidentally vindicatory, on the other hand, when the increased prestige it confers on its object is due to features that are relatively local, or of limited timespan. (Craig 2007, p. 183) Unlike subversive genealogies, vindicatory genealogies show us that we have good reasons to hold the beliefs and practices that we hold, either because the belief-forming process was reliable and truth-conducive, or else because the beliefs and practices present themselves as solutions to inherently important problems. Though this need not always be the case, vindicatory genealogies will typically confer a certain necessity and inevitability to their objects (‘they must be true’). By contrast, subversive genealogies will typically highlight the contingency of the beliefs and practices in question.43 (Recall Craig’s quote at the beginning of this section, distinguishing genealogies emphasizing contingency from genealogies emphasizing necessity.) How does the distinction between these two kinds of genealogy fare when applied to the conceptual genealogies of philosophical concepts that are the object of the present analysis? Here too, it seems that this is a useful distinction. A conceptual genealogy of this kind may be vindicatory if it shows that a given philosophical concept or doctrine has a ‘venerable pedigree,’ for example that it was developed and/or maintained by some of the great figures of our philosophical canon. One example that springs to mind is the enthusiasm with which proponents of the ‘Language of Thought’ hypothesis (e.g., Fodor) received historical analysis showing that this general idea had antecedents in Latin medieval philosophy, Ockham in particular (Panaccio 2004). In contrast, a conceptual genealogy may be subversive if it shows that the historical (and conceptual) grounds for a given notion or doctrine are either confused, shaky philosophical ideas, or else philosophical theses and doctrines to which we no longer want to commit. One example of the latter would be feminist critiques of logic (Plumwood 1993), among others. We have seen that, while he considers the possibility of a third, evaluatively neutral kind of genealogy, Craig then goes on to dismiss it (I will

96  Catarina Dutilh Novaes discuss his reasons for doing so shortly). If he is right, then any genealogy has an intrinsic evaluative component, and thus the vindicatory vs. subversive distinction will be exhaustive. However, in the next section I argue that this third category is not only viable; it is also quite promising, in particular from the point of view of the method of conceptual genealogy that I am sketching here. I will use the term ‘explanatory’ for my own characterization of evaluatively neutral genealogies of this kind. (Koopman 2013 also defends a third option, which he describes as ‘problematization’ and attributes it to Foucault.)

4.2 Genealogy as Explanatory Craig does consider the possibility of evaluatively neutral genealogies, which he describes in the following (rather dismissive) terms: There may also be neutral genealogies, which give us a history of X without either impugning or enhancing the standing of X. I doubt whether there can be such a thing as an intrinsically neutral genealogy [. . .]. But I also doubt whether this is a very interesting class for philosophy, and don’t propose to spend time or energy on it. (Craig 2007, p. 184) In other words, even if there can be such a thing as a neutral genealogy, it will not constitute an interesting enterprise, in any case not for the philosopher. Craig’s argument in support of this claim is based on functionalist considerations: [V]ery many genealogies work by ascribing functions to their objects, telling us what they are for. If the function is of some importance to us and the object performs it well, we have to that degree a recommendation, if we find the function in some way disreputable, then a critique. If the function really is one to which we are indifferent it becomes unclear what the genealogist can be aiming for. (Craig 2007, p. 184) The underlying assumption seems to be that what determines the value of a genealogy is the importance we attach to the function(s) attributed to the object by means of the genealogical analysis. It seems thus that the value of the function as such is not put under scrutiny; we’ve already assumed that it is valuable, or we are only interested in it because it is valuable for us. In response to Craig, firstly we may point out that a genealogy may lead precisely to a critical evaluation of the very function(s) we attribute to its object: from valuable to invaluable, from neutral to valuable, from invaluable to neutral, etc. A genealogy of monotheistic religion along Humean lines may uncover that the traditional, presumed function of religion, that of revealing truths about God, is not well-served by these practices, but it may nonetheless unearth other desirable functions performed by religious

Conceptual Genealogy for Analytic Philosophy  97 practices, such as social cohesion. So even the theorist who does not value the goal of obtaining divine truths may still view a genealogy of religion as a worthwhile enterprise if it reveals other (desirable) consequences and functions of religious practices. (Of course, the Nietzschean/Marxist genealogist will, by contrast, emphasize the undesirable consequences of religious practices.) Moreover, in the spirit of the broadly Nietzschean conception of genealogy adopted here, changes of functions and meanings through time are precisely one of the things that a genealogy aims at capturing. Secondly, and more importantly, Craig seems not to appreciate sufficiently the purely epistemic dimension of a genealogy, which emerges independently of whatever evaluative judgment one may wish to pass. As well put by Allen (2013) in her review of (Koopman 2013), [T]he point of genealogy is not only to demonstrate that our practices or concepts or norms or forms of life are contingent and therefore could be otherwise; genealogy also aims to show how those practices, concepts, norms, and forms of life have been composed through complex practices. Indeed, it is at least conceivable that one and the same genealogical narrative may be subversive for some and at the same time vindicatory for others, while both agree on the details of the narrative. (A genealogy of religion showing that it does not perform the function of revealing divine truths successfully, while also revealing other desirable social functions of religious practices, may be subversive for the revelation-seeker and yet vindicatory for the social theorist.) What this suggests is that the epistemic, explanatory component is not intrinsically tied to the evaluative component, and thus that a genealogy can also fulfill a purely explanatory function. Koopman (2013) defends what he describes as a ‘normatively modest’ conception of genealogy (as opposed to the ‘normatively ambitious’ conception of Nietzsche and Williams), which he refers to as ‘problematization,’ following Foucault. He emphasizes the Kantian, critical nature of Foucault’s project: Foucault’s transformative appropriation of Kant’s critical project is best understood as deploying critical inquiry for the purposes of the problematization of our historical present. Kant’s and Foucault’s projects are both properly critical in that they are inquiries into the conditions of the possibility forming the limits of our human ways of being. (Koopman 2013, p. 17) The key Kantian terms here are ‘conditions of possibility’ and ‘limits.’ Foucault’s main transformation consists in ‘historicizing’ the Kantian critical project; a problematizing genealogy consists in a history of the present, in a narrative of the historical conditions of possibility for things to be as they

98  Catarina Dutilh Novaes are in the present—how they came about. The challenge for Koopman is to equate the kind of transcendental necessity involved in the Kantian idea of ‘conditions of possibility’ with the emphasis on contingency that the historicist approach brings in. At any rate, it seems to me that the crux of Koopman’s claim to being ‘normatively modest’ in his conception of genealogy44 is the emphasis on the explanatory nature of genealogy thus understood. It seeks to make the present comprehensible, intelligible—to explain how the present came about—which subsequently may or may not lead to efforts of revision and transformation. My conception of the genealogy of philosophical concepts as essentially explanatory (thus also intended to be ‘normatively modest,’ if not outright ‘normatively neutral’) is related to the Foucault/Koopman approach in a number of ways, in particular given the common emphasis on the epistemic dimension of elucidating how things came about as preceding evaluative judgment. However, while Koopman emphasizes ‘conditions of possibility,’ I emphasize ‘underlying assumptions,’ in the sense of the idea of superimposition of layers of meaning detailed in section 3.3 above. The thought is that the history of a philosophical concept is marked by a number of branching moments in which theoretical choices are (tacitly or explicitly) made, and these choices constitute underlying assumptions which we do well to isolate (but not necessarily reject). By isolating them, we may conclude that these are assumptions we should continue to endorse, but with the benefit of having exposed them and made them explicit thanks to the genealogical process. (To be sure, there may well be non-genealogical ways to uncover assumptions as well.) Or we may conclude that these assumptions, which presumably seemed reasonable to philosophers at one point in time, are no longer plausible to us; in this case, the genealogy may lead to a subversive critique of the concept in question. Perhaps an example may be required to illustrate the point, so here follows a condensed version of the genealogy of the concept of logical form that I developed in a number of articles (Dutilh Novaes 2011, 2012a, b). The concept originates in Aristotle’s metaphysical (not logical) hylomorphism, which encompasses three tenets: uniqueness of form (each substance has only one substantive form, which is the principle of unity that keeps its parts together);45 a non-mereological understanding of the form-matter composite (form is not a part of the whole; rather, it is the principle of unity that keeps the parts together); and a sharp, principled distinction between what is form and what is matter (though they are intrinsically intertwined in substances). Aristotle, however, does not apply the form-matter distinction to logical objects such as arguments in any systematic way. The next step in the development of the concept of logical form occurs with the ancient commentators (second to sixth century A.D.), who went on to apply the form-matter distinction systematically to logical objects, in particular syllogistic arguments. They maintained the tenets mentioned

Conceptual Genealogy for Analytic Philosophy  99 above (uniqueness, non-mereologicality, principled distinction) by identifying the figure of a syllogism as its form. Early medieval authors then went on to identify two senses of form with respect to syllogistic arguments, namely figure and mood, thus modifying the uniqueness condition. In later authors, however, such as the fourteenth-century author John Buridan, the form of a syllogism is understood exclusively as its mood, and we thus arrive at a mereological understanding of logical form according to which the form of an argument—its logical, syncategorematic terms—is a proper part of the argument as a whole. But then we also return to the idea of uniqueness of form: each argument has exactly one, unique logical form. Thus, a genealogy of the concept of logical form unveils the deeply metaphysical presuppositions still underlying what I described as ‘logical hylomorphism as we know it’; these are presuppositions tied to a particular metaphysical theory of the nature of substances (Aristotle’s hylomorphism). At the same time, we observe that one of the components of the original matrix, the non-mereological understanding of the hylomorphic compound, is now lost. Having exposed these presuppositions, the theorist is now in a better position to evaluate them critically, but not necessarily to supplant or revise them. True enough, this kind of genealogy is not entirely neutral insofar as it shows that certain components of the concept of logical form, which might be viewed as necessary, constitutive features of the concept, are in fact somewhat contingent. For example, one may wonder what a non-mereological logical hylomorphism might look like, and whether it might be more compelling than its mereological counterpart, which we now by and large accept uncritically. But at this point, the analysis is largely evaluatively neutral in that it does now adjudicate between the different options. Similarly, the presupposition of uniqueness of form, which makes perfect sense within Aristotle’s framework and is essentially maintained in modern logical hylomorphism, could in theory be questioned: Why can’t an argument have a plurality of logical forms? The space of theoretical possibilities is thus enlarged once assumptions are isolated and it becomes clear that different positions with respect to each of them are viable.46 In a similar vein, a number of ‘intuitions’ pertaining to this concept can be explained as products of contingent, historical processes of ‘superimposed layers of meaning,’ and can thus be put under critical scrutiny rather than taken as unquestionable ‘hard data’ for philosophical analysis. In sum, the kind of conceptual genealogy of philosophical concepts that I defend is, as such, neither subversive nor vindicatory, though it can be used for both ends. In first instance, however, it seeks to produce a narrative that is above all explanatory by describing the theoretical turning points in the historical shaping of a philosophical concept through time (in the spirit of Canguilhem’s ‘shifts of meaning’). In doing so, it reveals assumptions underlying the concept in question that are often tacitly and uncritically accepted, thereby widening the space of theoretical possibilities.47

100  Catarina Dutilh Novaes

4.3 The Genetic Fallacy What remains to be discussed now is the most prominent objection put forward against genealogical explanations, namely the charge that they commit the so-called ‘genetic fallacy.’ The term itself is credited to Cohen and Nagel in their 1934 textbook Logic and Scientific Method (Honderich 2005). Generally speaking, it consists in the fallacy of confusing the causal origins of a belief with its justification: “any attempt to support or to discredit a belief, statement, position or argument based upon its causal or historical genesis, or more broadly, the way in which it was formed” (Klement 2002, quoted in Koopman 2013, p. 62). In other words, the genetic fallacy would amount to a conflation of the descriptive level of how something came about with the normative level of its justification (again, the good old Popperian distinction between context of discovery and context of justification). One reason to think that such explanations are indeed fallacious is the observation that past instantiations of a given phenomenon may be so fundamentally different from its current instantiations that reference to the former can in no way serve as justification for the latter. However, it has been argued that genetic reasoning is not always fallacious (Klement 2002). Koopman (2013) takes his response to the genetic fallacy to be precisely what distinguishes his Foucaultian conception of genealogy from vindicatory or subversive genealogies. Indeed, if a genealogy does not have the normative goal of justifying, supporting, or discrediting a belief or practice, it cannot be said to be confusing the normative level of justification with the descriptive level of origins. A genealogical analysis that takes itself to be essentially descriptive and explanatory makes no strong normative claims to start with. At the same time, if the ‘layers of meaning’ conception of philosophical concepts is correct, then retracing the historical development of a concept will tell us something important about its current instantiations as well. The same holds for the notion of conceptual genealogy of philosophical concepts presented here. Going back to the example of the genealogy of the concept of logical form, by retracing the Aristotelian, metaphysical origins of some of the components of current versions of logical hylomorphism, one neither justifies nor debunks these underlying assumptions. The fact that the form-matter distinction has venerable metaphysical origins does not mean that its application to logical entities is automatically justified; nor does it mean that it is unjustified, given that conceptual distinctions can be fruitfully transferred from one realm of investigation to another. But once these components have been isolated, the general normative question becomes: can these metaphysical principles, which seem at least prima facie plausible in metaphysical contexts, be transposed (mutatis mutandis) to the realm of logic? Are logical entities such as arguments the kinds of things to which we can attribute form and matter? On which version of the form-matter

Conceptual Genealogy for Analytic Philosophy  101 distinction? But it is precisely thanks to the genealogical analysis that these normative, systematic questions can be asked. There is another sense in which the broadly Nietzschean conception of genealogy endorsed here does not fall prey to the genetic fallacy: in the sense that the synchronic analysis of ‘superimposition of layers of meaning’ outlines precisely the changes in function and meaning in the practices and concepts being analyzed. Hence, from the start we have the recognition that past instantiations of the phenomenon in question may be fundamentally different from its current instantiations, so much so that reference to past instantiations may do no justificatory work. In other words, even if something started out as ‘good’ (or ‘bad’), this does not mean that its current instantiations will necessarily also be ‘good’ (or ‘bad’), as goodness (or badness) are not necessarily preserved and transmitted through shifts of meaning. (See section 3.3 for Nietzsche’s critique of the practice of projecting current meanings to past instantiations, and vice versa.) And yet, examining these shifts of meaning, and identifying traces of previous instantiations as well as the theoretical choices that led to the changes, greatly contributes to our understanding of the current instantiations of some of our beloved philosophical concepts. And thus, I conclude that the methodology of conceptual genealogy as formulated here does not fall prey to the genetic fallacy. 5. CONCLUSION In this paper, I’ve drawn inspiration from a number of authors to formulate the methodology of conceptual genealogy, which I claim is relevant for any kind of philosophical analysis (even those that view themselves as purely systematic and normative). The main inspiration comes from Nietzsche, but Foucault and Canguilhem have also contributed important ingredients. These are so-called continental authors, and yet I claim that historically informed analysis is just as important for analytic philosophers. The goal of this paper was to provide a more explicit articulation of the method, in its ins and outs, and perhaps convince others of its viability, fruitfulness, and relevance. But ultimately, the proof is in the pudding, i.e., the fruitfulness and relevance of the ‘conceptual genealogy’ approach can only be established by means of successful applications of the method: philosophical analyses that succeed in illuminating aspects of philosophically important concepts that remain otherwise difficult to account for. Some of my previous work on the history and philosophy of logic may serve as illustration for the method in action, but I take it that others have been relying on something similar to this approach in their own work. And, of course, I hope yet others will feel inclined to adopt some of these ideas in their own work, such

102  Catarina Dutilh Novaes that discussions within analytic philosophy may become more thoroughly informed by the history of philosophy.

ACKNOWLEDGMENTS This paper is living proof of the power of the Internet to foster collaborative work. Most of it was published in installments in the form of blog posts at NewAPPS, and I’ve greatly benefited from feedback from readers there. The whole project has also benefited from extensive discussions on Facebook with fellow philosophers. And thus, many people have contributed to the project in important ways, but I would like to thank in particular Leon Geerdink, Matthew Duncombe, Peter Tarras, and Colin Koopman.

NOTES  1 “Habermas (2003) suggests that continental philosophy, following G.W.F. Hegel, continually reflects upon, and brings up into its manifest content, its own sociohistorical situation (79–80). Thus, unlike analytic philosophy, continental philosophy has never turned away from culture, tradition, literature, psychodynamics, and political economy in the course of thinking about language and meaning. By contrast, analytic philosophy—especially in the wake of the so-called linguistic turn—has tended to think about language in abstraction from such matters” (Sachs 2011, p. 303). (I am quoting Sachs’ paraphrase instead of Habermas’s own text because—perhaps unsurprisingly—the paraphrase is much clearer and to-the-point than the original.) See also (Williamson 2014) for analytic philosophy as a “somewhat anti-historical tradition.”  2 References to my previous work will often be made to illustrate the points being made. This does not mean that I am the only (analytic) philosopher resorting to the method of conceptual genealogy, but obviously I know my own work better than the work of others.   3 More generally, the paper will only very selectively discuss the incredibly vast philosophical literature on genealogy. In effect, a comprehensive overview of this literature would require a book-length treatment, and so a number of pertinent works will have to be left out. In other words, this paper should not be seen as a systematic discussion of the concept of genealogy as such, but rather perhaps as a ‘methodological manifesto’.  4 As noted by Haslanger (2005), “our concepts and our social practices are deeply intertwined. Concepts not only enable us to describe but help structure social practices, and our evolving practices affect our concepts.” The main reason to restrict the method of conceptual genealogy as described here to philosophical sources and materials is essentially methodological; attention to other kinds of sources would require a kind of training and expertise that the philosopher (or even the historian of philosophy) typically does not receive/ possess. However, the philosopher engaged in conceptual genealogy can and should engage with a broader range of information, for example historical analyses of educational institutions, as studied by those with proper expertise such as historians.   5 See also (Koopman 2013) for a similar tripartite taxonomy of genealogies.   6 See (Westphal 2010) and (Sachs 2011).

Conceptual Genealogy for Analytic Philosophy  103   7 Carnap seems to have been a key figure in the movement away from history within analytic philosophy (Sachs 2011). Indeed, his fallout with Heidegger has been described as one of the seminal events in the schism between analytic and continental philosophy (Friedman 2002).   8 Sellars would be another example of an analytic philosopher who regularly engages with the history of philosophy in his analyses, to some extent from a ‘genealogical’ perspective. The same applies to those influenced by him, e.g., Robert Brandom. But perhaps the analytic philosopher having engaged most extensively with the history of philosophy for systematic analysis in recent years is Rorty, especially in his Philosophy and the Mirror of Nature (1979).  9 It is also interesting to notice that most of the genealogical projects in the analytic tradition tend to be of the vindicatory kind, whereas the continental projects tend to stick closely to the Nietzschean subversive spirit. The exception is again Hacking, who is, in any case, in a league of his own. 10 See also (Akehurst 2010, Introduction). 11 Haslanger (2005) helpfully distinguishes three kinds of analysis of X (say, of knowledge): conceptual, descriptive, and ameliorative. The first pertains to our concept of X; the second pertains to X as such, as a phenomenon in reality; the third pertains to the function of concept X. The genealogical approach to be described here focuses on conceptual analysis, since we are interested in how conceptions of X evolved over time among philosophers, but also on ameliorative analysis, as the functional role of concepts (possibly connected to social, cultural factors outside the practice of philosophy) will also be of paramount importances. (More generally, much of what Haslanger has to say on genealogy in her paper is eminently relevant for the present discussion. Unfortunately, I only became aware of these connections once the present paper was already in press, at which point substantive modifications were no longer possible.) 12 See also (Srinivasan ms.) on reasons why even the staunch analytic philosopher should not dismiss the genealogical perspective completely. 13 In what follows, I mostly rely on Geuss’s reading of Nietzschean genealogy, as it contains all the elements needed for the argumentation to be developed later on. 14 See (Geuss 1994, p. 275) for the five main characteristics of pedigree tracing. 15 Notice that Beiser (2011, p. 9) does not include Hegel among the historicists, chiefly because Hegel’s project was to turn history into a kind of science, with necessary laws not different from the laws of the natural world (hence his teleological conception of history). For this he (and Marx) was severely criticized by other nineteenth-century historicists. 16 Geuss (1994) details subsequent transformative steps in the development of Christianity, but the ‘Pauline step’ is perhaps the most illustrative one. 17 It seems to me that the term ‘essential’ is not the most appropriate to capture the stable component in a genealogy, but the general point still stands. 18 Leiter provides the following passage by Nietzsche in support: “We have to distinguish between two of its aspects: one is its relative permanence, a traditional practice [Brauch], a fixed form of action, a ‘drama’, a certain strict sequence of procedures, the other is its fluidity, its meaning [Sinn], purpose and expectation, which is linked to the carrying out of such procedures (GM II: 13).” 19 This idea bears some resemblance to the famous ‘thesis-antithesis-synthesis’ triad, typically—but wrongly—attributed to Hegel, minus the teleological component. 20 HPS manifesto, available at http://www.pitt.edu/~pittcntr/Events/All/Conferences/others/other_conf_2007–08/andHPS/andHPS_manifesto.htm 21 Canguilhem vehemently criticized what he described as ‘the virus of the precursor’ (Gutting 1989, p. 39), that is, the preoccupation with establishing who

104  Catarina Dutilh Novaes said what first. What matters for the historian and philosopher of science from Canguilhem’s perspective are not so much the actors, but the concepts produced and developed. 22 One rough way in which the distinction between concepts and theories can be cashed out is that concepts describe or interpret phenomena, whereas theories attempt to explain them. See (Gutting 1989, pp. 32–55). This kind of theoretical plasticity is also present in philosophical concepts, as we will see when discussing specific examples. 23 “[T]he determination of the concept of ‘‘normal’,’ especially in physiology and clinical medicine, is linked to specific laboratory equipments and sets of practices; in other words, that there is no absolute concept of the ‘‘normal’’ state in medicine or physiology.” (Méthot 2012, p. 119) 24 “Il s’agissait donc pour moi de construire une intrigue dont les protagonistes seraient non des personnes, mais des concepts, des problèmes, des règles et des arguments.” Interview for Actu Philosophia, available at http://www. actu-philosophia.com/spip.php?article77. The starting point for this movement towards ‘anonymization’ is de Libera’s own work on medieval logical texts, in particular the sophismata literature of the end of the twelfth century and the thirteenth century. As he describes it, 75% of the relevant texts have no clear authorship, and so obviously an author-centered approach would have been eminently unsuitable for the analysis of this material. Notice also that De Libera focuses on the concept of ‘archaeology’, not that of ‘genealogy’; we discuss this distinction later on. 25 It may be argued that the analogy does not hold water because Nietzschean genealogy is genealogy of practices, of real ‘forms of life’, while the kind of conceptual genealogy I am proposing is not. However, precisely because my notion of philosophical concepts is closely tied to practices and applications itself, the gulf between the two enterprises will be much less large than it might seem at first. 26 What follows is a description of each of these aspects; I don’t exactly offer arguments to support this account of philosophical concepts (as opposed to ahistorical, realist/essentialist conceptions). Indeed, what could count as an argument here? I take it that the explanatory success of analyses based on this conception (i.e., their ability to explain aspects and features of given philosophical concepts better than alternative approaches) is the ultimate ‘test,’ and so I do not expect an opponent of this approach to be thoroughly convinced solely on the basis of this description; instead, she is invited to examine case studies and applications of this methodology. 27 In fact, different philosophical traditions may emerge precisely if two or more groups take different paths with respect to core philosophical concepts. So, at a given point in time, it may well happen that more than one path is instantiated, but by different groups/traditions. This can be observed for example in the different interpretations given to some core Aristotelian concepts in the Latin and Arabic medieval traditions, respectively. 28 This is a Wittgensteinian idea: “One of my most important methods is to imagine a historical development of our ideas different from what has actually occurred. If we do that the problem shows us a quite new side.” (Wittgenstein 1998, p. 45) 29 I owe this caveat to Jacob Archambault. 30 A similar point holds of Canguilhem’s conception of the genealogy of concepts, which (on Méthot’s interpretation at least) is very much attuned to practical aspects such as the development of material techniques. According to Méthot, Canguilhem’s is an ‘embodied-embedded’ conception of concepts, contrary to what some critics have said. 31 See (Kusch 1991, Part I, chap. 2) for further comparison between what a ‘real’ archaeologist does and the Foucaultian archaeologist, as well as for the influence of psychoanalytic concepts. Kusch also documents the pervasiveness of the concept of archaeology among Foucault’s teachers and intellectual

Conceptual Genealogy for Analytic Philosophy  105 predecessors—in other words, the idea of archaeology for conceptual analysis is not a Foucaultian novelty. 32 Koopman (2013) emphasizes Foucault’s indebtedness to Kant, interpreting his genealogical method as first and foremost a critical project, concerned with the limits and conditions of possibility for “our human ways of being” (Koopman 2013, p. 17). (More on this point below.) 33 See (Koopman 2008) and (Koopman 2013, Chap. 1) for more detailed elaboration of these ideas. 34 A small autobiographical note: I also started thinking of this approach in terms of archaeology, but gradually came to realize that what is really required is the genealogical perspective. 35 Crane (2015) also emphasizes the importance of focusing on a tradition of philosophical texts, but introduces the idea of how these texts are read as equally tradition-defining. 36 I do make a plea for taking into account the larger historical and cultural contexts for the production of philosophical texts, in particular in connection with the embedding of the philosophical theories into broader institutional and social practices. In this sense, documentation going beyond purely philosophical texts is also relevant for the analysis. 37 This concern is what makes me less enthusiastic about ‘state-of-nature’ genealogical projects, such as those described in (Craig 2007). 38 Leiter (2002/2014, 134) reads this passage as suggesting that the Nietzschean genealogist is interested in essences, real things. I abstain from getting involved in a debate with Leiter’s essentialist reading of Nietzsche, but in any case I very much share Nietzsche’s concern for proper documentation and a rejection of ‘English hypothesis-mongering.’ I take it that Foucault’s focus on discourse is also in the spirit of concern for proper grounding and documentation. Indeed, as suggested by Koopman (2013, p. 61), despite his criticism of hypothesis-mongering, Nietzsche himself indulges in a fair amount of speculation when engaged in genealogy, while Foucault displays a much greater degree of archival breadth and attention for documentation. 39 The importance of these factors is also pointed out by authors such as Bruno Latour, Annemarie Mol, and Ian Hacking, among others. 40 Koopman (2013, Chap. 2) offers an in-depth examination of this question, and so will serve as the main source for the present discussion, along with Craig (2007). 41 To be honest, I am suspicious of conjectural genealogies in general, which seem to me to be detrimental to progress in a number of different disciplines by producing ‘just-so stories.’ But this is not the place for an extensive discussion of my reservations. 42 The connections between Nietzsche and Darwin have received much attention from scholars, a recent example being Johnson (2013). 43 However, I do not want to maintain that contingency is inherent to subversive genealogies, while necessity is inherent to vindicatory genealogies. It seems to me that subversive genealogies emphasizing necessity as well as vindicatory genealogies emphasizing contingency are both at least conceptual possibilities. 44 Notice though that he does not claim to be ‘normatively neutral,’ as a genealogy as problematization carries the potential for intervention: “if genealogy helps us to see how our present was made, it also thereby equips us with some of the tools we would need for beginning the labor of remaking our future differently.” (Koopman 2013, p. 130) 45 Medieval theories of the form-matter compound will later make room for a plurality of substantive forms, thus parting ways with Aristotle’s own version of hylomorphism. 46 Crane (2015) makes a similar point with respect to the philosophical questions we take to be central: “The second moral is that an awareness of the history

106  Catarina Dutilh Novaes of the question one is pursuing can make one sensitive to the contingency of the question, in the sense of the contingency of the philosophical concerns that give rise to it. For example, the formulation of the doctrine of materialism/ physicalism in analytic philosophy took a very specific form in the twentieth century—in terms of all truths being expressed in physical language—because of very specific ideas deriving from logical positivism about what philosophy can and cannot do. Reflection on these ideas and their effects can help us see that they are detachable from the core of doctrines like materialism; and dispensing with these assumptions can help us to see the questions in a new light. In some cases, it may help us move away from the questions altogether, and to pose new questions which make more sense to us today.” (Emphasis added.) 47 As for Craig’s charge that an evaluatively neutral genealogy will not be philosophically interesting, I refer the reader to my defense of descriptive metaphilosophy (a term introduced by Rescher, in opposition to prescriptive metaphilosophy) as philosophy: “Inspired by Nietzsche and Foucault, I call the enterprise of tracing the history of a given philosophical concept conceptual genealogy. It now seems to me that what I’ve been arguing for all along, in Rescher’s terms, is for the importance of descriptive metaphilosophy for the enterprise of prescriptive metaphilosophy, and thus for the claim that historically-informed descriptive metaphilosophy is indeed part of philosophy tout court, just as prescriptive metaphilosophy. More to the point, I’ve been claiming that prescriptive metaphilosophy desperately needs descriptive metaphilosophy.” ‘A plea for descriptive metaphilosophy as philosophy’, http://www.newappsblog.com/2014/03/a-plea-for-descriptivemetaphilosophy-as-philosophy.html

WORKS CITED Akehurst, T. L. 2010. The Cultural Politics of Analytic Philosophy: Britishness and the Spectre of Europe. New York: Continuum. Allen, A. 2013. Review of Koopman (2013). Notre Dame Philosophical Reviews. https:// ndpr.nd.edu/news/41851-genealogy-as-critique-foucault-and-the-problemsof-modernity/ Awodey, Steve and Erich H. Reck. 2002. Completeness and Categoricity. Part I: Nineteenth-century Axiomatics to Twentieth-century Metalogic. History and Philosophy of Logic 23 (1): 1–30. Beiser, Frederick C. 2011. The German Historicist Tradition. Oxford: Oxford University Press. Canguilhem, G. 1978. On the Normal and the Pathological. Dordrecht: Reidel. Castelnérac, B. & Marion, M. 2009. Arguing for Inconsistency: Dialectical Games in the Academy. In Giuseppe Primiero, ed., Acts of Knowledge: History, Philosophy and Logic. London: College Publications. Craig, E. 1990. Knowledge and the State of Nature: An Essay in Conceptual Synthesis. Oxford: Oxford University Press. Craig, E. 2007. Genealogies and the State of Nature. In Alan Thomas, ed., Bernard Williams. Cambridge: Cambridge University Press. Crane, T. 2015. Understanding the Question: Philosophy and its History. In John Collins and Eugen Fischer, eds. Experimental Philosophy, Rationalism, and Naturalism: Rethinking Philosophical Method. London: Routledge, 72–84. De Libera, A. 2007, 2008, 2014. L’Archéologie du Sujet (vols. 1–3). Paris: Vrin. Duncombe, M. 2014. Irreflexivity and Aristotle’s Syllogismos. Philosophical Quarterly 64 (256): 434–452. Dutilh Novaes, C. 2008. An Intensional Interpretation of Ockham’s Theory of Supposition. Journal of the History of Philosophy 46 (3): 365–393.

Conceptual Genealogy for Analytic Philosophy  107 Dutilh Novaes, C. 2011. The Different Ways in Which Logic Is (said to be) Formal. History and Philosophy of Logic 32 (4): 303–332. Dutilh Novaes, C. 2012a. Form and Matter in Later Latin Medieval Logic: The Cases of Suppositio and Consequentia. Journal of the History of Philosophy 50 (3): 339–364. Dutilh Novaes, C. 2012b. Reassessing Logical Hylomorphism and the Demarcation of Logical Constants. Synthese 185 (3): 387–410. Foucault, M. 1971. Nietzsche, la généalogie, l’histoire. In S. Bachelard et al., eds., Hommage à Jean Hyppolite. Paris: Presses Universitaires de France. Foucault, M. 1969. Archaeology of Knowledge. London: Routledge. Foucault, M. 1975. Surveiller et Punir. Paris: Gallimard. Foucault, M. 1976. Histoire de la sexualité, 3 volumes: La volonté de savoir, L’usage des plaisirs, and Le souci de soi. Paris: Gallimard. Friedman, M. 2002. Carnap, Cassirer, and Heidegger: The Davos Disputation and Twentieth Century Philosophy. European Journal of Philosophy 10 (3): 263–274. Geach, P. T. 1962. Reference and Generality. Ithaca, NY: Cornell University Press. Geuss, R. 1994. Nietzsche and Genealogy. European Journal of Philosophy 2 (3): 274–292. Gutting, G. 1989. Michel Foucault’s Archaeology of Scientific Reason. Cambridge: Cambridge University Press. Gutting, G. 2013. Michel Foucault. In E. Zalta (ed.), Stanford Encyclopedia of Philosophy. http://plato.stanford.edu/entries/foucault/. Hacking, I. 1995. The Emergence of Probability. Cambridge: Cambridge University Press. Hegel, G. W. F. 1820 (1991). Elements of the Philosophy of Right. Cambridge: Cambridge University Press. Honderich, T., ed. 2005. The Oxford Companion to Philosophy. Oxford: Oxford University Press. Hylton, P. 1990. Russell, Idealism, and the Emergence of Analytic Philosophy. Oxford: Oxford University Press. Johnson, D. R. 2013. One Hundred Twenty-Two Years Later. Journal of Nietzsche Studies 44 (2): 342–353. Kahane, G. 2011. Evolutionary Debunking Arguments. Noûs 45 (1): 103–125. Kapp, E. 1975. Syllogistic. In Jonathan Barnes, Malcolm Schofield, and Richard Sorabji, eds., Articles on Aristotle. London: Duckworth, 1–35. Klement, K. C. 2002. When Is Genetic Reasoning Not Fallacious? Argumentation 16 (4): 383–400. Koopman, C. 2008. Foucault’s Historiographical Expansion: Adding Genealogy to Archaeology. Journal of the Philosophy of History 2 (3): 338–362. Koopman, C. 2013. Genealogy as Critique: Foucault and the Problems of Modernity. Bloomington, IN: Indiana University Press. Kornblith, H. 2011. Why Should We Care About the Concept of Knowledge? Episteme 8 (1): 38–52. Kusch, M. 1991. Foucault’s Strata and Fields: A Study in Archaeological and Genealogical Science Studies. Dordrecht: Kluwer. Kusch, M. 2013. Naturalized Epistemology and the Genealogy of Knowledge. In Martin Lenz and Anik Waldow, eds., Contemporary Perspectives on Early Modern Philosophy. Berlin, Springer, 87–97. Leiter, B. 2002/2014. Nietzsche on Morality. London: Routledge. MacFarlane, J. 2000. What Does It Mean to Say That Logic Is Formal? Dissertation, University of Pittsburgh. Méthot, P.-O. 2012. On the Genealogy of Concepts and Experimental Practices: Rethinking Georges Canguilhem’s Historical Epistemology. Studies in History and Philosophy of Science Part A 44 (1):112–123.

108  Catarina Dutilh Novaes Mulligan, K., P. Simons, and B. Smith. 2006. What’s Wrong with Contemporary Philosophy? Topoi 25 (1–2): 63–67. Nietzsche, F. W. 2007. On the Genealogy of Morality. Cambridge: Cambridge University Press. Packer, M. 2010. The Methodology of Qualitative Research. Cambridge: Cambridge University Press. Panaccio, C. 2004. Ockham on Concepts. Burlington, VT: Ashgate. Plumwood, V. 1993. The Politics of Reason: Towards a Feminist Logic. Australasian Journal of Philosophy 71 (4): 436–462. Robinson, H. 2014. Substance. In E. Zalta, ed., Stanford Encyclopedia of Philosophy. http://plato.stanford.edu/entries/substance/Stanford Rorty, R. 1979. Philosophy and the Mirror of Nature. Princeton: Princeton University Press. Sachs, C. B. 2011. What Is To Be Overcome? Nietzsche, Carnap, and Modernism as the Overcoming of Metaphysic. History of Philosophy Quarterly 28 (3): 303–318. Srinivasan, A. 2011. Armchair v. Laboratory, review of Tamar Szabo Gendler’s Intuition, Imagination, and Philosophical Methodology, London Review of Books 33(18) (22 September): 17–18. Srinivasan, A. (ms.). The Archimedean Urge. Street, S. 2006. A Darwinian Dilemma for Realist Theories of Value. Philosophical Studies 127 (1): 109–166. Westphal, K. R. 2010–11. Analytic Philosophy and the Long Tail of Scientia: Hegel and the Historicity of Philosophy. The Owl of Minerva 42 (1–2): 1–18. Williams, B. 2002. Truth and Truthfulness: An Essay in Genealogy. Princeton, NJ: Princeton University Press. Williamson, T. 2014. How Did We Get Here from There? The Transformation of Analytic Philosophy. Belgrade Philosophical Annual 28: 7–37. Wittgenstein, L. 1998. Culture and Value. Chicago, IL: University of Chicago Press.

Part II

Truth and Meaning

This page intentionally left blank

5 Truth and Epoché The Semantic Conception of Truth in Phenomenology David Woodruff Smith

In the present essay, I explore a model of truth in intentional experience. I work within a broadly Husserlian model of intentionality, central to Husserl’s conception of phenomenology, while drawing in a broadly Tarskian model of truth, central to Tarski’s conception of semantics in logical theory. In this synthetic model, an experience is veridical, its content true, if and only if appropriate conditions in the world actually obtain. (We may consider, as well, truth conditions in alternative possible worlds.) In this way, we synthesize the patterns of semantic structure appraised by Tarski and Husserl, looking toward the foundations of phenomenology from a “semantic” point of view. Husserl’s method of epoché, integral to his conception of transcendental phenomenology, can then be seen as a practice of semantic ascent. Thus, we ascend from the intended object in the world—assuming veridical intentionality—to the semantic or noematic content through which the object is intended in consciousness. This practice involves moving in and out of lived conscious experience, moving from everyday experience into phenomenological—phenomeno-logical—reflection on the intentionality of that form of experience. This approach to the theory of truth in phenomenology offers a fresh perspective on truth itself. Is there a bona fide higher-order perspective on truth, or is the title “true” a redundant way of positing reality from within experience, as “deflationist” views would hold? And this approach to phenomenology via truth-theory offers a fresh perspective on the elusive methodology of epoché. We emphasize the role of ideal meaning in intentional acts of consciousness. And so we consider epoché as a “semantic” movement from experience in object-level consciousness to phenomenological reflection in meta-level consciousness, thereby abstracting meaning from experience. 1. TRUTH AND PHENOMENOLOGY: FROM HUSSERL TO TARSKI AND ONWARD How does truth play in phenomenology? It may seem obvious that truth per se is “bracketed” in the practice of phenomenological reflection via epoché.

112  David Woodruff Smith For, if the world is in brackets, truth itself is bracketed, as we turn from the external world to our consciousness as if of things in that world. However, that view is misleading, and for interesting reasons we’ll explore here. Indeed, the problem of truth weaves its way through the foundations of phenomenology, apparent in the historical development of the discipline. In 1933, the logician and mathematician Alfred Tarski developed a mathematical model of truth for a formalized symbolic language (Tarski 1933/1936). In 1944, in a more intuitive philosophical mode, Tarski published an article titled “The Semantic Conception of Truth and the Foundations of Semantics” (Tarski 1944/2001). Tarski’s work in model theory led the development of what is now called semantics in logical theory. Gottlob Frege’s early logic of Begriffsschrift remained a syntactic system, though supplemented later by his theory of Sinn and Bedeutung (as observed in Hintikka and Hintikka 1986). Tarski’s style of semantics defined the truth of a sentence in a model, where a model (for the truth of a sentence in a specified language) is a set-theoretic structure consisting of a domain of entities and a range of relations or properties defined on that domain. By the 1960s, models were seen as an abstraction of possible worlds (notably in Hintikka 1962, 1969), adding more philosophical substance to model-theoretic semantics. On this approach, a sentence is true in a world if and only if the world is as the sentence specifies, and the sense or meaning of the sentence is explicated in terms of its truth conditions, the condition(s) or situation or state of affairs that make it true in a given world. Thus did Tarski’s work lead inter alia into contemporary semantic theory. In a parallel vein, Donald Davidson argued (in Davidson 1967) that a Tarskian truth-definition (sans possible worlds, sans facts or states of affairs, and sans Fregean Sinn) yields all we need for a theory of meaning. And today it is a commonplace in philosophy of mind and perception that the content (or meaning) in a thought or perception is defined by appropriate truth or satisfaction conditions. Tarski was familiar with Edmund Husserl’s Logical Investigations of 1900–01 (Husserl 1900–01, revised 1920). Husserl’s prior work was in mathematics (the calculus of variations) and philosophy of arithmetic, but in the Investigations, Husserl developed the foundations of phenomenology, built around his theory of intentionality. In the Prolegomena to the Investigations, Husserl sketched a vision of logic that correlated three levels of categories: categories of thought or experience, categories of ideal meaning (Sinn), and categories of objects in the world (including states of affairs, Sachverhalten). On Husserl’s theory, language expresses thought or experience, which carries ideal meaning, which directs the experience toward an appropriate object in the world. And logic defines the correlations among those three levels of categories, that is, categories of thought or experience, meaning or content, and objects “intended.” Husserl’s model of these logical correlations—mathematical in spirit but expressed in philosophical prose rather than set-theoretical formulations—framed his conception of phenomenology. Thus, an experience with a certain form has a meaning-content of a

Truth and Epoché  113 certain form that directs the experience toward an object of a certain form: this structure defines the intentionality of the experience. And the theory of intentionality was central to Husserl’s conception of the new discipline of phenomenology. (The present perspective on Husserl’s progression from logic into phenomenology is detailed in Chapters Two and Three of Smith 2007 and in the 2nd ed., 2013.) In Ideas I (Husserl 1913/1981), Husserl amplified his conception of phenomenology as the “science of the essence of consciousness,” i.e., consciousness as experienced from the first-person perspective (see §§27–33). There, Husserl approached phenomenology via his famous method of epoché or “bracketing.” In order to focus on my consciousness of a given object, I am to suspend or bracket my assumption of the existence of that “intended” object in the surrounding world. In reflection, I then analyze the structure of that conscious experience and its intentionality, regardless of whether its “intended” object exists. There, in the experience, I find its ideal “noematic content,” or “noema,” a form of ideal sense (Sinn, the same term Frege used for the sense of an expression in a logically defined language). Husserl’s logical theory of intentionality is amplified as his mature phenomenological theory of consciousness centered on intentionality. (In Smith 2007, 2013 I have elaborated the ways in which Ideas I follows the path set out in the Logical Investigations.) Truth appears in Husserlian phenomenology, then, as the question of whether an act of consciousness is successfully directed via its noematic content toward an appropriate object or state of affairs in the world. And that question, I am saying, is brought to the fore through epoché. With a Tarskian twist, we may say, truth is articulated by the conditions under which the act’s content would be true: whether or not the appropriate object or state affairs actually exists—hence within the space of epoché. 2.  TRUTH AND INTENTIONALITY With intentionality in hand, I propose, we thus should see truth as consisting in a successful intentional relation, thus in the veracity of an intentional experience. An experience is intentionally directed, by virtue of its content or sense, toward an appropriate object in the world; and if successful, or veridical, it is intentionally related to that object. Characteristically, the paradigm of truth concerns the veracity of a thought or belief in representing something in the world, especially a state of affairs. An act of thinking that p is intentionally directed toward a state of affairs that p, and the thought that p is true if and only if the state of affairs that p actually exists. Let us note in passing how truth applies to other forms of intentionality. First, a belief that p, we should specify, is not an occurring act of consciousness, but rather an enduring subconscious state: a propositional attitude that grounds the disposition to consciously think that p. As an overt act of

114  David Woodruff Smith thinking may be intentionally related to an existing state of affairs, which renders the thought true, so the belief that p may be intentionally related to an existing state of affairs that p, which renders the belief true. Second, we note that visual perception is Husserl’s familiar paradigm. We may say a visual experience is veridical, or “true” to the world, if its content successfully or veridically “intends” an existing object or situation, whence the experience is in a successful intentional relation to that object, by virtue of the content in the experience. We’ll focus on the truth of an experience of thinking, but it is helpful to bear in mind the parallel case of veracity in perception. I shall assume that the content of an act of thinking that p is a proposition or propositional sense

. An act of thinking that p, then, has the content

: a propositional sense or Sinn in Husserl’s idiom, a thought or Gedanke in Frege’s idiom. In a Husserlian model, the “noematic” sense or content

, in an act of thinking that p, “intends” the (putative) state of affairs [p]. Alternatively, by virtue of its content the act “intends” that state of affairs, that is, in the world beyond consciousness per se. Some philosophers collapse “propositions” into putative states of affairs. But I’ll assume a sharp distinction between propositions and sates of affairs. Then: The proposition

is true if and only if the state of affairs [p] exists, i.e., there exists a corresponding state of affairs [p]. And accordingly: An act of thinking that p is true if and only if its content

is true, whence the act is intentionally related to an existing state of affairs [p]. In this model, what makes the act of thinking that p true, and what makes its content

true, is the way in which the state of affairs [p] satisfies the truth or veracity conditions formulated in the above biconditionals—that is, the satisfaction of the content

in the act by the existing state of affairs [p] “intended” by the proposition

where entertained in the act of thinking that p. Given this intentionalist perspective on truth, I propose, we should see the practice of epoché as pulling back from the mundane assumption of veracity, in order to focus on the structure of the experience of thinking—and thus to consider its truth conditions. To assess the veracity, the truth, of an experience of thinking, we must then reverse the bracketing of the question of the existence of the posited state of affairs. And so we determine, by whatever means we have, whether such a state of affairs actually exists or obtains. I’ll elaborate on this view of epoché as we move along.

Truth and Epoché  115 (On relevant details that I draw from Husserl, compare Smith 2013, 2nd edition, amplifying the 1st edition of 2007. On current approaches to truth, see Lynch 2001. On the intentionalist approach to truth, linking Husserlian and Tarskian frameworks, see Smith 2005.) 3.  CONDITIONS OF TRUTH A logical theory of truth in a Tarskian formulation is designed to articulate truth conditions for a sentence in a specified language. In Tarski’s famous constraint on a theory of truth, called Convention T: (T)  The sentence “p” in a language L is true in L if and only if p, i.e., it is the case that p. This condition is a schematic convention that Tarski proposed as a condition of the adequacy of a formal definition of truth for the given language L. On Tarski’s theory, the formulation (T) for a given sentence “p” is itself a sentence in a metalanguage M about the object language L. Thus, “ ‘p’ is true in L” is a sentence in M, and not a sentence in L. Separating the metalanguage from the base object language helps Tarski solve the problem of the liar paradox (“This sentence is false”). Here we simply note the two levels of language, in Tarski’s model. The details of Tarski’s theory of truth—historically innovative details in logical theory—involve specific conditions on the several basic forms of expression in a formally defined language L, notably his treatment of variables and quantifiers in a first-order quantified language. Those details lie beyond our current project. Tarski himself assumed an ontology of individuals and sets, a nominalist framework, avoiding the complications of “facts” (in Russell’s ontology) or “states of affairs” (in Husserl’s ontology). A traditional “correspondence” theory of truth might say the sentence “p” is true if and only if it represents or corresponds to and so is made true by the existing state of affairs that p. Tarski’s formulation of truth conditions avoided appeal to states of affairs or such a “correspondence.” At any rate, our study here goes beyond gospel Tarski as we draw in aspects of intentionality, and in due course we’ll allow a role for states of affairs as noted. Tarski offered his theory of truth as a foundation for formal semantics. As noted earlier, the title of his 1944 essay was “The Semantic Conception of Truth and the Foundations of Semantics.” In the 1930s, logical theory worked largely with syntax, as in Rudolf Carnap’s conception of “logical syntax,” but after the work of Tarski—and David Hilbert and Kurt Gödel—logic embraced semantics per se, mapping the ways in which expressions relate to things in the world. In fact, the terms “syntax” and “semantics” (and “pragmatics”) were not used in logical theory until the years after Tarski’s work, taking firm root in the 1940s, and leading through model theory into possible-worlds semantics by the 1960s.

116  David Woodruff Smith Whence, in modern-day possible-worlds semantics, conditions of truth for a sentence are defined relative to an appropriate type of possible world, i.e., an alternative situation or state of affairs, relevantly alternative to the actual world. Thus: (TW)  The sentence “p” in a language L is true in L in a world W if and only if in W it is the case that p. Truth simpliciter is then defined as truth in the actual world. Today the “semantic” conception of intentionality looks to the intentional relations of experience to things in the world. Without the idiom of “semantics,” however, in his Logical Investigations Husserl clearly envisioned a mapping of forms of experience and forms of “sense” onto forms of things in the world. Without the name “semantics,” Husserl launches the Investigations with a vision of the logical theory of how expressions and meanings signify things in the world, and from there Husserl launches his theory of intentionality, a theory of how experiences and their meaning-contents intend things in the world. (See Smith 2007, 2013 on this interpretation of Husserl’s development of phenomenology. See Smith and McIntyre 1982 on details of the correlation between logical semantics and intentionality theory, drawing on Jaakko Hintikka’s style of possible-worlds semantics for a language of perception or belief, especially in Hintikka 1969.) We’ll look to a model of truth for an act of thinking, as opposed to truth for specified forms of sentence in a given language. If you will, we turn to the “language” of intentional consciousness, i.e., the structure of intentional experiences bearing structured meaning-contents. Thus, I propose, we may consider the significance of a neo-Tarskian theory of truth conditions for an experience of thinking, assuming a broadly Husserlian theory of intentionality. 4.  TRUTH CONDITIONS FOR THINKING Looking to Tarski’s condition (T), assuming Husserl’s basic theory of intentionality, we formulate a similar neo-Husserlian condition (T*) on the structure of intentionality in thinking: (T*)  An act of thinking that p, with the noematic sense

, is true if and only if p, i.e., where the sense

designates the state of affairs [p]. A full theory of intentionality would appraise various forms of content or sense, and the patterns of intentional relation to appropriate things in the world, where such things actually exist. For our purposes, however, we

Truth and Epoché  117 focus on a simple form of thinking, say, where (in the first person) I think that “Husserl was Moravian.” For this act of thinking, then, the condition (T*) defines the form of truth conditions for this simple thought: My act of thinking that Husserl was Moravian, with the content , is true if and only if Husserl was Moravian. The full analysis of truth-conditional structure would take its place in a wider theory of intentionality, encompassing a rich variety of forms of intentional experience. For our purposes, however, we focus on the form of this simple act of thought, its content or sense, and the intentional relation of the act, by virtue of its sense, to an appropriate structure in the world, namely, a state of affairs consisting of the individual Husserl instantiating the property of being Moravian. A possible-worlds variation on (T*) brings out the intentional force of an act’s content (adapting the style of Hintikka 1969). Thus: My act of thinking that Husserl was Moravian, with the content , is true in a world W if and only if in W Husserl was Moravian. This style of truth-conditional analysis homes in on the noematic content of my act of thinking (see Smith and McIntyre 1982). For the case of truth simpliciter, however, we stay with truth in the actual world. Following this neo-Tarskian-Husserlian theory of truth, then, the structure of truth for this act of thinking is depicted in Figure 5.1. For the record, Husserl explicitly used quotation marks to abstract the noematic sense in an act of consciousness (see Ideas I, §89). As we refocus on the expression “Husserl was Moravian” and therewith on the sense , we ascend from the intended object of my thought to the content of my thought: that is, from the state of affairs [Husserl was Moravian] to the sense . This form of semantic ascent lays the foundation for the assimilation of Husserl’s theory of intentionality to semantic theory, including his own style in the Logical Investigations. (That assimilation is a running theme in Smith 2007, 2013.) When we look only to the Tarskian condition (T) or to the neo-Husserlian condition (T*), the statement of truth conditions looks utterly trivial. And, indeed, in a way it is. For in a simple case of presumably veridical thought, well, the thought simply “intends” and reaches out to the obvious candidate. Thus, my thinking that Husserl was Moravian truly reaches out to the situation or state of affairs that Husserl was indeed Moravian. What could be simpler? What more is there to say about truth?

118  David Woodruff Smith Proposition < Husserl was Moravian >

represents

[Husserl was Moravian] state of affairs

TRUTH

< Husser l was Mor avian> noema

[Husserl was Moravian]

act

WORLD

Figure 5.1 The Structure of Truth as a Successful Intentional Relation

5.  TRUTH AND THE PROBLEM OF REDUNDANCY Tarski’s convention (T) has led to an interesting opposition. On one view, the predicate “true” is simply redundant, as (T) purportedly shows. On that “deflationist” view, there is nothing more to be said: further questions, say, about the ontology of meanings and objects like “facts” or states of affairs—well, these issues are simply otiose. For the predicate “true” is simply redundant, on a basic deflationist view: to say “ ‘p’ is true” is just to say “p,” that is, with assertoric force. On another view, however, the predicate “true” is constrained by (T) in such a way that we may then expand our purview into ontology and, in the current enterprise, into phenomenology. “Semantics” in logic then appraises the relations among sentences and things in the world—and ideal meanings, if we go with the likes of Husserl and Frege. And the “semantic” model of intentionality, following Husserl in basic outlines, appraises the intentional relation between acts of

Truth and Epoché  119 consciousness and their appropriate objects in the world—with ideal meanings playing the role of designating said objects. Now, the intentionalist convention (T*) raises a problem parallel to the redundancy problem just noted about the truth conditions for a sentence. What is the difference between my thinking, assertively, that Husserl was Moravian and my thinking, reflectively, that my thought is true, i.e., that indeed Husserl was Moravian? Here we turn to the practice of epoché. In the mundane attitude, I think assertively that Husserl was Moravian, and in fact (I take it) he was. In a deflationist spirit, that is all there is to “truth”: as articulated in the truth conditions specified above for my act of so thinking. However, in the transcendental attitude, achieved by epoché, I reflectively observe that my act of so thinking “intends” the state of affairs that Husserl was Moravian, and so, bracketing the question of the existence of the “intended” state of affairs, I analyze the noematic sense (looking to its role in my experience, drawing on a horizon of background meaning). In this transcendental spirit, I move from the surrounding world into my consciousness directed toward the worldly state of affairs that Husserl was in fact Moravian. Here is the problem of truth and epoché: the movement, in reflection, between experience and the world experienced. (On the parallel between Tarskian and Husserlian views, see Smith 2005, 2013, pp. 261–262. On the role and status of ideal meaning in Husserl’s phenomenology, see Smith 2013, pp. 414–433.)

1. The Experience of Truth As we look into this neo-Tarskian, neo-Husserlian conception of truth conditions à la (T*) for intentional acts of thinking, I should like to cast a wider net, considering the form of truth for other forms of experience. When I judge that p, when I think that p with an assertive positing character, I have an experience of truth—from the first-person perspective. When I say “p,” in assertoric mode, I assert or posit that it is the case that p. And so far as my speech act goes, my assertion reaches out to the world and, from my first-person perspective, veridically represents the state of affairs that p. Similarly, in the absence of speech, when I think assertively that p, my thought-that-p intends the state of affairs that p. And so far as my thought goes, my consciousness reaches out to the world and, from my first-person perspective, intends the posited state of affairs—presumably veridically. At this level of positing reality, skeptical considerations do not arise. Of course, I may be wrong, but in the milieu of everyday evidence, I posit and so experience the state of affairs that p. So far as my intentional activity is concerned, from the first-person perspective, I experience truth, the veracity of my thought or judgment (or assertion). Husserl famously emphasized the sense of evidence, called “intuitive fullness,” thus “intuition” (Anschauung), that we may experience in perception

120  David Woodruff Smith and judgment. That first-person sense of evidence is part of the assertoric force of a typical visual experience or an assertive thought or judgment. To be clear, I am assuming that the successful intentional relation does not reduce to the intuitive or evidential character in experience. That is, truth does not reduce to the phenomenological character of seeming-veracity. Now, at a different level of intentionality, I step outside my thought or judgment that p, and in a higher-order thought I judge that my thought-thatp is true. In the spirit of Tarskian theory, my meta-judgment then explicitly posits the truth of my initial thought. Thus, in a basic experience of thinking, I think that p, and in a separate act of reflection, I think that my thought-that-p is true. This meta-thought I execute in reflection upon my original thought. Here is an explicit experience of truth: positing the success of the intentional relation between my thought-that-p and its intended object, the state of affairs that p. This reflective higher-order thought enters the realm of phenomenological reflection. We might say this reflective thought is executed from a higher-order first-person perspective, that is, where: I think that (the act wherein I think that p is true). The first “I  think” in this structure is at a higher level than the second “I think,” rather as if I am observing myself from a meta-first-personal perspective. That new perspective is not properly a third-person perspective, as it is still “I” at work, but my new, phenomenological perspective is a perspective now outside the basic experience wherein I think that p. That phenomenological perspective we’ll explore further as we proceed. At this point I want to emphasize that truth can be experienced also in a more direct way: in my positing the existence of the state of affairs that p, when I think assertively that p—in a circumstance of appropriate evidence with the presumption of knowledge. From a logical point of view in Tarski’s wake, we do not usually say we represent (or “experience”) truth until we ascend from the object language (saying “p”) to the metalanguage (saying “the sentence ‘p’ is true”). That is the Tarskian meta-logical perspective. Similarly, from a phenomenological point of view, we do not usually say we “intend” truth per se until we ascend from mundane experience to phenomenological reflection on everyday experience. That is the Husserlian “transcendental” perspective. From that reflective perspective, we can assess the evidence supporting my thought (or judgment) that p. This perspective is central in Husserl’s phenomenological theory of knowledge, in which a thought or judgment is

Truth and Epoché  121 supported or “fulfilled” through appropriate patterns of evidence (in perception, background beliefs, intersubjective confirmation, etc.). And if we bring in external circumstances, as in today’s externalist models of knowledge, well, then we move back and forth between internal and external support for a claim to the truth of my thought that p and, further, a claim to my having knowledge that p. (Our present concerns are with truth as opposed to knowledge; knowledge requires, in addition, appropriate evidence, circumstance, intersubjective collaboration, etc.) But now consider perception, in contrast with thought or judgment. In an everyday case of visual perception, I see that the sun is shining over the Pacific. Skeptical considerations do not apply; I simply look out at the ocean and see that the sun is shining. In this way, from my first-person perspective, I directly experience the sun’s shining: visually. In my experience of seeing that the sun is shining, I directly experience or “intend” the state of affairs that the sun is shining. So far as my experience goes, I thus experience the veridicality of my perception. We might even say I “feel” my perception’s successfully intending that state of affairs. In Husserl’s idiom, my visual experience is a fusion of visual sensation and meaningful interpretation (a fusion of sensory hyle and intentional morphe), and so my visual experience is a direct “intuition” of what I see. In visual intuition, then, I directly experience the veracity or “truthiness” of my visual experience of the sunshine over the Pacific. (For future generations: the term “truthiness” was coined by the American television artist Stephen Colbert. Truthiness does not itself guarantee truth, the latter-day Husserl notes.) In Husserl’s wake, seeking a new account of “transcendence” toward beings in the world, Heidegger characterized truth (Wahrheit, in German) as “unconcealment,” or “uncovering” what appears. Drawing on the etymology of the Greek aletheia, Heidegger sought a more primordial sense of truth than that of the correspondence of a proposition with a state of the world. On one level, Heidegger was seeking the basic ways in which we experience and deal with things in the world, prior to anything like the logician’s model of truth as correspondence or representation. The notion of truth as disclosedness comes alive, I believe, when we consider everyday perception. Heidegger himself did not emphasize the direct sense of reality in perception per se, as his focus was on the sense of things around us in, for example, hammering a nail. But hammering a nail is a complex activity of seeing and hammering the nail, using one’s hand to wield the hammer. In such an activity we are in direct contact with hammer and nail, directly experiencing that part of the world in which we dwell and perceive and act. (Cf. Wrathall 2011 on Heidegger’s conception of truth as “unconcealment”: see pp. 40–56 on how truth conditions play in a Heideggerian framework. I thank Paul Livingston for noting, in discussion, that direct perception could be understood as a form of unconcealment à la Heidegger.) As we noted, Husserl characterized perception as “intuition” (Anschauung): that is, a direct experience of something. A simple visual experience

122  David Woodruff Smith thus presents me immediately with the reality of (say) the nail I am hammering. In that way, I experience an immediate intentional relation to the nail and my hammering the nail. The success or veridicality of my visual and tactile and sensory-motor “intention” of the nail is not something I experience only as I step back and form the higher-order judgment that my perception is veridical (and my hammering successful). Rather, in the flow of everyday life, I immediately experience the veridicality of my perception of the nail (and the success of my hammering it). That is to say, the phenomenology of everyday perception is such that I experience truth—in the normal case where there is no real question of hallucination or systematic illusion. The structure of my perceptual experience is such that, to the best of my evidential situation, in my perception I am in a veridical intentional relation to what I perceive. The point of this excursus into the experience of truth is to draw us from the Tarskian logical model of truth to the Husserlian phenomenological model of truth. That is: assuming that “truthiness”—veridicality of intentionality—is a feature of our everyday experience, already familiar in a simple visual or tactile experience prior to reflection thereon. Again, in a neo-Husserlian vein, we may say that we experience truth—notably in “direct” perception—from a first-personal perspective. But we may also say, in that vein, that we experience truth in a further way—in phenomenological reflection—from a meta-first-personal perspective as we consider truth conditions per se. And then we may return to the world in search of further evidence for the satisfaction of an experience’s truth conditions.

2. Meaning and Truth from a Phenomenological Point of View Turning explicitly to epoché, I should like to sharpen the outlines of the way in which the theory of intentionality shapes the “semantic” approach to truth in experience. As Tarski assumed a semantic approach to logic in developing model theory per se, so Husserl had assumed a semantic approach to intentionality theory, an approach crucial to Husserl’s ever-widening conception of phenomenology. And in the “Prolegomena” to the Logical Investigations, Husserl explicitly outlined a vision of what became model-theoretic metalogic in Tarskian form. Thus, as noted earlier, Husserl outlined a system correlating categories of thought or experience, meaning, and objects—assuming the content or meaning of a thought is expressible in language. I hasten to remind us that the term “semantics” did not enter the vocabulary of contemporary logic (launched by Frege) until Tarski’s work. So we are tracing in hindsight the development of an approach to logic and to phenomenology that we today may call “semantic”—precisely because we are appraising meaning in experience. The fundamentals of Husserl’s theory of intentionality are laid out in his Logical Investigations, distinguishing act, content, and object of intentional experience. By the time of Ideas I, Husserl had found the method of

Truth and Epoché  123 bracketing, or epoché. There is much to be said about Husserl’s theory and practice of epoché, along neo-Kantian lines concerned with whether we ever reach a world beyond the “phenomena,” or things as they appear, things as they are “intended” in our consciousness. The key point I want to stress, however, is the way that epoché fits into Husserl’s theory of intentionality. Only within that theory do we explain what epoché does. By “bracketing” the question of the existence of the surrounding world, I turn my focus from the objects of my experience to my experiences themselves and, in particular, to the noematic contents or meanings entertained in my experiences. (Husserl introduces the method of bracketing, or epoché, in Ideas I, §§27–33; he details the turn to noematic contents then in §§89–91.) Husserl’s paradigm is an act of visual perception, but for present purposes we focus on an act of thinking. Suppose I think that Husserl was Moravian. In the practice of epoché I now bracket the question of the existence of what my thinking posits, i.e., the fact that Husserl was Moravian, the existing state of affairs that Husserl was Moravian. Where do I now stand, as it were? I have suspended the standpoint of the natural attitude, as Husserl put it. And that leaves me living in my act of thinking that Husserl was Moravian—but in abeyance of the actual fact of Husserl’s origins. As I reflect on my so thinking, I appreciate the content of my thinking, viz., the sense . I  understand that meaning, that propositional sense, and so in reflection I may formulate the conditions under which that proposition and so my thinking would be true. Following the convention (T*) as above, I say in transcendental phenomenological reflection: My act of thinking that Husserl was Moravian, with the content , is true if and only if Husserl was Moravian—i.e., in fact, in the actual world. This observation I make in the attitude of epoché. As the ancient skeptics said, I withhold judgment, but, Husserl says, I thereby reflect on my experience of so thinking. And the truth conditions above show the force of the content of my thinking. (Around 1907, Husserl took his “transcendental turn,” incorporating neo-Kantian terminology à la transcendental idealism and moving toward the methodology of epoché. See Mohanty 2008, 2011; Moran 2005; Smith 2007, 2013. Scholars are divided over whether this “turn” is a radical break with his prior realism. I see continuity in Husserl’s development.)

3. The World in and out of Brackets According to a common interpretation of Husserl’s practice of epoché, the technique of bracketing leads us into the “phenomena” of consciousness

124  David Woodruff Smith as we leave behind the “mundane” world. In a Kantian twist, we come to understand how the objects of our experience are shaped by the noematic meanings that inform our consciousness. And we thus enter a transcendentally enlightened sphere of consciousness, as we leave the noumenal world behind and concern ourselves only with the phenomenal world. Phenomenology is the science of consciousness, and with phenomenology we transcend familiar forms of naturalism (the world is but particles in the void) and historicism (the world is but a cultural artifact), coming to understand consciousness itself and its role in our dealings with “the world” qua phenomenon. Accordingly, in this view, the world is transformed into phenomena of consciousness, and a strong form of idealism fits naturally with that perspective. Still, a more nuanced version of the “phenomenal” reading of Husserlian epoché is possible. On my account, the practice of phenomenology via epoché presupposes certain ontological structures (e.g., parts-in-wholes, essences, manifolds, etc.). Alternatively, John Drummond holds, all of ontology is “reinscribed” in Husserl’s transcendental phenomenology. On this approach, the world enters phenomenological brackets, and we come to understand that the world—in all its ontological structure—just is the world as perceived and thought and dealt with. Importantly, this “reinscription” view of transcendental phenomenology is not committed to any form of subjective idealism. For, if I may put it so, things in the world are constituted partly by the properties they present in our consciousness of them, and those things are perfectly real—our consciousness does not bring them into existence. (See Smith 2013, Chapter 9 on discussion of this theme of reinscription.) Now, the semantic conception of intentionality—on the present construction—takes a different view of the significance of epoché. Bracketing the object of consciousness is a technique for turning attention to the act of consciousness and its content. On this view, epoché effects an ascent from the object of consciousness to the content of consciousness, that is, to the noematic sense or meaning entertained in consciousness. The truth conditions for an act of thinking, in accord with convention (T*), correlate the act’s content—an ideal noematic meaning—with the act’s “intended” object. That correlation, on this view, presupposes appropriate ontological structures: from the object intended to the ideal meaning to the act of consciousness within the stream of consciousness—and on to the successful intentional relation of veridicality, in which the intended object exists and satisfies the ideal content according to specified truth conditions. Husserl himself, we noted, emphasizes the “quotation” of noematic meaning. And he occasionally speaks of a zig-zag methology. The neo-Husserlian convention (T*) allows that with epoché we move from object back to the meaning through which we experience the object, and then, noting the truth conditions, we move back to the object. In our running example, following (T*), we note the relation between the proposition

Truth and Epoché  125 —the noematic sense of my thinking that Husserl was Moravian—and the “intended” and existing state of affairs [Husserl was Moravian]. In the practice of phenomenology, on this “semantic” approach, we move back and forth between everyday experience and phenomenological reflection. In the attitude of everyday experience, we posit the existence of various things in the world around us, in “the surrounding world of everyday life,” or “the life-world.” In the attitude of phenomenological reflection, we bracket the question of the existence of the things we typically posit in everyday experience, and thereby we come to reflect on the structure of our experience. In the practice of transcendental phenomenology, on this view, we move back and forth between the level of everyday “positing” experience and the level of “transcendental” reflection on the structure of everyday experience. We do not leave the surrounding world behind, entering into a realm of pure consciousness where the “real” world comes to appear as mere illusion. Rather, we step back from the concerns of everyday life, we step out of the attitude of everyday life, and now we reflect on the structure of everyday experiences of perception, thought, action, etc. And in this “transcendental” attitude we analyze the structure of intentionality, and in reflection, in analysis, we find that every intentional experience involves an appropriate noematic content or sense. Here we appraise consciousness and the range of meaning or sense characteristic of different types of experience of worldly things, as in seeing a sailboat gliding on the calm waters of the Pacific or thinking that Husserl was Moravian. Hence, in the practice of epoché, we come to adopt two related attitudes. We normally live in familiar types of experience; Husserl called this the “natural attitude.” In phenomenological reflection, however, we live in experiences directed toward the modes of everyday consciousness and their ranges of meaning or sense; Husserl called this the “transcendental-phenomenological attitude.” When we understand these two levels of experience, we may move back and forth between these levels, as we do in considering (T*) style truth conditions!

4. From Epoché to “Semantic” Levels of Consciousness Husserl felt that the method of epoché was his greatest achievement. But, from today’s perspective, what’s the big deal? What was so revolutionary? If we follow the “semantic” approach to intentionality in phenomenology, can we see what was so stunning to Husserl in his day? Mathematical logic began with Frege, Cantor, et al., but formal semantics did not take shape until Tarski. With an eye to semantics, Jaakko Hintikka has contrasted two views of logic and of language, drawing on a distinction articulated by Jean van Heejinoort. One view of logic assumes that modern symbolic logic (viz., first-order quantifier logic) is a universal medium, in which all theories can be formulated with due precision. The other view

126  David Woodruff Smith assumes that logic is a calculus, where one formal language can be interpreted within another. Frege held the first view as he put forth his Begriffsschrift. Tarski assumed the second view, where a given object language is interpreted within truth conditions formulated in a metalanguage. Hintikka generalized the opposition from logic to language in general. In one view, language is a universal medium: what I can express in language I can express in (especially) my mother tongue. Ludwig Wittgenstein became the champion of this sort of view, following Frege and Bertrand Russell in logic. (W. V. Quine held a similar view, where ordinary language is best “regimented” in accord with first-order quantifier logic.) In an alternative view, language is a “calculus”: what is expressible in one language can be expressed in another language, and one language can specify what another language says. This view of language inhabits the metalanguage style assumed by Tarski. And the model-theoretic style Tarski initiated evolved into the possible-worlds style of semantics developed by Hintikka and others. The philosophical significance of this opposition between language as universal medium and language as calculus is detailed by Merrill B. Hintikka and Jaakko Hintikka in their Investigating Wittgenstein (1986; see pp. 1–29). Logicians assuming the universalist stance were “semanticists without semantics,” as the Hintikkas put it. For these theorists, meaning is captured implicitly in “logical syntax” (Rudolf Carnap’s term) or philosophical “grammar” (Wittgenstein’s term). Following the Hintikkas’ lead, Martin Kusch took the universal/calculus opposition into phenomenology (in Kusch 1989). Kusch explicitly relates Husserl’s phenomenology with the meta-logical, model-theoretic tradition: Following [mathematician David] Hilbert, Husserl accepts early on the idea of metalogical proof. Central to his whole enterprise is the notion of acts that relate to one another as object- and meta-acts. . . . Husserl argues for the possibility of a transcendental language as a true metalanguage with respect to natural ordinary language. To safeguard this distinction, he makes use of a special quotation convention, noema-quotation, and he speaks of Sinne of different order. (Kusch 1989, p. 133, summarizing a lengthy analysis in pp. 11–134) Kusch argues at length that Husserl thus assumes a conception of language as a “calculus,” whereas Heidegger assumed a conception of language as a “universal medium.” In the calculus view, we can move from one language to another, where one language interprets another language, just as Tarski uses a metalanguage to specify truth conditions for an object-language. By contrast, in the universalist view, we are always working within our home language. That language is the only place where we understand meaning. For Heidegger, we know, “Language is the house of being,” and phenomenology’s concern is “the meaning of being.”

Truth and Epoché  127 Now I should like to move from logic to language to consciousness per se. On one view, consciousness is itself a universal medium of meaning. Perhaps Kant already assumed such a view, insofar as reason itself Kant held to be informed by a universal scheme of conceptual categories of the understanding, without which cognition would not be possible. By contrast, as modern logic was taking shape in Husserl’s day, the logical calculus scheme was coming into view. One domain of cognition or (better) consciousness features our everyday dealing with the world, in perception and thought and action. But in phenomenological reflection, another domain of cognition or consciousness takes shape. With the practice of epoché, we pull back from our mundane experience of the world in thought and perception and action, and we focus on the structure of those forms of everyday consciousness. And then, as we return to our everyday roles in the world, we appreciate the role of meaning in our experience. Now bring in the semantic perspective on consciousness itself. In reflection we practice “meta”-consciousness, reflecting on our own level of everyday consciousness, the “object”-level consciousness toward which our meta-level consciousness is directed. If logic requires a revolutionary model of metalanguage, envisioned by Tarski, then transcendental phenomenology requires a revolutionary model of meta-consciousness, envisioned by Husserl. Tarski’s truth-convention (T) divided formal languages into object-language and metalanguage in the formulation of truth conditions. And, in retrospect, Husserl’s theory of intentionality divided the domain of phenomenology—the field of consciousness—into the everyday object level and the transcendental noematic-meaning level. With epoché, the experience of meaning and truth—expressible in truth conditions—became apparent in consciousness itself. Thus, Husserl’s semantic conception of intentionality, by any other name, defined Husserl’s conception of phenomenology. And, accordingly, the neo-Husserlian convention (T*) articulates the semantic conception of truth as veridical intentionality. POSTSCRIPT The technical virtues of Tarski’s formulation of truth conditions include his treatment of the satisfaction of open sentences (with variables) and the recursive formation of conditions of truth for logically complex sentences, including occurrences of quantifiers (‘for some x,’ ‘for all x’) and connectives (‘and,’ ‘or,’ ‘not,’ ‘if . . . then’). We have focused here only on a simple predicative sentence such as ‘Husserl was Moravian,’ which would express the content in an act of thinking that Husserl was Moravian. How logically complex propositional contents are satisfied by appropriate entities in the world is a complex problem, with many proposals in the literature. I leave those issues aside here, in order to focus on a

128  David Woodruff Smith simple experience of thinking and its conditions of truth. In Smith (2005) I considered a radical proposal that might be drawn from Husserl, in which a complex structure called a “manifold’ might serve to make true a complex proposition expressible in a logically complex sentence. Enough said, here. WORKS CITED Davidson, D. 1967. Truth and Meaning. In M. P. Lynch, ed., The Nature of Truth: Classic and Contemporary Perspectives. Cambridge, MA: MIT Press. Originally published in Synthese 17: 304–323 (1967). Hintikka, J. 1962. Knowledge and Belief: an Introdution to the Logic of the Two Notions. Cornell U. Press. Hintikka, J. 1969. Models for Modalities. Dordrecht: Springer. Hintikka, M. B. and J. Hintikka. 1986. Investigating Wittgenstein. Oxford: Basil Blackwell. Husserl, E. 1913 [1981]. Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy: First Book. F. Kersten (trans.) Dordrecht: Kluwer. Originally published in German as Ideen zu einer reinen Phänomenologie und phänomenologishen Philosophie, I. Buch: Allgemeine Einführung in die reine Phänomenologie. (Halle: Max Niemeyer, 1913). Kusch, M. 1989. Language as Calculus vs. Language as Universal Medium: A Study in Husserl, Heidegger and Gadamer. Dordrecht: Springer. Lynch, M. P., ed. 2001. The Nature of Truth: Classic and Contemporary Perspectives. Cambridge, MA: MIT Press. Mohanty, J. N. 2008. The Philosophy of Edmund Husserl: A Historical Development. New Haven: Yale University Press. Mohanty, J. N. 2011. Edmund Husserl’s Freiburg Years: 1916–1938. New Haven: Yale University Press. Moran, D. 2005. Edmund Husserl: Founder of Phenomenology. Cambridge, MA: Polity Press. Smith, D. W. 2005. “Truth and Experience: Tarski vis-à-vis Husserl”. In M. E. Reicher and J. C. Marek, ed., Experience and Analysis. Erfahrung und Analyse. The Proceedings of the 27th International Wittgenstein Symposium. Vienna: öbv a hpt, 270–284. Smith, D.W. 2007. Husserl. Second Edition. London and New York: Routledge. Smith, D. W. 2013. Husserl. 2nd ed. (revised). London: Routledge. First edition, 2007. Smith, D.W. and R. McIntyre 1982. Husserl and Intentionality: A Study of Mind, Meaning, and Language (Dordrecht and Boston: D. Reidel Publishing Co., 1982). Tarski, A. 1933/1936, “The Concept of Truth in Formalized Languages,” in A. Tarski (ed.), Logic, Semantics, Metamathematics. Oxford University Press. 152--278 (1936) Tarski, A. 1944/2001. The Semantic Conception of Truth and the Foundations of Semantics. In M. P. Lynch, ed., The Nature of Truth: Classic and Contemporary Perspectives. Cambridge, MA: MIT Press, 331–364. First published in Philosophy and Phenomenological Research 4 (1944). Wrathall, M. 2011. Heidegger and Unconcealment: Truth, Language, and History: Philosophical Essays Volume 5. Oxford: Oxford University Press.

6 From Difference-Maker to ­Truthmaker (and Back) Jeffrey A. Bell

Metaphysical questions have increasingly become the focus of a number of philosophical debates among both analytic and continental philosophers. Among the theories that have given rise to metaphysical questions and debates is truthmaker theory. The following essay will explore the metaphysical implications of truthmaker theory and show how Deleuze’s metaphysical project provides for the realism that truthmaker theory calls for. This shift towards metaphysical questions as are found in truthmaker theory is somewhat surprising, given the dominance of the post-Kantian approach to philosophy one finds in both analytic and continental traditions and its attendant concern with deducing the conditions that constitute legitimate claims to know. This effort, to paint with a rather broad brush, is performed without concern for the nature of the reality that is known, or for the real that is independent of the conditions whereby the real comes to be known. Reality serves, as Peirce argued, as the object represented by the opinion we are all fated to agree upon at the end of the road of inquiry,1 a point Hilary Putnam reaffirms when he argues against “metaphysical realism.” Metaphysical realism becomes “incoherent,” Putnam claims, “at a particular point,” namely, “just at the point at which it claims to be distinguishable from Peircean realism.”2 In other words, to speak of reality as it is in itself will entail the best scientific theories and opinions we have settled on so far. To appeal to metaphysical realism and a reality distinct from that posited by Peircean realism, however, involves appealing to a reality independent of the opinions inquiring minds have settled upon, and thus it is incapable of being known or expressed in a way others may grasp or understand. Metaphysical realism is therefore incoherent. Within the continental tradition in philosophy one finds a similar tendency to dismiss the efforts of those who would investigate the nature of reality as it is in itself and independent of those who would come to know it. This approach also dates to the post-Kantian tradition and one can trace its lineage from Husserl through Heidegger to Derrida, and to the many other philosophers we have come to associate with continental phenomenology and post-structuralism. Despite Husserl’s claim to return to the things themselves, the real as it is in itself is for Husserl simply the correlate

130  Jeffrey A. Bell of the intentionality of consciousness itself, and the reality these intentional objects may correspond to is bracketed and set aside in what he calls an époche. Heidegger and Derrida will each, in their separate ways, continue in much this same vein and adopt a largely anti-realist stance that leaves aside metaphysical questions regarding the nature of reality.3 The recent turn to metaphysical questions regarding the nature of reality is subsequently a rather stark departure from the approach to philosophy that has been mainstream for decades, and it is a departure one finds in the work of both analytic and continental philosophers. Among analytic philosophers, for instance, David Lewis is unquestionably the most influential philosopher to approach traditional metaphysical questions, and his arguments regarding the reality of possible worlds gets applied to traditional metaphysical problems and arguments such as the problem of the one and the many, the ontological argument, and truthmaker theory.4 With respect to the latter, for instance, Lewis argues that “the most promising, if not the most prominent, among the grand theories of truth are the theories that somehow require what’s true to depend on the way the world of existing things is, or on the way some part of that world is.”5 The theory of truthmakers, therefore, opens the door for pursuing a metaphysics of the nature of these truthmakers, as Lewis does by arguing for them in terms of a possible worlds analysis.6 Hilary Putnam is skeptical of the truthmaker approach, and for reasons already mentioned—namely, it presupposes a reality that is “distinguishable from Peircean realism.”7 Other philosophers, as we will see, have accepted the challenge that truthmaker theory presents and have pursued the metaphysical questions regarding the nature of the truthmakers themselves. In continental philosophy, there has also been a return to metaphysics. After decades of pursuing a post-Kantian styled anti-realism, as found most notably in the Heideggerean and post-structuralist critique of the metaphysical project, metaphysical problems have come to be seen as legitimate, if not outright essential to the ongoing work of a number of contemporary philosophers. Quentin Meillassoux, for example, has argued that what should be of pressing concern to philosophers today is, among other things, to address the problem of correlationism, by which he means the view whereby it is understood that the nature of reality is always correlated with something else, whether this be a human subject (transcendental ego, Dasein, etc.) or a linguistic, textual, and vitalist reality. In each case, the real never appears as it is in itself but only as correlated with this other reality. Meillassoux’s problem, therefore, is metaphysical in that he sets out to determine whether and how one can come to know the absolute, that is, the real as it is in itself and independent of any correlation. Gilles Deleuze is similarly concerned with metaphysical questions. As Deleuze puts it in an interview shortly after he published his book on Leibniz, for him “[p]hilosophy is always a matter of inventing concepts. I’ve never been worried about going beyond metaphysics or any death of philosophy.”8 For Deleuze in particular,

From Difference-Maker to Truthmaker (and Back)  131 his concern was to invent concepts that enable us to think the real as difference in itself, or as difference-makers as we will discuss it here, and this serves as the principle of sufficient reason for the determinate realities that are the truthmakers that truthmaker theory calls for. Deleuze’s project is thus fundamentally metaphysical in nature, and it bears in important ways on truthmaker theory, as we will see. In the first section of this essay, we will turn to the arguments of Mulligan, Simons, and Smith in their now classic 1984 essay, “Truth-Makers.” We will focus more precisely on what is argued to be a key advantage of adopting a truthmaker theory—namely, the separation between ontology and logic that truthmaker theory presupposes. It is this separation that will provide the explanatory gap that is thought necessary to provide the constraints for our knowledge claims about the world, for without such a gap between propositions and the ontology that makes them true, we cannot say, so truthmaker theory argues, that our explanations extend to anything but other explanations rather than to the reality that is being explained. Moreover, we will extend Michael Morris’s argument that if this gap is to be up to the task and show how a truthmaker can provide proper constraints for the truth of propositions, then it must be a non-propositional reality. To motivate this move to a non-propositional reality, we will turn in the second section to discussing David Armstrong’s truthmaker theory. By arguing for states of affairs as the fundamental truthmakers of propositions, Armstrong believes he has provided a counter to the charge that truthmaker theory runs afoul of the Humean challenge that there should be no unexplained necessary connections. In the third section, we will show that Armstrong’s approach, if it is to be successful in avoiding the Humean challenge, must adopt what Jonathan Schaffer and Ross Cameron have called priority monism. The fundamental truthmaker, on this approach, is the “one totality fact.”9 This approach, however, does not provide sufficient restraint, for it too continues within the vein of the propositional model, and thus in the fourth section, following the lead of Jessica Wilson’s arguments for fundamental determinables, we will argue that Deleuze’s metaphysics of multiplicity offers us a non-propositional reality that successfully avoids the concerns of the Humean challenge and Putnam’s critique of metaphysical realism, concerns that will have been highlighted in this essay. I Early in their essay “Truth-Makers,” Mulligan, Simons, and Smith refer to Russell, Wittgenstein, and Husserl, in recognition that the truthmaker theory they are about to set forth can trace its roots to seminal figures in both analytic and continental traditions. What motivates their move to truthmaker theory, and what they find most enticing about it, is that while Tarksi’s work may have provided an important theory of truth, Tarski

132  Jeffrey A. Bell and subsequent defenders of Tarksi’s theory have largely abandoned “the complex and bewildering difficulties of the relations between language and the real world, turning instead to the investigation of more tractable set-theoretic surrogates.”10 In doing this, however, they have largely left the relationship between truth, language (or propositions), and the real world unexplained.11 Rather than attempt to grapple with issues such as the correspondence between propositions and any reality that may make such propositions true, the tendency instead, Mulligan et. al. argue, is to rephrase the problem in terms of set-theoretic language (à la Quine), with the resulting problem being a matter of establishing relations of quantification that are internal to the language itself, ignoring the sticky question of whether the language actually corresponds to the real world. It is at this point that appeal is made to the Peircean claim that reality is simply the object represented by the final opinion we come to settle upon. Truthmaker theory, by contrast, is thus seen by Mulligan et. al. as an effort to return to and address the sticky question regarding truth and reality, and they do so by taking on “a philosophically non-neutral correspondence theory” of truth (288-89). In short, in contrast to Tarksi’s approach, where “ ‘This cube is white’ is true iff this cube is white,” Mulligan et. al. will argue instead that If ‘This cube is white’ is true, then it is true in virtue of the being white (the whiteness) of this cube, and if no such whiteness exists, then ‘This cube is white’ is false.12 In other words, Mulligan et. al. will not account for truth in terms of a relationship between a metalanguage and an object language, as Tarski does, but will embrace a correspondence theory that calls upon something in reality—“the being white of this cube”—in virtue of which the propositions is made true. Key to this approach for Mulligan et. al., is the acceptance of a gap between the ontological and the logical. By calling for such a gap, Mulligan et. al. admit that they “are consciously departing from a dogma that has characterized much of analytic philosophy since its inception: the dogma of logical form.”13 What they are departing from, in short, is the notion that there is “a perfect parallelism of logical and ontological complexity,” such that “ontologically complex objects (those having proper parts)” must also in some way be “logically complex.”14 This is exemplified in Wittgenstein’s Tractatus when Wittgenstein argues that when we picture reality, our ability to do so presupposes that the picture itself, “of whatever form, must have in common with reality, in order to be able to depict it—correctly or incorrectly—in any way at all, is logical form, i.e. the form of reality.”15 Whenever we say something that is true, the truth of our proposition is true, for Wittgenstein, because the proposition successfully mirrors

From Difference-Maker to Truthmaker (and Back)  133 the logical form of reality itself.16 It is this mirroring that Mulligan et. al. deny, and hence they propose to “uphold the independence of ontological from logical complexity.”17 For example, Mulligan et. al. argue for the role of moments as truthmakers and argue that, given this theory, one may see Dick kick Harry, where such a kick is simply “their somewhat unusual way of greeting each other,”18 and account for the fact that the same ontological reality—that is, moment—may be false according to one logical form (that is, given one extensional relationship between the terms) and true given another. The only way to do this under traditional assumptions, they argue, would be by “creating a new extensional position for a term designating something (i.e., some moment) that is both a kick and a greeting, and this is to concede defeat,” for this move presupposes the independence of the moment as an ontological reality distinct from their logical descriptions and in virtue of which these descriptions are made true. Mulligan et. al.’s arguments, however, are unpersuasive, for they do not account for the explanatory gap between ontological moments and the logical structure of propositions that represent these moments, but rather they simply assume this gap. In his recent work, Michael Morris has addressed this issue and has extended Mulligan et. al.’s truthmaker strategy and has offered an argument that accounts for the gap between the ontological and the logical. For Morris, what this entails is that the truthmaker must be non-propositional, for with this move we have the explanatory gap between the ontological (non-propositional) and the logical (propositional) and we have the beginnings of an account of how the non-propositional makes propositions true. Guiding Morris’s account is what he sees as the challenge philosophers such as John McDowell raise in their attempt to detail that which serves as the rational constraint upon our experience so that our conceptual accounts of the world do not become a “frictionless spinning in a void.” As McDowell states the problem, We need to conceive this expansive spontaneity [of conceptual understanding] as subject to control from outside our thinking, on pain of representing the operations of spontaneity as a frictionless spinning in a void. The Given seems to supply that external control.19 The given, however, is not sufficient to provide the necessary friction, since McDowell accepts and extends Wilfrid Sellars’s “Myth of the Given” argument, which concludes, through Sellars’s “inconsistent triad” argument, that there is no such thing as non-conceptual content.20 What, then, is to provide for the rational constraint upon the conceptual formulations of our experience? This is the problem that Morris takes on board from McDowell, and for Morris the problem, as he puts it, is “to describe how the world, which is not in itself propositional, can constrain description, which is propositional.”21 In making this move, however, is Morris not simply resurrecting the Myth of the Given that McDowell, Sellars, and others had sought to

134  Jeffrey A. Bell avoid? For Morris, this will depend upon how the explanatory gap he calls for is understood. McDowell, however, cannot take the route of arguing that the world itself is a non-propositional constraint on experience because McDowell, Morris argues, is forced to concede that “there is no explanatory gap between conceptual world and conceptual experience,” and hence there is nothing other for our conceptual formulations to do than further detail the conceptual form of the world through our conceptual, propositional accounts. As McDowell has argued more recently, all our experiences, including our perceptual experiences, are “world-disclosing,” and what this entails is that “any aspect of its content hangs together with other aspects of its content in a unity of the sort Kant identifies as categorical.”22 In line with the analytic tradition, therefore, there is a parallel between our propositions and the “world-disclosing” experience that serves as the constraint that assures that our propositions will all hang together. The world and our propositions share the same logical form. In extending Mulligan et. al.’s arguments, Morris sets out to argue that, if there is to be an explanatory gap between our concepts and propositions and the reality that makes these propositions true, then this reality is non-propositional. The subsequent challenge for Morris will then be to address McDowell’s likely charge that he is left with an ineffable pointing, since how else can one describe the non-propositional except through propositions?! The reason for the difficulty people such as McDowell have with pursuing the line of thought, Morris proposes, is precisely the fact that they embrace the parallelism of logical and ontological form. Following Mulligan et. al., Morris drops this assumption, or what he calls the isomorphism thesis, whereby it is assumed that “sentences and facts have the same kind of structure (propositional structure).”23 Morris rejects this thesis and argues that in doing so one is not left with an ineffable pointing. To the contrary, Morris readily admits that a cup, for example, “can correctly be described as an object [but] describing it as an object is not describing it as it is in itself.”24 This claim appears to be contradictory, Morris argues, only when we confuse a metalinguistic claim with the nature of the world as it is in itself. In other words, within language a singular term refers to an individual or an object, and “for something to be an object,” Morris adds, “a property, a fact, or any of the rest, is always a (partially) linguistic matter.”25 It is simply a claim about language itself, or a metalinguistic claim, to say that the terms and words of a language refer to the objects, properties, and facts of the world, but for Morris this does not entail that the world itself consists of objects, properties, and facts. Far from it—for Morris the world is non-propositional, and any frustration we may have regarding Morris’s thesis is simply the result, he claims, of our “hope for a system of representation which is not a system of representation.”26 What Morris has not given us, however, is an account of how the non-propositional reality serves as the constraint in virtue of which our propositions become true propositions. He has shown that the call for a

From Difference-Maker to Truthmaker (and Back)  135 non-propositional world is not incoherent, but not how it relates to the truth of propositions. As we will see in section 4, Deleuze does provide such an account. To anticipate these arguments, as well as prepare for our discussion of David Armstrong’s truthmaker theory, we can say for the moment that what is critical to Deleuze’s account is his critique of representation. Despite Morris’s reaffirmation of Mulligan et. al’s break with the tendency among traditional analytic philosophy to accept a parallelism between the logical structure of our propositions and the ontological structure of the world as it is in itself, Morris continues to embrace a parallelism of identity between our descriptions and representations and that which is represented. To see what this entails for Morris, we can turn to the example he offers at the conclusion of his essay. In referring to a painting as an example of a non-propositional representation, Morris argues that: “Since painting does not use propositional structure, it does not represent things in terms of the traditional ontological categories.” The implication, however, is that the painting nonetheless accurately represents the non-propositional nature of the world, and it is precisely this process of representation that Deleuze rejects, for it presupposes the identity of that which is presented and given and then represented in the painting—namely, the identifiable objects with their extensive properties and qualities—and it is this identity that allows for the possibility of re-presentation. Identity, however, is for Deleuze what needs to be explained rather than presupposed, and it is the non-propositional nature of reality that provides this account. Before turning to Deleuze’s account, however, and in order to motivate it, let us turn first to Armstrong’s truthmaker theory. II Leaving aside for the moment the difficulties associated with explaining the relationship between a non-propositional truthmaking world and the propositions that describe this world, we can begin this section by noting that a central premise of truthmaker theory is that the truthmaker necessitates the truth of the proposition. Mulligan et. al. state this premise as the “first principle of truth-making”; namely, the principle that “what is made true is true.”27 Armstrong endorses this “first principle of truth-making,” and in response to questions he poses as to whether “truthmakers actually necessitate their truths, or is the relation weaker than that,” and whether “all truths have truthmakers,” Armstrong says yes to the first—“the relation is necessitation, absolute necessitation”—and yes to the second as well— “every truth has a truthmaker.”28 Armstrong will refer to these aspects of truthmaker theory as Truthmaker Necessitarianism and Truthmaker Maximalism. But it is at this point that problems quickly begin to emerge, for, as Armstrong admits, the truthmakers in virtue of which propositions are true are not always propostions themselves, and hence the problem is to

136  Jeffrey A. Bell explain how the truthmaker necessitates the truth of the proposition if the truthmaker is at least in part non-propositional.29 This brings us to a version of the problem we discussed in the previous section—how does the non-propositional world constrain and serve as truthmaker for descriptions of the world (propositions)? For Armstrong, the solution to this problem entails arguing for an “internal relation” between propositions and the truthmakers that make these propositions true, and by internal relation Armstrong means that “given just the terms of the relation, the relation between them is necessitated.”30 An internal relation is to be contrasted with an external relation, or a relation of supervenience whereby the relation is external to the terms.31 A further advantage of this approach, as we will see, is that by arguing for internal relations, Armstrong is able to avoid Hume’s skepticism concerning necessary relations. As Hume famously argued, there are no necessary connections between distinct elements but only a regularity of constant conjunctions. What carries the weight of Armstrong’s argument for the necessity of internal relations between truthmakers and the propositions these truthmakers make true are states of affairs. A state of affairs, such as “John is funny,” for instance, entails a relationship between a subject and a predicate, or involves a relation between a property and that which bears this property. The property of being funny, for example, is something we can truthfully say about John, and thus in describing a world that is non-propositional, to continue with Morris’s argument from the previous section, we are saying something about the world, attributing properties, objects, etc. to this world that, as we saw Morris argue, is in itself not a property, object, etc. What makes this possible? For Armstrong, it is the nature of a state of affair to entail a relation between a property and that which bears this property, including that which is non-propositional; consequently, it is a state of affairs, John’s being funny, that is the truthmaker for the proposition that John is funny. Central to this argument is Armstrong’s acknowledged break with Quine’s extensional approach, and with Quine’s claim that talk of predicates “gives us ‘ideology’ rather than ontology.”32 Armstrong, by contrast, argues that a “great advantage, as I see it, of the search for truthmakers is that it focuses not merely on the metaphysical implications of the subject terms of propositions but also on their predicates.”33 In other words, rather than follow the Quinean path of accepting only extensional relations between a quantifier and the range and scope of objects that fall under this quantifier, Armstrong adopts, with his use of states of affairs, an intensional relationship between a subject and its predicate. Armstrong is forthright on this point: “States of affairs have constituents (particulars, properties and relations) that seem to be parts of states of affairs, but not to be mereological parts.”34 That is, the constituents of a state of affair such as “a’s being F [e.g., John’s being funny] and a’s having R to b,” are not proper parts—they are not mereological (i.e., extensional) parts—even though they can, as Armstrong admits, “be truthmakers on their own just as much as the states of affairs that contain

From Difference-Maker to Truthmaker (and Back)  137 them.”35 John, for instance, can also be tall, handsome, and smart, and other people may also be funny. In serving as a truthmaker, however, the constituents of a state of affairs—a’s (John’s) being F (funny)—are in an internal relation with each other and are not extrinsically related. It is in this way, Armstrong believes, that he is able to circumvent Humean skepticism, for states of affairs do not presuppose the extrinsic relation between distinct elements that gets the Humean arguments off the ground. Is Armstrong’s strategy successful? Armstrong’s approach has some noteworthy skeptics. David Lewis, for instance, finds the whole notion of “ ‘unmereological composition’ profoundly mysterious,” and, even if one grants it, Lewis finds it incapable of explaining the idea of necessary connection.36 We can begin to understand Lewis’s reasons for doubting Armstrong’s approach to truthmakers by turning to one of Armstrong’s own examples—to wit, methane. As a mereological composition, we can say that methane, following Lewis, consists of four suitably bonded carbon atoms and one hydrogen atom. As Armstrong points out, however, the existence of these five atoms, suitably bonded, does not necessarily yield a methane molecule, for there is “the possibility of a larger molecule containing as a proper part the atoms that make up a methane molecule bonded just as they are within the methane molecule.”37 The butane molecule, for example, is composed of the same proper parts as the methane molecule, but it consists of more carbon and hydrogen atoms. How, then, do we differentiate the butane molecule from the methane molecule if we do not do so by way of a mereological composition of proper parts? Armstrong’s solution is to propose adding a totality fact to the composition that individuates methane and hence distinguishes it from butane. As Armstrong puts it, “a totality state of affairs is included in the concept of methane”; that is, a methane molecule “must contain five suitable atoms, four hydrogen and one carbon suitably bonded, and, furthermore, that must be all the atoms that it contains.”38 We can now make sense of Lewis’s complaint that Armstrong’s use of states of affairs has “in no way explained or excused”39 the necessary connection that characterizes the truthmaker relation. By incorporating a “totality state of affairs” into a’s (suitably bonded molecules) being F (methane), Armstrong has not overcome Hume’s skepticism regarding necessary relations between distinct, extensional terms; to the contrary, Armstrong has simply incorporated extensionality into the nature of states of affairs without explaining or excusing the necessary relation between the parts that make a “totality state of affairs” unique—e.g., methane rather than butane. In short, Armstrong is unable to account for the difference between states of affairs without incorporating the notion of a limit or totality fact into the very nature of these states of affairs. With this move, however, the Humean skepticism regarding the necessity of extensional relations resurfaces, for as states of affairs become incorporated into ever larger states of affairs, we end up presupposing, rather than explaining, the necessary relation between extensional terms as we continue to incorporate ever larger totality facts.

138  Jeffrey A. Bell At this point, I would argue that there are two paths one could take to address the concerns that have arisen thus far. The one path is to bite the bullet and argue that all states of affairs presuppose, and necessarily presuppose, a fundamental totality fact, or what Ross Cameron calls the “one totality fact.”40 In short, if one is to uphold truthmaker theory and avoid the Humean challenge, then one must embrace a form of priority monism whereby all distinct facts and states of affairs presuppose, as a logically prior precondition, the “one totality fact” that is logically prior. As we will see in the next section, however, priority monism may resolve the issues we have seen concerning Armstrong’s understanding of states of affairs as truthmakers, but it does not resolve our earlier problem concerning the effort to explain how a non-propositional truthmaker can make propositions true. Moreover, rather than accounting for or explaining the manner in which propositions represent and describe the non-propositional, it presupposes the identity of that which is described and given, an identity that allows for the possibility of re-presentation. For this reason, we will argue for a Deleuzean philosophy of difference and the metaphysics of an absolute, infinite substance that it presupposes, in order to provide the account of identity that will, in the end, explain how propositions are made true by a non-propositional reality. To further motivate this move, let us first turn to the arguments in favor of priority monism. III The person most associated with arguing for priority monism is Jonathan Schaffer. In a series of articles, Schaffer makes the case for accepting a neo-Aristotelian metaphysics in order to address a number of philosophical debates, and in his essay “The Least Discerning and Most Promiscuous Truthmaker,” he argues that if one is going to accept truthmaker theory, then one will be led, if one is to avoid Humean skepticism among other problems, to embrace priority monism. As Schaffer understands monism, it consists of there being one fundamental substance that is logically prior to distinct facts and states of affairs that are logically posterior to, and ontologically dependent upon, this one substance.41 “The substances,” or the one substance, as Schaffer will argue, “are the ground of being.”42 Schaffer is therefore not arguing for the existence of one thing, which is a “crazy view,”43 according to Schaffer, but rather he is arguing that the one substance is in a grounding relation to the beings that are posterior and hence “derivative, dependent, explicable, reducible, abundant, and secondary”44 to this substance. The truthmaking relation is an example of such a grounding relation “between substance and truth.”45 By developing this line of argument, Schaffer is able to employ priority monism to handle a number of problems that had plagued truthmaker theory, such as the truth of negative existentials. The problem with negative

From Difference-Maker to Truthmaker (and Back)  139 existentials is to account for what it is in the world that makes it the case that the negative existential is true—what is it that makes the proposition “there are no unicorns” true? If we adopt priority monism, however, Schaffer argues that we can do “without any negative facts at all”46 if we take the world as a whole, the one substance, as that upon which our truths are ontologically dependent. As a result, what makes it true that there are no unicorns is precisely the fact that the world, in its entirety, lacks unicorns. It is this one world, the one “fundament” as Schaffer puts it, that thus provides the ground for the truths of our propositions, including the truth of negative existentials, and the world also serves as the constraint (as Schaffer admits47) that allows us to avoid the frictionless spinning in the void. This does not mean that we may not be mistaken, or that a healthy Humean skepticism may not lead us to rethink the truth of our propositions, or even rethink what may or may not exist. For Schaffer, however, whatever propositions are taken to be true or false are made so by the final arbiter which is the one world, the one “totality fact.” But is this one world not simply the same world Peirce claimed is the ultimate cause of our all having the same opinions at the end of the road of enquiry? For Schaffer, it appears that it is, for as he concludes his essay he argues that “priority monism allows for the existence of you and me, tables and chairs, and pebbles and planets, inter alia . . . Priority monism only says that all these partialia are fragments of the cosmos, shards of an integrated whole.”48 Whatever claims we make along our way on the road of enquiry, therefore, are claims about the partialia of the world, and the truth of these claims depends, in the end, on the one real substance, the one “integrated whole.” But has priority monism helped us along our way to develop a metaphysical realism? With Schaffer’s arguments regarding the one “integrated whole” are we not simply back at a version of Putnam’s internal realism, with the difference between our propositions and the reality that makes these propositions true elided in favor of the former rather than the latter? One of the benefits of truthmaker theory, as we saw Mulligan et. al. argue, was that it calls for a gap between the ontological and the logical, or between the propositional and the non-propositional, as Morris developed this argument. With priority monism, it appears that we have lost sight of this gap, and although the reality of the one substance may be presupposed as a necessary condition for the dependent posteriors that are the primary subject of our propositions, it is nonetheless the priority relation of the world itself as a determinate totality fact to the determinate, individuated facts that serves as the truthmaker for our propositions, and hence the parallelism between the determinate ontological and logical continues to be maintained. Perhaps this is as it should be and Putnam is right that metaphysical realism is incoherent. We will argue that Putnam is not right, but first let us return to the overarching motivation for the turn to priority monism—namely, it sought to address Hume’s skeptical challenge regarding necessary relations between distinct terms.

140  Jeffrey A. Bell As we have seen, a basic premise of truthmaker theory is the necessitating relation between truthmakers and the propositions these truthmakers make true. If the relationship between components of truthmaker as state of affairs and that proposition that represents this state of affairs is understood, as Armstrong does, then the relationship between the components is necessary and intensional, and in this way we circumvent Humean skepticism regarding necessary extensional relations between distinct terms. Armstrong’s approach, we saw, was in the end unsuccessful unless one accepts priority monism. Ross Cameron makes the move to priority monism and does so precisely to address the Humean problem. The solution Cameron proposes to resolve the Humean problem is to rely on a benign form of necessary relation; otherwise, as Cameron puts it, “if the truth of a proposition necessitates the existence of a thing distinct from that truthmaker, as it will in the case of contingent intrinsic predications, then we have a necessary connection between distinct contingent existents,”49 and with this the Humean challenge becomes relevant. If we are going to say that the truth of the proposition that the ball is blue necessitates the existence of a blue ball, then this case of contingent intrinsic predication is necessitating that the relationship between distinct existents for the ball could have been green, red, etc., and there are other things that can be blue. As Kris McDaniel puts it, “Humeans love recombination. Provided that x and y are distinct contingent beings, x could exist in a world without y and vice versa.”50 The ball could exist in the world without being blue and there could be blue things without there being balls, but if the proposition “the ball is blue” is true then we have a necessary connection between these apparently distinct contingent existents. By arguing in favor of states of affairs as truthmakers, the correct insight Armstrong was developing, Cameron argues, was to make the necessary relation between distinct terms a case of benign necessity. Cameron gives the following example of a benign necessary relation: “For all possible things x, Necessarily, x exists if and only if the singleton of x exists.”51 That is, within the context of ZF set theory, no set is an element of itself and thus the element x and the unit set or singleton of x—{x}—are distinct; however, the relationship between them is necessary, as Cameron points out. This necessity is benign, moreover, and thus if the singleton {x} exists, then necessarily x exists, and this is precisely how Cameron understands truthmaker necessity: It’s just part of the meaning of talk about sets that we don’t let anything deserve the name ‘the singleton of x’ unless it has x as its sole member, and it’s part of the theory of truthmaking under offer that this can’t be the case unless x makes it true that that set exists.52 From here, Cameron follows the path to priority monism that we have sketched above—namely, to avoid incorporating extensional relations between distinct terms into states of affairs, what becomes necessary is that what serves as the truthmaker for propositions is one total state

From Difference-Maker to Truthmaker (and Back)  141 of affairs, the world or “the big totality fact,”53 as Cameron puts it. It is this “totality fact” that is the higher-order reality upon which the truth of propositions depends. The problem, however, as Cameron sees it, is that “a higher-order truthmaker must, of course, necessitate the existence of each of the first-order truthmakers, otherwise it’s not doing the job it was introduced to do.”54 If the higher-order truthmaker did not necessitate the existence of the lower-order or ontologically dependent truthmakers, then we would end up with unexplained necessary connections whenever there was a contingent claim made (e.g., the ball is blue) that did not depend upon the higher-order truthmaker. It is to avoid this consequence that one must accept priority monism. Cameron is explicit on this point: “So if No Unexplained Necessary Connections is true, the first-order truthmakers had better all exist in virtue of the higher-order truthmaker. But that’s to say that the first-order truthmakers are all ontologically dependent on the higher-order truthmaker; which is to say that there’s only one fundamental existent: the higher-order truthmaker.”55 The reliance of Cameron and Schaffer upon a notion of ontological dependence whereby the higher-order totality fact grounds the lower-order truthmakers is a view that is not without its critics. Jessica Wilson, for instance, argues that the grounding relation, “like supervenience, is too coarse-grained to do the work of appropriately characterizing metaphysical dependence on its own, failing to distinguish importantly different (eliminativist, reductionist, non-reductionist, emergentist) accounts of such dependence . . . ”56 It is too coarse-grained, for instance, to tell us how normative facts and states depend upon naturalistic facts and states, but it is precisely this relation that naturalists attempt to explain.57 In Wilson’s own account, she will argue for a mutual grounding of fundamental determinables—what Wilson calls the powers-based subset strategy58—that provides a better account of metaphysical dependence. Due to constraints of space, we cannot detail Wilson’s arguments, but what is relevant for our purposes here is Wilson’s claim that what is most problematic about most of the accounts that rely on the grounding relation is that what is taken to be fundamental is a determinate fact, the totality fact, for instance, whereas Wilson argues that what is fundamental are determinables.59 The advantage of this approach, Wilson claims, is that it provides an adequate account of the modal fact that a determinable may be multiply realized—for instance, the determinable red may be realized in the determinates scarlet, crimson, etc.—and this modal fact regarding multiply realizable determinables cannot be accounted for if we attempt to ground it on the basis of determinate facts, including the “big totality fact.” The problem with this latter approach, as Wilson states it, is that “grounding determinable modal facts in determinate facts [ultimately] requires appeal to a comparatively unnatural entity, presumably involving a higher conjunctive ‘world property’ corresponding to the global pattern at issue. By way of contrast, grounding the determinable modal facts in the determinable itself has the advantages of simplicity and naturalness.”60

142  Jeffrey A. Bell But has the move to fundamental determinables helped us with the Humean concern regarding necessary relations between distinct existents? While Wilson is not moved by the concerns other philosophers have for Hume’s skepticism regarding necessary connections between distinct existents,61 we do find the effort to account for the relation between the non-propositional truthmaking reality and the propositions this reality makes true to be an effort worth pursuing. Moreover, Armstrong, Schaffer, and Cameron find the effort worth pursuing as well, but each, in the end, departs from this effort when they call for a determinate totality fact or state of affairs to be the ground and fundament for the lower-order truthmakers and the determinate propositions they make true. With Wilson’s move to fundamental determinables, the determinates come to be accounted for in terms of determinables rather than determinables being grounded in determinate facts. This provides an opening for metaphysical indeterminacy, as Wilson herself admits,62 and yet the determinables themselves are metaphysically indeterminate only insofar as they are multiply realizable—that is, they are indeterminate precisely in that it is indeterminate which determinates the determinable provides the ground for. Indeterminacy is thus understood not as the ground or sufficient reason for identity itself but rather it is indeterminate precisely for the reason that there is a multiply realizable relation between the determinate entities and their determinable grounds.63 Yet as we have argued, it is precisely the nature of identity itself that needs to be accounted for if we are to explain the relationship between a non-propositional reality and the propositions this reality makes true. With his theory of difference-makers, as we will call them, Deleuze provides the account we are looking for. It is to this that we now turn. IV Central to Deleuze’s philosophical project is the claim that the extensive properties and qualities of things presuppose an intensive difference that the extensive properties and qualities mask or hide in what Deleuze calls a “true cover up.”64 Stated differently, the extensive properties and qualities that allow us to differentiate between distinct elements presuppose a difference-maker that is irreducible to these properties and qualities. As Deleuze puts it, It is not surprising that, strictly speaking, difference should be ‘inexplicable’. Difference is explicated, but in systems in which it tends to be cancelled . . . It [difference, or the difference-maker] is cancelled in so far as it is drawn outside itself, in extensity and in the quality which fills that extensity. However, difference creates both this extensity and this quality.65

From Difference-Maker to Truthmaker (and Back)  143 If we take the proposition “this ball is blue,” the truthmaker for this proposition is assumed, as we have seen, to be a determinate part of reality, even if a non-propositional part, that necessitates the truth of this proposition. For Deleuze, what accounts for the relationship between reality and the proposition this reality makes true is not the set of extensive relations and qualities shared by reality and the proposition that describes and is made true by this reality, but rather it is the difference-maker that creates the extensity and quality that individuates the reality and hence allows for the possibility of describing and representing this individuated reality by way of propositions. With Deleuze’s theory of difference-makers, therefore, we have a true gap between the ontological and the logical in that the difference-maker provides the sufficient reason for the individuated and distinct elements that become the subject of logical propositions, while the difference-maker itself, or difference-in-itself as Deleuze also puts it, is not to be confused with that which is described and represented by propositions. The difference-maker is “cancelled” and hidden, in a “true cover up,” by the individuated and determinate properties propositions represent. With Deleuze’s philosophy, as we will see, we have an argument for metaphysical indeterminacy that claims that the indeterminate is metaphysically fundamental and, moreover, fundamental in a way that does not subordinate the difference-maker to the individuated, determinate identities—that is, truthmakers—for which the difference-maker is the sufficient reason. The key move in Deleuze’s argument is the concept of multiplicity, multiplicity being understood by Deleuze to be a substance that is irreducible to extensive, determinate properties and qualities. By taking this move, Deleuze’s metaphysics is also able to avoid the Humean challenge against necessary relations between distinct existents, for such existents are not primary but presuppose the indeterminate substance or multiplicity as their sufficient reason. So, how does Deleuze arrive at the concept of a multiplicity? The first step is to set up an infinite series. For example, in Difference and Repetition Deleuze argues that “Every phenomenon refers to an inequality by which it is conditioned.”66 Every determinate phenomenon, in other words, or all that is given as sensible phenomena and then described and represented by way of propositions, is a composite or multiplicity, Deleuze claims, for it involves “a system [that] is constituted or bounded by at least two heterogeneous series . . . [of which] each is itself composed of heterogeneous terms, subtended by heterogeneous series which form so many sub-phenomena.”67 Deleuze refers to this process as intensity (i.e., difference-making) and intensity, Deleuze claims, “is the form of difference in so far as this is the reason of the sensible.” In clarifying this claim, he adds: Every intensity is differential, by itself a difference. Every intensity is E—E’, where E itself refers to an e—e’, and e to ε—ε’ etc.: each intensity is already a coupling (in which each element of the couple refers

144  Jeffrey A. Bell in turn to couples of elements of another order), thereby revealing the properly qualitative content of quantity. We call this state of infinitely doubled difference which resonates to infinity disparity. Disparity—in other words, difference or intensity (difference of intensity)—is the sufficient reason of all phenomena, the condition of that which appears.68 Deleuze returns to this point twenty-five years later in his book on Leibniz. Here Deleuze finds an ally in Leibniz’s philosophy as he reaffirms his claim that every determinate phenomenon presupposes an infinite series—an infinitesimal differential relation. For example, with “the color green,” Deleuze argues, “yellow and blue can surely be perceived, but if their perception vanishes by dint of progressive diminution, they enter into a differential relation (db/dy).”69 The color green, in other words, presupposes the differential relation of blue and yellow, db/dy, and yet, as Deleuze continues, “nothing impedes either yellow or blue, each on its own account, from being already determined by the differential relation of the two colors that we cannot detect.”70 And so on in infinitum. We can also turn to the trailblazing work of Jose Benardete, especially his use of what he calls the Zeno procedure, for further clarification of Deleuze’s argument.71 Benardete, moreover, draws upon the Zeno procedure in order to do much of what Deleuze sets out to do with his concept of multiplicity—that is, to set forth an infinite series or multiplicity that is neither reducible to an extensive point or atom from which other extensive properties and qualities are derived, nor is an extensive infinite series or substance.72 As Benardete sets up the Zeno procedure, he offers the example of a cylindrically shaped stick of wood with a Z written on the right-hand end of the stick. The Zeno procedure consists of dividing the stick in half infinitely—in other words, it consists of revealing the disparity that is “the sufficient reason of all phenomena, the condition of that which appears.” Benardete presents it as follows: Is it at all possible to sever the infinitesimal tail from the rest of the stick? The answer is, yes. Let us first cut the stick in half and inscribe the number 1 on the right-hand end of the butt-end of the left-hand piece. Now let us throw the left-hand piece into a box, leaving the right-hand piece on the table. Next, we shall cut this latter piece again in half, inscribing the number 2 on the right-hand butt-end of the new left-hand piece . . . Once more we shall cut the Z piece in half, &c. ad infinitum. At the end of the minute we shall have an infinite number of pieces in the box, each being of some rational length, and lying on the table there will be a single infinitesimal chip with the Z plainly marked on one side.73 Most significantly for our purposes, Benardete adds that, while we have the infinitesimal lying on the table, “there was no specifiable cut which severed the infinitesimal chip from the rest of the stick.”74 In other words, Benardete

From Difference-Maker to Truthmaker (and Back)  145 is arguing that “there is no actual point of separation (be it rational or irrational) between the infinitesimal tail [that is, the Z chip] of any closed continuum and the main body,” and yet, as his example is intended to show, we nonetheless have a separable infinitesimal. The conclusion to draw, according to Benardete, and Deleuze draws the same conclusion, I argue, is that the infinitesimal or difference-maker is not to be confused with the processes and procedures we identify with extensive magnitudes—e.g., with specifiable cuts and points of separation, or with extensive properties and qualities. The way to understand infinitesimals and difference-makers, therefore, is as the intensive magnitudes that are real but are not the extensive magnitudes of the stick or the specifiable, determinate cuts of the stick. It is this infinitesimal or difference-maker that offers us the alternative Deleuze himself seeks,75 and what we have sought as well in our effort to detail the relationship between a non-propositional reality and the propositions that are ultimately made true by this reality. We arrive at this alternative, for the infinitesimal is not a minimal extensive magnitude, an irreducible and determinate point that brings an infinite regress to an end (it is what is left over after an infinite regress and not the end point that is included within the regress itself); and the infinitesimal is obviously not an extensive continuum or totality of infinite steps, for the infinitesimal (the Z chip) is precisely what is eliminable from the stick after an infinite totality of cuts! Moreover, it is this infinitesimal difference-maker that is the non-propositional reality that is the sufficient reason for all determinate, identifiable existents that are the truthmakers of propositions. What Deleuze offers us with his theory of difference-makers is, I would argue, a way to give full weight to the gap between the ontological and the logical and to account for the necessity of the truthmaking relation without being reducible to the terms of that relation. A Deleuzean metaphysics of multiplicity and difference also offers an account of the restraint our conceptual formations require. In contrast to McDowell, Schaffer, and Cameron, however, this restraint is not a “big totality fact” or a “world-disclosing” experience; to the contrary, Deleuze offers a non-propositional restraint in the form of the difference-maker. The difference-maker serves as the restraint for our conceptual formations and propositions not by being the already constituted terms, qualities, and properties that delineate in advance what propositions will or will not be true in the end. For Deleuze, difference-makers provide the non-propositional restraint in that they are the conditions presupposed by the Peircean ideal theory, the very theory Putnam argued was the only coherent form of realism one could accept. The problem with metaphysical realism, as we saw Putnam argue, was that one cannot coherently argue for a “real” that is independent of and outside any logical, propositional description, for we can only know the real by way of such descriptions. Deleuze would agree. The difference-maker cannot be known outside a logical, propositional description and theory, nor is it known by such a theory, but it is the reality or process of heterogenesis (difference-making) that all

146  Jeffrey A. Bell individuated identities and phenomena presuppose—it is the “sufficient reason of all phenomena”—and hence the theories and descriptions that represent these phenomena and are made true by them. The difference-maker also guarantees the transformation of such identities and the emergence of new truthmakers and truths. The reason for this is that the difference-maker is multiply realizable, and thus is metaphysically indeterminate in the sense Wilson argued for, but this indeterminacy is not a matter of the determinate not yet being actualized, or not yet knowing which of the indeterminate ways it grounds the determinate; rather, the difference-maker is the indeterminate condition for all determinate phenomena, and for the process of further differentiation and transformation of phenomena. We thus move from metaphysically indeterminate difference-maker to determinate truthmaker in a process that covers up and cancels the difference-maker, and yet the difference-maker is inseparable from the truthmaker and, for this reason, we return, inevitably and necessarily, to the difference-maker again and again. There is no final, determinate point upon which all our truths will settle. There is no determinate fact or world that will, in the end, make for the final list of true propositions. There is, as Benardete has shown, only an intensive, infinitesimal difference; or, as Deleuze argues, there is only a series of intensive, infinitesimal differences, an infinitely doubled difference and disparity. It is difference all the way down. NOTES   1 See Peirce (1955), p. 38: “The opinion which is fated to be ultimately agreed to by all who investigate, is what we mean by the truth, and the object represented in this opinion is the real. That is the way I would explain reality.”   2 Putnam (1977), p. 489.   3 Heidegger, for instance, will, in his theory of Dasein, remain true to the Kantian theme of the limits of human knowledge, implying therefore that access to reality as it is in itself is always mediated by and through Dasein. Similarly for Derrida, his famous claim that “There is nothing outside of the text” ([il n’y a pas de hors-texte] in Derrida (1976, p. 158) in turn leads one to the natural conclusion that we are incapable of accessing reality as it is in itself but are fated to a textual mediation of reality. For more on the anti-realist legacy in continental philosophy, see Braver (2007).   4 For Lewis on the problem of the one and the many, see Lewis (1993); for the ontological argument, Lewis (1983), and for truthmaker theory, see Lewis (2003).   5 Lewis (2003), p. 603.   6 David Armstrong will argue that truthmaker theory provides an excellent window onto the metaphysics of a philosopher for it entails pointing to that which one takes to be the reality that makes our propositions true. See Armstrong (2004), p. 23.  7 As Putnam (1977) puts it, “If one cannot say how THE WORLD is theory-independently, then talk of all these theories as descriptions of THE WORLD is empty,” p. 492. In other words, if one cannot say anything about the truthmaker that makes a theory true independent of a theory, then the

From Difference-Maker to Truthmaker (and Back)  147 theory begs the question regarding the nature of reality in itself and is thus empty.   8 Deleuze (1977), p. 136.   9 Cameron (2010), p. 195. 10 Mulligan et. al. (1985), p. 288. 11 Mulligan et al. (1985), pp. 288–89: “Putnam has argued that Tarski’s theory of truth, through its very innocuousness, its eschewal of ‘undesirable’ notions, fails to determine the concept it was intended to capture.” 12 Ibid., p. 297. 13 Ibid., p. 298. 14 Ibid. 15 Wittgenstein (1998), 2.18. 16 Richard Rorty (1979) will also criticize this mirroring of the logical and ontological in his famous book Philosophy and the Mirror of Nature. 17 Mulligan et al. (1985), p. 298. 18 Ibid., p. 308. 19 McDowell (1996), p. 11. 20 Due to constraints of space, we cannot address the critique of the Myth of the Given here. For those unfamiliar with the arguments, the canonical reference is Wilfrid Sellars’s (1997) Empiricism and Philosophy of Mind. For Sellars’s “inconsistent triad” argument, see pp. 21–22. 21 Morris (2005), p. 61. 22 McDowell (2007), p. 346. 23 Morris (2005), p. 51. 24 Ibid., p. 63. 25 Ibid., p. 64. 26 Ibid., p. 65. 27 Mulligan et al. (1985), p. 313. 28 Armstrong (2004), p. 5. 29 As Armstrong (2004), pp. 5–6 recognizes, the principle of Truthmaker Necessitarianism cannot involve necessity in the logical sense of an entailment relation for “an entailment relation,” Armstrong argues, “must be [between] propositions, but the truthmaking term of the truthmaking relation is a portion of reality, and, in general at least, portions of reality are not propositions.” 30 Ibid., p. 9. 31 Armstrong correctly notes that this is Lewis’s approach. Lewis is a regularity theorist when it comes to causal relations, and this is due to the fact that he argues, along Humean lines, that causal relations are not real in themselves but are to be accounted for in terms of the regularities of distinct phenomena. It is the regularity and patterned relations of distinct phenomena upon which causal relations supervene. Armstrong, by contrast, argues for intrinsic causal relations, or states of affairs. 32 Ibid., p. 23. 33 Ibid. Armstrong is referring to Quine’s passage from Ways of Paradox (Quine 1976), p. 232, where Quine argues not for an ontology of predicates but rather for a relationship of quantification over sets of extrinsic objects: “quantifiers may be said to have been interpreted and understood only insofar as we have settled the range of their variables.” 34 Armstrong (2004), p. 19. 35 Ibid., p. 18. 36 Lewis (1991), p. 57. Lewis will later admit to understanding “unmereological composition” but adds the latter criticism in Lewis (2003), p. 613: “Somehow, this state of affairs is said to be constructed—neither mereologically nor set-theoretically—from its ‘constituents’ F and A. I used to complain that I didn’t know what this unmereological composition was. But now I think

148  Jeffrey A. Bell I  understand it well enough, because I  can define it in terms of necessary connections . . . My only remaining complaint is that this necessary connection between seemingly distinct existences has been in no way explained or excused.” 37 Armstrong (2004), p. 61. 38 Ibid., pp. 61–62. 39 Lewis (2003), p. 613. 40 Cameron (2010), p. 195. 41 Schaffer (2010), p. 309: “To begin with, I work within a neo-Aristotelian framework, which posits substances and posteriors related by ontological dependence.” 42 Ibid. 43 Ibid., p. 324. 44 Ibid. 45 Ibid., p. 310. 46 Ibid., p. 313. 47 Ibid., p. 315: “So I do agree that constraints are needed, but am offering a rival account of these constraints: do no require minimality, instead require only the world.” 48 Ibid., p. 324. 49 Cameron (2010), p. 180. 50 McDaniel (2009), pp. 252–253. 51 Cameron (2010), p. 184. 52 Ibid., p. 186. 53 Ibid., p. 195. 54 Ibid., p. 191. 55 Ibid., p. 192. 56 Wilson (2014), p. 6. 57 Ibid., p. 12: “Hence it is that naturalists almost never rest with the bare grounding claim, typically schematically expressed using the ‘nothing over and above’ locution, but rather go on to stake out different positions concerning how, exactly, the normative goings-on metaphysically depend on the naturalistic ones.” 58 See Wilson (2011). 59 See Wilson (2012). 60 Wilson (2012), p. 14. 61 Admittedly, for Wilson this is not a problem that warrants the concern it has received. See Wilson (forthcoming), where she argues that “Hume’s dictum,” as she puts it—to wit, the claim that “There are no metaphysically necessary connections between distinct, intrinsically typed entities” (p. 150)—is one of the three dogmas that unjustifiably guides metaphysical methodology today. Wilson claims that she has been “looking for good reasons to believe HD (Hume’s dictum), on the assumption that you aren’t Hume. So far, I haven’t found any.” (p. 152) In the next section we will provide a Deleuzean argument that does address the Humean problem but in a way that embraces Wilson’s call for a metaphysical indeterminacy. 62 See Wilson (2013). 63 This is especially clear in Barnes and Williams (2011). In this essay Barnes and Williams argue that “Naturally, if the actual world doesn’t contain any indeterminacy, then there will only be one world in the space of precisifications,” but they argue for metaphysical indeterminacy and this entails embracing the claim that something, p, “is indeterminate if some worlds in the space of precisifications represent that p and some represent that ¬p” (p.112). In other words, they argue that “When matters are metaphysically indeterminate, it is

From Difference-Maker to Truthmaker (and Back)  149 indeterminate which world is actualized,” (ibid.) and hence which world determinately is; and yet the determinate itself, identity, is presupposed and left unaccounted for. We continue to find the mirroring of the ontological and the logical. 64 Deleuze (1994), p. 166. 65 Ibid., p. 228 66 Ibid., p. 222. 67 Ibid. 68 Ibid. 69 Deleuze (1992), p. 88. 70 Ibid. 71 See Benardete (1964), whose work anticipates the more recent and widely shared moves by analytic philosophers to pursue metaphysical problems. 72 In Smith (2012), Smith argues for the claim that Deleuze adopts Spinoza’s view that “simple bodies are actually infinite,” and he does so for, “On the one hand, it says that there are indeed ultimate or final terms that can no longer be divided—thus it is against the indefinite; but on the other hand, it says that these ultimate terms go to infinity—thus they are not atoms but rather terms that are ‘infinitely small’, or as Newton would say ‘vanishing terms’ ” (p. 249). Similarly for Benardete, his use of the Zeno procedure is to avoid what he sees as the false metaphysical dilemma of there either being fundamental minima or atoms or a continuum of infinite regress. See Benardete (1964), p. 202. 73 Ibid, p. 272–274. John Hawthorne and Brian Weatherson (Hawthorne and Weatherson 2004) have recently addressed Benardete’s arguments as they relate to this example. In the context of discussing the supertask of cutting gunk to infinity, the problem they uncover is one that arises if we attempt to marry a theory of gunk (that is, an infinitely divisible substance) to the supertask of cutting a piece of matter that always occupies an extended volume of space. If the infinitesimal does not occupy an extended, specifiable volume, as Benardete argues, or if it is an intensive difference and differential as Deleuze will argue, then gunk is compatible with Benardete’s arguments. As they argue: “If we are allowed parts that lack finite extent, then it will be consistent to adopt Benardete’s picture of the outcome.” (Hawthorne and Weatherson 2004, p. 342, emphasis mine). I argue elsewhere that the Deleuzean interpretation of Benardete offered here ultimately rejects gunk. See Bell (forthcoming). 74 Ibid, p. 274. 75 Benardete will see his own argument as an effort to find a third alternative between the view of reality as constituted of basic minim that are irreducible extensive points and a continuum that is infinitely divisible (i.e., gunk).

WORKS CITED Armstrong, D. M. 2004. Truth and Truthmakers. Cambridge: Cambridge University Press. Barnes, E. and J. R. G. Williams. 2011. A Theory of Metaphysical Indeterminacy. In Karen Bennett and Dean Zimmerman, eds., Oxford Studies in Metaphysics. Oxford: Oxford University Press, 103–138. Bell, J. A. Forthcoming. Gilles Deleuze and Félix Guattari’s What Is Philosophy? A Critical Introduction and Guide. Edinburgh: Edinburgh University Press. Benardete, J. A. 1964. Infinity: An Essay on Metaphysics. Oxford: Oxford University Press. Braver, L. 2007. A Thing of This World: A History of Continental Anti-Realism. Evanston: Northwestern University Press.

150  Jeffrey A. Bell Cameron, R. 2010. From Humean Truthmaker Theory to Priority Monism. Noûs 44 (1): 178–198. Deleuze, G. 1977. Negotiations. Trans. Martin Joughin. New York: Columbia University Press. Deleuze, G. 1992. The Fold. Trans. Tom Conley. Minneapolis, MN: University of Minnesota Press. Deleuze, G. 1994. Difference and Repetition. Trans. Paul Patton. New York: Columbia University Press. Derrida, J. 1976. Of Grammatology. Trans. Gayatri Spivak. Baltimore: Johns Hopkins University Press. Hawthorne, J. and B. Weatherson. 2004. Chopping Up Gunk. The Monist 87 (3): 339–350. Lewis, D. 1983. Anselm and Actuality. In Philosophical Papers, Vol. 1. Oxford: Oxford University Press, 21–25. Lewis, D. 1991. Parts of Classes. New York: Basil Blackwell. Lewis, D. 1993. Many, but Almost One. In Ontology, Causality, and Mind: Essays on the Philosophy of D.M. Armstrong. Cambridge: Cambridge University Press, 23–38. Lewis, David. 2003. Truthmaking and Difference-Making. Noûs 35 (4): 602–615. McDaniel, Kris. 2009. Structure-Making. Australasian Journal of Philosophy 87 (2): 251–274. McDowell, John. 1996. Mind and World. Cambridge, MA: Harvard University Press. McDowell, John. 2007. What Myth? Inquiry 50 (4): 338–351. Morris, Michael. 2005. Realism Beyond Correspondence. In Helen Beebee and Julian Dodd, eds., Truthmakers: The Contemporary Debate. Oxford: Clarendon Press, 49–66. Mulligan, K., P. Simons, and B. Smith. 1985. Truth-Makers. Philosophy and Phenomenological Research 46 (3): 287–321. Peirce, C. S. 1955. Philosophical Writings of Peirce. New York: Dover. Putnam, H. 1977. Realism and Reason. In Proceedings and Addresses of the American Philosophical Association, 483–498. Quine, W. V. O. 1976. Ways of Paradox. Cambridge, MA: Harvard University Press. Rorty, R. 1979. Philosophy and the Mirror of Nature. Princeton, NJ: Princeton University Press. Schaffer, J. 2010. The Least Discerning and Most Promiscuous Truthmaker. The Philosophical Quarterly 60 (239): 307–324. Sellars, W. 1997. Empiricism and Philosophy of Mind. Cambridge, MA: Harvard University Press. Smith, D. 2012. The Conditions of the New. In Essays on Deleuze. Edinburgh: Edinburgh University Press, 235–255. Wilson, J. 2011. Non-Reductive Realization and the Powers-Based Subset Strategy. The Monist 94: 121–154. Wilson, J. 2012. Fundamental Determinables. Philosophers’ Imprint 12: 1–17. Wilson, J. 2013. A Determinable-Based Account of Metaphysical Indeterminacy. Inquiry 56 (4): 359–385. Wilson, J. 2014. No Work for a Theory of Grounding. Inquiry 57 (5–6): 535–579. Wilson, J. Forthcoming. Three dogmas of metaphysical methodlogy. In Hatthew Haug, ed., New Essays on Metaphysical Methodology. New York: Routledge. Wittgenstein, L. 1998. Tractatus Logico-Philosophicus. New York: Dover Publications.

7 Reasons, Epistemic Truth, and History Foucault’s Criticism of Putnam’s Anti-Realism Lee Braver Although Michel Foucault and Hilary Putnam may seem an unlikely pair with which to begin a dialogue between continental and analytic philosophy, they actually stake out startlingly similar positions. This confluence seems more natural once we note their parallel intellectual influences. Both locate the starting point of modern thought, including their own work, in Kant’s rejection of realism, i.e., the view that the world possesses an inherent, defined essence with the concomitant definition of knowledge as the mind passively copying this structure.1 This rejection shifts the topic of epistemology and metaphysics from the intrinsic organization of mind-independent reality to the organizing principles of experience and knowledge, thus founding what has become known in analytic circles as anti-realism.2 Foucault and Putnam inherit and modify this Kantian move in similar ways, the former under the influence of Nietzsche, Heidegger, and the structuralists; the latter under that of Quine, the later Wittgenstein, and the pragmatists. Although it would be absurd to suggest that these lines of development have no significant differences, their criticisms of and alterations to Kant’s thought run in parallel lines, as I will explain. Their shared origin in Kant is unsurprising given the fact that, as observed by Putnam’s former student and later editor James Conant, Kant is the most recent common figure to whom these two traditions [analytic and continental—LB] can trace themselves back. He represents the crossroads at which the history of Western philosophy branches. . . . One of Putnam’s motivations for returning to Kant, and for taking his philosophical bearings from Kant’s formulations of the traditional problems, would appear to be to heal this rift: to find a piece of nonaligned ground, somewhere within earshot of both sides.3 Kant’s position relative to the divide is unique; Anat Biletzki puts it nicely in “Bridging the Analytic-Continental Divide”: “in the beginning there was Kant.”4 As I have done on a larger scale in my first book, A Thing of This World: A History of Continental Anti-Realism, we can use Kant’s work to “commensurate” Putnam and Foucault’s divergent terms and methods since

152  Lee Braver his influence bequeathed both traditions the same, or at least overlapping, issues. Demonstrating that they are working on similar topics can enable dialogue and show how desirable such interactions would be, since each tradition can employ the distinctive insights and strengths they have developed to criticize and contribute to the other on topics both are interested in.5 I will show in the first part of my paper that when examined in this light, Putnam’s internal realism and Foucault’s genealogy turn out to be remarkably similar. Once commensurated, I will begin the exchange between the two in the second part of my paper, showing in particular how Foucault’s work offers a powerful criticism of Putnam’s thought. 1. KANTIAN COMMENSURATION Putnam describes his middle period—in which he holds the view he calls “internal realism”—as a “demythologized Kantianism,”6 i.e., Kant’s importantly correct insights purified of his unfortunate mistakes. Here is his diagnosis of Kant’s thought in a nutshell: even though Kant’s recognition that [1] our description of the world is shaped by our own conceptual choices, which are in turn shaped by our nature and interests, is marred and flawed—marred, on the one hand, by [2] the notion of the Ding an sich, and, on the other, by [3] the notion that our conceptual choices are fixed once and for all by some kind of thick transcendent structure of reason—it does seem to me that Kant made a decisive advance over all previous philosophers in [4] giving up the idea that any description of the world can be simply a copy of the world.7 Putnam applauds Kant’s [4] rejection of the correspondence theory of truth because it cannot survive the anti-realist insight that [1] our descriptions, thoughts, and experiences are always mediated by concepts. The fact that our concepts shape our experience of the world means that we can never simply reflect the world as it is, or compare our thoughts with the unadulterated original. But he rejects Kant’s commitment to [2] noumena and [3] a single conceptual scheme embedded in a “thick transcendent structure of reason,” two ideas that he considers incompatible with Kant’s other views. Internal realism is Putnam’s attempt to understand Kant better than he understood himself by preserving his essential insights while cutting away his errors.8 According to Putnam, a consistent Critical philosophy would never have accepted [2] noumena in the first place. By dispensing with that illicit notion, internal realism is the system that Kant should have had, similar to the way Hegel thought of his own idealism as an internal development and correction of Kant’s.

Reasons, Epistemic Truth, and History  153 What we can know—and this is the idea that Kant himself regarded as a kind of Copernican Revolution in philosophy—is never the thing in itself, but always the thing as represented. And [4] the representation is never a mere copy; [1] it always is a joint product of our interaction with the external world and the active powers of the mind. The world as we know it bears the stamp of our own conceptual activity. If this is right, then the question immediately arises: [~2] what sense does it make to talk of “things in themselves” at all? A world of unknowable “things in themselves” seems a world well lost. And thus most philosophy that has followed Kant in insisting on [1] our own conceptual contribution to constituting what we call “the world” has refused to follow Kant in [~2] postulating a world of Things in Themselves.9 Putnam raises two objections to noumena: first, his pragmatic intuitions lead him to doubt the meaningfulness of any belief in an entity that is in principle inaccessible. The second objection is more Wittgensteinian,10 viewing the significance of concepts as dependent on their role in our practices. It is these practices that determine what meaning concepts have—even, paradoxically, the concept of “things as they exist independently of our thoughts and practices.” Putnam writes, “we don’t have notions of the ‘existence’ of things or of the ‘truth’ of statements that are independent of . . . the procedures and practices that give sense to talk of ‘existence’ and ‘truth.’ ”11 A claim like “noumena exist” tries to break free of the realm of human thought to describe wholly extra-experiential reality, but it cannot escape the gravitational field of our practices without completely losing meaning. It is we who fix the meaning of “exist,” and who determine where and how to apply it, making it relative to our thoughts and practices. Kant accepts [4] correspondence when applied to phenomena, and Putnam’s internal realism allows a similar “trivial” correspondence once a conceptual scheme has constituted the relevant realm of objects and set up the way that signs relate to them. Signs do not intrinsically correspond to objects, independently of how those signs are employed and by whom. But a sign that is actually employed in a particular way by a particular community of users can correspond to particular objects within the conceptual scheme of those users. ‘Objects’ do not exist independently of conceptual schemes. We cut up the world into objects when we introduce one or another scheme of description. Since the objects and the signs are alike internal to the scheme of description, it is possible to say what matches what.12 Although Putnam considers his view to be a kind of realism in that he fully allows that objects exist, this is an internal realism because reality is always viewed from within or as organized by a particular conceptual scheme.13 Even the most basic concepts like “real” or “thing” are relative to specific

154  Lee Braver schemes, rendering all levels of reality phenomenally real but transcendentally ideal in Kantian terminology. The implications for truth are that “the metaphysical realist insists that a mysterious relation of ‘correspondence’ is what makes reference and truth possible; the internal realist, by contrast, is willing to think of reference as internal to ‘texts’ (or theories), provided we recognize that there are better and worse ‘texts.’ ‘Better’ and ‘worse’ may themselves depend on our historical situation and our purposes; there is no notion of a God’s-Eye View of Truth Here.”14 Despite his frequent ridiculing of Derrida,15 Putnam’s position actually agrees with Derrida’s (in)famous assertion that “there is nothing outside of the text.”16 In Putnam’s terms, this says that we cannot intelligibly discuss reality independently of conceptual schemes.17 Putnam rejects Kant’s [3] insistence on a single permanent set of experience-organizing concepts.18 Instead of seeking the one necessary set of concepts that would apply the same way everywhere at all times, Putnam embraces conceptual relativism: “the logical primitives themselves, and in particular the notions of object and existence, have a multitude of different uses rather than one absolute ‘meaning.’ ”19 A unique and necessary set of concepts is no longer tenable once the transcendental subjectivity Kant used to shelter his concepts from the flux of time has been dissolved into history by the loss of the noumenal realm. We must take the changing history of science seriously in the wake of Kuhn (Hegel performed a similar service for continentals). Our notions of rationality and of rational revisability are not fixed by some immutable book of rules, nor are they written into our transcendental natures, as Kant thought, for the very good reason that the whole idea of a transcendental nature, a nature that we have noumenally, apart from any way in which we can conceive of ourselves historically or biologically, is nonsensical. Since our notions of rationality and of rational revisability are the product of our all too limited experience and all too fallible biology, it is to be expected that even principles we regard as ‘a priori’, or ‘conceptual’, or whatever, will from time to time turn out to need revision.20 Following Quine, Putnam takes a pragmatic view of inquiry in which our goals and interests decide which conceptual scheme to employ in a particular situation, since they also determine what we would accept as a satisfying, relevant answer.21 With neither transcendental nor transcendent harbors to protect our concepts, they become as subject to change as everything else. Putnam takes the idea of the “contextually a priori”22 from Quine to describe this new conception. The contextually a priori is the set of theories that organize experience and knowledge, and which would require extraordinary counterevidence to be overthrown. These central views undergo transformations at

Reasons, Epistemic Truth, and History  155 irregular intervals in history, creating revolutions that introduce incommensurable concepts and modes of justification, as Kuhn shows. Sounding a bit like Deleuze, Putnam writes that, “an essential part of the ‘language games’ that we play in science, in morals, and in the law is the invention of new concepts; . . . new concepts carry in their wake the possibility of formulating new truths.”23 Putnam’s favorite example24 of this is the extraordinarily successful but deeply strange quantum mechanics, which shows “how with the development of knowledge our idea of what counts as even a possible knowledge claim, our idea of what counts as even a possible object, and our idea of what counts as even a possible property are all subject to change.”25 The concepts that, constituting experience, were themselves impregnable to revision by empirical data for Kant can now be re-formed by experience; a one-way determination has become a feedback loop. Turning to Foucault, it turns out that he too takes Kant’s rejection of realism as the essential starting point of modern philosophy.26 He writes of his own contentious relationship to philosophy that if he “is indeed perfectly at home in the philosophical tradition, it is within the critical tradition of Kant.”27 He means by this that he follows Kant’s rejection of knowledge as [4] discovering or copying independent reality, since “we should not imagine that the world presents us with a legible face, leaving us merely to decipher it.”28 Instead of naively trying to examine mind-independent reality, Foucault’s work pursues the anti-realist analysis of the [1] concepts that constitute objects. “What, in short, we wish to do is to dispense with ‘things’. . . . To substitute for the enigmatic treasure of ‘things’ anterior to discourse, the regular formation of objects that emerge only in discourse.”29 In a phrase reminiscent of Putnam’s contextual a priori, Foucault calls the fundamental concepts of a particular period its “historical a priori”30 in his early work. Like Kant’s forms of intuition and concepts of the understanding, they structure a specific “epistemological field”31 within which certain kinds of entities can appear and be known in certain ways, thereby forming the condition for the possibility of positive sciences. This a priori is what, in a given period, delimits in the totality of experience a field of knowledge, defines the mode of being of the objects that appear in that field, provides man’s everyday perception with theoretical powers, and defines the conditions in which he can sustain a discourse about things that is recognized to be true.32 Although prior to mundane experience and knowledge of entities because they are constitutive of them, Foucault agrees with Putnam against Kant in claiming that this a priori changes profoundly over the course of time: “it is, from beginning to end, historical.”33 He neatly summarizes his relationship to Kant here: “if the critical question is knowing ‘the general conditions for the subject to possess truth,’ the question I would like to pose is, rather, ‘to what specific and historically definable transformations has the subject

156  Lee Braver had to submit for there to be an injunction to speak the truth about the subject?’ ”34 Whereas Putnam generally remains at a fairly abstract level, only briefly discussing specific concepts embedded in historical events such as the advent of quantum mechanics, Foucault’s work is consumed with detailed analyses of the causes and effects of just these kinds of changes in the conceptions of what counts as reason, as knowledge, as normal, as ill, etc. at those points of transition when they are “stirring under our feet.”35 Similar to Putnam’s reaction, Foucault denounces the crisis in which we have been involved for so long, and which is constantly growing more serious: a crisis that concerns that transcendental reflexion with which philosophy since Kant has identified itself . . . which concerns an anthropological thought that orders all these questions around the question of man’s being, and allows us to avoid an analysis of practice . . . which, above all, concerns the status of the subject.36 Kant opened up the space for these questions by [1] denying realism and focusing instead on the conditions for the possibility of knowledge,37 but he immediately compromised it by [3] insisting on a unique conceptual scheme embedded in a permanent human nature.38 Although Kant may have awoken from a dogmatic slumber, Foucault says that he put modern thought into an “anthropological sleep,”39 which must now be broken. Foucault’s denial of transcendental subjectivity takes the form of declaring the “death of man”—a considerably more dramatic rhetorical style than Putnam’s (he is French, after all), but basically the same idea. He proclaims generally that “all my analyses are against the idea of universal necessities in human existence.”40 In a specific criticism of Kant’s Table of Categories, Foucault writes that his own work does not “enable us to draw up a table of our distinctive features, and . . . sketch out in advance the face that we will have in the future. . . . It deprives us of our continuities; it dissipates that temporal identity in which we are pleased to look at ourselves when we wish to exorcise the discontinuities of history.”41 Eliminating this commitment to an unchanging human nature lets philosophy study how concepts respond to empirical events, which constitutes the necessary step beyond Kant: “the void left by man’s disappearance . . . is nothing more, and nothing less, than the unfolding of a space in which it is once more possible to think.”42 Kant’s single set of forms of intuition and concepts of the understanding anchored in the transcendental subject melts into a proliferating succession of concepts we can only discover by studying their actual historical employment. Once noumena have been eliminated, reality can only be the phenomena,43 and these get constituted differently by the various concepts that form at different points in history. Without a noumenal remainder preserving an identity in the face of empirical fluctuation, Foucault is committed to claiming that reality itself fundamentally alters at those points of epochal

Reasons, Epistemic Truth, and History  157 conceptual revolution, when “everything that had been presented to view, now takes on a new mode of being.”44 The Order of Things—the book in which he first works out the implications of this view—analyzes the changes in the historical a priori of what we blithely call economics, biology, and philology over the last few centuries. Without the assumption of constant entities beneath our studies and experiences of them, alterations in the historical a priori amount to “mutations that suddenly decide that things are no longer perceived, describe[d], expressed, characterized, classified, and known in the same way, and that it is no longer wealth, living beings, and discourse that are presented to knowledge in the interstices of words or through their transparency, but beings radically different from them.”45 The subjects of the sciences of language, life, and nature fundamentally change with the concepts, which explains some of the shocking claims he is famous for, such as: before the end of the eighteenth century, man did not exist—any more than the potency of life, the fecundity of labour, or the historical density of language. He is a quite recent creature, which the demiurge of knowledge fabricated with its own hands less than two hundred years ago: but he has grown old so quickly that it has been only too easy to imagine that he had been waiting for thousands of years in the darkness for that moment of illumination in which he would finally be known.46 Each of these words—man, life, labour, and language—is being used here as a technical term for a specific epochal instantiation of what we usually think of as an unchanging cross-epochal object. Foucault’s claim is that they fundamentally change with each new episteme, so much so that it is misleading to speak of the same object’s persistence across epistemic shifts. Although the objects we find in our present episteme seem mind-independent and permanent, this very appearance depends on an episteme: “it is only in the blank space of this grid that order manifests itself in depth as though already there, waiting in silence for the moment of its expression.”47 Metaphysics as the study of the unchanging constituents of reality gives way to what Foucault calls “historical ontology,” which examines the phenomena that change along with the historical concepts. “The history of the ‘objectification’ of those elements which historians consider as objectively given (if I dare put it thus: of the objectification of objectivities), this is the sort of circle I want to try and investigate.”48 Thus, on both the side of the subject and object (to use a distinction he wants to undermine), Foucault believes that his “essential task was to free the history of thought from its subjection to transcendence.”49 In his middle or genealogical period, Foucault expands his analyses to show how constitutive concepts are structured and maintained by complex webs of power relations, institutions, and practices in addition to the discourses that had been the focus of his earlier archaeological work. He

158  Lee Braver sometimes uses the term “apparatus” to refer to this web of words, practices, and relations that “crystallize into institutions, . . . inform individual behaviour, . . . act as grids for the perception and evaluation of things.”50 Lacking transcendental protection, concepts are permeated by the forces of the world like everything else; without a metaphysical realm to provide the invidious comparison, the entities formed by these shifting concepts are simply reality itself, a view he inherited from Heidegger’s phenomenological ontology.51 2. INTERNAL CRITICISMS ACROSS THE DIVIDE We are now in a position to see how these thinkers are engaged in similar projects. Both start from the rejection of what Putnam famously calls the God’s-eye view of reality. The very idea of a world in-itself wholly independent of our experience and organizing concepts is either incoherent or wholly irrelevant to our epistemology and metaphysics. More than just useless for practical purposes, claims about such a realm cannot connect with the practices that are the source of all sense; both thinkers would agree with Wittgenstein’s dictum that “a wheel that can be turned though nothing else moves with it, is not part of the mechanism.”52 I want to argue in this part of the paper that, while both accept this far-reaching idea, Foucault works out its implications more thoroughly and more consistently than Putnam. Instead of forming an external criticism that can be shrugged off, my framing of the issue presents Foucault’s insights as internal criticisms of Putnam: these are problems that Putnam ought to acknowledge by his own lights, making the engagement far more interesting and important. If all vestiges of noumena have been eliminated, then, on the one hand, reality is exhausted by what we can in principle experience and, on the other hand, truth can only be about this reality. There is no other world and no other truth. This is what Foucault means by his epigrammatic statement, “truth is a thing of this world,” and Putnam strongly endorses this view, at least during the time period we are examining, by making truth epistemic.53 This means that truth is explicated entirely in terms of ideas and processes that are at least in principle within our reach. Now, if truth can only be about those entities constituted by the contextual or historical concepts formed within empirical existence, Foucault draws the conclusion that whatever affects the formation of these concepts and justificatory practices plays a role in determining reality and truth, and so has a rightful place in ontology and epistemology. In particular, what he calls power influences the formation of concepts—“the processes of objectification originate in the very tactics of power and of the arrangement of its exercise”54—and so must be acknowledged and analyzed instead of ignored or castigated because it shouldn’t wield any influence. Losing the God’s-eye view means that we can no longer take up the position, claimed

Reasons, Epistemic Truth, and History  159 by thinkers from Plato to Habermas, that would allow us to sift through actual epistemic practices from the outside, calling some (those we like) free, truth-producing, or ideal, while dismissing others (those we don’t like) as distortive, coercive, or falsifying. We have lost the external position from which we can make this distinction, since all sorting takes place within, and is affected by, this ever-changing process. In other words, Foucault’s claim isn’t just that truth is about this world but that it itself is an occurrence within this world, and thus cannot escape the “profusion of entangled events” that make up history.55 We have no access to a truth that would transcend this all-too-human realm; all we have are the shifting claims and practices embedded in institutions that are formed by, among other things, power effects. “Truth itself forms part of the history of discourse and is like an effect internal to a discourse or a practice.”56 Instead of seeking ideal de jure truth or epistemological practice, Foucault uncompromisingly insists on empirically analyzing the de facto. What I mean by the term [historical a priori] is an a priori that is not a condition of validity for judgements, but a condition of reality for statements. It is not a question of rediscovering what might legitimize an assertion, but of freeing the conditions of emergence of statements . . . the a priori of a history that is given, since it is that of things actually said. . . . It has to take account of the fact that discourse has not only a meaning or a truth, but a history, and a specific history that does not refer it back to the laws of an alien development.57 Foucault studies what has actually been written in order to discover the pattern that has been followed without appealing to what should have been said or done. Nothing that actually influences the formation of concepts can be alien to the true practices of knowledge. If it forms a part of the process, then it’s as legitimate as any other part. These views are summed up in Foucault’s famous phrase, “power/knowledge.” The conjunction does not mean that the two are identical; if that’s all he meant, he would hardly need entire books to express it. Nor does it mean that power corrupts knowledge, since corruption requires a pure knowledge that then got sullied. Thinking of power as corruptive or repressive presupposes a realist conception of things’ innocent preexistence, “the idea that underneath power with its acts of violence and its artifice we should be able to recuperate things themselves in their primitive vivacity.”58 Without the God’s-eye view positing things independent of the unclean process of everyday existence, power-events infuse everything from start to finish. “Power/knowledge” indicates that the two cannot be extricated from one another.59 In two interrelated moves, Foucault reconceives power as productive rather than repressive and knowledge as an effect of empirical events that always include power-effects: “we must cease once and for all to describe the effects of power in negative terms. . . . In fact, power produces;

160  Lee Braver it produces reality; it produces domains of objects.”60 Without a pure truth to provide an invidious comparison, the involvement of power in the production of truth does not invalidate it; it is only a Platonic hangover that renders us suspicious. The fact that a person’s perspective and social position influence the epistemological status of their utterances “does not mean that what the person says is not true, which is what most people believe. When you tell people that there may be a relationship between truth and power, they say: ‘So it isn’t truth after all!’ ”61 If truth is not correspondence to mind-independent reality but instead is epistemic, then we have no sense of “really” true that is fully extricated from what is accepted as true, and this is inevitably implicated in the play of power. My argument is that this position is the logical consequence of the elimination of noumena. Without a transcendental safe-house, power partially determines concepts, which then constitute the objects that get known, since, as a good anti-realist, Foucault does not believe that objects simply exist on their own, pre-sorted into knowable sets: “knowledge follows the advances of power, discovering new objects of knowledge over all the surfaces on which power is exercised.”62 If we have given up the realm of ideal, pure reality and, in David Wiggins’s term, “self-identifying objects,” we cannot call this power-infused world corrupt or false or fallen; it is the place where truth resides. By taking these power effects into account, Foucault’s epistemology more consistently applies the rejection of noumena and the God’s-eye view than Putnam, who still seeks the good and right epistemological processes. It is important to Putnam to preserve the “realist intuition”63 that truth is incorrigible; he wants to escape relativism and any view of truth that makes it temporal and changing, though without resorting to realism’s linking beliefs to an extra-experiential reality. His solution is to distinguish between, on the one hand, “justified” claims that are indexed to a particular time and conceptual system and that can be overturned and “true” claims, on the other hand, which are justified under ideal epistemological conditions and thus eternally correct.64 This distinction allows him to maintain that truth is epistemic or conceptually bound to human justificatory practices but not to any particular practice, so that we can always say that “p is justified now, but it may still turn out to be false.”65 Like others, I  find this compromise unsatisfactory on both counts. On the side of preserving the epistemic status of truth, it ends up shrinking the connection that truth bears to our actual justificatory conditions down to an idealized future point about which Putnam can only offer unhelpful platitudes;66 this constitutes an attenuation to a virtually noumenal point. If this ideal state cannot have any effect on, and possesses no knowable relationship to, how we know things here and now, then the same pragmatic intuitions that lead Putnam to reject noumena should also eliminate these. Furthermore, as he realizes in his 1994 Dewey Lectures, this conception of truth as what is true under idealized or “sufficiently good epistemic circumstances”

Reasons, Epistemic Truth, and History  161 does not actually avoid a realist reliance on mind-independent reality; it simply pushes its role back one step from “reality directly determining whether p is true” to “reality determining whether the conditions are good enough for me to know whether p is true.” As he states, “the world was allowed to determine whether I actually am in a sufficiently good epistemic situation or whether I only seem to myself to be in one—thus retaining an important idea from common-sense realism.”67 As for the other side of Putnam’s balancing act, the encounter with Foucaultian anti-realism raises a glaring question: why work so hard to retain this “realist intuition” about a timeless truth in the first place? John McCumber has defined the central disagreement between analytic and continental philosophers in terms of their attitudes towards time: continentals embrace it and consider its effects ubiquitous, while analytics seem, one could almost say, allergic to it.68 In many cases, this distinction amounts to a neutral description or at most an external criticism of analytic philosophy, but if Putnam has given up the God’s-eye view and made truth epistemic, if it’s just us humans down here endlessly squabbling, then how can truth be exempt from time and history?69 Continental philosophers generally deny deeply temporal beings like us the right to anything eternal, even one eternally deferred like Putnam’s idealized truth conditions. I’m arguing that Foucault’s epistemology is more consistent in holding that without noumena or the God’s-eye view, truth cannot but be temporal and worldly through and through. I think he would read Putnam’s commitment to this “realist intuition” in terms of Nietzsche’s diagnosis: “God is dead; but given the way of men, there may still be caves for thousands of years in which his shadow will be shown.—And we—we still have to vanquish his shadow.”70 On this reading, Putnam is still fighting with the shadows of the God’s-eye view. Instead of looking for ideal conditions, Foucault studies the de facto historical justificatory practices that have actually occurred, believing that truth cannot be located anywhere else than within actual knowledges or scientific practices. One isn’t assessing things in terms of an absolute against which they could be evaluated as constituting more or less perfect forms of rationality, but rather examining how forms of rationality inscribe themselves in practices or systems of practices, and what role they play within them. . . . I would like, in short, to resituate the production of true and false at the heart of historical analysis and political critique.71 This shows how Putnam misunderstands Foucault when he accuses him of treating earlier periods and practices as irrational.72 ­Foucault describes this kind of reaction when he says that “one has often tried to ­blackmail all criti­ cism of reason and every critical test of the history of r­ ationality so that one either recognizes reason or casts it into irrationalism.”73 His purpose is to show that there have been many different rationalities in history—an idea

162  Lee Braver Putnam is committed to, as we saw above—but evaluation of them would require a single authoritative form of reason. There is no position outside of them from which to render a final, decisive verdict. Throughout, Foucault tries to show that practices that look irrational to our eyes are actually operating according to a thoroughly organized “logic” of their own that he does not hesitate to call reason, or rather, a reason. To say once and for all what is rational and what is not requires the position of a perfect reason that escapes history and power to judge all other ways of thinking from the God’s-eye perspective. Foucault’s analyses appear to expose irrationality only if one is operating with a conception of what constitutes true and proper reason that is defined once and for all. He himself accepts all of these instantiations as forms of reason: I am trying to analyze the forms of rationality: various proofs, various formulations, various modifications by which rationalities educe each other, contradict one another, chase each other away, without one therefore being able to designate a moment in which reason would have lost its basic design or changed from rationalism to irrationalism.74 Given Putnam’s own claims that advances in science alter “our idea of what counts as even a possible knowledge,” that “new concepts carry in their wake the possibility of formulating new truths,” that “talk of ‘existence’ and ‘truth’ [can only have meaning] within” a conceptual scheme, and “that our notions of rationality evolve in history,” Foucault’s position can be considered a demythologized Putnam, i.e., a more consistent version of Putnam’s own views. Furthermore, expanding epistemology to include power effects gives Foucault’s epistemology greater explanatory power than Putnam’s. Foucault’s works analyze the ways in which the conjunction of historical forces and events at particular times account for the emergence of new concepts and practices, using such mundane facts as the invention of the rifle or new methods of investing capital to explain facts traditionally considered transcendent and eternal, such as “the history of the modern soul.”75 Despite caricatures as an irrationalist, Foucault’s method allows him to claim that history “is intelligible and should be susceptible to analysis down to the smallest detail—but this in accordance with the intelligibility of struggles, of strategies and tactics.”76 Putnam is aware that “(to use Kantian language) one thing physics cannot do is account for its own possibility,”77 since it requires conceptual and semantic facts that admit of no physical explanation. Furthermore, it is a common view in analytic philosophy, upheld by such figures as Quine, Kuhn, and Wittgenstein, that reality or sensory data radically underdetermine our concepts and so cannot account for the specific ones that we have. But Foucault, unlike Putnam, has an alternate account of how these conceptual schemes come about.78

Reasons, Epistemic Truth, and History  163 Another point that the two agree on is that epistemology and ethics are deeply connected. Putnam believes that values and ethical ideals such as human flourishing necessarily play a role in our supposedly disinterested search for truth, and he employs a Deweyan argument to generate ethical criteria out of the inquiry itself, criteria he identifies roughly with Habermas’s values of communicative ethics.79 Foucault’s genealogical works also consider the notions of human normalcy and health—modern scientific versions of flourishing—central to knowledge claims and justificatory practices, but he adds an important element to this analysis. Genealogical works like Discipline and Punish and The History of Sexuality show in detail how easily these kinds of ideals and values can be manufactured and how often they mask oppression. This leads to the ethical implications of anti-realism, a topic Foucault explored in great detail but that has played virtually no role in analytic discussions of the subject.80 For instance, Michael Dummett gives a well-known reading of Wittgenstein that places him fairly close to Foucault as calling it “vain and presumptuous to attempt to see reality through God’s eyes: all we can do is to describe our own practices as we can view them through our own eyes. . . . We cannot intelligibly talk about what is true independently of what we recognize as true.”81 Dummett goes on to show how this position amounts to a kind of idealism that merges with realism since, without the contrast of reality in-itself, the world we experience simply is the world. He concludes that “if realism and idealism, or, as I am here preferring to say, externalism and internalism, coincide, there can be no harm in being an externalist. The externalist has an advantage over the internalist, in that, of certain things that we say, it is our practice to claim that they represent how things are in themselves, independently of how we apprehend them.”82 In other words, if the literal content of the two positions converges, then there is no real difference between them, and realism enjoys the tie-breaker of having common sense on its side. Foucault, on the other hand, would follow J. L. Austin here in claiming that, since we are analyzing actual instances of sentences rather than idealized abstractions such as “The cat is on the mat,” we must look beyond their mere content to their actual effect: “once we realize that what we have to study is not the sentence but the issuing of an utterance in a speech situation, there can hardly be any longer a possibility of not seeing that stating is performing an act.”83 And the actions committed by claims of expertise, according to Foucault, are a form of oppression and enforced conformity. In general, the power that experts like psychologists wield in our society is underwritten by our faith in their knowledge of a realist self. This is neatly summarized in Plato’s argument that, since philosophers are the ones who know the Forms of humanity and the city, they ought to rule, and anyone with enlightened self-interest wants to be ruled by the knowledgeable. Power ought to follow knowledge, even if it seldom does. It is the psychologists’ claim to know the mind’s true essence that induces us to obey them for

164  Lee Braver our own good, in the courtroom, the classroom, and the bedroom. In turn, part of normal behavior will be to obey the law, pay taxes, and continue going to analysis.84 These knowledge claims, which are maintained and legitimated by realism, have far-reaching and often devastating effects. The large, ill-defined, and confused family of “abnormal individuals,” the fear of which haunts the end of the nineteenth century, does not merely mark a phase of uncertainty or a somewhat unfortunate episode in the history of psychopathology. It was formed in correlation with a set of institutions of control and a series of mechanisms of surveillance and distribution, and . . . it gives rise to laughable theoretical constructions that nonetheless have harshly real effects.85 Obviously the category of abnormality is parasitic on the category of normality, which itself is simply a form of the true nature of humanity that we all want to achieve and are in need of expert guidance in order to do so; the whole structure is grounded on realism. Although Putnam is right to see that ideals like human flourishing play an important role in the ways humans pursue knowledge, his pragmatic attempt to base epistemological criteria on them treats this concept much too casually. The response to Dummett’s equanimity about the choice between realism and anti-realism is that realist claims to know mental illnesses and human nature are what empower experts in the human sciences to lock people up. At the very least, many grow up convinced that they are horribly abnormal.86 This is the harm done by realism that Foucault’s studies try to expose and so disarm; if there is no human nature, then actions taken on its basis lose their justification. Foucault’s descriptions of the vast diversity of conceptual schemes that science has employed throughout its history, combined with the anti-realist elimination of a true reality that could determine which one is right, have the effect of undermining our faith in these authorities: “historical analysis is a way of avoiding the sacralization of theory: it allows one to erase the threshold of scientific untouchability.”87 Without the legitimation of independent reality, incommensurable views—such as whether someone is insane or a seer, homosexual or deviant—lack a neutral means of conclusive adjudication. The victory of one view over the other is neither the discovery of independent reality nor the outcome of free rational discussion, given that a common rationality is lacking. This is why many postmodernists believe that the reconciliation of incommensurable ways of thinking entails violence,88 a violence that Foucault’s anti-realist analyses try to reveal.89 While this view might initially appear to be the kind of melodramatic leftism typical of continentals, we can find at least partial counterparts in analytic philosophy. Besides Kuhn’s incommensurable paradigms that are

Reasons, Epistemic Truth, and History  165 influenced by such forces as the internal politics of physics departments or the relative longevity of the proponents of different theories, Wittgenstein agrees that heterogeneous language-games cannot be resolved rationally, since rationality exists only within particular language-games. He calls the process that changes the way someone thinks a kind of “conversion” brought about by “persuasion” rather than autonomous, rational discourse. While freely admitting that reasons would be given, he asks, “but how far do they go? At the end of reasons comes persuasion. (Think what happens when missionaries convert natives).”90 The point is that for them to count as reasons, the interlocutor must already share a form of rationality or argument language-game, which in turn cannot be imparted by reasoning on pain of infinite regress. Wittgenstein even employs violent imagery to make the point: “is it wrong for them to consult an oracle and be guided by it?— If we call this ‘wrong’ aren’t we using our language-game as a base from which to combat theirs?”91,92 NOTES   1 I am confining my discussion here to Putnam’s twenty-year internal realist period. Before 1976, Putnam was a realist, while in the mid-nineties he began to distance himself from internal realism (Clark and Hale 1994, p. 242; Putnam 1994a, p. 456).   2 As is often noted, there are many forms of anti-realism; see Braver 2007 for more. I will be focusing on Putnam’s version rather than, say, Dummett’s, which defines it in terms of the bivalence of truth.   3 Putnam (1990), p. xxxiii. It is because Putnam traces his lineage so directly to Kant as the common source of both branches of contemporary philosophy that he describes himself as “think[ing] about questions which are thought to be more the province of ‘continental philosophy’ than of ‘analytic philosophy’, for instance, . . . the fact that our notions of rationality evolve in history” (Putnam 1983, p. vii). Compare this last issue with Foucault’s description of his own work near the end of his life: “how is it that thought, insofar as it has a relationship with the truth, can also have a history? That is the question posed” (Foucault 1989, p. 456).   4 Biletzki (2001), p. 291.   5 For a Gadamerian analysis of this kind of dialogue, see my “Davidson’s Reading of Gadamer: Triangulation, Conversation, and the Analytic-Continental Divide” in Malpas (2011).   6 Putnam (1978), p. 6.   7 Putnam (1995), pp. 29–30, bracketed numbers added; see also 1990, p. 41.   8 See, e.g., Putnam (1981), p. x. Many commentators have noted and discussed Putnam’s attempts to place himself directly in Kant’s line of descent: Norris (2002), p. 15, Moran (2000), pp. 65–7; Van Kirk (1984), p. 13, Abela (1996), p. 47, Brown (1988), p. 145, Allais (2003), p. 375.   9 Putnam (1990), p. 261, bracketed numbers added; see also 1987, p. 33; 2002, p. 109. 10 On the influence of Wittgenstein’s later work, see Putnam (1990), pp. xxxiv, xl–xli.

166  Lee Braver 11 Putnam (1983), p. 230; see also 1994a, p. 513. 12 Putnam (1981), p. 52; see also Putnam (1988), p. 114, (1983), pp. 206, 272, (1987), pp. 17, 32–33, 35–36, 40, (1990), p. 28, (1992), p. 120, Clark and Hale (1994), p. 253, Quine (1969), pp. 48–49, (1980), pp. 10, 16. 13 Putnam actually claims that he introduced “internal realism” to refer to his earlier scientific realism, which he was abandoning, but readers confusedly used it to refer to his new position, and he acquiesced to this usage (Putnam 1994a, p. 461, n.36). It seems perfectly appropriate to this middle period position to me. 14 Putnam (1990), p. 114; see also p. 121. 15 See Putnam (1990), p. 113, (1992), pp. 72; 109, (1995), p. 75, (2002), p. 180. 16 Derrida (1976), p. 158. 17 There are other elements to Derrida’s idea here, including the relation of writing to thought, but this “internal realist” claim seems prominent. For more on the connection of Derrida to analytic philosophy, see Samuel C. Wheeler III’s excellent Deconstruction as Analytic Philosophy (Stanford: Stanford University Press, 2000). 18 Securing the complete set of forms of intuition and concepts of understanding is very important to Kant, however (see Kant 1965, Axx, Bxxiii–xxiv, Bxxxviii, Axiv, B23, A13/B27, A64/B90, B79, A83/B109, A148/B187–8, A338/ B396, A403, A462/B490, A591/B619, A676/B704, A805/B833, A847/B875). 19 Putnam (1987), p. 19, in bold in the original; see also Clark and Hale (1994), p. 246. 20 Putnam (1981), p. 83; see also 1978, p. 138; 1981, pp. x, 73, 79; 1983, pp. 95, 179, 206, 230–1; 1987, pp. 17–19, 35–36; 1988, p. 112; 1992, p. 120; 1994a, p. 448; 1995, p. 17. 21 See Clark and Hale 1994, p. 291, n.20; Putnam 1983, pp. 128, 201, 215; 1990, p. 29. Here is Quine’s version of the substitution of pragmatic standards for correspondence once realism has been shown to be incoherent: “it is meaningless, I suggest, to inquire into the absolute correctness of a conceptual scheme as a mirror of nature. Our standard for appraising basic changes of conceptual scheme must be, not a realistic standard of correspondence to reality, but a pragmatic standard” (Quine 1980, p. 79). In his famous “Two Dogmas” paper, he spells out this pragmatic standard by which we evaluate conceptual schemes as “the degree to which they expedite our dealings with sense experiences” (ibid., p. 45). Quine, in turn, got this idea from Carnap’s distinction between internal and external questions which concern entities within a conceptual framework and the framework itself, respectively. 22 Putnam (1983), p. 95. Presumably these would be the beliefs towards the center of Quine’s “web of belief” or those in Wittgenstein’s river-bed rather than river (Wittgenstein 1969, §96–99). Goodman posits a similar idea in his scheme or system which lets us “decide whether things are alike. . . a network of labels that organizes, sorts, or classifies items in terms of the types of diversity to be recognized” (Goodman and Elgin 1988, p. 8; see also Goodman 1978, pp. 8, 13). 23 Putnam (2002), p. 109. 24 Another frequent example of this conceptual relativism is mereology, the Polish mathematical-logical system that counts objects in a fundamentally different way than traditional math, logic, and common sense do. See, e.g., Putnam (1987), pp. 18–21. 25 Putnam (1994a), p. 451. 26 See Foucault (1970, pp. 340–1; 1977, p. 38). For more on Foucault’s relationship to Kant, see Braver (2007), pp. 344–8.

Reasons, Epistemic Truth, and History  167 27 In Gutting (1994), p. 314. 28 Foucault (1972), p. 229. 29 Foucault (1972), p. 47. 30 Foucault (1970), p. xxii. He also calls it an episteme. 31 Foucault (1970), p. xxii. 32 Foucault (1972), p. 117; see also 1970, pp. 158, xix, 74, 208, 217. 33 Foucault (1970), p. xxii; cf. pp. 22; 119. See Todd May’s comment: “Foucault’s writings have as one of their central purposes to show how concepts, principles, assumptions, and methods that we take for granted as the givens for all thought are themselves subject to a history that can be described” (May 1993, p. 31). 34 Foucault (2005), p. 253 note. 35 Foucault (1970), p. xxiv. 36 Foucault (1972), p. 204. 37 And by his attempt to understand his own period, as described in “What is Enlightenment?” 38 See Foucault (1977), p. 38. 39 Foucault (1970), p. 340. 40 Martin, Gutman, and Hutton (1988), p. 11. 41 Foucault (1972), p. 131; cf. pp. 156; 183; Rabinow (1984), pp. 87; 118. 42 Foucault (1970), p. 342; see also pp. xiii–xiv; 200; 275; 345. 43 See Foucault (1972), pp. 45, 72 44 Foucault (1970), p. 264. 45 Foucault (1970), p. 217. 46 Foucault (1970), p. 308; see also pp. 72; 128. 47 Foucault (1970), p. xx, italics added; see also p. 322; Veyne in Davidson (1997), p. 160. 48 Rabinow (1984), p. 46; see also Foucault (1989), p. 286. 49 Foucault (1972), p. 203; see also Rabinow (1984), p. 93. 50 Foucault (1989), p. 282; see also 1980, pp. 196–197. 51 See Braver (2007), pp. 181–98, 266–73, 348–360. 52 Wittgenstein (2001), §271. 53 Rabinow (1984), p. 72; see also Putnam 1978, pp. xxx, 126; 1988, p. 115; 2002, pp. 180, 110. I say “most of his career” because he did not hold this view in his early work. He is still committed to it at least through 2002. 54 Foucault (1995), p. 102; see also 1978, pp. 72, 94. 55 Rabinow (1984), p. 89; see also pp. 72–73; 1989, p. 280. 56 Faubion (2000), pp. 253–254. 57 Foucault (1972), p. 127; see also pp. 74; 117. 58 Foucault (1989), p. 221; see also pp. 198; 201; 1972, p. 68; 1980, pp. 108, 118–119. 59 See Foucault (1989), p. 462. 60 Foucault (1977), p. 194. 61 Foucult (1989), p. 446. 62 Foucault (1978), p. 204; see also p. 23; 1989, pp. 157, 285; 1995, pp. 27, 102. 63 Putnam (1981), p. x. 64 See Putnam (1981), pp. 55–5; 1983, pp. xvii, 84; 1988, p. 115; 1990, pp. 41, 115. 65 See Putnam (1983), p. 86. 66 See e.g. Clark and Hale 1994, p. 291, n.26; Putnam 1983, p. xvii; 1988, p. 115. 67 Putnam 1994a, p. 462; see also Rorty 1998, pp. 51–4 and Dummett in Clark and Hale 1994, p. 63 for excellent objections to Putnam on this point. Note

168  Lee Braver that Putnam seems to view this as a strength, since his recent position maintains this common-sense realism. As far as I can tell, however, he is simply sidestepping the problems realism has always had—for himself as well—rather than disarming them. 68 See McCumber (2003, 2005). Peter Hylton describes a “general repudiation of the historical mode of understanding within analytic philosophy. In particular, analytic philosophy seems to think of itself as taking place within a single timeless moment” (Hylton 1990, p. vii; cf. 4, 16). 69 Rorty is right to point out the “cautionary” use of truth in the sense that we might later come to reject what we now hold to be true, but what is gained by saying “p is justified but not true because it might later turn out to be false” instead of “p is now true but might later be false?” (Rorty 1998, p. 60; see also 1991, pp. 27, 128). 70 Nietzsche (1974), §108. 71 Foucault (1989), p. 280. 72 See Putnam (1981), pp. 156, 162, 167. 73 Foucault (1989), p. 353. 74 Foucault (1989), p. 355; see also pp. 276; 280; Faubion (2000), p. 225; Faubion (1998), p. 324. 75 Foucault (1995), p. 23; see also pp. 87; 163; 1989, pp. 276–277. 76 Rabinow (1984), p. 56. In this way, Foucault’s genealogical expansion of his earlier focus on discourse in his archaeological phase replays Marx’s criticism of Hegel’s emphasis on Geist, i.e. the cultural or intellectual features of a society. “Thus, for instance, if in England a machine is invented which deprives countless workers of bread in India and China, and overturns the whole form of existence of these empires, this invention becomes a world-historical fact. Or again, take the case of sugar and coffee, which have proved their world-historical importance in the nineteenth century by the fact that the lack of these products, occasioned by the Napoleonic Continental System, caused the Germans to rise against Napoleon, and thus became the real basis of the glorious Wars of Liberation of 1913. From this it follows that this transformation of history into world history is by no means a mere abstract act on the part of ‘self-consciousness’, the world spirit, or of any other metaphysical spectre, but a quite material, empirically verifiable act” (Marx 1998, p. 58–59, see also Marx 1992, p. 80–81). 77 Putnam (2002), p. 106. 78 Given Foucault’s epistemology, of course, he cannot claim ultimate correspondence truth for his views, but they can possess the empirical truth discussed above, as well as strategic utility. 79 See Putnam 1981, pp. 128; 134–137; 201; 2002, pp. 142, 104–105. 80 Of course, the issue of moral realism has received a great deal of attention ever since Moore, but the ethical implications of epistemological or ontological anti-realism have not, as far as I am aware. 81 Clark and Hale (1994), p. 55–57. 82 Ibid. 83 Austin (1975), p. 139. Todd May puts this point well: “the emergence of a discourse is treated as an ‘event’. . . something to be investigated not in the depths of its reference, but in the horizonality of its relations with surrounding discourses and practices” (May 1993, p. 27). 84 See Foucault (1980, p. 121; 1989, p. 235). 85 Foucault (2003, p. 323), italics added. Note the tone of barely suppressed outrage when he discusses these subjects: “the repressive role of the asylum is well known: people are locked up and subjected to treatment—chemical

Reasons, Epistemic Truth, and History  169 or psychological—over which they have no control; or they are subjected to the nontreatment of a straitjacket. But the influence of psychiatry extends beyond this to the activity of social workers, professional guidance counselors, school psychologists, and doctors who dispense psychiatric advice to their patients—all the psychiatric components of everyday life which form something like a third order of repression and policing.” Language, Counter-Memory, Practice: Selected Essays and Interviews. Ed. Donald F. Bouchard. Trans. Donald F. Bouchard and Sherry Simon. Ithaca: Cornell University Press, 1977, pp. 228–229. 86 For a fascinating account of this phenomenon in conjunction with a good analysis of Foucault, see Ladelle McWhorter’s Bodies and Pleasures: Foucault and the Politics of Sexual Normalization. 87 Foucault (1989), p. 198. 88 See, e.g., Lyotard 1988; Levinas 1969, pp. 72–73; 1996, p. 23; Derrida 1976, p. 127. 89 Of course, this line of argumentation also seems to rob Foucault of positive norms, including any that he might need for his own criticisms. This forms one of the most common and important criticisms of his work, the classic statements of which appear in Fraser (1981); Habermas in Hoy (1986), Habermas (1987, p. 286), and Taylor in Hoy (1986). Although I believe that he has the resources to say a great deal more than this, within the confines of this paper I must leave it at Foucault’s negative point that without realist justifications, “it is good, for ethical and political reasons, that the authority that exercises the right to punish should always be uneasy about that strange power and never feel too sure of itself” (Faubion 2000, p. 461). 90 Wittgenstein (1969), §612. 91 Wittgenstein 1969, §609, italics in original; see also §92; §217; §233; §262. For more on this, see Braver 2012, Chapter 4. 92 I would like to thank Ken Alpern, Colin Anderson, and John Koritansky for their helpful comments and criticisms of this paper.

WORKS CITED Abela, P. 1996. Putnam’s Internal Realism and Kant’s Empirical Realism: The Case for a Divorce. Idealistic Studies 26 (1) (Winter/Spring): 45–56. Allais, L. 2003. Kant’s Transcendental Idealism and Contemporary Anti-Realism. International Journal of Philosophical Studies 11 (4): 369–392. Austin, J. L. How To Do Things with Words. 2nd ed. Cambridge: Harvard University Press, 1975. Biletzki, A. 2001. Bridging the Analytic-Continental Divide. International Journal of Philosophical Studies 9 (3): 291–294. Braver, L. 2007. A Thing of This World: A History of Continental Anti-Realism. Evanston, IL: Northwestern University Press. Braver, L. 2012. Groundless Grounds: A Study of Wittgenstein and Heidegger. Cambridge, MA: The MIT Press. Brown, C. 1988. Internal Realism: Transcendental Idealism? Midwest Studies in Philosophy 12: 145–155. Clark, P. and Bob H., eds. 1994. Reading Putnam. Cambridge: Blackwell. Davidson, Arnold I., ed. Foucault and His Interlocutors. Chicago: University of Chicago Press, 1997. Derrida, J. Of Grammatology. Trans. Gayatri Chakravotry Spivak. Baltimore: The Johns Hopkins University Press, 1976.

170  Lee Braver Goodman, Nelson, and Catherine Z. Elgin. Reconceptions in Philosophy and Other Arts and Sciences. Indianapolis: Hackett, 1988. Goodman, Nelson. Ways of Worldmaking. Indianapolis: Hackett, 1978. Faubion, J. D., ed. 1998. Essential Works of Foucault 1954–1984: Aesthetics, Method, and Epistemology. Vol. 3. Series ed. Paul Rabinow. New York: The New Press. Faubion, J. D. 2000. Essential Works of Foucault 1954–1984: Power. Vol. 3. Series ed. Paul Rabinow. New York: The New Press. Foucault, M. 1970. The Order of Things: An Archaeology of the Human Sciences. New York: Vintage Books. Foucault, M. 1972. The Archaeology of Knowledge. Trans. A. M. Sheridan Smith. New York: Harper Colophon Books. Foucault, M. 1977. Language, Counter-Memory, Practice: Selected Essays and Interviews. Ed. Donald F. Bouchard. Trans. Donald F. Bouchard and Sherry Simon. Ithaca: Cornell University Press. Foucault, M. 1978. The History of Sexuality: An Introduction. Trans. Robert Hurley. New York: Vintage Books. Foucault, M. 1980. Power/Knowledge: Selected Interviews and Other Writings 1972–1977. Ed. Colin Gordon. Trans. Colin Gordon et al. New York: Pantheon Books. Foucault, M. 1988. Politics, Philosophy, Culture: Interviews and Other Writings 1977–1984. Ed. Lawrence D. Kritzman. Trans. Alan Sheridan et al. New York: Routledge. Foucault, M. 1989. Foucault Live: Collected Interviews, 1961–1984. Ed. Sylvere Lotringer.Trans. Lysa Hochroth and John Johnston. New York: Semiotext(e). Foucault, M. 1995. Discipline and Punish: The Birth of the Prison. Trans. Alan Sheridan. New York: Vintage Books. Foucault, M. 2003. Abnormal: Lectures at the Collège de France, 1974–1975. Ed. Valerio Marchetti and Antonella Salomoni. Trans. Graham Burchell. New York: Picador. Foucault, M. 2005. The Hermeneutics of the Subject: Lectures at the Collège de France 1981–1982. Ed. Frédéric Gros. Trans. Graham Burchell. New York: Palgrave Macmillan. Fraser, N. 1981. Foucault on Modern Power: Empirical Insights and Normative Confusions. Praxis International 1 (October): 272–287. Gutting, G., ed. 1994. The Cambridge Campanion to Foucault. New York: Cambridge University Press. Habermas, J. 1987. The Philosophical Discourse of Modernity: Twelve Lectures. Trans. Frederick Lawrence. Cambridge, MA: MIT Press. Hoy, D. C. 1986. Foucault: A Critical Reader. Cambridge: Blackwell Publishers. Hylton, P. 1990. Russell, Idealism, and the Emergence of Analytic Philosophy. New York: Clarendon Press. Kant, I. 1965. Critique of Pure Reason. Trans. Norman Kemp Smith. New York: St. Martin’s Press. Levinas, E. 1969. Totality and Infinity. Trans. Alphonso Lingis. Pittsburgh: Duquesne University Press. Levinas, E. 1996. Basic Philosophical Writings. Ed. Adriaan T. Peperzak, Simon Critchley, and Robert Bernasconi. Bloomington: Indiana University Press. Lyotard, J-F. 1988. The Differend: Phrases in Dispute. Minneapolis: University of Minnesota Press. Marx, Karl. The Poverty of Philosophy. International Publishers, 1992. McCumber, J. 2003. Just in Time: Toward a New American Philosophy. Continental Philosophy Review 36: 61–80. McCumber, J. Reshaping Reason: Toward a New Philosophy. Bloomington: Indian University Press, 2005.

Reasons, Epistemic Truth, and History  171 Malpas, J., ed. 2011. Dialogues with Davidson: Acting, Interpreting, Understanding. Cambridge, MA: The MIT Press. Martin, L. H., Huck G., and Patrick H. H., eds. 1988. Technologies of the Self: A Seminar with Michel Foucault. Amherst: The University of Massachusetts Press. Marx, K. and Engels, F. 1998. The German Ideology. Amherst, New York: Prometheus Books. May, T. 1993. Between Genealogy and Epistemology: Psychology, Politics, and Knowledge in the Thought of Michel Foucault. University Park, PA: The Pennsylvania State University Press. McWhorter, L. 1999. Bodies and Pleasures: Foucault and the Politics of Sexual Normalization. Bloomington, IN: Indiana University Press. Moran, D. 2000. Hilary Putnam and Immanuel Kant: Two ‘Internal Realists’? Synthese 123: 65–104. Nietzsche, F. 1974. The Gay Science. Trans. Walter Kaufman. New York: Vintage Books. Norris, C. 2002. Hilary Putnam: Realism, Reason, and the Uses of Uncertainty. New York: Manchester University Press. Putnam, H. 1978. Meaning and the Moral Sciences. Boston: Routledge & Kegan Paul. Putnam, H. 1981. Reason, Truth, and History. New York: Cambridge University Press. Putnam, H. 1983. Realism and Reason: Philosophical Papers vol. 3. New York: Cambridge University Press. Putnam, H. 1987. The Many Faces of Realism. LaSalle, IL: Open Court. Putnam, H. 1988. Representation and Reality. Cambridge: The MIT Press. Putnam, H. 1990. Realism with a Human Face. Ed. James Conant. Cambridge: Harvard University Press. Putnam, H. 1992. Renewing Philosophy. Cambridge: Harvard University Press. Putnam, H. 1994a. Sense, Nonsense, and the Senses: An Inquiry into the Powers of the Human Mind. The Journal of Philosophy 91 (9) (September): 445–517. Putnam, H. 1994b. Words and Life. Ed. James Conant. Cambridge: Harvard University Press. Putnam, H. 1995. Pragmatism: An Open Question. Cambridge: Blackwell Publishers. Putnam, H. 2002. The Collapse of the Fact/Value Dichotomy and Other Essays. Cambridge: Harvard University Press. Quine, Willard van Orman. 1969. Ontological Relativity and Other Essays. New York: Columbia University Press, Quine, Willard van Orman. 1980. From a Logical Point of View: Nine Logico-Philosophical Essays. 2nd rev. ed. Cambridge, Mass.: Harvard University Press, Rabinow, P., ed. 1984. The Foucault Reader. New York: Pantheon Books. Rorty, R. 1991. Philosophical Papers Volume 1: Objectivity, Relativism, and Truth. New York: Cambridge University Press. Rorty, R. 1998. Philosophical Papers Volume 3: Truth and Progress. New York: Cambridge University Press. Van Kirk, C. A. 1984. Kant’s Reply to Putnam. Idealistic Studies 14: 13–23. Wittgenstein, L. 1969. On Certainty. Eds. G.E.M. Anscombe and G.H. von Wright. Trans. Denis Paul and G.E.M. Anscombe. New York: Harper Torchbooks. Wittgenstein, L. 2001. Philosophical Investigations. 3rd ed., rev. Trans. G. E. M. Anscombe. Madsen, MA: Blackwell Publishers, 2001.

8 Metaphor without Meanings Derrida and Davidson as Complementary Samuel C. Wheeler III

A metaphor ascribes to an object a property it does not really have. How are metaphors meaningful and informative, if what they say is not literally true? The very word “metaphor,” “transport across,” obviously is itself a metaphor for a way of using language that departs from the model of ascribing properties to objects that have them. “Trope” is likewise a metaphor, of a word turning. Various accounts explain how a meaning fitting one object can be carried across to another object to which it does not properly apply. For instance, perhaps there is a systematic way meanings are transformed into metaphorical meanings when they are applied to different categories of objects. Such accounts of metaphor posit meanings that attach to words and apply properly to some objects and improperly to others. Derrida and Davidson are both anti-metaphysical. They reject meanings as entities that could be transferred and solve or reject the issues such a view fosters. Neither holds that term- or sentence-meanings match properties or that facts to yield truth. What they have to say about metaphor is therefore quite different from accounts that view meanings as trans-linguistic entities. Derrida’s (1971) “White Mythology” and Davidson’s (1978) “What Metaphors Mean” are the best works on metaphor I know of. While the issues about metaphor these essays discuss are quite different, these essays agree on some fundamental theses about language and meaning. Furthermore, as I will argue, the arguments of each philosopher supplement and strengthen the arguments of the other. While almost certainly neither philosopher ever read anything the other wrote, these essays from mutually alienated traditions complement one another. The first section of this essay is a discussion of some aspects of the argument in “White Mythology.” “White Mythology” is a complex, 62-page essay with many strands, and I omit discussion of many of those strands. My discussion focuses on topics that connect Derrida’s discussion with Davidson’s. Derrida’s primary concern is with metaphor as it shapes metaphysical concepts. The second section summarizes Davidson’s discussion of metaphor. Davidson is interested primarily in how metaphors comport with his conception of meaning as truth-conditions.

Metaphor without Meanings  173 The third section discusses how Derrida’s discussion of the origin and basis of philosophical concepts could be incorporated into Davidson’s thinking. This section examines how Davidson’s understanding of metaphor and his anti-metaphysical stance might be enriched by this particular treatment of exactly these metaphors. The fourth section examines how Derrida’s discussion of metaphor can be supplemented with Davidson’s account of what metaphor is, and how metaphors convey meaning. Briefly, Davidson can supply non-metaphysical theories of phenomena that Derrida needs to have some account of. 1. “WHITE MYTHOLOGY: METAPHOR IN THE TEXT OF PHILOSOPHY” Derrida’s thesis, very roughly, is that philosophy is founded on a metaphor, that of the intellectual light, with a network of related metaphors such as clarity, properly applied meanings, and meanings that are before the mind in this light. But this very founding is problematic, because the concept of metaphor itself depends on the notion that words carry meanings, which meanings are intellectual attachments to material signs. Metaphor is the borrowing of a meaning proper to one thing or kind and the transfer of that meaning to another. But the idea of meanings as distinct from their material bearers has been developed on the basis of the metaphor of the light of the intellect as like regular photon-constituted light. So, the very remark that philosophy is founded on a metaphor is a remark that presupposes philosophical concepts. Derrida regards metaphysics as a pervasive set of notions that dominate our thought on almost every topic. So, the task of philosophical critique is quite subtle for Derrida. Every critique will presuppose the applicability of, by employing, exactly the notions and suppositions that are being attacked.1

A) Exergue The text suggests an account of metaphor and its function in shaping philosophy. However, all such projects are going to turn out to be subtly misconceived. If metaphor is in the text of philosophy, then a philosophical account of metaphor will include an explanation of itself. So, the account will explain the very terms that it uses. Derrida develops the problematic consequences of this consideration at length. A theme in many discussions of philosophy has been the claim that philosophical concepts are abstractions from the sensory. A particular version of this theme is that philosophical concepts are metaphors that have lost their literal meaning and have come to mean the philosophical concept. Hegel supposed that this removal of sensory meaning and its recovery and

174  Samuel C. Wheeler III transformation into another meaning was an advance to a higher plane, an Aufhebung. The sensory literal meaning is a necessary ladder to the philosophical meaning. Anatole France has a less flattering account of philosophy’s relation to metaphor. The removal of the sensory meaning, once enlivened by sensory content, is now erased and devoid of real, sensory meaning, rather than being a higher sublime replacement. Nietzsche, whom Derrida quotes,2 reflects another part of the philosophical attitude toward metaphor, that metaphor is inferior to speech that actually presents reality. Nietzsche, in saying that truth is merely an army of figures, himself accepts the idea that real truth would be expressed in terms that reflect reality. Derrida’s relation to these theses about metaphor and philosophy is complex. On the one hand, he thinks the thesis that philosophy, in particular metaphysics, is metaphor is sort of right. On the other hand, the relation of metaphor to metaphysics is more complex. Both Hegel’s idea of recovery in a higher plane and France’s critique that philosophy is washed out sensory ideas presuppose a system of philosophical concepts established by metaphors. To anticipate: If the figural connections of predicate to predicate and application to application are not governed by “natural” and therefore governed resemblances but by dissemination,3 then metaphysical notions are suspect as scientific or appropriate descriptions of anything. “Dissemination” pictures linguistic connections as determined by words as words, rather than conceptual connections. “Dissemination” applies both to predicates connecting with other predicates and to application of predicates. Neither process is controlled by language-transcendent meanings. “Metaphor in the text of philosophy,”4 as a list of programmatic questions, is exactly what Derrida’s kind of analysis transcends. Exactly how “metaphor seems to involve the usage of philosophical language in its entirety”5 is what “White Mythology” spends many pages explaining. Derrida substitutes “usure” for “usage” and takes the economic and physical senses of “usure” to be related, a characteristic move following from taking words as marks to be basic.6 “Usure” as losing and regaining with a profit corresponds to Hegel’s notions on the topic, while “usure” as wearing away corresponds to France’s. An important thesis of the essay is encapsulated in the claim, “Usure does not overtake a tropic energy otherwise destined to remain intact.”7 Derrida denies meanings as trans-linguistic signifiers, but the treatment of metaphor as “usure” supposes that there are such meanings that get dissipated and perhaps recovered in a higher form. If dissemination, un-rule-governed drift, is what actually happens in linguistic connection among predicates and application of predicates, the founding conceptions of philosophy are figural uses of terms of ordinary language unsupported by fit with the phenomena. So there is no literal explanation of their significance. Metaphysics is constituted out of a series of developments of this “forced” metaphor.

Metaphor without Meanings  175

B)  Plus de Metaphore The core thesis of the entire essay is in the long and dense paragraph that begins this section:8 “The exergue effaced, how are we to decipher figures of speech, and singularly metaphor, in the philosophic text?” This paragraph tries to show the “condition for the impossibility of this project” by some intricate considerations: a) . . . Metaphor remains, in all its essential characteristics, a classical philosopheme, a metaphysical concept. It is therefore enveloped in the field that a general metaphorology of philosophy would seek to dominate. The metaphor of “domination” corresponds roughly to the analytic metaphors of “reduction” and to the common philosophical metaphors of “priority.” A subject matter or discourse is dominated by another if it can be accounted for by that other, without loss. So, for instance, physics would dominate biology if all biological phenomena could be expressed and explained in physical terms. Biology would then presuppose physics, so that an explanation of physical phenomena in biological terms would be presupposing what is to be explained. Is this presupposition really a problem? The phenomenological notion of analysis assumes that it is. But Quine’s (1960) thesis is that we start from where we are. We think there are physical objects, so those can be the basis of our theorizing. Even though it may be the case that all our evidence for physical objects is sensory, physical explanations of sensory experience are legitimate. Quine rejects the phenomenological project of constructing an epistemological foundation for our knowledge of the world. Derrida’s project, though, is not that of examining attempts to ground knowledge of something clearly legitimate, such as our knowledge of the external world. Metaphysics, as he conceives it, is a system of notions with pretensions to being a science that allows the mind to master the world, to get it right. Those notions and the field they constitute are not obviously legitimate. I have argued9 that Davidson dissolves metaphysical problems by rejecting exactly the central metaphysical concepts that Derrida is attacking. So, Derrida is presenting an argument that these concepts are not rationally legitimate. If the analytical concepts of metaphysics are themselves founded on metaphors without natural propriety, that is, on catachreses, then metaphysics cannot give an account of metaphor that actually explains metaphor, in the sense of giving a straight explication of what a metaphor is and how it works. In particular, if the central metaphysical idea of meanings as logoi inspectable by the mind’s eye is an unfounded metaphor, then

176  Samuel C. Wheeler III explaining metaphor using such meanings cannot be a genuine explanation. Derrida continues this argument: b) Metaphor has been issued from a network of philosophemes which themselves correspond to tropes or to figures, and these philosophemes are contemporaneous to or in systematic solidarity with these tropes or figures. This stratum of ‘tutelary’ tropes, the layer of ‘primary’ philosophemes (assuming that the quotation marks will serve as a sufficient precaution here), cannot be dominated. It cannot dominate itself, cannot be dominated by what it itself has engendered, has made to grow on its own soil, supported on its own base. Therefore, it gets ‘carried away’ each time that one of its products—here, the concept of metaphor—attempts in vain to include under its own law the totality of the field to which the product belongs.10 Philosophy cannot be understood as a special case of effaced, erased metaphors, or in any other reduction to “metaphorology.” In brief, philosophical explanatory concepts are themselves figures. Philosophical terms are non-literal. If a term is defined in terms that essentially involve itself, then the definition is circular, and so no grounding has been supplied. But why can’t there be an account of the role of metaphor in philosophy in philosophical terms just because metaphor itself is a concept that springs from philosophical terms? What is wrong with a concept “dominating itself?” The next two sentences clarify the problem: c) If one wished to conceive and to class all the metaphorical possibilities of philosophy, one metaphor, at least, always would remain excluded, outside the system: the metaphor, at the very least, without which the concept of metaphor could not be constructed, or, to syncopate an entire chain of reasoning, the metaphor of metaphor. This extra metaphor, remaining outside the field that it allows to be circumscribed, extracts or abstracts itself from this field, thus subtracting itself as a metaphor less.11 “Metaphor” is itself a metaphor, and so cannot be accounted for by an account of metaphor in philosophy. What is the “so” here? Why cannot the metaphor without which the concept of metaphor cannot be constructed be included in this account of how metaphor functions in philosophy? Metaphysics, if it is a theory that masters reality, uses terms whose definitions are precise. “In the order of concepts [that is, in metaphysical theorizing] . . . when a distinction cannot be rigorous or precise, it is not a distinction at all.”12 Derrida’s critiques show that metaphysical distinctions cannot be made precise, and so do not meet metaphysical standards. An account of the functioning of metaphor within philosophy would be a literal

Metaphor without Meanings  177 philosophical explanation, using rigorous and precise concepts, of how figuration figures in the concepts of philosophy. It would use the concept of metaphor, and explain it. The rest of the essay shows why a philosophical account of metaphor is impossible. The next several paragraphs give a short version of the argument. First, an account of the functioning of metaphor ought to be a straight and literal account, since the functioning of metaphor is going to be explained scientifically. If it is explained, it covers the whole functioning of metaphor within the field of philosophy. But the very concept of metaphor is itself (as Derrida will argue) established by metaphors and, by the theory of metaphor, established in a way that does not permit literal paraphrase. The catachreses (unfounded, unnatural metaphorical application of terms) that establish the concepts of metaphysics and that set up the very notion of meanings, the logoi that permit the idea of the transfer of meaning, must be covered in this account of the way metaphor functions in philosophy. But this means that the account being given is not literal, not “straight,” since the very term “metaphor” is itself a metaphor, and so accounts using it are themselves metaphorical. That is, without the logoi that are established by the theory developed from the light-of-the-intellect metaphor, the notion of the transfer of logoi cannot be understood. So, suppose we take the metaphorical, figural account and say what it means literally. That is, the metaphor of “metaphor” “transferring” or “bearing across” can just be replaced by a literal account of what is going on with meaning in this kind of figure. But if meaning, logoi, and so literalness are all founded only on catachreses, metaphors with no natural, literal equivalents, then there is no straight account available. If the philosophical notions in terms of which metaphorical figuration is to be explained are metaphors that cannot be rationally explained, then there is no straight philosophical account of things. So, the metaphors that are philosophical concepts are metaphors left out of the account. The fundamental such metaphors are those related to the light of reason. Establishing that these metaphors are catachreses according to the classical accounts of metaphor is the project of two later sections, “The Ellipsis of the Sun” and “The Flowers of Rhetoric,” discussed below. Derrida summarizes what he will argue: A philosophical account of metaphor would organize itself along oppositions (proper/improper; sensible/ intelligible, but to use these distinctions “one would have to posit that the sense aimed at through these figures is an essence rigorously independent of that which transports it, which is an already philosophical thesis, . . . the thesis which constitutes the concept of metaphor, the opposition of the proper and the improper, of essence and accident, . . . of thought and language, of the intelligible and the sensible.”13 Derrida’s thesis is that in fact all of these distinctions are founded on a philosophical metaphor, the one that takes “light” in an intellectual sense, thereby inventing the intellect, intelligibility,

178  Samuel C. Wheeler III the thinkable sameness that links two particulars. This “founding metaphor” of philosophy is not built into the nature of things. Derrida returns to the contrast between Hegel and France on “usure.”14 Hegel and France take the sensory to have a purely transparent present meaning. France thinks of the reduction of sensory meaning as wearing away and washing out, while Hegel thinks of it as an increase in value. Hegel regards this loss and recovery as Aufhebung, sublimation. A sensory term takes on a metaphorical usage (i.e., is displaced from its proper domain) and thereby acquires a spiritual meaning (a higher value), which then becomes the literal meaning of the term, which has thus come back to its proper home (literal use) but now is worth more. Both views incorporate the metaphors of philosophy. Derrida argues against these proposals in the next two sections. First, the sensory cannot be an original meaning-supplier. Second, non-meaning-regulated dissemination is a better model of philosophical concept-construction, rather than either Hegel’s conception of the recovery and transformation of meaning that creates philosophical meanings for terms or France’s conception of wearing-away of given sensory meaning. The next sections of the essay complete Derrida’s argument. I will address two central topics: First, a straight account is impossible. Second, accounting for metaphorical figuration of the very terms in which metaphor is explained is impossible, since the explaining terms are themselves metaphorical.

C) The Ellipsis of the Sun: Enigmatic, Incomprehensible, Ungraspable This section and the next develop Aristotle’s account of metaphor, which Derrida analyzes as a work on metaphor whose “entire surface is worked by a metaphorics.”15 Derrida shows this by a detailed exhibition of that surface being worked. These metaphors construct the subject matter of proper and improper, terms which presuppose logoi and which constitute the theory that allows metaphor to be distinguished from extending a predicate properly to a new case of its extension. Derrida explores the relation of this account to the sensory, discussing Aristotle’s example of the sun sowing rays as seeds. There is nothing clearly in the real world that makes the relation between the rays of the sun more analogous to the sowing of seeds than to an anus, as in Bataille (1927). “Where has it ever been seen that there is the same relation between the sun and its rays as between sowing and seeds?”16 This metaphor is effective, as Davidson will put it, but not accurate in any literal sense. Because they result from derived meaning, such metaphors require a proper term to be the non-metaphorical meaning-supplier from which derived meanings get derived. We drift along linking analogy to analogy, as “wipe out” goes from erasure to killing a military unit to winning a football

Metaphor without Meanings  179 game and getting very drunk. What is the proper here? If “all the terms in an analogical relation already are caught up, one by one, in a metaphorical relation,”17 then what supplies meaning? Neither the sun itself nor the sense-impression of the sun can supply a meaning, according to Aristotle’s own views. The sun itself is a term, not a proper meaning, one which would designate the sun and only the sun, by its very meaning. So the sun is not a proper meaning-supplier.18 More generally, sensory items, whether sense-data or their sources, are not meanings.19 So, neither Hegel’s nor France’s conception of “usure” can be right.

D)  The Flowers of Rhetoric: The Heliotrope The demonstration that Aristotle’s classic account of metaphor is a work on metaphor whose “entire surface is worked by a metaphorics” continues. Derrida discusses “the proper” and notes a difficulty raised by Aristotle’s distinguishing “good” metaphors from bad ones. The good ones have to get something right. So, good metaphors have to convey a property the object has, even though that property does not define the object. 1) “Properness,” as in “proper name,” is there being a single name for a single thing. Generalized to predicates, on the metaphysical model, words can have several meanings, but those meanings must be finite so that, in principle, the one word-one meaning model would be preserved. Derrida refers to Aristotle’s discussion of contradiction in Metaphysics Γ 4 and following, where Aristotle demands that a speaker mean exactly one thing. Dissemination, as unlimited drift of meaning, would count as not meaning at all, and would be sophistry. 2) The proper is different from the essence. A property that only one kind of animal always has, for instance, is proper to that kind of animal, but may not be its essence, what it is to be that animal. Good metaphors convey such properties. In this kind of case, these properties are repeatable and can be expressed by meanings. 3) Now Derrida returns to the sun and notes that, as a sensory object, not only can the sun not have a repeatable meaning designating itself as entity, as Aristotle notes in Metaphysics Z15, but it cannot even have a property, in Aristotle’s sense of something that is special to it. For Aristotle, “properties” are ontological attachments, albeit of a dependent sort. So, “being a fiery object above the Earth in the daytime” does not qualify as a property. So, by Aristotle’s own lights, sun metaphors are always imperfect metaphors, metaphors that do not present a property of the object. Furthermore, the sun itself is not properly meant in discourse, but only in an imperfect metaphor. So by the classical account of metaphor, we do not get at meanings

180  Samuel C. Wheeler III by starting from sensation, as Hegel’s and France’s accounts supposed. Seeing delivers particulars, which cannot be reduced to predicates, that is, general terms having meanings. So, the idea that metaphorical meaning could derive from sensory meaning is an illusion. Leaving out much detail, Derrida’s conclusion is that “everything in the discourse on metaphor, that passes through the sign eidos, with its entire system, is articulated with the analogy between vision of the nous and sensory vision, between the intelligible sun and the visible sun. The determination of the truth of Being in presence passes through the tropic system.”20 So “intellectual vision” is a metaphor with only metaphorical foundation, which metaphor is required in order to have Being as presence. That is, without the intellectual light, we could not have essences “before the mind,” by analogy with having something before the eyes. So, (254) “Philosophy, as theory of metaphor, will first have been a metaphor of theory.” Derrida’s first works were discussions of Husserl, who conceived of meanings as items that were present to the mind. Following Dagfinn Føllesdal (1958), and noting that Husserl and Frege were both working on the project of refuting psychologism in logic and mathematics, I think of Husserlian essences as like Fregean senses—non-linguistic items that the mind apprehends, which are the meanings of linguistic expressions. Derrida takes Husserl’s careful analyses to be the most sophisticated versions of logoi, and takes the problems he finds in Husserl’s analysis to be a reason to take the presence before the mind of such meanings to be an illusion. “White Mythology” is a further argument that “presence,” as meanings are supposed to be “present before the mind,” is unsupportable. There are no such meanings. Derrida’s later discussions are examinations of the consequences he sees of there being no transcendental signifiers, meanings that can be transferred and expressed in a variety of linguistic systems. They are also demonstrations by particular cases of the un-logos-controlled “dissemination” that actually occurs when words are connected to other words without trans-linguistic meanings to restrict their connections to the actually valid ones.

E)  La Metaphysique—Releve de la Metaphore For Hegel, as Derrida discussed above, sensory concepts are sublimated, erased, and recovered in a higher form, that is, aufgehoben, to become philosophical concepts. From Derrida’s point of view, given his analysis of metaphor, what happens is not sublimation but “taking off” in dissemination.21 Dissemination is not constrained by logoi. It links terms to terms with logical support.22 Since there are no natural resemblances between the eye of vision and the “eye” of the mind, the sublimation of the intake of photons is not a scientifically or rationally supported move. So, Derrida says, what happens with metaphors that become philosophical concepts is a “releve” in the sense of “taking off” rather than recovery in sublimation.23

Metaphor without Meanings  181 Derrida shows how Plato and Descartes, for all the differences in their particular metaphors, have essentially the same metaphysical conception of meanings as ideas before the mind illuminated by the light of reason. He then concludes with several paragraphs that summarize the discussions of the previous sections. “Metaphor, therefore, is determined by philosophy as a provisional loss of meaning, an economy of the proper without irreparable damage, a certainly inevitable detour, but also a history with its sights set on, and within the horizon of, the circular re-appropriation of literal, proper meaning. This is why the philosophical evaluation of metaphor always has been ambiguous: metaphor is dangerous and foreign as concerns intuition (vision or contact) concept (the grasping of proper contact of the signified,) and consciousness (proximity or self-presence); but it is in complicity with what it endangers, is necessary to it in the extent to which the de-tour is a re-turn guided by the function of resemblance (mimesis or homoiosis) under the law of the same. . . . The teleology of meaning, which constructs the concept of metaphor, coordinates metaphor with the manifestation of truth . . . as presence without veil . . . with the appropriation of a full language . . . with the vocation of a full nomination.”24 Derrida takes metaphysics to be a would-be science based on a metaphor that has no basis in nature. There is no reason to model understanding on seeing and no reason to take metaphysics seriously. 2.  “WHAT METAPHORS MEAN” Davidson’s essay25 “What Metaphors Mean” has more limited goals. Davidson has a prima facie difficulty with what metaphors mean because he treats meaning as truth-conditions. Truth-conditional semantics, as Davidson conceives it,26 gives the meaning of a sentence by using a sentence in the interpreter’s language. So, in the Tarskian27 formula “ ‘Snow is white’ is true if and only if snow is white,” the first clause applies the predicate “true” to a mentioned object, a sentence, and the second clause says what has to be the case for that predicate to apply to that object. Analogous formulas describe true-of conditions of predicates: “ ‘White’ is true of an object a if and only if a is white.” The sentence explains what “white” means by saying what a thing has to be for that adjective to apply. A theory of meaning for a language is an empirically established counterfactual-supporting set of generalizations about terms in the language that says what the world has to be like for each sentence in the language to be true. Such a theory shows how the truth-conditions or satisfaction-conditions of complex expressions depend on those of component expressions. For each simple predicate, there is a clause, such as “ ‘Flies’ is true of an object θ if and only if θ flies.” For every compounding device, there is a clause such as the one for “and”: “For all sentences X and Y, X concatenated with “and” concatenated with Y is true if and only if X and Y are true. There is nothing about “fit” of meanings

182  Samuel C. Wheeler III to anything beyond trivialities such as that “Joe is a frog” fits the world just in case Joe is a frog.28 A theory of meaning, for Davidson, is a recursive theory about a person at a time such that, for every sentence of the person’s language, there is an account of what the world is like when that sentence is true. Like Derrida, Davidson does not think that there are meanings as self-interpreting items that attach to signs. He argues that meanings as entities are useless in a theory of meaning. There are meaningful expressions, but that cannot be understood as expressions that are full of meaning. The difficulty posed by metaphors for Davidson is that they are meaningful, but their meaning cannot be understood as their truth-conditions. Davidson examines a number of suggestions—that metaphors are compressed similes, that there is some kind of context-dependent transformation of meanings that generate the meaning of a metaphor, and so forth. He argues that they all are incorrect. “The central mistake I shall be inveighing against is the idea that a metaphor has, in addition to its literal sense or meaning, another sense or meaning.”29 This is in accord with Derrida’s view that metaphors are not governed by any kind of meaning-transforming logic. The difference, which turns outs to be only on the surface, as I argue below, is that Davidson assigns meanings to sentences. As we will see, this is only a surface difference. Davidson’s positive account of metaphor is rhetorical.30 A central thesis of his philosophy of language is that semantics, that is, truth-conditions, should be sharply separated from what speakers are doing with sentences. Sentences are presented with their truth-conditions for an unlimited variety of purposes. Interpretation of speech acts is a special case of interpretation of actions. For rather Derridean reasons, Davidson thinks it is impossible to get a systematic semantics by starting from speech-act potentials and deriving meanings from these. There is an unlimited and non-rule-governed variety of things that could be done with a sentence. Metaphors differ from literal speech, not in their truth-conditions, but in their force, in the purpose for which they are presented.31 “I think metaphor belongs exclusively to the domain of use.” Most metaphors are in fact false sentences. The Earth is not really a floor, even though Dante says it is. The literal meaning is the only meaning the sentence has. A metaphor achieves its objective by using its literal meaning. Since a reasonable interpreter knows the sentence is false, the interpreter will interpret the speaker as having something else in mind in uttering it. Davidson imagines that a Saturnian is learning English by being shown examples of the extension of “floor.” For the Saturnian to appreciate that calling the Earth a floor is metaphorical, he has to know that the Earth is not just another kind of floor.32 The Earth is not actually a floor. Metaphors, when they are effective, lead the hearer to see what they are applied to differently. “How many facts or propositions are conveyed by a photograph? None, an infinity, or one great unstatable fact? Bad

Metaphor without Meanings  183 question. A picture is not worth a thousand words, or any other number. Words are the wrong currency to exchange for a picture.”33 There is no “literal” paraphrase of what the metaphor conveys. Davidson denies “the thesis that associated with a metaphor is a definite cognitive content that its author wishes to convey and that the interpreter must grasp if he is to get the message.”34 Davidson thus would agree with Derrida’s characterization of the metaphors he discusses as “catachreses.” Since there is no literal paraphrase, there is nothing that can be put in general terms, that is, expressed in predicates, that makes the metaphor correct or appropriate. Metaphors are presented to produce effects. Metaphors can be effective or not, and the effects can be good or bad, but the evaluation of metaphorical inscriptions as true or false is not appropriate. The apparent difference between Davidson’s and Derrida’s thinking about metaphor is that they have different conceptions of truth. Derrida assumes that “truth” would be the correspondence of a meaning expressed by a sign to reality. Derrida had argued before35 on phenomenological grounds that there are no such meanings. Such meanings, or logoi, have been part of every metaphysical system. “White Mythology” is part of his continuing attack on metaphysical notions, such as Truth, which rest on meanings. Derrida concludes that there is no such thing as Truth as metaphysics conceives of it. Davidson likewise rejects meanings as they have come down in his tradition. He agrees with Quine that there is no empirical justification for logoi and observes that they are useless for semantics anyway.36 Davidson, though, also rejects the classical correspondence account of truth. On Fregean grounds, all true sentences can only refer to the same thing, if they refer at all. There are no entities corresponding to sentences as facts are supposed to. For Davidson, this does not mean that there are no true sentences. He takes truth to be a primitive concept intimately tied to our understanding of ourselves and the world, whose structure is illuminated by the Tarski37 truth definition.38 A truth-definition, though, does not give the meaning of a sentence in any more illuminating terms than the interpreter’s language. One way to express what is in common to Davidson and Derrida is to say that language is as explicit as meaning gets. All there is to meaning is human applications of language in the world. Derrida holds that this does not deliver correspondence truth. Davidson agrees. So, on the issue of whether there is truth, then, there is no clear disagreement between Davidson and Derrida. 3.  DERRIDA SUPPLEMENTING DAVIDSON Derrida thinks of words in terms of their connections with other words, which connections cannot be systematically divided into the logical and the rhetorical, given the absence of logoi. The discussion of the significance of philosophical terms by tracing their origins is a topic in Hegel that continues

184  Samuel C. Wheeler III in the discussion quoted in the Exergue section of “White Mythology.” Derrida, following Hegel and Heidegger, thinks of the origins of concepts as continuing to be relevant as those concepts are used historically. Davidson is not primarily interested in particular predicates. His project is understanding how language, as an infinite capacity to understand new sentences, it possible. By and large, since the understanding of predicates is by clauses such as “ ‘Flies’ is true of Theaetetus if and only if Theaetetus flies,” particular predicates are not generally a topic. On a few occasions, Davidson did discuss particular predicates. In Davidson (1991), on James Joyce, Davidson describes something like dissemination. However, the dissemination of particular words, and the fact that we can interpret such dissemination, was not a major interest. Davidson’s understanding of interpretation has affinities with Derrida. Davidsonian interpretation interprets the other in a way that maximizes the degree to which the other is a “believer in the true and lover of the good.”39 Such maximization will often yield more than one optimum, so that interpretation is sometimes indeterminate. As Davidson (1986) argues, there is no algorithm for interpretation. No rules tell us how to understand someone who says he has prostrate difficulties. More thinking about word-connections would have given Davidson more arguments for indeterminacy of interpretation. Davidson never discussed the fact that many terms are metaphors or other figures that became “literal” in the sense that most users of the language have no idea that the word has that figural origin. In fact, as I have argued elsewhere,40 this phenomenon, since it takes place gradually, creates an indeterminacy in which truth-definition characterizes the person’s language. That is, if it is indeterminate whether “crush” is literal or metaphorical in a sentence like “The Seahawks crushed the Broncos in the 2014 Superbowl,” then it is indeterminate whether the truth-definition of the speaker has one clause corresponding to the word “crush,” so that the utterance is a metaphor, or more than one, so that the utterance is literal. Davidson characterized metaphors as analogous to pictures—getting us to conceive of things in a certain way. Davidson did not discuss the possibility that some metaphors yield deceptive or misleading pictures and did not discuss the possibility of interlocking networks of metaphors. Derrida’s argument in “White Mythology” shows that the picture of meaning and language that dominates metaphysical thinking is a network of terms organized by vision metaphors—meanings are “before the mind,” we’re enlightened by clear explanations, and so forth. Derrida points out how the proper meaning of terms, which is required for the notion of metaphor to have purchase, is itself supported by logoi.41 Given his view that metaphors present pictures, Davidson could accept that the interlocking network of metaphors that Derrida shows to be the core of metaphysics is a complex picture of a thinking and meaning. The complex picture does not decode into any list of propositions, and so is

Metaphor without Meanings  185 not strictly speaking a theory. The picture nevertheless shapes the way people think about thinking and meaning, just as the network of metaphors in pastoral poetry shapes a false and perhaps anti-feminist conception of romantic love. Although Davidson’s use of rhetorical categories is focused on differences of force with which sentences with truth-conditions are presented, Davidson should also be open to the idea of non-logical, “rhetorical” connections among predicates. Davidson, in denying meanings and the analytic-synthetic distinction, also denies a sharp line between logical and rhetorical connections among primitive predicates, as argued in Wheeler (1986) in the case of Quine. Davidson should recognize that, even though in his terms it is true that an explanation clarifies an argument, the connection between this clarity and light strongly suggest the mistaken but natural-seeming theories that have dominated philosophical thinking. Davidson, that is, could adopt Derrida’s explanation for the attraction of metaphysics. Davidson could interpret Derrida’s discussion in “White Mythology” as an account of why it is that philosophers and non-philosophers are intuitively drawn to accounts of language use, translation, and communication that appeal to language-transcendent entities. Derrida’s analysis of the metaphorics of light, sight, seeing, and thinking can supplement Davidson’s explanations of how humans communicate without such entities. Derrida’s genealogy of the conception of meanings as entities shows that something that appears natural and obvious is neither.42 Davidson could endorse this genealogy and explain philosophers’ resistance to the idea that there are no meanings and so no analytic truths that depend on meanings. The alternative explanations of language use and understanding he proposes would be given negative support. 4.  DAVIDSON SUPPLEMENTING DERRIDA Deconstruction is popularly characterized as a kind of relativism holding that ordinary life and communication is impossible. In fact, Derrida thought that people could communicate and that “France is roughly hexagonal” is right in a way that “France is roughly triangular” is not.43 But he supplies no account of how language actually manages to communicate, how people get things right, and what it is for people to mean things. Derrida’s rejection of metaphysical notions argues that those notions are not equipped to master the world. Derrida did not think that. Rather, he thought that the metaphysical picture by which meanings control real connection and distinguish logical connection from rhetorical connection is incoherent. The theory of logoi was to guarantee determinacy and so intellectual mastery, but depends on theory-apt distinctions. Dissemination is another, less misleading, picture. Dissemination, though, is not an

186  Samuel C. Wheeler III explanatory theory of how we can, for instance, write essays that people understand. Davidson has such an account. Davidson’s approach to philosophical issues is different. The uselessness of meanings in semantics leads him to do semantics differently, in order to explain why, without such meanings, we can understand one another and understand new utterances without language-transcendent meanings. Davidson offers a replacement for the metaphysical picture that accommodates the failure of determinacy and the fact that we get along pretty well in the world and understand novel communications. Davidson recognizes that language use, despite not having language-transcendent meanings, still requires theory. A person with a language has an unlimited ability to understand new sentences. So there must be a recursive structure that a language-learner acquires. This is entirely compatible with the dissemination of every primitive predicate in the language, which Davidson (1991) suggests. A person can understand that “wiped out” is true of an entity if and only if it is wiped out, even on first hearing the term applied to a surfing outcome. As an interpreter, she will understand what is meant, without having a rule by which this new application is determined. Derrida could adopt Davidson’s accounts of truth and meaning, as well as his account of what metaphors mean. Derrida agrees with Davidson that there are no meanings as entities. Davidson agrees with Derrida that, given that there are no meanings as logoi, some indeterminacy is inevitable. Davidson’s analyses of this indeterminacy could be supplemented by some of Derrida’s considerations. Derrida could remain a correspondence theorist about Truth, one who says that correspondence is required for Truth, but that Truth does not exist, while recognizing that there is something else, truth, which has the centrality to the intentional sphere Davidson ascribes to it. Derrida should not object that it is sometimes indeterminate whether an utterance is true. Davidson’s is not a metaphysical theory. Some distinctions that cannot be made sharp in other terms are acceptable for human discussions, if not for theorizing. Derrida recognizes that there are value distinctions, but such distinctions are not suitable for metaphysics as a mastering theory. “What philosopher . . . ever renounced this axiom: that in the order of concepts (for we are speaking here of concepts and not of the colors of clouds or the taste of certain chewing gums) when a distinction cannot be rigorous or precise, it is not a distinction at all.”44 Davidson’s account of truth, his account of interpretation as maximization of agreement rather than capturing meanings, and the probabilistic way the terms of the intentional scheme are connected via truth place the outputs of his theory, but not the theory itself, in the cloud-and-chewing gum category. For instance, truth-definitions are precise and determinate. What is indeterminate is which truth-definition characterizes a particular speaker at a time.

Metaphor without Meanings  187 There is also no reason that Derrida, given that he accepts that Davidsonian interpretation permits an understanding of metaphor that accounts for ordinary speakers’ intuitions, could not accept Davidson’s account of what metaphors are. That is, given that “literal” expressions have the feature Davidson calls “truth,” which Derrida also accepts as characterizing something, Derrida can accept that metaphors are false in Davidson’s sense. Derrida and Davidson could agree. 5. CONCLUSION Derrida and Davidson follow distinct but complementary anti-metaphysical strategies in these essays and in other works. Derrida argues that the central metaphysical concepts are conceptually ungrounded. His intricate discussion portrays metaphysics as misleading metaphors engendering concepts that cannot be made precise. Davidson argues that the important topics metaphysical notions are supposed to deal with are better dealt with in other ways, which he provides. Metaphysical apparatus is not needed. Davidson shows especially that truth, meaning, and predication need no metaphysical aid. It is too bad that there is yet no attack on the pseudoscience of metaphysics combining the textual traditions and arguments of both thinkers.

NOTES   1 The difficulty is akin to that of liberation discourses having to speak in the language of the oppressor. Derrida has to use such terms “under erasure.”   2 Derrida (1971), p. 217.   3 Derrida (1981) elaborates this notion.   4 Derrida (1971), p. 209.   5 Derrida (1971), p. 209.   6 “Taking words as marks to be basic” is a topic of Derrida (1976).   7 Derrida (1971), p. 209, third paragraph.   8 Derrida (1971), p. 219.   9 I argue this in detail in Wheeler (2014). 10 Derrida (1971), p. 219. 11 Derrida (1971), pp. 219–220. 12 Derrida (1988), p. 123. 13 Derrida (1971), pp. 228–229. 14 Derrida (1971), p. 225. 15 Derrida (1971), pp. 231–232. 16 Derrida (1971), pp. 242–243. 17 Derrida (1971), p. 243. 18 There are two considerations Aristotle gives showing that sensory objects cannot be known by definition. The first, from the Topics V, 3, 131b20–30 depends on observing that sensory objects can pass out of sight and so out of certain existence. This same argument is repeated in Metaphysics Z 1040a2–8. The second consideration, in Metaphysics Z 15 1040a28-b5, is that the sun cannot be defined as the unique bearer of its attributes because it could continue

188  Samuel C. Wheeler III to exist without those attributes. Also, another entity could have had them. If truth and meaning consist of matching a meaning to an object, there cannot be a meaning, in the sense of a repeatable iterable nature that entities instance, which is the meaning of a name. If sensory items are referred to by names, the only way they could refer by having a meaning fitting an object would be if their meaning were a definite description. Precisely the same point, with a similar argument, much elaborated, is made by Kripke (1980). Aristotle and Kripke agree that names are not definite descriptions—that names do not have meanings that could be transferred. Derrida does not mention this stronger consideration by which Aristotle undermines his own account of metaphor. 19 Thus Derrida shows that Aristotle agrees with Sellars (1963). 20 Derrida (1971), pp. 253–254. 21 Alan Bass, the translator, notes that “releve” means both “aufheben” and “taking off.” 22 Thus Derrida rejects the distinction between logic and rhetoric, as Wheeler (1986) argues that Quine should, given his rejection of the analytic-synthetic distinction. 23 I omit discussion of the next part of the section, which investigates the possibility of a metaphilosophical account of metaphor, a kind of description from the outside of how metaphors work. 24 Derrida (1971), p. 270. 25 I have discussed this essay in Wheeler (1989, 2004, 2007, 2011, 2013). 26 The classic exposition is Davidson (1967) 27 Tarski (1956). 28 A more thorough exposition and defense of this conception of semantics is given in Chapter 1 of Wheeler (2014). 29 Davidson (1978), p. 246 30 Davidson’s account is thus akin to de Man’s discussion of Rousseau in de Man (1979). Davidson, given his view that most of what most people believe is true, regards “normal” applications of predicates to entities as true, whereas de Man’s Rousseau treats predication of, say, “frog,” to another frog as a figure. See Wheeler (1989). 31 Davidson (1978), p. 247. 32 Davidson (1978), p. 252. 33 Davidson (1978), p. 263. 34 Davidson (1978), p. 262. 35 In Derrida (1973). 36 Davidson (1967), p. 21. “Postulating meanings has netted nothing.” 37 Although Tarski (1956), p. 155, note 2, characterizes his definition of truth as agreeing with Aristotle’s, Tarski’s is a correspondence theory only in a trivial sense. A true sentence is satisfied by all sequences, which does not distinguish more than one “truth maker” for all true sentences. The only entity corresponding to a true sentence is the set of all sequences. 38 See Davidson (1967, 2000). 39 Davidson (1970), p. 222. 40 Wheeler (2011), for instance. 41 The metaphysical metaphor of the light of the intellect that enables “clear” thinking is more than just a philosopher’s invention. Plato is not inventing the metaphor of the Sun. Rather, Indo-European languages encapsulate such metaphors. In Greek, the normal way to talk about knowing is the “have seen.” In English, “wit” and “wisdom” are cognate with “video.” Perhaps the metaphysical idea is natural to humans—we naturally think of thinking as like seeing, so that the meanings of our words are “before the mind” as objects are before our eyes. If that conjecture is true, however, it does not show

Metaphor without Meanings  189 much. Cultures that have been separated for tens of thousands of years share a conception of nature as populated by gods who control natural phenomena. Likewise, the belief in a soul that persists after death occurs to many human groups that have had no contact. The naturalness in the sense of “natural for humans” is not a compelling argument that some number of gods exist and that there is an afterlife. 42 A suspect genealogy, of course, does not show that a notion is wrong. The discovery of the structure of benzene, after all, was due to a dream. But the ring structure was an accurate model of the benzene molecule. 43 I talked with him on many occasions in the mid-eighties. He was an amiable, reasonable guy, eager to talk to an analytic philosopher who appreciated his work. 44 Derrida (1988), p. 123.

WORKS CITED Aristotle. 1966. Metaphysics. Trans. H. G. Apostle. Bloomington, IN: Indiana University Press. Bataille, G. 1927/1985. The Solar Anus. In Alan Stoekl, ed., Visions of Excess. Minneapolis, MN: University of Minnesota Press, 5–9. Davidson, D. 1967. Truth and Meaning. In D. Davidson (1984), 17–36. Davidson, D. 1970. Mental Events. In D. Davidson (1984), 207–224. Davidson, D. 1978. What Metaphors Mean. In D. Davidson (1984), 245–264. Davidson, D. 1984. Inquiries into Truth and Interpretation. Oxford: Oxford University Press. Davidson, D. 1986. A Nice Derangement of Epitaphs. In Richard Grandy and Richard Warner, eds., Philosophical Grounds of Rationality: Intentions, Categories, Ends. Oxford: Clarenden Press, 157–174. Davidson, D. 1991. James Joyce and Humpty Dumpty. Midwest Studies in Philosophy 16: 1–12. De Man, P. 1979. Allegories of Reading, New Haven: Yale University Press. Derrida, J. 1971. White Mythology: Metaphor in the Text of Philosophy. In J. Derrida (1982), 208–271. Derrida, J. 1973. Speech and Phenomena and Other Essays on Husserl’s Phenomenology. Evanston, IL: Northwestern University Press. Derrida, J. 1976. Of Grammatology. Baltimore: Johns Hopkins University Press. Derrida, J. 1981. Dissemination. Chicago, IL: University of Chicago Press. Derrida, J. 1982. Margins of Philosophy. Chicago, IL: University of Chicago Press. Derrida, J. 1988. Limited Inc. Evanston, IL: Northwestern University Press. Føllesdal, D. 1958. Husserl und Frege: ein Beitrag zur Beleuchtung der Entstehung der phanomenologischen Philosophie. Avhandlinger utgitt av det Norske videnskaps-akademi i Oslo. 2:Hist.-filos. klasse, 1958, no. 2. Oslo: I kommisjon hos Aschehoug. Kripke, S. 1980. Naming and Necessity. Cambridge, MA: Harvard University Press. Plato. 1974. Plato’s Republic. Trans. G. M. A. Grube. Indianapolis: Hackett. Quine, W. V. O. 1960. Word and Object. Cambridge, MA: Harvard University Press. Sellars, W. 1963. Empiricism and the Philosophy of Mind. In W. Sellars, ed., Science, Perception, and Reality. London: Routledge and Kegan Paul, 127–196. Tarski, A. 1956. The Concept of Truth in Formalized Languages. In Logic, Semantics, Metamathematics. Oxford: Oxford University Press, 152–278. Wheeler, S. C. III. 1986. The Extension of Deconstruction. The Monist 69 (1) (January): 3–21. In S. Wheeler (2000), 36–55.

190  Samuel C. Wheeler III Wheeler, S. C. III. 1989. Metaphor in Davidson and DeMan. In Reed Way Dasenbrock, ed., Redrawing the Lines. Minneapolis, MN: University of Minnesota Press, 116–139. Wheeler, S. C. III. 2007. Wahrheit, Metapher und Unbestimmheit. In F. J. Czernin and T. Eder, eds., Zur Metapher. Berlin: Wilhelm Fink Verlag. Wheeler, S. C. III. 2011. Davidson, Derrida and Difference. In J. Malpas, ed., Dialogues with Davidson. Cambridge, MA: MIT Press, 29–42. Wheeler, S. C. III. 2013. Davidson and Literary Theory. In E. Lepore and K. Ludwig, eds., Companion to Donald Davidson. Malden, MA: Blackwell, 371–391. Wheeler, S. 2014. Neo-Davidsonian Metaphysics: From the True to the Good. New York: Routledge.

Part III

Metaphysics and Ontology

This page intentionally left blank

9 Why Is Time Different from Space? John McCumber

It is often held that a proper Aufhebung of the analytical/continental split would apply the clarity and rigor of analytical philosophy to the kind of urgently relevant issues favored by continental philosophers. Both rigor and relevance are necessary for good philosophy, but the following observations suggest that there are limits to the Aufhebung model: (1)   That time and space differ from one another is perfectly obvious. (2)   Just how they differ is somewhat harder to state. (3)   And why they differ is a question so bizarre as to look misshapen. Observations (2) and (3) above comment on questions that I will call “2’ ” and “3” respectively. As my title indicates, I am concerned with 3; but it is hardly a standard problem for continental philosophers, let alone their analytical confrères. This means that while I will seek to draw on both traditions, the result will resemble the “neither-nor” of demarcation more than the “both-and” of Aufhebung (see McCumber 2013, pp. 121, 208–210). But is 3 really ill formed? If so, I should end here. If not, why does it look that way? Exploring this issue will be useful because one reason 3 looks so strange is that it is embedded within a thicket of other questions, and indeed paradoxes, in such a way that how we understand it depends on how we answer and disarm them. I will begin by sketching some of these other issues. 1.  PLACING THE QUESTION One of them is suggested by 2: it would seem that how space differs from time needs to be settled before we can ask why. For the former is clearly a matter of description, while the latter is one of explanation, i.e., of giving causes, and it seems only obvious that you must know what something is before you can identify what causes it. Whether this is always the case is dubious, for the wrong explanation of something can skew its description; if I think a spell has made my uncle sick, I will likely describe his symptoms

194  John McCumber in highly misleading terms. In any case, looking at some of the features that have been claimed to distinguish space from time can yield a sense of the further entanglements among the preliminary issues. Kant, to begin with, claims that time and space are “forms” of our faculty of intuition, abstract principles of order that contrast with each other in two ways: (a)   Space is the form of outer representations, while time is the form of all representations. (b)   Space is the ordering of compatibles, while time is the ordering of incompatibles. (Kant 1996, pp. 76–77, 87) (a), it should be noted, does not mean that the content of external representations is temporally ordered: the representation of a set of stationary objects—of a “state of affairs”—does not exhibit temporal order, because all its contents are simultaneous. What is ordered in time is our apprehension of that content. Such apprehension is an act of the mind—of the Kantian “subject”—and as such occurs before and after other acts of that subject; Kant moves, we may say, from the representation to the representing. At least four other obvious differences between time and space have been obvious enough to be noted by philosophers: (c)    Time seems to flows; space does not. (Prior 1967, 1996; Oaklander 1994, pp. 281) (d)    The future is far less knowable than the past and present, but the three dimensions of space are in principle equally knowable (Horwich 1987, pp. 77–90). (e)    We always act for the sake of the future. It is obvious that we do not always act for any spatial dimension (Horwich 1987, pp. 177–198). (f)     Time gives rise to an A-series, while space does not: time can be ordered not only by earlier-than, but also in terms of past, present, and future; space cannot. It has been argued, often and in many ways, that this is not the case: that the A-series is somehow reducible to the B-series, or even that it is an illusion.1 Even if it is, (3) remains: why does time give rise to that illusion, while space does not? These considerations raise two more preliminary issues. (c), (d), and (e) concern experiential differences between space and time, and if we adopt the “illusory” account of the A-series, (f) does as well. What, then, is the relationship between how space and time appear to us and how they really are? Are space and time really different from each other or do they just look that way? In either case, question (3) remains: if space and time only seem to be distinct, perhaps because of some neurological feature or cultural presupposition, then we have to explain why those features and presuppositions are there.

Why Is Time Different from Space?  195 The final preliminary issue is raised by all six differences: what kinds of difference distinguish space from time? Kant’s account assumes that their distinction is relatively simple, in that space and time contrast with one another on only two axes: inner/outer and compatible/incompatible. C. D. Broad makes do with one.2 But what if there are more? What if space and time differ not merely in six, but in a hundred or a thousand ways? Would those differences be organized so that some (even, perhaps, Kant’s two) are basic to the rest? Or would they constitute merely a disconnected string of contrasts? This issue has methodological bearing: if we try to formulate the distinctions, can we classify or must we list? In order not to beg the questions of complexity and organization, I will refer to the set of differences between space and time as the “distinction” between them. We need not pursue these preliminary matters yet, because the present point is merely that the issue of how, and so that of why, space differs from time is embedded in a variety of other, and deep, issues. Understanding why space and time are different may require tackling such other preliminaries as whether their distinction is apparent or real, why representations are sorted into inner and outer ones, why space and time can or must be ordered serially, what action is, and what it means for components of space and time to be compatible and incompatible with each other. In addition to these issues (and there are probably more of them), common sense tells us that a distinction between two things cannot exist unless those two things do, and contemporary physics tells us that space and time did not exist before the Big Bang. Space, time, and so the distinction between them, have therefore all come to be. What comes to be, presumably, does so in time. But how can time itself come to be in time? Here we confront a dilemma. If we say that time did not come to be, then the Big Bang theory is wrong. If we say that it did, then it did so in time, and time already existed. One horn of our dilemma is an empirical falsehood, while the other is a paradox. This dilemma is not tied to Big Bang theory, however. I have elsewhere identified the view that things must be treated as if they came-to-be as a counsel of philosophical prudence (McCumber 2011, pp. 390–391). It is therefore philosophical prudence, not the arcane mathematics and delicate observations of physics, that suggests we choose the paradoxical horn of the dilemma. 2.  DISARMING THE PARADOX This quick tour of the thicket of questions and paradoxes in which the question of why time differs from space is embedded shows us, I think, why that question looks so bizarre. The first step [in answering it is to disarm the paradox. This can be done in a couple of ways, for non-temporal explanations

196  John McCumber of temporal phenomena are possible. If we look back to Aristotle, for example, and in particular to Metaphysics XII.7, we find that the prime mover is wholly outside of time—wholly eternal and so unchanging—yet causes the order in the universe, which takes it for its telos. Aristotelian final causes can thus operate as causes without being in time. If we can say that the difference between time and space has a final cause, then perhaps we can explain why time arose as f and became differentiated from space in terms of some atemporal order of finality. This way of disarming the paradox does not presuppose the existence of time, but runs up, to be sure, against a discrepancy between our modern views and Aristotle’s ancient one. Aristotle located final causes within nature (physis) and thereby dissociated them from consciousness, but modernity has restricted them to the conscious operations of minds. If the difference between time and space has a final cause, it seems, some being must consciously have taken that distinction as its purpose. What kind of a being could do that? What sort of purpose might it have? Our disarmament verges on theology. We can also disarm the paradox by claiming that time has come to be, not from something wholly atemporal, but from a “situation” (to use the vaguest, and so least committal, term) that, without being fully temporal, is sufficiently time-like to contain comings-to-be. One of these comings-tobe, we could argue, produced what we know as time. I will accept this and argue that it produced space as well, though differently, and so is responsible for their distinction. In characterizing this situation, I will take as a clue the general truth that Plato called “antapodosis”: nothing comes to be from itself, for if it did, it already existed when it supposedly came to be (see Phaedo 70e-71e). This applies elegantly to our paradox and suggests that the distinctness of time and space must have been preceded by a situation in which they were indistinct. Let is call this situation of indistinctness “α.” The question now is how to characterize α. 3. INTO THE (PHENOMENOLOGICAL) CENTER OF THE THICKET As a second clue to he nature of α, I will take it that we have experiences of space and time.3 They are not, to be sure, sensory objects the way lions and doorknobs are; that is why Kant could designate them as “forms,” rather than objects or contents. But they are, somehow, aspects of how lions, doorknobs, and other such objects (or contents) are perceptually arranged. This means, as it did for Kant, that even though space and time are not objects of experience in any straightforward way, we know some things about them through experience. We can check our statements about them as we check our ordinary experiences of sensory objects: by testing what we say against our experience. If experience can teach me that a working doorknob cannot

Why Is Time Different from Space?  197 be sixty feet across, it can also teach me that if item A in my visual field is horizontally smaller than item B and to its left, it cannot also be to B’s right in my visual field.4 Space and time are thus known, or known about, empirically. But what about α? Is it something that is given in our experience, like space and time (and doorknobs)? If so, we would have α as a kind of experience in which space and time are not distinct; they would fall out of α under certain circumstances, which the explanation would have to specify. If not, then α, if it is to be known at all, must be something that we infer from experience. In that case, α would be like the subatomic world, in that it would be a region that we cannot experience, but which is necessary to account for what we do experience. This possibility can never be ruled out; new empirical properties of space and time can always be discovered that require appeal to non-empirical entities to account for them. But we should begin empirically, for only so can we formulate the distinction clearly and check our explanation of it. I will, then, look first to see whether an empirical account can be given of α. Is there a way of experiencing things in which space and time are indistinct? This explanatory strategy, note, does not commit us to viewing space and time as subjective phenomena. It says only that we must begin understanding their distinction from our experiences of them. Even if we succeed in explaining that distinction without appeal to anything beyond experience, it will remain open whether space and time, as “subjectively” experienced, correspond to features of an “objective” world. “Objective” space and time are not denied but bracketed, and this brings us to phenomenology—not the eidetic transcendentalism of the classic Husserl, but to a humbler approach in which we try to describe α, i.e., try to describe what it is like to experience things without experiencing them in terms of space and time. This approach itself has a unique temporality that is often misunderstood: the crucial descriptions are not the ones I give here, but the ones readers (if any) will give when they apply the concepts I introduce below to their own experience. The humble phenomenologist (as opposed to the transcendental kind) does not describe, but suggests, concepts for future descriptions; those descriptions come later, with the readers. This humble approach uncovers the shoulders of some giants, for the phenomenological tradition, beginning with Heidegger and Merleau-Ponty, has in fact explored such a way of experiencing. Heidegger treats it as what he calls “concernful circumspection” (umsichtig Besorgen, Heidegger 1962, pp. 98, 135–148), and Merleau-Ponty in terms of what he calls “motility” (Merleau-Ponty 1945, pp. 324–344). But Heidegger and Merleau-Ponty need not become chief occupations, for Edward S. Casey has carefully and critically appropriated the lessons of their work. I will therefore be chiefly occupied with his investigations of what he calls “place” (Casey 2009, hereinafter GBP).

198  John McCumber 4.  THE UNITY OF PLACE As Michael Treisman has pointed out, our ordinary experience of space and time is not Newtonian: We do not experience ourselves as phenomenologically located at a point in an infinite homogenous space extending equally in all directions. Rather, we perceive ourselves as contained in a bubble of space whose extent is delimited by our viewpoint and the reach of our senses, and which moves and adjusts itself with us as we peregrinate through our personal worlds. (Treisman 1999, p. 220) Place, this moving and adjusting bubble, has, at this point, the status of a candidate for α. If it is a successful candidate, α will be an experienced realm, as opposed to an inferred one, in which space and time are somehow indistinct and out of which they can be shown to fall under specifiable circumstances. I will argue here that Casey’s more detailed phenomenological account of place, with some modifications, meets these criteria and so coincides with α. One of Casey’s central points is that place, insofar as it is experienced by us, is experienced through our bodies (GBP 26, 48). And, as Casey pursues the embodied experience of place to its phenomenological roots, it becomes obvious that the body’s relation to place is not that of a passive observer: we experience place through our body’s active and complex “emplacement” of us. The body does this, Casey shows, through a variety of binaries, which are in turn derived from the binary structure of our bodies: front and back, left and right (GBP 48). We may put the matter this way: because I have eyes and ears on both sides of my body, sensory messages reach me twice: what my right ear hears and my right eye sees differs from what my left ear and left eye respectively sense. The small differences between when a sound hits my left ear and when it hits my right ear tell me whether it comes from my right, my left, or from directly in front or back of me. The ocular unification of the slightly different messages delivered by my left and right eyes gives me depth perception.5 This sort of unification places objects in a particular region with respect to my body: it “localizes” them. Localization as just described does not suffice for emplacement, however, for these diverse localized sensory messages do not establish me within a unified place. It is one thing to say that the door is to my left and the couch is to my right, and another to say that I am here between the door and the couch. For emplacement to happen, the multiple sensory messages already localized must somehow be combined into a single surrounding environment. Neurological details of how this happens need not concern us; we are looking for the principle by which unified places recurrently come about. There must be such a principle, it seems, for a random set of unifications

Why Is Time Different from Space?  199 could not lead to any sort of enduring unity, and places are relatively stable. Moreover, since the difference between a unified place and a mere set of localized messages is itself perceived, this principle must be an empirically evident one. Casey’s account of it is stunningly original. The place we occupy contains things at varying distances from us, and, as Casey argues, the basic formula for such distance is Here and There (GBP 55). The concrete information I receive about what is in front of me and what is behind, what is to my left and right, is thus brought together into a unified sense of place, not by a single principle that applies to all messages (such as relative distance from my body), but by a binary that cuts across them and so applies to all. 5.  HERE/THERE AS A UNIFYING BINARY What distinguishes an enduring place from a mere congeries of sensory messages is the here/there dyad, and to do this, that dyad must for Casey have a number of characteristics. It must be absolute, in the sense that hereness and thereness are ultimate and so cannot be increased (as Casey notes, there are no comparatives in English for “here” and “there:” GBP 57); exclusive, in that that nothing can be both here and there; and exhaustive, in that everything I perceive is either here or there: The sectors of here and there, taken together, are coextensive with the experiential field as a whole, leaving no remainder. Everything in the field, even the minutest details, is designatable either as “here” or as “there.” (GBP 55) Here/there is thus not in fact a mere “dyad” but what I will call a “binary.” It governs its field by dividing that field exhaustively and exclusively; and because it is absolute, it cannot be deconstructed or otherwise gotten beyond. The characteristics of absoluteness, exhaustiveness, and exclusivity are not adventitious. Exhaustiveness and exclusivity are both required for the unification of various perceptual messages into a unified place to be totalistic. If here /there were not exhaustive, there would be perceptions that did not come under its rubric at all, that were neither here nor there. They would not be integrated with the rest of the perceptual field, and so it would not be wholly unified. Similarly, if it were not exclusive, there would be perceptions that were somehow both here and there. They would then belong to the perceptual field in virtue of some other characteristic than of being sorted by the binary, which in that case would not be doing its unifying job. Finally, if the binary were not absolute, we could make heres here-er, and theres there-er. Here and there would not be basic to place: some deeper principle or activity, the one that sets their degrees, would govern the binary’s differing deployments.

200  John McCumber 6.  PROBLEMS WITH THE HERE/THERE BINARY The here/there binary thus seems to function, as Casey points out (GBP 58), as a sort of corporeally grounded version of a Kantian transcendental. Like Kantian categories such as cause and effect, here and there can apply to anything whatsoever that I perceive, and applying them is necessary to my perceiving of anything whatsoever. But Casey’s binary, unlike Kant’s, is not rooted in the necessary structures of the mind, or “faculties”; rather, it grows from the constitution of the body. This is a salutary emendation to the great German, for his treatment of the faculties has long been recognized, beginning with Friedrich Schiller in 1801, as one of the most problematic parts of his philosophy (see Schiller 1967). But does it work? Can appeal to the body show the origin of here/there? In fact, I suggest, Casey’s own analysis shows that it cannot. First, as he points out, the body itself cannot be stabilized as either here or there. As he puts it: Wherever the exact center of the here may be, when localized in this discrete way it is capable of being opposed to another part of my body taken as there: say, my feet. (GBP 52) If my feet can become there, so can the rest of me: even the area behind my eyes, where I usually “locate” myself (GBP 52), can become a there—when I have a headache, for example. But how a something that can become there found an absolute here? Moreover, anything experienced must, in virtue of exclusivity, be either here or there. Since the body is both, it cannot be a “thing”; it must somehow be fragmented, and indeed constantly fragmenting and refragmenting as body parts that are here (such as my headache) become there and vice versa. What is here and what is here is thus a shifting matter: The more I try to pin down the here—to grasp it as a this here—the more I find it dissolving between my philosophical fingers. . . . My body and it here alike—even my body taken as an “absolute here”—are designated by “shifters” that, despite their univocal meaning, are empty in their semantic cores. (GBP 70). The body’s role in sustaining the unity of place via its status as the absolute “here” also undergoes, Casey notes, a number of other breakdowns (GBP 51). Some of these are unusual states of consciousness: mystical experiences, disorientation, euphoria, fugue, emotional stress. Some are pathological: Korsakoff’s syndrome, temporal lobe epilepsy, Parkinson’s disease. Others are not so unusual, such as distraction, to which I will return, or the situation of “low tension” between here and there in which they seem to be continuous with each other (as when I automatically go to the shelf, where

Why Is Time Different from Space?  201 I know a book I need is waiting: GBP 56). Still other failures of corporeal emplacement are standing features of it, which means that they are not “breakdowns” at all. Following Whitehead, Casey argues that the body has a non-simple location, in that it is never wholly within one specifiable locus; it is always already on the way to somewhere else and is “a moving locus that relates in a constantly changing way to its various theres” (GBP 65–66). The here constituted by the body is thus always already impeached, and the body is Not a simple source but an origo that is always already split between taking its own stance as definitive for orientation and taking its cues from the environing world. (GBP 87–88) The body, I suggest, is too fractured to be responsible for the absolute, exhaustive, and exclusive binary of here/there binary. But if the absolute hereness and thereness that unify place thus cannot be due to my body, where might they come from? In Casey’s account, the body constitutes place as the agent of the ego. Casey speaks, throughout his discussion, of the body as where I am—“It is by my body—my lived body—that I am here” (51; see also the opening paragraphs of the discussion on p. 50). And as Casey treats it, this ego, while embodied, remains distinct from the body: in Stephen Erickson’s words, it is in but not of the body (Erickson 1999). Their separateness is evident, for example, when Casey takes up Husserl’s doctrine that I cannot have knowledge of the mental lives of others: that the egos of other people are, cognitively speaking, “actively resistant” to mine (GBP 53–54). If people’s bodies as we observe them tell us, in principle, nothing about what their minds are doing, then, as far as we can know, the mental and physical happenings in a given individual have nothing to do with one another. Each domain can then be explained only out of itself, and the ego in particular becomes a sovereign realm in which things happen without affecting, or being affected by, the body (as for Husserl; see McCumber 2011, pp. 141–143). I take it that after Heidegger (and Ryle: see Ryle 1949), such a view of the ego is phenomenologically suspect.6 But if the self-contained ego goes, the exhaustiveness and exclusivity of the here/there binary go as well. For it is the ego, not the body, that contributes them. 7.  NEAR/FAR AS A UNIFYING BINARY The contribution is unneeded, for there is in fact no warrant to describe place in terms of a here/there binary that is asserted to be absolute, exhaustive, and exclusive. The description can start from another binary that Casey also discusses: that of near and far. Strictly speaking, near and far is not even a dyad, much less a binary, but a continuum; unlike here and there,

202  John McCumber near and far are matters of degree (GBP 57). Nor is it exclusive: the same thing can be both near and far. An example of this is depth, “the indeterminate envelopment of nearness by farness and of farness by nearness” (GBP 69). An example of depth, in turn, is a pathway: its beginning with respect to us is near, while its end is far (GBP 60). It makes no sense, then, to view pathways as “primarily spatial”; what matters in them is the amount of stretching or reaching one has to do to traverse them, and this is a matter of space and time indifferently: the grocery store is both half a mile and a 10-minute walk away. Depth itself, the mutual enveloping of near and far occurs “in accordance with the bodily actions that bring one into the vicinity of the other” (GBP 69). What makes a bodily movement into an action is, traditionally, the presence of an intention or purpose, so what is near and what is far, as subject to depth, are defined in terms of my actions and purposes. The near in general is just what I can reach (GBP 57) and what I want to reach (GBP 60)—it is a function of my body as desirous (GBP 59), where what I desire defines the project I am currently engaged in. Projects and desires, moreover, are culturally and historically inflected (for which see GBP 64)—very few people, a hundred years ago, had either the desire or the project of learning how to drive a car. The near and the far, as culturally determined, are not imposed by my individual body in any quasi-Kantian “transcendental” way. They differ not only from moment to moment but historically and culturally; they are what Casey calls “shifters,” but without the overarching formal matrix of here and there. These constant changes mean that what is near is constantly becoming far and what is far is constantly becoming near. Motion toward and away from me—approach and retreat—is thus a fundamental feature of place. 8.  THE DYNAMISM OF NEAR AND FAR It does not have that status on Casey’s account, where the exclusivity of here and there means (as we saw) that anything I perceive must be either here or there and cannot be both. Though here and there cannot, as we saw, be defined by their content, they serve as formal loci for that content—all of it: given exclusivity, anything in my perceptual field must have, we saw, first and foremost a stable location either within the here or within the there. There is thus a strange temptation to say that what is there cannot become here, and what is here cannot become there. Casey does not succumb to this temptation, but Husserl, whom he quotes in a footnote, skirts it when he phenomenologically envisages motion as, first and foremost, rotation around a stationary embodied ego: All worldly things here for me continue to appear to me to be oriented around my phenomenally stationary, resting organism. That is, they are oriented with respect to here and there, right and left, etc., whereby a

Why Is Time Different from Space?  203 firm zero of orientation persists, so to speak, as absolute here. (GBP 387–388n26) Or, as Husserl puts it in Cartesian Meditations: [Nature for me] is constituted as an identical unity of my manifold modes of givenness—an identical unity in changing orientation around my animal organism (the zero body, the body of the absolute Here). (Husserl 1960, p. 123, also quoted at GBP 387–388n26.) The here/there binary thus applies, in the first instance, to things that are not moving toward or away from me. The most basic kind of motion they appear to have is rotation around my body, as the “body of the absolute here.” Approach and retreat are derivatives of this. This means that, from the point of view of the here/there binary, things are classified by their enduring location: they are primarily spatial and not temporal. This amounts to a privileging, by the here/there binary itself, of space over time: as Casey puts it, the here/there binary is “primarily spatial” (GBP 58). It is wise, however, if we are seeking a situation where space and time are indistinct, not to privilege either of them from the start. When approach and retreat become basic to near and far, motion ceases to be derivative. Rather, it becomes basic; what is derivative is perceiving something as being at a fixed distance from me, whether because it is stationary or because it is rotating around me (i.e., moving across my visual field). I find phenomenological warrant for this reversal of priority in the fact that we find our attention drawn automatically to things approaching or receding from us, while we have to fix our attention to see stationary objects.7 The priority of approach and retreat is presumably associated with their connection to the most basic phenomena of life itself: pursuit and avoidance. We could thus say that place is, in the first instance, a dynamic set of approaches and retreats that are determined as such with respect to a sentient body. But approach and retreat, as well as pursuit and avoidance, are not appropriate terms for what I am after, for they all denote actions. All four thus imply a distribution of active and passive such that one set of entities is doing the approaching/retreating/pursuing/avoiding, while another set is undergoing them. A full account of why this is wrong will have to await my discussion of what I call “co-occupancy” below; but wrong it is, and so I introduce the terms “nearing” and “faring.” To say something “nears” me leaves it open whether I am moving toward it, it toward me, or both. In a case of nearing, the distance separating a number of bodies decreases, whether one of those bodies, some of the others, or all of them are doing the moving. Similarly for “faring.” Nearing and faring so defined have, if we reflect on them, both spatial and temporal dimensions: to “near” is to approach in space over time; to “far” is to recede in space over time. But in nearing and faring, space and

204  John McCumber time are indistinct, in that neither can be experienced without the other. If they were given as distinct, place would be given as some sort of compound of space and time—a compound that could not become place, as defined here, until we had also, and mysteriously, introduced motion. Rather, by taking motion, in the forms of nearing and faring, as empirically basic, we will be able to show how and why space and time fall out of them. Place, characterized as a set of nearings and farings, rather than as a moving “bubble” or as a compound of here and there, reveals itself to be the α we have been seeking. 9.  NEARING/FARING VS. HERE/THERE Not everything, to be sure, is nearing and faring—some things are indeed experienced as rotating around me, and others as stationary. And something can both near and far at once—as when I perceive how long a pathway is as I step onto it. Unlike here/there, nearing/faring is thus neither exclusivistic nor exhaustive. But absoluteness of a sort is preserved, in a way that shows us how here and there derive from nearing and faring: what is “here” is just what is at the end point of a nearing, and what is “there” is at the end point of a faring. What is here cannot get any nearer, simply because it is at the end point of its nearing; similarly, what is there is at the end point of a faring and cannot get any farer. The here itself, as a distinct region, is then the end point of all nearings. Because it is the end point of all nearings, it is not given empirically—but neither need we bestow on it the status of a Kantian a priori. We can say, more humbly, that it is “imagined.” The there is, correspondingly, the imagined end point of all farings. The unity of place, insofar as it is dependent on the here/there binary, is then imagined as well. It is, of itself, not enduring but intermittent, existing only when it is imagined; not all places, in other words, are enduringly unified. What occasions this imagining is (as we will see) a certain sort of convergence among the nearing and farings that constitute my place. The unity of place, when it appears, is thus grounded in nearings and farings. We thus do not have to accept here and there as “transcendentally” basic, or as grounded in some foundational Husserlian ego, but can give an empirical, phenomenological account of their origins. Place, considered as the set of nearings and farings relative to a given body, provides the experiential basis both for overcoming the here/there binary’s privileging of space over time and for avoiding its suspect appeals to Husserlian “transcendental” egos. Whether my account actually captures our experiences is humbly left to the reader. 10. CO-OCCUPANCY The question of why time differs from space can now be phrased as: how is it that nearing and faring give rise to the two distinct forms of space

Why Is Time Different from Space?  205 and time? To begin answering this, we can note that there is one aspect of place that is not considered by Casey (or, for that matter, by Heidegger or Merleau-Ponty in the places cited): that any given place is normally co-occupied, i.e., shared with others. Something “co-occupies” a place with me when I experience a number of nearings as converging on it. This was not the case on Casey’s account: because I am always here, and because the here/there binary is exhaustive and exclusive, there can be only one basic here in my perceptual field, the one where I am; the heres of others, as we saw, are opaque and, in Casey’s words, I necessarily “rely on my body as the primary agent in the landscape” (GBP 17). But near and far do not have that kind of absoluteness. Objects in my perceptual field can be seen to near a given thing without that other thing having to be conceived as invested with an ego that is like mine but opaque to it. When I see ants converging on a fallen cookie, for example, I do not have to attribute an ego to the cookie in order to understand what is going on. But the cookie is, nonetheless, a “co-occupant” of my place. If places are co-occupied, I am not the “primary agent” in them, partly because I am not primarily an agent at all: as suggested above, things near and far me as much as I near and far them. The situation, we may say, is Einsteinian: if nearings and farings are the basic constituents of place, all that matters is whether the distance between bodies is increasing or decreasing. The cookie nears the ants as much as the ants near the cookie. To say that one body is at rest while the others move is a step beyond, requiring the imposition of an absolute metric based on the Husserlian “absolute here.” In a co-occupied place, nearings and farings are with respect to multiple centers, none of which is privileged. Agency in the place belongs equally to others, who are the centers of their own nearings and farings. I am not the single center of a co-occupied place. Co-occupancy is normal for humans. In an absolutely private space, as in a private language, I would be solitary at the center; all nearings and faring would be relative to me. Not only is that usually not the case, but even approximating it—as does solitary confinement—is so unnatural as to be immediately threatening to sanity.8 11.  A LITERARY EXAMPLE OF CO-OCCUPANCY As an example of co-occupancy and how it works, let us consider a passage from Graham Greene’s The Honorary Consul (Greene 1973). Doctor Eduardo Plarr, a South American physician in early middle age, is visiting a brothel with his novelist friend, Doctor Saavedra. I quote the passage at length: An airy patio about the size of a tennis court was surrounded by small cells. . . . Two open doors faced him, when he had taken a seat . . . Each possessed a little shrine with a lighted candle which gave to the

206  John McCumber tidy interiors the atmosphere of a home rather than of a place of business. A group of girls sat at a table apart, while two talked with young men, leaning against the pillars of the verandah which surrounded the patio. . . . One man sat alone over a glass, and another, dressed like a peón, stood by a pillar, watching the girls with an unhappy, envious expression, (perhaps he hadn’t the means to buy even a drink). A girl named Teresa came immediately to take the novelist’s order (“Whiskey, he advised, “the brandy is not to be trusted”), and afterward sat down with them unasked. ”Teresa comes from Salta,” Doctor Saavedra explained, leaving his hand in her care like a glove in a cloakroom. She turned it this way and that as though she were looking for holes. “I am thinking of setting my next novel in Salta. . . . For the first time,” Doctor Saavedra said, “I am proposing to write a political novel.” “Political?” Doctor Plarr asked with some surprise. A cell door opened and a man came out. He lit a cigarette, went to a table, and drank from an unfinished glass. In the glow of the light, below the saint’s shrine, Doctor Plarr could see a thin girl who was straightening the bed. She arranged the coverlet with care before she came out and joined her companions at their communal table. An unfinished glass of orange juice awaited her. The peón by the bar watched her with his hungry envy . . . “I was telling you about my political novel.” Doctor Saavedra spoke with irritation.n(Greene 1973, pp. 55–57) No fewer than eleven people, their locations, and their movements, are described here with extreme precision. By contrast, no mental contents—no thoughts, beliefs, emotions, desires, or the like—are even mentioned with the single exception, in the last line, of Dr. Saavedra’s irritation. But even that is presented as heard, rather than inferred; it does not lie behind his words, in some egological realm, but accompanies them (“with irritation”). The positions and movements of the bodies tell us directly what the minds are doing. It is not necessary to say that Dr. Saavedra’s hand, “in the care [of Teresa] like a glove in a cloakroom,” manifests his negligence and her perfunctory concern; it is part of those. Similarly, the stillness of the peón, together with his fixated gaze, constitutes his impotent desire; all that remains is to speculate on is the causes of his impotence. Contra Husserl and Casey, then, there is no intrinsic opacity of one mind to another. 12.  SCRIPTS AND THE OPACITY OF OTHERS This is not intended to deny that other people are, occasionally, mysterious to us. There is a lot we don’t know about the physiology that makes ants converge on a cookie, but we would hardly call their convergence

Why Is Time Different from Space?  207 mysterious. The fact that some convergences are mysterious suggests that more is going on with nearing and farings than mere change of location. We can explore this by looking at one mysterious person in the passage from Greene: Clara, the thin girl. She comes out of her cell, which in terms of place constitutes a nearing to the seated Eduardo: the distance between their bodies decreases. But this nearing exists only for Eduardo: Clara goes back to her companions and her orange juice without even noticing him (she will not recognize him when they meet again: Greene 1973, p. 69). From her point of view, Clara has no more neared Eduardo than she neared the pillars of the verandah. So there must be something to nearing (and faring) over and above physical movements of bodies. What might this be, if not a Husserlian ego? Clara’s actions, from making her bed to returning to her orange juice and her friends, are routinized: she is following what we may call a “script.” A “script” in this sense is a familiar sequence of actions and events. It is because they are familiar that scripts allow us to read the significance of a person’s (or an ant’s) actions directly from those actions themselves. Thus, because we know the scripts that concern gloves and cloakrooms, we can tell that Dr. Saavedra’s leaving his hand in Teresa’s “like a glove in a cloakroom” is part of his negligence of her. A script may be followed mindlessly, merely from habit, such as making a bed and walking back to your orange juice. Or it may be followed as a conscious project, as with Eduardo’s conversation and girl-watching. In both cases, scripts come under what Heidegger calls “contexts of involvement” and embed in the overarching teleology of world (see McCumber 1999, pp. 227–229). As is already clear, I eschew such wider embedding, and have elsewhere referred to scripts as “parameters” (McCumber 2005, pp. 145–153). Each script, moreover, has its own nearings and farings. “Display yourself” nears the clients; “rejoin your companions” nears the girls at the table; and “return to your orange juice” nears the glass on it. But not all scripts need involve an overall change of bodily location; Teresa, examining the hand of Dr. Saavedra, nears it in accordance with a script that requires her to demonstrate solicitude for him—but its distance from her does not decrease. The peon, too, does not move, and no one approaches him; but his facial expression as he watches the girls shows that he is nearing them in a very intense way. What makes Clara mysterious is that she is following two, or perhaps three, scripts at once. First and most obviously, she is following the script “rejoin your companions,” which includes various rules for walking over to them and resuming her conversation. We may also want to say that she is following a second script, “get back to your orange juice,” depending on whether she views drinking the orange juice as merely part of being with her companions or as a separate activity; we do not know which is the case. And third, she is a prostitute and her job is get potential clients to notice her (indeed, she will go back to her cell with the peón). She is—whether

208  John McCumber she wants to or not—therefore following the script “display yourself for potential clients.” Part of the fascination her movements arouse, not least in Eduardo, is that this last script seems no more important to Clara than the other two: her potential clients matter no more to her than her orange juice. Eduardo, too, is following a plurality of scripts, specifically two: “conversing with a friend” and “watching a young woman.” The fascination here, for the reader, is that he does not appear to be following the most obvious script: “deciding which woman to take to bed.” 13.  SCRIPTS, EGOS, AND THE UNITY OF PLACE In the situation as Greene has described it, there is thus something more than bodies and their movements and, indeed, something more than bodies and the scripts they are following. This “something more” is not a mysterious set of minds or egos, however, but resides in the fact that any given body may be following more than one script at the same time. The opacity of others thus does not require the postulation of an ego over and above their bodies; all it requires is that the movements of a given body be compatible with a variety of scripts. When we do not know what relative priorities a particular body gives to its various scripts, that body becomes mysterious. The open evidence of bodies following scripts then replaces Husserl’s problematically opaque ego. The scripts in which all of the characters in this scene except Clara and Eduardo are engaged are so evident from the descriptions of their bodies that the characters are, we might say, one with their bodies. Clara too is one with her body, and so is Eduardo, but their movements are ambiguous in that they are consonant with more than one script. In Clara’s case, we cannot see which is more important. It is because of this that we must speculate about what is in Clara’s mind; she will remain mysterious to Eduardo even after he has fallen in love with her and gotten her pregnant. Eduardo, for his part, is following two scripts when he ought to be following a third. Only they, then, are mysterious enough to have “minds.”9 We see that the nearings and farings that constitute a given place are functions, not only of corporeal location and movement but also of the scripts in play. Indeed, we may say that more basic to scripts than the nearings and farings of physical movement is the nearing/faring immanent to the script itself: as we work through it we approach its last line and distance ourselves from its first. Every script has an archê, which elicits the first motion made when a person engages in it. And it has a last line, a telos to which it moves step by step; a script thus has the basic structure of what Kant would call “internal teleology” (Kant 1987, pp. 244–247). If a place is determined by its nearings and farings, and if these in turn are provoked by the scripts in play, then for two people to be in the same place they must be engaged in the same script (though they may be playing different roles in it). Eduardo and Saavedra are both engaged in “conversation

Why Is Time Different from Space?  209 with a friend,” and to that extent they are in the same place. But Eduardo is also engaged in a different script, that of watching a girl, and to that extent his place is different from Dr. Saavedra’s. With Eduardo and Clara, the situation is more complex: they are both engaged in the same script of “choosing a woman,” insofar as she is displaying herself to him. But she is engaged, and perhaps solely engaged, in “rejoin your companions” and “return to your orange juice,” and to that extent she and Eduardo are not in the same place at all. Place, as a set of nearings and farings, does not have the kind of enduring unity that the here/there binary claims to give it. My overall “place” is a function of the various nearings and farings that are going on around me, and these in turn are functions of the various scripts I am following 14.  HOW SPACE IS DIFFERENT FROM TIME When we drop the oddly “transcendental” here/there binary from our account of place, we can conceive of it as fundamentally co-occupied. Co-occupancy entails that place contains two sorts of nearing: those that are nearing me (“egocentric” nearings) and those that are not. Time, I will argue, emerges from the former of these, space from the latter. At least in one sense they do. For the birth of space and time from place is, we will see, a two-step process, resulting in two different kinds of each. In the first, minimal step, space and time become distinct, but do not leave place behind: they are identified as distinct empirical aspects of place. These first versions of space and time will provide, in turn, the materials for the “imaginative” construction of the space and time in the senses given them by Kant (and so many others), which “purifies” them by separating them from each other and so from place itself. The first version of space, empirical space, arises from nearing. Any script I follow, I noted above, is a case of nearing. What it nears most basically is its own last line, which is a state of my being—the state of having completed that script. It is in this sense that Heidegger says that every context of involvement ends in a Seinsmöglichkeit, a “possibility of Being,” of Dasein (Heidegger 1962, p. 116). Sometimes, however, the nearing of a script is averted. I may, for example, become aware of a body moving according to a script that, when consummated, actualizes a possibility, not of my being, but of someone else’s. Such averted nearings can be called “allocentric.” Clara’s pursuit of her orange juice, as observed by Eduardo, is an example; its consummation has nothing to do with him. Allocentricity plays a role in one way that space “falls out” of place.10 Greene gives an example of how this can work. When Eduardo notices Clara, we expect him to be following a script. That script (“choose a woman to take to bed”) is a common one in a brothel. It begins with Eduardo nearing Clara by noticing her; this (according to the script) is the first step

210  John McCumber towards physical intimacy. She, however, is not following Eduardo’s script; her script is allocentric to his, as she shows by going directly to her orange juice and her colleagues. She remains over there, at a distance, and Eduardo is pulled back into his conversation with Doctor Saavedra. It is in ways like this that allocentric nearings give birth to space, to the experience of things being at a distance. We may characterize such space, indeed, as absolute distance—not in the sense of infinite distance, but in that of distance that cannot be overcome because nearing is averted. When an overall place contains nearings that are not converging on me, space “falls out” of place. This, of course, is the usual case: since space is normally co-occupied, some things are usually nearing me, while others are nearing other people. If space arises from allocentric nearings, then time, as its empirical counterpart, arises from nearings that are not averted, which means that they are consummated: the object that is nearing me actually reaches me. When such a nearing is scripted, it also reaches me in the sense that it reaches the internal telos of that script, as what we have seen Heidegger call a “possibility of my being.” We may characterize the first version of time, empirical time, as the process of consummation itself, by which nearings coincide with me physically. 15.  SPACE, TIME, AND SUBJECTIVITY So considered, time and space are both purposive—but not theologically so. Time is, in Kant’s sense, intrinsically purposive because it presents the achievement of a telos—of the possibility of my being, which is the internal telos of a script. Space, by contrast, is extrinsically purposive: things are presented spatially, not because it is their inherent nature to be so, but because it is useful for us to become aware of the aversions around us at any moment. The sensory messages that strike my left and right ears or eyes exhibit such external purposiveness. In and of themselves, they are just waves propagating somewhere; it is my sensory organs that make use of them and give them a purpose. Empirical time and space are thus explained in terms of purposes, as suggested above, but the purposes involved are not theological. Nor are they part of a timeless order, as is the humanity under the moral law that Kant seeks to instate as the purpose of nature (Kant 1987, pp. 312–323). They are human. But they are not subjective, for they generate the “subject” itself. We can see this by returning to Kant’s formulation of the contrasts between them: (a)   Space is the form of outer representations, while time is the form of all representations. (b)   Space is the ordering of compatibles, while time is the ordering of incompatibles. 

Why Is Time Different from Space?  211 Kant draws both contrasts, I suggest, with an absoluteness that verges on dogmatism: representation for Kant is either inner or outer (or as Casey might say, here or there), without degrees. This absoluteness is hardly unique to Kant: it is very much in the spirit of Aristotle’s definition of “limit” (peras) as the first point beyond which it is not possible to find any part, and the first within which every part is. (Metaphysics V.17, 1022a4–6) Limits, including the limits of the mind investigated in Kantian critique (Kant 1996, p. 8), are thus what I call exclusivistic. Applying this view of limits to space and time is to view them as the sides of a dichotomy, rather than as the terminal points of a continuum. In such a view of time, nearing is abstracted away, and the result is a second sort of time—a “pure” time that contains only the final state of consummation. This is a time without space, and so in itself a disembodied, or as I will say an “imagined,” phenomenon. Space in the Kantian sense is then the complement of this: an imaginary domain in which all nearings (and so all farings) are allocentric—a realm in which nothing nears or fars anything and so, at the limit, an external domain of stationary objects. “Imaginary” here has a special meaning. Where we actually live is place, the experienced set of nearings and farings where aversion and consummation, and so space and time, are both present to various degrees. It is possible for us to be, certainly for short periods, in a place where nothing is nearing, or where all nearings are egocentrically consummated. But to take such exceptional circumstances as basic is to picture something beyond experience, to imagine something exceptional to be the standard case. Since pure time is a set of consummations, and since consummation is the coinciding of a nearing with me, such time is, as Husserl recognized, the dynamic substrate of the “subject” itself (Husserl 1960, pp. 73–80). Since nearing (and farings) have been cut out, it is wholly without the possibility of co-occupancy; subjective mind is absolutely private, opaque, and “actively resistant” to the consciousness of others. But, as we saw above, it is also imaginary, because place—and so time—is basically co-occupied. The absolute denial of space to the mind, however congenial to Kant, is subject to the aporias and inconsistencies that Derrida finds in Husserl: “the temporalization of sense [meaning] is, from the outset, a ‘spacing’ ” (Derrida 1973, p. 86). Thus, just as for Wittgenstein there cannot be a private language, in the sense of one for which I make up all the rules (Wittgenstein 1958, pp. 88–96), so there cannot be an absolutely private place—one in which all nearings and farings occur with respect to me. Mind itself can be said to be “spatial” in that the moment thoughts are expressed in language—the moment they become thoughts—they are given over to an externality that I cannot wholly control.11 We submit them to a

212  John McCumber community: they near the others who hear them, and even our private thoughts are allocentrically co-occupied. We would do well, then, to do to Kant what we did with Casey: to replace his “inner” and “outer” with nearing and faring. The more something nears us, the more temporal (and so subjective) it becomes.12 The less something nears us, the more spatial and objective it remains. When we see space and time as termini of a continuum of nearings and farings, rather than as sides of a dichotomy, the four other differences between space and time discussed above can be explained. Because pure time is a succession of consummations that are states of a single subject, they constitute not a set but a single series: they are linearly ordered. When time is defined as a series of consummations, nothing between any two consummations counts as temporal: the flow appears to be continuous. When we see the points in a time series as consummations of processes of nearing and faring, and when those nearings are scripted, we get an A series: the present is the line I am currently executing, the past is the lines I have already executed, and the future is the lines I have yet to execute. The A-series thus does not arise from the B-series; rather, the B-series, Kantan time, is what is left when nearing (and faring) are abstracted from. The pure future is less knowable than the past or present in pure time because when a script is complete, nothing follows on it—except the activation of some other script. The empirical future is not less known than the empirical past because, as I have argued elsewhere, there is always a plurality of paths by which the present may have come to be (McCumber 2005, pp. 35–36; 2013, pp. 203–207); contra Horwich, who suggests that “the past is fixed” but does not argue the point (Horwich 1987, p. 778), there is no one way the past, as far as we can know it, has to be. Finally, we always act for the future because actions have purposes, and purposes are determined by scripts. Actions are script-based, and we act so as to move from the present line to a future one. But if we always act for a future, not all futures are acted for: when a script is complete, the future is unknown, and so we cannot act for it. Thus, in pure time there is no action 16.  WHY SPACE DIFFERS FROM TIME Time differs from space because place, their common origin, contains not only nearings that are consummated, but others that are allocentric. Allocentric nearings give rise to empirical space as their common feature, and consummated ones similarly give rise to empirical time. It is only when aversion and consummation are absolutized, so that one or another of them is all there is, that we come to have pure space and pure time as Kant and others have characterized them. Space differs from time because not everything converges on me.

Why Is Time Different from Space?  213 APPENDIX: ANALYTICAL AND CONTINENTAL APPROACHES TO TIME Viewing time and space as originating in place, and place itself as a dynamic set of nearings and farings, does more than allow us to discuss a single odd-looking question. It provides insights into both literatures, insights that can highlight important features of analytical and continental thought taken more widely. The issues come out more clearly in the case of time than in that of space, so these concluding remarks will focus on that. One striking difference between the analytical and continental literatures on time is the respective weight they give to our experiences of time. Analytical philosophers, inspired presumably by physics, tend to downplay the issue of time as we experience it.13 They begin from various “puzzles” about time and seek, in solving them, to understand time as physicists do: in terms that are mind-independent or “objective.” They are concerned with the issue of whether the results of their endeavors are objective, subjective, or, as Clifford Williams has argued, both at once (Williams 1994a, p. 358). This does not mean, however, that analytical philosophers of time entirely abandon issues of how we experience it. Arthur Prior, who was an analyst, claims that J.M.E. McTaggart, who was not, paid acute attention to “what might be broadly called the phenomenology of time” and that this was one of his singular merits (Prior 1967, p. 1). Evidently Prior, to whom we owe tense logic, thought that understanding our experience of time was important to understanding time itself. Other analytical philosophers confronting the issue of how we experience time rely on the work of their continental colleagues, as when Paul Horwich presents our experience of time by simply repeating Husserl as expounded by Izchak Miller (Horwich 1987, p. 33); and many rely on what are taken to be relatively uncontroversial presuppositions about our experiences of time, two of which will be important here: that we experience time primarily as observers, and that time passes or “flows.”14 These two commonplaces may be connected in a suspicious way: the river I observe may exhibit a smooth flow, but when I jump into in it I find myself buffeted by a congeries of eddies and currents. These appeals to subjective experience on the part of philosophers who aim for objective accounts of time and its puzzles presumably derive from the fact that it is much easier for a philosopher to claim that the propositions she puts forward are about time if those propositions bear some more or less obvious relation to time as we experience it. Absent such a relation, philosophical accounts of time stand to be no more about time than string theory is about strings. The continentals, inspired presumably by phenomenology, tend to begin from our experiences of time. With exceptions such as Hans Meyerhoff (Meyerhoff 1955), they do not so much seek subjectivity as reject the issue of objectivity vs. subjectivity altogether. They seem to view “objective” and “subjective,” not as a conceptually exclusive dichotomy, but as termini of a

214  John McCumber continuum, between which stand any number of intermediate states ranging from sheer idiosyncrasy to universal consensus. This view is related to the peculiar temporality of phenomenological discourse noted above: propositions in such discourse are validated, not when they are propounded, but later, when hearers or readers verify those propositions in their own experience. The more of them do so, the more universal, and so the more “quasi-objective,” the propositions become. Phenomenological statements thus make a sort of epistemic ascent from subjective idiosyncrasy to universal consensus.15 This pull to objectivity presumably derives from the fact that abandoning consensus even as a goal would allow subjective insight full rein; accounts of time would become entirely arbitrary. Both the analytical and approaches thus occupy an uneasy middle ground between giving wholly subjective accounts, which would simply convey one’s personal experiences of time, and giving wholly objective ones that eschew discussion of those experiences altogether and are couched solely in mind-independent terms. While analytical approaches seek objectivity as mind-independence, they cannot wholly abandon appeals to our experiences. While phenomenological approaches can never reach the full objectivity of mind-independent truth, it certainly seeks to come as close to it as its starting point in individual experiences allows. Both, I suggest, run into problems. If analytical approaches are not going to abandon experience altogether it seems that they should pay close attention to the ways in which we experience time and, in particular, to the two presuppositions I noted above. If the continentals or phenomenologists, for their part, want to make their accounts of time as objective as possible, they need to deal with a problem they have long avoided: If a theory of time is more desirable the more human beings assent to it, it should be still more desirable to go beyond human consciousness, to explain how non-human entities are in time. A theory that can be validated only by consensus will have trouble explaining how it can extend to pre-human entities such as those preserved in fossils, and indeed non-conscious beings in general.16 The “idealistic” strain in continental philosophy surfaces here as a refusal to consider such extensions other than as afterthoughts, at least as cursory as analytical accounts of our experience of time. It has been challenged only recently by Quentin Meillassoux, Graham Harman, and other “speculative realists” (Meillassoux 2010, pp. 16–27). Classificatory schemes usually have exceptions—and this one is no exception. There are analytical philosophers, such as Williams, who offer careful construals of our experiences of time, and continentals, such as Husserl, who accept the two presuppositions I have adduced. But if what I have presented here is merely a couple of strands within the analytical and continental literatures, rather than the entirety of those literatures, they are major, indeed classic, strands; and they have not yet, as far as I can see, been coherently challenged. Both analytical and continental literatures have their problems, then; but they also have their uses. To see this, we must understand them in ways that

Why Is Time Different from Space?  215 they are not usually understood. The continental or phenomenological literature does not establish truths about time, but merely suggests them. Their validity does not extend beyond the set of human beings who verify them in their own experience, and we must either complete the phenomenological ascent by another path or leave the fossils themselves. Analytical philosophy, for its part, is often understood, not as producing actual solutions to the puzzles it discusses, but as illuminating possible solutions; for theories are advanced as solutions to problems, and only observation can tell us which of a variety of theories is actually correct. Philosophy being largely bereft of empirical data, it cannot help us decide among competing theories. We should not look to the philosophical literature on time travel, for example, to tell us whether it is possible, because the final decision on that will require a lot of experimentation; but we can look to it to tell us whether a given account of time travel is logically consistent or not, and to clarify the implications and commitments of those that pass this test. Such clarification, operating on an entirely logical level, does not require any empirical data; it treats, as David Lewis suggests, of possible worlds—not the actual one (Lewis). This relatively standard view should be replaced by one that takes account of the fact that many analytical philosophers of time are not, in fact, doing purely logical analysis, but rely on accounts of our experience of time that have certain presuppositions. This makes it important to see where those presuppositions come from. A helpful turn here is to the pre-analytical source of much contemporary analytical thinking about time: J.M.E. McTaggart. McTaggart’s view of our experience of time—of “time as it appears to us prima facie” (McTaggart 1993, p. 95)—is that it contains different positions, which appear to be orderable in two ways: by before and after (the “B-series”) and by past, present, and future (the “A-series”). McTaggart wishes to reject the latter ordering and to retain the former, while denying that it is truly temporal because positions in it never change (if event x is earlier than event y in the B-series, it will stay that way forever: McTaggart 1993, pp. 96–97). Since these are the only options, there is no such thing as a truly temporal ordering of experience; hence McTaggart’s title: “The Unreality of Time.” McTaggart’s motivation for this is not merely to correct various views of Bertrand Russell, as Prior’s treatment suggests (Prior 1967, p. 3). It is, ultimately, to vindicate a mysterious someone whom McTaggart thinks is Hegel.17 For the B-series, or what Prior calls time as tapestry, is nothing other than time as it would appear to God (Prior 1967, p. 48)—or to the Hegelian Absolute. McTaggart’s argument is thus advanced, not in the scientific spirit of the analysts who would appropriate it, but in the service of a monistic ontology according to which: (1)   Reality (the Absolute) never changes in any way. (2)   Therefore the parts of the Absolute do not change.

216  John McCumber (3)   But everything is a part of the Absolute. (4)   Therefore nothing ever changes, and time is unreal. Paul Horwich has argued that McTaggart stacks the deck by formulating the A-series in such a way as to facilitate his refutation of it. His crucial move is to claim that the movement of an event from future to present to past constitutes a “genuine change,” i.e., the acquisition and loss of non-relational properties of futureness, presentness, and pastness; in Horwich’s words, “this claim is implausible and never really substantiated” (Horwich 1987, p. 25; also see pp. 19–21). What I want to suggest is that McTaggart’s characterization of the B-series is also in the service of his “Hegelian” monism, for the B-series, we have seen, is time as it would look to the Absolute or to God time sub specie aeternitatis. All events are ordered by before and after (McTaggart, p. 95) and so positioned along a single line so that any event is earlier than, later than, or simultaneous with any earlier one. Time is thus a totality of events knowable all at once, which is how an “absolute” mind would know it. The two main presuppositions of analytical accounts of time that I identified above thus turn out to have origins in a deeply suspect metaphysics. The idea that we know time through observation correlates with the B-series, in which nothing ever changes and all objects are fixed. The idea that time flows applies to the A-series, which sees that flow as universal and, therefore, homogeneous. The present account is designed to eliminate these metaphysical presuppositions by (a) paying closer attention to our experience than do the analysts I have adduced, and (b) locating those experiences in a wider, non-human realm. The crucial move in this is to see that nearing and faring need not be with respect to us; they can be motions of non-human, and indeed non-sentient, bodies. It turns out that time is not known through observation but through activity. Not always, to be sure, our activity: it is movements in place, nearing and faring, that introduce us to what becomes time when we absolutize it in various imaginative ways. Nearings and farings also do not flow, for the set of them is not unified into a single series. Some, to be sure, are related to each other as earlier, later, and simultaneous; but others are not. (What non-imaginary sense does it make, we might ask with the Einsteinians, to say that Clara’s return to her orange juice is earlier than, later than, or simultaneous with a solar flare on Betelgeuse?) Our experiences of time and space are not ignored, but are argued to arise as the results of a particular, sentient way of coping with nearing and faring. *** One final remark can bring out the full importance of the analytical literature. The relatively thin experiential base we find in much of it probably debars it, I have suggested, from actually solving the puzzles it discusses. Doing so would require choosing among a set of logically consistent competing solutions of those puzzles (tensed accounts of time vs. non-tensed;

Why Is Time Different from Space?  217 presentism vs. growing universe, etc.); and that would require the kinds of robust matters of fact that are foreign to philosophy in general, not merely its analytical varieties. But science, from which this approach takes its inspiration, is supposed to solve problems, not merely state or clarify them. What, one may ask, is the use of discussing issues that cannot be decided? This, I suggest, is what strikes so many outsiders as alienating about much analytical literature. Professor Y writes an article defending theory T by complicating T so as to take account of objections raised by Professor X, an adherent of theory T’. X then writes a rejoinder that complicates T’ so that Y’s emendation of T is unsuccessful. The result is a series of theories of increasing subtlety and complexity, but no actual decision as to which of them is, finally, to be accepted. Such escalation of nuance can be taken as an end in itself; but it is much more than that. If analytical philosophers have relatively little to proffer in the way of accepted theories that scientists like, they can provide a great deal of illumination to their fellow philosophers—the ones who are trying to understand, not how things are independently of us, but how they are when we are around. This illumination, however, is sidelong: it is not of the issues that analysts themselves actually discuss, but of other ones. When analysts reason about the implication of the distinction between the A-series and the B-series, they are, if my account of McTaggart is right, asking about what distinguishes God from man—an important question even for atheists. When they attempt to relate the perceived direction of time to thermodynamics, they are asking about how we experience death—what resonances does the coming heat death of the universe have in our vanishingly local lives? When they ask whether time travel is logically coherent, they are raising the question of why we are unable to move in time as we do in space. In illuminating these questions, analytical philosophers show us the widest implications—the logical ones—of their possible answers. In this way, they shed important light on the only thing philosophy can illuminate: the lives we lead as temporary denizens of a universe knowable only through science. NOTES   1 For a collection of some of these ways, see Oaklander and Smith (1994). The question is most vigorously pursued in that anthology by Clifford Williams (1994b).   2 Broad’s formulation of the distinction, 150 years after Kant’s, can be viewed as specifying (b): “In the temporal series of experiences, which constitutes a person’s mental history, there is a genuine dyadic relation which is intrinsic to the series and involves no reference to any term outside the latter. This is the term “earlier than.” In the linear spatial series there is no intrinsic direction (Broad 1938, pp. 267–268).” The specification lies in the fact that being “earlier than” something is one way of being incompatible with it. Broad’s formulation here does not address Kant’s point that space is restricted to “outer”

218  John McCumber representations, while time is not; indeed, it does not follow from it that space and time are overall “forms” of anything.   3 For a discussion of this issue, see Treisman 1999, pp. 218–219.   4 I may not, of course, need experience to teach me this; it could be some sort of a priori truth, like the law that every event has a cause. But experience also teaches me that. That events have causes is inculcated, we may say, by our ordinary experiences of things. We do not leave the lessons of experience behind, as Kant argued, until we say that all events have causes (see Kant 1996, B3–4).   5 If the intervals between the arrival of two auditory messages, or the discrepancies between the arrangements my two eyes see, had to be understood temporally and spatially, we would encounter some obvious problems of circularity. But I see no reason why this should be the case. If space and time are kinds of order—on which point I will agree with Kant—then intervals and discrepancies are what is most basically ordered. Other principles of order are possible for them, as we will see, and in any case the road is long from the mutual interrelations of auditory intervals and visual discrepancies to overall conceptions of space and time.  6 Gilles Deleuze’s grounding of time in a “contemplative,” “passive synthesis” indicates that in addition to the Bergsonian provenance of his account, which is often emphasized, there is a Husserlian one as well—or rather counter-Husserlian: for we may say that while Husserl arrives at the transcendental by bracketing everything outside the ego, Deleuze does so by bracketing, precisely, the ego (Deleuze 1994, pp. 74, 76).   7 If my attention is drawn to something moving across my visual field, like a dog running on a lawn, it is presumably because it has the potential of veering from its present course and approaching or receding from me.   8 Co-occupancy can, however, be hidden. In Lee Daniels’s The Butler, the protagonist, a White House butler, is told: “When you are in a room, the room should look empty.” When someone can make the others in a room disappear, or can cause them to make themselves disappear, that person becomes the sole ordering force in a fixed location, thus assuming what I call “ousiodic dominance” over that bounded place (see McCumber 1999). The co-occupancy of place thus has political valences, to be explored elsewhere.  9 What is not in play here is the possibility advanced by Heidegger in “The Origin of the Work of Art” (Heidegger 1971): that someone may appear to be following a script with which we are unfamiliar. That a script of some sort is there is clear; but what that script is remains even more mysterious than is the case here. 10 Another way is when a nearing is “stymied:” when it is not allocentric but ends before it actualizes a possibility of my being, thus failing to reach its internal telos. Discussion of this here would unduly lengthen my investigation; Heidegger has given a useful account at Heidegger 1962, pp. 102–107. 11 Hegel also emphasized this. See Hegel 1979, p. 187. 12 If the here is the end point of all nearings, as I suggested above, then absolute hereness, which we can only imagine, would be time itself as absolute subjective innerness. But this is not to be confused with time as experienced, which is an aspect not of the subject but of place. 13 Thus, the Stanford Encyclopedia of Philosophy treats the experience of time in a separate article from the one on the philosophy of time (Le Poidevin 2009; Markosian 2014). 14 L. Nathan Oaklander characterizes the view that time flows as “an impression [i.e., experience] deeply felt by all of us (Oaklander 1994, p. 289); this, of course, has nothing to do with its truth. Further cases of these presuppositions

Why Is Time Different from Space?  219 are too numerous even to begin to document, but for salient examples of the first, see Godfrey-Smith 1979 passim; Skow 2009; Williams 1994b, p. 372; of the second see Mellor 1998, pp. 1, 66–67; Prior 1996, p. 45. For what it is worth, I accept both of these as true of time, but not of place. A third common presupposition is that we can change the future but not the past. As Arthur Prior puts it: One of the big differences between the past and the future is that once something has become past it is as it were, out of our reach‑once a thing has happened, nothing we can do can make it not to have happened. But the future is to some extent, even though it is only to a very small extent, something we can make for ourselves. (Prior 1996, p. 48) This view structures Prior’s account of time, for it is in order to salvage what he calls “real freedom” that he rejects tenseless (B-series) views of time (ibid.). 15 This view has origins in Kant’s account, in the Prolegomena, of the distinction between judgments of perception, which claim validity only for the person uttering them at the time of the utterance, and judgments of experience which claim objective validity because they are formed through the categories, which are the same for all humans: All our judgments are at first merely judgments of perception . . . Judgments of experience take their objective validity not from the immediate knowledge of the object (which is impossible) but from the condition of universal validity of empirical judgments. (AA IV: 298–299) 16 Indeed, as Quentin Meillassoux points out (Meillassoux 2010, pp. 16–27), it has trouble explaining how such beings could even have existed, let alone how they did so in time. 17 McTaggart simply asserts that Hegel treats time as unreal (McTaggart 1991, p. 94). For Hegel’s real view see McCumber 2011, pp. 31–56.

WORKS CITED Broad, C. D. 1938. An Examination of McTaggart’s Philosophy. Volume II. Cambridge: Cambridge University Press. Casey, Edward S. 2009. Getting Back into Place. 2nd ed. Bloomington, IN: Indiana University Press. Deleuze, Gilles. 1994. Difference and Repetition. Trans. Paul Patton. New York: Columbia University Press. Derrida, Jacques. 1973. Speech and Phenomena and Other Essays on Husserl’s Theory of Signs. Trans. David B. Allison. Evanston, IL: Northwestern University Press. Erickson Stephan A. 1999.The (Coming) Age of Thresholding. Dordrecht: Kluwer. Godfrey-Smith, W. 1979. “Special Relativity and the Present,” Philosophical Studies 36 (3): 233–244. Greene, Graham. 1973. The Honorary Consul. Harmondsworth: Penguin. Hegel, G. W. F. 1979. Phenomenology of Spirit. Trans. A. V. Miller. Oxford: Oxford University Press. Heidegger, Martin. 1962. Being and Time. Trans. John Macquarrie and Edward Robinson. New York: Harper & Row. Marginal pagination. Heidegger, Martin. 1971. The Origin of the Work of Art. In M. Heidegger, ed., Poetry, Language, Thought. Trans. Albert Hofstadter. New York: Harper & Row, 15–88.

220  John McCumber Horwich, Paul. 1987. Asymmetries in Time. Cambridge, MA: MIT Press. Husserl, Edmund. 1960. Cartesian Meditations. Trans. Dorion Cairns. The Hague: Martinus Nijhoff. Kant, Immanuel. 1987. Critique of Judgment. Trans. Werner S. Pluhar. Indianapolis, IN: Hackett. Kant, Immanuel. 1996. Critique of Pure Reason. Trans. Werner S. Pluhar. Indianapolis, IN: Hackett. Le Poidevin, Robin. 2009. The Experience and Perception of Time. Substantively revised article. Stanford Encyclopedia of Philosophy. 2011 ed. http://plato. stanford.edu/cgi-.in/encyclopedia/archinfo.cgi?entry=time-experience. Accessed February 3, 2015. Markosian, Ned. 2014. Time. Substantively revised article, Stanford Encyclopedia of Philosophy. 2014 ed. http://plato.stanford.edu/entries/time/. Accessed February 3, 2015. McCumber, John. 1999. Metaphysics and Oppression. Bloomington, IN: Indiana University Press. McCumber, John. 2005. Reshaping Reason. Bloomington, IN: Indiana University Press. McCumber, John. 2011. Time and Philosophy. Durham: Acumen. McCumber, John. 2012. The Baffling ‘Nature’ of Time. Parrhesia 15: 14–23. McCumber, John. 2013. On Philosophy: Notes from a Crisis. Stanford, CA: Stanford University Press. McTaggart, John Ellis.1993. The Unreality of Time.”In J. Westphal and C. Levinson, eds., Time. Indianapolis, IN: Hackett Publishing Company, 94–111 (originally published 1908). Meillassoux, Quentin. 2010. After Finitude. Trans. Ray Brassier. London: Continuum. Mellor, D. H. 1998. Real Time II. New York: Routledge. Merleau-Ponty, Maurice. 1945. Phénoménologie de la perception. Paris: Gallimard. Meyerhoff, Hans. 1955. Time and Literature. Berkeley: University of California Press. Oaklander, L. Nathan. 1994. Introduction: The Problem of Our Experience of Time. In L. N. Oaklander and Q. Smith, eds., The New Theory of Time. New Haven, CT: Yale University Press, 289–292. Oaklander, L. Nathan and Smith, Quentine, eds. 1994. The New Theory of Time. New Haven, CT: Yale University Press. Prior, A. 1967. Past, Present, and Future. Oxford: Oxford University Press. Prior, A. 1996. “Two Essays on Temporal Realism: ‘A Statement of Temporal Realism’ and ‘Some Free Thinking About Time.’ ” In B.J. Copeland, ed., Logic and Reality: Essays on the Legacy of Arthur Prior. Oxford: Oxford University Press, 43-51. Ryle, Gilbert. 1949. The Concept of Mind. Chicago: University of Chicago Press. Schiller, Friedrich. 1967. On the Aesthetic Education of Man in a Series of Letters. Trans.Elizabeth M. Wilkinson and L. A. Willoughby. Oxford: Clarendon (first publication in 1801). Skow, B. 2009. “Relativity and the Moving Spotlight,” Journal of Philosophy 106 (12): 666-678. Treisman, Michael. 1999. The Perception of Time: Philosophical Views and Psychological Evidence. In Jeremy Butterfield, ed., The Arguments of Time. Oxford: Oxford University Press, 217–246. Westphal, Jonathan and Carl Levenson, eds. 1993. Time. Indianapolis, IN: Hackett Publishing Company.

Why Is Time Different from Space?  221 Williams, Clifford. 1994a. The Phenomenology of A-Time. In L. Nathan Oaklander and Quentine Smith, eds., The New Theory of Time. New Haven, CT: Yale University Press, 351–359. Williams, Clifford 1994b. The Phenomenology of B-Time. In L. Nathan Oaklander and Quentine Smith, eds., The New Theory of Time. New Haven, CT: Yale University Press, 360–372. Wittgenstein, Ludwig, 1958. Philosophical Investigations. 3rd ed. Trans. G. E. M. Anscombe. New York: Macmillan.

10 Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein Thinking Language Bounding World Paul M. Livingston

This is a tale of two readings, and of a non-encounter: the missed encounter between two philosophers whose legacy, as has been noted, might jointly define the scope of problems and questions left open for philosophy today. In particular, I will discuss two remarks, one by Wittgenstein on Heidegger, and the other by Heidegger on Wittgenstein. The first is one of only two (as far as I know) recorded remarks by Wittgenstein about Heidegger, and the second is one of only two (again, as far as I know) by Heidegger about Wittgenstein.1 As readings, both remarks that I shall discuss are, at best, partial, elliptical, and glancing. Interestingly, as I shall argue, each is actually a suggestive misreading of the one philosopher by the other. By considering the two misreadings, I shall argue, we can understand better the relationship between the two great twentieth-century investigators of the still obscure linkages among being, language, and truth. And we can gain some insight into some of the many questions still left open by the many failed encounters of twentieth-century philosophy, including what might be considered the most definitive encounter that is still routinely missed, miscarried, or misunderstood, the encounter between the “traditions” of “analytic” and “continental” philosophy, which are still widely supposed to be disjoint. I I begin with the remark by Wittgenstein on Heidegger. It comes in the course of a series of discussions between Wittgenstein and members of the Vienna Circle held in the homes of Friedrich Waismann and Moritz Schlick and later collected under the title Wittgenstein and the Vienna Circle. The remark dated December 30, 1929, reads: On Heidegger: I can very well think what Heidegger meant about Being and Angst. Man has the drive to run up against the boundaries of language. Think, for instance, of the astonishment that anything exists [das etwas existiert]. This astonishment cannot be expressed in the form of a question,

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  223 and there is also no answer to it. All that we can say can only, a priori, be nonsense. Nevertheless we run up against the boundaries of language. Kierkegaard also saw this running-up and similarly pointed it out (as running up against the paradox). This running up against the boundaries of language is Ethics. I hold it certainly to be very important that one makes an end to all the chatter about ethics—whether there can be knowledge in ethics, whether there are values [ob es Werte gebe], whether the Good can be defined, etc. In ethics one always makes the attempt to say something which cannot concern and never concerns the essence of the matter. It is a priori certain: whatever one may give as a definition of the Good—it is always only a misunderstanding to suppose that the expression corresponds to what one actually means (Moore). But the tendency to run up against shows something. The holy Augustine already knew this when he said: “What, you scoundrel, you would speak no nonsense? Go ahead and speak nonsense—it doesn’t matter!”2 The remark was first published in the January 1965 issue of the Philosophical Review, both in the original German and in an English translation by Max Black. In that version, in both the German and English texts, Waismann’s title, the first sentence, and the last sentence were there omitted, so that the remark as a whole appeared to make no reference either to Heidegger or to Augustine.3 Whatever this might indicate about the analytic/continental divide at the time of that publication, the remark itself shows that Wittgenstein had at least some knowledge of the contents of Being and Time (which had appeared just two years before he made it) and that he held its author at least in some esteem. The comparison with Kierkegaard, whom Wittgenstein also greatly respected, shows that he recognized and approved of the marked “existentialist” undertone of Being and Time and understood the deep Kierkegaardian influence on Heidegger’s conception there of Angst, or anxiety, as essentially linked to the possibility of a disclosure of the world as such. Indeed, in Being and Time, Heidegger describes Angst as a “distinctive way in which Dasein is disclosed” and as essentially connected to the revealing of the structure of being-in-the-world, which is, in turn, one of the most essential structures of Dasein. Thus, for Heidegger, it is Angst that first discloses the joint structure of Dasein and being-in-the-world as such.4 Since Angst is not fear before an individual or individuals, but a kind of discomfort toward the world as a whole, “the world as such is that in the face of which one has Angst,” according to Heidegger, and this is evidently, thus, close to the experience that Wittgenstein calls “astonishment that anything exists.” It is an index of the extraordinary diversity of Wittgenstein’s philosophical influences (as well as evidence against the often-heard claim that he either did not read the history of philosophy or did not care about it) that

224  Paul M. Livingston he manages in this very compressed remark, to mention approvingly, in addition to Heidegger and Kierkegaard, two philosophers whose historical contexts and philosophical methods could hardly be more different: G. E. Moore and St. Augustine. The concern that links Augustine, Kierkegaard, Moore, and Heidegger across centuries of philosophical history and despite obviously deep differences is something that Wittgenstein does not hesitate to call “Ethics,” although his own elliptical discussions of the status of ethics and its theory are certainly anything but traditional. Some years earlier, in the Tractatus, Wittgenstein had described “ethics” very briefly and elliptically as “transcendental,” holding that “it is impossible for there to be propositions of ethics” and that “ ethics cannot be put into words.”5 The position expressed in this brief passage is further spelled out, though, in the brief “Lecture on Ethics” that Wittgenstein had delivered to the “Heretics Society” in Cambridge on November 17, 1929, six weeks before the remark on Heidegger.6 In the “Lecture,” Wittgenstein considers the status of what he calls, partially following Moore, “absolute judgments of value,” judgments that something simply is valuable, obligatory, or good in itself, without reference to anything else that it is valuable for. His thesis is that “no statement of fact can ever be, or imply, a judgment of absolute value.”7 This is because all facts are, in themselves, on the same level, and no fact is inherently more valuable than any other (cf. TLP 6.4). In a book written by an omniscient author and containing descriptions of all bodies and their movements as well as all human states of mind and thus containing “the whole description of the world,” there would nevertheless be no ethical judgments, or anything implying one, for even statements of relative value or descriptions of human states of mind would themselves simply be descriptions of facts. It follows that there can be no science of Ethics, for “nothing we could ever think or say should be the thing.”8 Nevertheless there remains a temptation to use expressions such as “absolute value” and “absolute good.”9 What, then, is at the root of this inherent temptation, and what does it actually express? Speaking now in the first person, Wittgenstein describes “the idea of one particular experience” that “presents itself” to him when he is tempted to use these expressions. This experience is, Wittgenstein says, his experience “par excellence” associated with the attempt to fix the mind on the meaning of absolute value: I believe the best way of describing it is to say that when I have it I wonder at the existence of the world. And I am then inclined to use such phrases as ‘how extraordinary that anything should exist’ or ‘how extraordinary that the world should exist.’10 A paradigmatic experience of ethics for Wittgenstein is thus the experience that one might attempt to express by saying that one wonders at the existence of the world; nevertheless, as Wittgenstein immediately points out, the expression necessarily fails in that it yields only nonsense. For although

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  225 it makes sense to wonder about something’s being the case that might not have been, or might have been otherwise, it makes no sense to wonder about the world’s existing at all. It is thus excluded at the outset that what one is tempted to describe as the “experience” of such wonder can be meaningfully expressed, and it is a kind of paradox that any factual or psychological experience should even so much as seem to have this significance. And if someone were to object that the existence of an experience of absolute value might indeed be just a fact among others, for which we have as yet not found the proper analysis, Wittgenstein suggests that it would be possible to see “as it were in a flash of light,” that every possible attempt to describe absolute value would yield only nonsense, rooted in the desire “to go beyond the world and that is to say beyond significant language.”11 What is nevertheless expressed in metaphors such as these—metaphors such as that of the vast structural correspondence of language and world, the coming into existence or creation of the world itself, or (as Wittgenstein also suggests in the “Lecture”), the “great and elaborate allegory” that represents God as seeing everything, but also as “a human being of great power whose grace we try to win”?12 As Wittgenstein notes in the “Lecture,” the curious peculiarity of metaphors of this type is that, while metaphors more generally are metaphors for something, these cannot be replaced with the “literal” description of the facts they are metaphors for, since there are no such facts.13 That they nevertheless arise at the point of the temptation that also yields the incoherent attempt to mark the place of absolute value might then be thought to indicate that their attempt is also the one that thought makes in trying to touch a point of the absolute, the real corresponding to the totality of the world or its grasping as a whole from a point beyond it, the point at which the value of the world—if it has value—could be assayed.14 The price of this attempt, however, is the admission of its necessary failure, the impossibility of anchoring thought at such a point of the real without contradiction, paradox, or the nonsense of metaphors that cannot be cashed in for their literal meaning, since what they stand for is, literally, nothing. Returning to the remark of December 30, Wittgenstein’s suggestion here is, then, that all of the philosophers he mentions (Moore, Augustine, and Kierkegaard as much as Heidegger) can in fact be read, in different ways, as having understood this impossibility for ethics or ethical propositions to come to expression. The theory of ethics is futile, in that the attempt to establish ethics as a positive knowledge or science, to determine the existence and nature of values, or even, as Moore had suggested, to define the Good itself, can yield only the “chatter” of a continually renewed nonsense that perennially fails to recognize itself as such. At the same time, however, it is in this essential failure to be expressed or expressible that Wittgenstein suggests (echoing the central distinction of the Tractatus between all that can be said and what, beyond the boundaries of language, can only be shown) the real yield of all attempts at ethical thought might ultimately be found. This is because of the link between the “tendency to run up against

226  Paul M. Livingston the boundaries of language” and what we should like to call the radical experiences of our relation to the world as such, including even the feeling that we may express as our astonishment that anything exists at all. Something very similar is indeed suggested by Heidegger’s notorious discussion of being and nothingness in the Freiburg inaugural lecture “What is Metaphysics?” delivered on July 24, 1929.15 Here, the experience of “the Nothing” [das Nichts] by means of which it is first possible for us to “find ourselves among beings as a whole” thereby allows “beings as a whole” to be revealed, even if “comprehending the whole of beings in themselves” is nevertheless “impossible in principle.”16 In the moods or attunements of boredom and anxiety, we are brought “face to face with beings as a whole” and the very unease we feel in these moods towards being as a whole also brings us a “fundamental attunement” that is “also the basic occurrence of our Dasein,” as exhibited in an experience of Nothing and nihilating in which “Dasein is all that is still there.”17 This experience also gestures toward a kind of dysfunction of speech and logos: “Anxiety robs us of speech” and “in the face of anxiety all utterance of the ‘is’ falls silent.”18 And, notoriously, Heidegger holds that in the encounter with “the nothing,” logical thinking itself must give way to a more fundamental experience: If the power of the intellect in the field of inquiry into the nothing and into Being is thus shattered, then the destiny of the reign of ‘logic’ in philosophy is thereby decided. The idea of ‘logic’ itself disintegrates in the turbulence of a more original questioning.19 It would thus not be amiss to see Wittgenstein’s invocation of this sense of wonder at existence, in both the remark on Heidegger and in the “Lecture on Ethics,” as suggesting significant parallels to the thought of the philosopher whose signature is the question of being and the disclosure of its fundamental structures, including the basic “experiences,” such as that of Angst, in which the being of the world as such—here, the totality of beings—may be disclosed. Yet as a reading of Heidegger’s actual position in Being and Time, the main suggestion of the passage—that these experiences are to be found by “running up against” the boundaries of language—is nevertheless essentially a misreading. For Being and Time contains no detailed or even very explicit theory of language as such, let alone the possibility of running up against its boundaries or limits. And insofar as Being and Time discusses language (die Sprache), the discussion is almost wholly subordinated to the discussion of Rede or concretely practiced discourse, something that does not obviously have boundaries at all.20 In Being and Time, Heidegger’s brief and elliptical discussion of language emphasizes its secondary, derivative status as founded in discourse and the fundamental ontological possibility of a transformation from one to the other. Thus, “The existential-ontological foundation of language is discourse [die Rede].”21 Language is “the way discourse gets expressed.”22

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  227 Discourse is itself the “articulation of intelligibility” and, as such an articulation, is always separable into isolated “significations” or “meanings” [Bedeutungen].23 Nevertheless the “worldly” character of discourse as an “articulation of the intelligibility of the ‘there’ ” means that it yields a “totality-of-significations” [Bedeutungsganze], which can then be “put into words” or can “come to word” (kommt zu Wort).24 Language can then be defined as a totality of (spoken or written) words; in this totality “discourse has a ‘worldly’ Being of its own.”25 It thus may subsequently happen that language, the totality of words, becomes something in the world that we can “come across as ready-to-hand” [Zuhanden] or indeed break up analytically into objectively present “world-things which are present-at-hand.”26 Language’s specific way of manifesting being-in-the-world, or of disclosing the worldly character of the beings that we ourselves are, is to appear in the world as a totality of words ambiguously experienced as tools of use or objective “word-things.” Discourse itself, Heidegger goes on to say, supports the ever-present possibilities of “hearing” or “keeping silent.”27 These possibilities, as possibilities of discursive speech, disclose “for the first time” “the constitutive function of discourse for the existentiality of existence.”28 But they are not in any direct way connected to the ontological structure of language itself, which must, Heidegger says, still be worked out.29 Whatever else it may be, the story of the existential significance of words in Being and Time is not, therefore, the document of an inherent human tendency to “run up against the boundaries of language” that ultimately, even in being frustrated, can yield a transformative indication of the limits of the world as such. The worldly character of language is, here, not a matter of its actual or possible correlation to the totality of facts or situations in the world, but rather of its tendency to appear within the world as an objectively present totality of signs or of “word-things,” abstracted and broken up with respect to the original sources of their meaning in the lived fluidity of discourse. This is not a subjective “running-up against the boundaries of language” but something more like a falling of meaning into the world in the form of its capture by objective presence. There are, to be sure, distinctive dangers here. Heidegger will go on to suggest that it is in this tendency to interpret language as an objectively present being that the traditional and still dominant conception of logos remains rooted, a conception that yields an insufficiently radical understanding of meaning and truth, one that the present, more penetrating, existential analytic must deconstruct. But there is no suggestion that any part of this analysis involves recognizing the boundaries of language as such, or considering how the tendency to speak beyond them issues in nonsense. Moreover, although the possibility of keeping silent does indeed bear, for Heidegger, a primary disclosive significance, what it tends to disclose is not the limits of the world beyond which it is impossible to speak, but rather, quite to the contrary, the inherent positive structure of Dasein’s capability to make the world articulate and intelligible. This is not

228  Paul M. Livingston the obligatory silence, which concludes the Tractatus, beyond the bounds of language where nothing can be said, but rather the contingent silence that results from a “reticence” of which Dasein is always capable, and which is indeed at the root of Dasein’s strictly correlative capability of “having something to say.”30 Heidegger’s remarks on the Nothing and anxiety in “What is Metaphysics?” were famously the basis for Carnap’s dismissive rejection, in the 1932 article “The Overcoming of Metaphysics through the Logical Analysis of Language” of Heidegger’s whole project as “metaphysical” and as violating the very conditions for the meaningfulness of any possible language.31 Part of what motivated Carnap in his ire may have been Heidegger’s visible contempt for the attempt to structure language logically; in the inaugural address, as we have seen, he describes the experience of the Nothing as leading to a “disintegration” of logic, and the remarks on language in Being and Time are dedicated to a “task of liberating grammar from logic.”32 From the perspective of Carnap’s logical empiricist project, which was dedicated to the elimination of dangerous and idle metaphysics by means of a clarification of the underlying logical structure of meaningful language as such, these suggestions could only seem to represent the most misleading kind of obscurantism. Yet as recent scholarship has emphasized, it would be a mistake simply to identify Wittgenstein’s conception of logical structure with that of Carnap, for whom Wittgenstein also had little sympathy. For whereas the point of identifying the bounds of language for Carnap is consolidation of science and objectivity by means of the identification and elimination of the “pseudo-sentences” that lie beyond them, the point is for Wittgenstein just about directly the opposite. As Wittgenstein famously wrote later, the whole point of the Tractatus was “ethical,” presumably in the sense that it was to bring us to a self-conscious experience of those limits beyond which we cannot speak: here was not, then, the excessive “beyond” of meaninglessness grounded in the violation of fixed logical rules but the very possibility of a “mystical” or “aesthetic” vision of the world, the vision sub specie aeterni of the world “as a limited whole.”33 So although it would certainly be wrong to say that the problem of the limits of language stands or falls with the rigid, deterministic conception of the structure of language that Carnap imposed in his critical remarks on Heidegger, there is, it seems, between Wittgenstein and Heidegger a significantly broader and more general question of the relationship of language and world that remains open, and probably remains with us even today. This can be put as the question: What does the very existence of language have to do with the nature of the world it seems to bound? And what does it mean that the structure of language, which seems to set the very boundaries of the possibilities for speaking of facts and objects and hence to determine what we can understand as the world, can again be thought (whether logically, grammatically, or historically) and even experienced within the world thus bounded? Without overstatement, it would be possible to say that this

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  229 is the question that links twentieth-century linguistic philosophy, in its specificity, to all that has formerly been thought under the heading of transcendence and the transcendental. And since it is not obvious where solutions may lie, it seems that this question remains very much with us today. II Heidegger’s remark on Wittgenstein comes almost forty years later, in one of Heidegger’s very last seminars, the last of three seminars the aging philosopher held in Le Thor, France.34 The seminar as a whole is ostensibly directed to the elucidatory discussion of Kant’s pre-critical work “The Sole Possible Proof for a Demonstration of the Existence of God,” and more specifically to its first chapter, on “existence as a whole” (Vom Dasein Überhaupt), though there is actually little explicit discussion of Kant anywhere in the seminar.35 The mention of Wittgenstein, however, comes early in the course of the first seminar session, on September 2, 1969: So we pose the question: what does the ‘question of being’ mean? [was besagt “Frage nach dem Sein”?] For, as a question, the question of being already offers numerous possibilities for misunderstanding—something confirmed by the continual failure to understand the book Being and Time. What does ‘the question of being’ mean? If one says ‘being’, from the outset one understands the word metaphysically, i.e. from out of metaphysics. However, in metaphysics and its tradition, ‘being’ means: that which determines a being insofar as it is a being [was das Seiende bestimmt, sofern es Seiendes ist]. As a result, metaphysically the question of being means: the question concerning the being as a being, or otherwise put: the question concerning the ground of a being [die Frage nach dem Grund des Seienden]. To this question, the history of metaphysics has given a series of answers. As an example: energeia. Here reference is made to the Aristotelian answer to the question, “What is the being as a being?” [“Was ist das Seiende als Seiendes?”]—an answer which runs energeia, and not some hypokeimenon. For its part, the hypokeimenon is an interpretation of beings and by no means an interpretation of being [die Auslegung des Seienden und keineswegs des Seins]. In the most concrete terms, hypokeimenon is the presencing [das Anwesen] of an island or of a mountain, and when one is in Greece such a presencing leaps into view. Hypokeimenon is in fact the being as it lets itself be seen [das Seiende in seiner Lage, so wie es sich sehen läβt], and this means: that which is there before the eyes, as it brings itself forth from itself [das, was da ist, vor den Augen, wie es da von sich selbst her sich hinzieht]. Thus the mountain lies on land and the island in the sea.

230  Paul M. Livingston Such is the Greek experience of beings. For us, being as a whole [das Seiende im Ganzen]—ta onta—is only an empty word. For us, there is no longer that experience of beings in the Greek sense. On the contrary, as in Wittgenstein, “The real is what is the case” [bei Wittgenstein heiβt es: “Wirklich ist, was der Fall ist”] (which means: that which falls under a determination, lets itself be established, the determinable), actually an eerie [gespenstischer] statement. For the Greeks, on the contrary, this experience of beings is so rich, so concrete and touches the Greeks to such an extent that there are significant synonyms (Aristotle, Metaphysics A): ta phainomena, ta alethea. For this reason, it gets us nowhere to translate ta onta literally as “the beings” [das Seiende]. In so doing, there is no understanding of what is being for the Greeks [hat man kein Verständnis für das eröffnet, was für den Griechen das Seiende ist]. It is authentically: ta alethea, what is revealed in unconcealment [das Offenbare in der Unverborgenheit], what postpones concealment for a time; it is ta phainomena, what here shows itself from itself [was sich von sich selbst her zeigt].36 As he often does at this late stage in his career, Heidegger couches his remarks as a kind of retrospective of his own work, giving a prominent place to the “question of Being” raised by Being and Time while complaining, as he often did, of that book’s failure ultimately to communicate the sense and significance of this question. In fact, though, as Heidegger clarifies, the relevant “question of Being” here is not simply the one formulated in Being and Time, which concerns the “meaning” or “sense” of Being, but rather (by way of a decisive shift) the question of the “ground of Being,” of what it means to think the various bases that determine, in each of the various conceptions or epochs that comprise the history of metaphysics, beings as such and as a whole.37 The problem, as Heidegger explains elsewhere, is now one of thinking what variously determines the historical ways—now in the plural—that beings as such and as a whole can be thought and experienced, and thinking this determining instance now, no longer itself in terms of beings but only in terms of being itself. This is the question of the historical truth of being, or of how being itself gives or grants itself in or as its event of Ereignis, by grounding—in various and historically variable ways—the possibility of conceiving of and experiencing beings in their totality. From the initial Platonic understanding of beings as determined by the “being-ness” of the idea, through the Medieval conception of God as the summa ens and the modern metaphysics of subjectivity and up to the contemporary regime of dominant technology, this question of grounding, according to Heidegger, receives a series of different answers in the metaphysical tradition.38 But all of these answers are ways of determining the character of beings as a whole by opening or projecting their sense.

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  231 How, though, does Heidegger here understand sense? The conception of sense as a projective grounding of entities as a whole is continuous with Being and Time’s definition, and consideration, of sense as projective comportment toward entities on the ground of possibilities.39 In Being and Time, this projective comportment is itself intimately related to Dasein’s capacity or structure of world-disclosure and to the way in which this structure allows entities to appear as unconcealed or in truth.40 With the “turn” from Being and Time’s analytic of Dasein to the later, “being-historical” project, truth, world-disclosure, and projective sense are no longer grounded simply or exclusively in Dasein, but rather in the prior structure of an “open” or “clearing” [Lichtung] that is itself a precondition for Dasein as well.41 The structure of this clearing, and its relation to Being itself—now understood primarily not as “the being of beings” but rather in independence of entities—allows for the clearing and grounding relation that multiply determines, over the course of the epochal determinations of metaphysical thought and practice, the character of beings as a whole. In particular (returning to the 1969 passage), whereas Aristotle thinks of the ultimate ground of beings as a whole as energeia, or as active, actual occurrence, a being or entity itself is here thought of as hypokeimenon, or as substance. The experience of the hypokeimenon, as it is thought and undergone by the Greeks, is one of the being of a being in its “let[ting] itself be seen,” its basic presencing and being revealed in truth. This “experience of beings in the Greek sense” permits and is permitted by, Heidegger suggests, an experience of “what . . . being is” for the Greeks, namely presencing and disclosure, the truth of what shows itself from itself as it itself is. Such an experience of beings not only remains faithful to their underlying character as it shows itself but is also, Heidegger says, “so rich” and “so concrete” that its synonyms in Greek connect it to the underlying meanings of truth (aletheia, or unconcealment) and indeed to the very meaning of what it is to be a phenomenon at all. At the same time, however, understanding the individual being or entity as hypokeimenon already essentially involves understanding it in relation to the structure of a logos: in particular, the hypokeimenon is something about which we speak, what is named in the grammatical subject of a sentence. For Aristotle and the Greeks more generally, the definitive character of individual entities is thus understood on the logical basis of the structure of the assertoric sentence, the legein ti kata tinos (i.e., a saying of something about something). This understanding already portrays the character of the being—here, the substantial bearer of the reference of the grammatical subject—as that which is logically or linguistically determinable by means of a predicative sentence that ascribes to it properties or determinations. This is the occasion for Heidegger’s reference to the contemporary conception that he attributes to Wittgenstein, according to which all that exists is the real in the sense of the logically and predicatively “determinable” or

232  Paul M. Livingston “determined,” and there is, he suggests, accordingly no possibility any longer of anything like a comparable insight into the character of the ta onta, what opens or grounds beings as a whole. However this may be, Heidegger’s reading of Wittgenstein is nevertheless a misreading, in an even more direct way than is Wittgenstein’s earlier reading of Heidegger. For the sentence that Heidegger here attributes to Wittgenstein is a direct misquotation. The first sentence of the Tractatus reads, “The world is all that is the case” [Die Welt ist Alles, was der Fall ist.] Heidegger misquotes this as “The real is what is the case” [Wirklich ist, was der Fall ist]. And this is, in fact, no innocuous substitution. We can begin to see why by considering the gloss that Heidegger immediately gives on what he takes the position that he attributes to Wittgenstein to imply. That all and only what is real (Wirklich) for Wittgenstein is all and only what “is the case” means, according to Heidegger’s gloss, that all that is the case, all that exists as an actual fact or real state of affairs, is what “falls under a determination, lets itself be established” or is “determinable.” This gloss is almost certainly Heidegger’s interpretation of the very next proposition of the Tractatus, 1.1, which holds that “The world is the totality of facts, not of things.” In its context, this proposition has the effect of denying that it is possible to consider the world as a whole simply as a collection or totality (however vast) of individual things or (in the Heideggerian jargon) beings (or entities), without the further structure given by their logical articulation and formation into facts and states of affairs. For, according to Wittgenstein, “the world divides” not into things or beings but into “facts” (1.2) and “the facts in logical space are the world.” (1.13). Facts, moreover, are not individual objects but “combinations” thereof, essentially structured in such a way that they are apt to be expressed by full assertoric sentences rather than individual names.42 Synthesizing all of this, then, it is clear that Heidegger takes it that, for Wittgenstein, for anything to be real at all is for it to be determined or determinable as a fact, to “stand under a determination” or to “let itself be established” as the case. This is the “determination” of a subject by a predicate, an individual by a “universal,” or an object by a concept, which is (variously thought) the underlying grammatical basis of the possibility of any assertoric sentence. To say that something is the case is then, according to Wittgenstein as Heidegger reads him, quite simply to say that an object or entity allows itself to be determined in such a way, to have the characteristic asserted to hold of it by a true proposition, or to allow such a proposition to be established and asserted as the truth. What, though, is involved in this determination or determinability, or in the submission of the nature of entities in general to their ability to serve as substrates or objects for predication in an assertoric sentence? Heidegger here echoes a critique of the assumption of the primacy of the assertion that has deep roots in his own thought, extending back the historical investigations into the original constitution and meaning of sentential logic that he

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  233 undertook already in the series of lecture courses immediately preceding the writing and publication of Being and Time: Here it is important to make a fundamental distinction in regard to speaking, namely to distinguish pure naming (onomazein) from the assertion [Aussage] (legein ti kata tinos). In simple nomination, I let what is present [das Anwesende] be what is. Without a doubt nomination includes the one who names—but what is proper to nomination is precisely that the one who names intervenes only to step into the background before the being. The being then is pure phenomenon. With the assertion, on the contrary, the one asserting takes part in that he inserts himself into it—and he inserts himself into it as the one who ranges over the being in order to speak about it. As soon as that occurs, the being can now only be understood as hypokeimenon and the name only as a residue of the apophansis. Today, when all language is from the outset understood from out of the assertion [von der Aussage her], it is very difficult for us to experience naming as pure nomination, outside of all kataphasis and in such a way that it lets the being presence as pure phenomenon.43 As Heidegger goes on to suggest, the determination of the being or phenomenon—what it is for something to appear in presence—in terms of the assertion and its logical structure of “saying something about something” is already on the way to the representational distinction between subject and object that is introduced explicitly by Descartes and reaches the highest point of its development in Kant and Hegel. Whereas, for the Greeks, “things appear,” for Kant, “things appear to me;” this is the ultimate consequence of Descartes’ identification of an absolute ground in subjective consciousness and his placement of the human “in his position as representer.”44 After Kant, Hegel then furthers the articulation of the distinction between subject and object by insisting on the subjective as mediation and thus as the essential dialectical “core” of objectivity. The whole development involves an ever-greater distancing from the original “Greek experience of being as phenomenon,” insofar as it means that the subject of consciousness is, in an ever more thorough way, placed at the basis of the representative presentation of objects, and the simple experience of beings presencing of themselves is thereby rendered more and more inaccessible. But the ultimate historical basis for this is to be found in the original conception whereby the beings are, as such, thought as determinable on the basis of the logical structure of the assertoric sentence itself, which is, on the early Wittgenstein’s telling, by stark contrast, formally the very basis for any conceivable possibility of meaning and truth. What, though, about the substitution that makes Heidegger’s quotation a misquotation of the Tractatus, the substitution of “the real” for “the

234  Paul M. Livingston world”? Coming as it does right in the midst of a passage devoted to discussing the historical possibilities for taking into account the nature of the whole—ta onta or everything that is, in Heidegger’s terms, “beings as a whole”—this substitution is far from philosophically innocuous. Indeed, it bears directly on the question of totality that is at issue in a different way, as we saw above, between Wittgenstein and Heidegger already in 1929. The German word “Wirklich” that Heidegger substitutes for “Welt” (world) here indeed means “real” and “actual,” but also has important connotations of effectivity and efficiency; what is “Wirklich” is not only what is real or is in being in the sense of simply existing, but also what is productive, energetic, or pro-active.45 Elsewhere, Heidegger had read the progressive historical determination of the nature of beings in terms of a series of transitions in the interpretation of the nature of beings as such, beginning with the ancient Greeks and culminating in modern times. The last stage in this progression, which Heidegger identifies with Nietzsche’s metaphysics of the will to power and absolute, self-positing subjectivity, indeed culminates, according to Heidegger, with the determination of beings in general as “real” in the sense of Wirklichkeit and effectiveness, a kind of technological regime of general, leveled effectiveness that treats all beings only in terms of their general calculability and their capacity instrumentally to cause and bring about determinate effects.46 This is nothing other, of course, than the universal reign of the thought and practice arising from the dominance of what he calls Gestell or enframing, the essence of modern technology. Within this dominance, it is no longer possible, Heidegger goes on to assert, to experience the “overabundance” or “excess” of what presences that the Greeks saw in the “coming-forthout-of-concealment” characteristic for them of physis and definitive of the phenomenal character of the phenomenon. Rather, “In extreme opposition to this [Greek overabundance], one can say that when the astronauts set foot on the moon, the moon as moon disappeared. It no longer rose or set. It is now only a calculable parameter for the technological enterprise of humans.”47 However, is Wittgenstein’s understanding of the way the structure and form of the world is determined by the logical structure of the sentence—the Tractatus conception of the world as the totality of articulate facts—really an example of an “enframing” attitude in this sense, a positioning of the subject or its will as the ultimate agency of effective representational control, calculation, and production? There are many reasons to doubt it. To begin with, Wittgenstein’s concern in the Tractatus is not primarily with the assertion [Aussage] in the sense of the subjective or agentive act of asserting something about something, but rather with the structure of the “assertoric” sentence [Satz], which itself and independently of any specific agent says that something is the case. Sentences are connected to the worldly states of affairs they describe not by means of any kind of subjective act of asserting, positing, or the like, but by their correspondence in logical form with

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  235 the possible situations they describe. The possibilities of this correspondence or non-correspondence, though they are not dependent in any obvious way on subjective decision, activity, or willful determination, nevertheless give the sentence its sense (4.2). “The subject” enters as such only very differently from this, as the transcendental “limit of the world” (5.632) and as what is manifest in the fact that “the world is my world” and “the limits of language (of that language which alone I understand) mean the limits of my world” (5.62). In this sense, “There is no such thing as the subject that thinks or entertains ideas.” (5.631). Moreover, since all facts are contingencies and there is thus no logical necessity to the determination of the result of a willed action, the attempt to produce the good by controlling and calculating effective outcomes must be futile as well: “The world is independent of my will.” (6.373). For all of these reasons, Heidegger’s suggestion that Wittgenstein’s conception of sentential articulation involves a criterion or structure of determination that itself is one of subjective agency appears misplaced, and his seeming assimilation of Wittgenstein’s logical conception of truth and meaning to the outcome of what is, for Heidegger, a vast and complex historical process of logical and technological enframing and determination is, accordingly, misleading. What, though, of the contrast that Heidegger draws between the articulate structure of the sentence or assertion—the structure of the legein ti kata tinos or the “saying of something about something”— and the simple and direct naming, which by contrast (for Heidegger) “lets the being be” in its direct and immediate presence? One might certainly see parallels between Heidegger’s underscoring of the distinction and the two levels of connection between language and the world that the Tractatus’s official theory of truth and reference maintains: those of names, on the one hand, and descriptive sentences, on the other. Here, in particular, it is necessary for the sense of the sentence, and especially for its ability in general to be true or false, that it be ultimately composed of simple names that are, by contrast, directly correlated to objects and that these objects accordingly must exist in order for sense to be possible at all.48 Both accounts then would seem to bear problematic witness to the possibility of a level of appearance or manifestation, beyond all facts and beings, that gives rise to the very sense with which all facts and worldly beings are endowed. Nevertheless, the Tractatus’s posing of the requirement for the objects correspondent to simple names or namings, far from simply “letting beings be,” immediately raises the question of the basis for the institution of the names of things. This is the problem of logical form as the ground of the necessity of their presence, what appears in the Tractatus as the basis of the very substance of the world.49 This is, however, one of the central problems that is taken up in a renewed fashion, alongside the critical interrogation of the necessitating force of logic and of the efficacy of rules in the determination of their instances, in the Philosophical Investigations. Here, the later Wittgenstein’s critique

236  Paul M. Livingston of the metaphysical picture of the Tractatus takes the initial form of an interrogation of the assumption of what “has” to exist on the level of the bearers of simple names, and a correspondent investigation of the idea that the speaking or understanding of a language consists in the operation of a calculus according to definite rules.50 The two skeins of critical interrogation that define the main argumentative movement of the Investigations, the so-called “private language argument” and the “rule-following considerations,” develop these lines of inquiry in a more general and broadly indicative way, critically investigating and raising deep problems, respectively, for the assumption of a privileged site for the naming of being in the subjective presence of self to self, and that of a guarantee of the unitary sense of a word in the effective repeatability of the rule as a self-identical structure unto infinity. If the critical investigation of these problems necessarily displaces the theoretical tendency to assume in advance that there must be only one essence of language or one way of characterizing or understanding the structure of the determination of sense in the instances of our lives, it is nevertheless possible to see both lines of critique as embodying complementary challenges to the unitary configuration of assumptions broadly characteristic of a form of life that is, today, recognizably “ours.” This is the configuration of, on the one hand, the capable subject of lived experience that makes the possibility of representational experiencing the criterion for the actuality of any object and, on the other, the effectiveness of regular and rule-defined processes of calculation, regulation, modification, and control in handling, producing, and economizing beings.51 What, then, finally is the force and effectivity of logic in relation to the relevant actions and events of this life? This problem is not only the one that led Frege somewhat problematically to introduce into his strictly logical vocabulary for the expression of inferential relations among contents a distinct symbolism for the separate operation of (assertoric) force or judgment.52 More broadly and in terms of the significance that both Heidegger and Wittgenstein ultimately see it as having, it is also the problem of the effectivity of systematic, mechanical, formal, or calculative rules in general in determining the totality of their instances and ensuring the smooth functioning of the techniques, institutions, practices, and systems governed by them. Correspondent to this figure of the effective functioning of rules on the level of the subject is the figure of the human speaker of language as their capable agent. But if the role of practices in our life were to be thought beyond the figure of human capacity—if, in other words, the possibilities of our world or life, as we meet them in the language that we speak and grasp them in the kind of being that we are, were determined otherwise than on the basis of the subjective capacities of an effective human agent, to what kind of domain for reflective consideration might this lead? In closing the seminar session of September 2, 1969, Heidegger again looks back at the problematic of Being and Time. In that book, Heidegger says, the question of being is already posed primarily not as the question

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  237 “What is a being?” but rather as “What is this ‘is’ ”? The question, thus posed, immediately runs into the “difficulty” that “if the ‘is’ is, then it is a being”; whereas if it is not, then it appears to be “bare and empty,” even perhaps (Heidegger suggests) to be nothing more than the “empty copula of a judgment.” It was in order to “come out of this aporia,” Heidegger suggests, that Being and Time posed the question of the “is” “from the perspective of the sense of Being,” understanding “sense” specifically as a “project” of “understanding.” But this way of formulating the question was, Heidegger now says, “inappropriate” in that it made it “all too possible to understand the ‘project’ as a human performance” and as a “structure of subjectivity”—this is indeed specifically, Heidegger suggests, how Sartre understands it in his own philosophy, which bases itself (he suggests), in this respect, upon Descartes. Nevertheless, beyond or before the assumption of such a constitutive subject of sense, it is possible, Heidegger suggests, to take up the problem of the “is” rather as a problem of truth, here understood as the pre- or non-subjective structural basis for opening and grounding disclosure of being: In order to counter this mistaken conception [of the “project” as subjective-PL] and to retain the meaning of “project” as it is to be taken (that of the opening disclosure), the thinking after Being and Time replaced the expression “meaning of being” with “truth of being.” And in order to avoid any falsification of the sense of truth, in order to exclude its being understood as correctness, “truth of being” was explained by “location of being” [Ortschaft]—truth as locality [Örtlichkeit] of being. This already presupposes, however, an understanding of the place-being of place. Hence the expression topology of be-ing [Topologie des Seyns]. The problematic of the topology of being to which Heidegger here gestures is, then, nothing other than the consideration of the structure through which being grants the sense and truth of beings, or (in other terms) how it grants the possibility of a world. III I have suggested that the missed encounter between Wittgenstein and Heidegger conceals foundational problems about logic, sense, meaning, and the totality of the world that are still substantially unsolved today. One of these is the ancient problem of the nature and force of the logos, which subsumes both the more local twentieth-century philosophical inquiry into language and the methods of formal and symbolic logic that have simultaneously defined many twentieth-century approaches. Here, as we have seen, there is co-implied a broad problem of force: what is the basis and status of the

238  Paul M. Livingston force of logical rules and laws, or of their actual effectiveness in relation to the activities, techniques, and practices that increasingly characterize life on the planet today? Another, closely related problem, as we have repeatedly seen, is the problem of the totality: the problem of our relation (if such there be) to the totality of the world or the being of all that is, or to whatever sets its limits or determines its extent. Without diminishing the difficulty of the issues involved in the formulation, both of these problems might also be put, it seems, as problems of finitude: that is, as problems about how a properly finite being has whatever problematic relationship it can have, whether in thought or in action, to what we might understand as the “infinite” dimension of sense, up to and including the sense of the world.53 Seen in this way, the problematic of the logic and sense of the world is, to begin with, a problem of limits. How is a being in the world nevertheless capable of grasping something of the world as a whole in which it exists, if it is capable of doing so at all? That both Heidegger and Wittgenstein see the question of the totality of the world as one that is linked to the question of the expressive or descriptive powers of language suggests, as we have seen, that it is to be addressed only by means of a renewed reflection on language’s forms, and on the historical and contemporary structures in which they allow themselves to be thought or presented. But that both also mark the problem as one in which language positively confronts its own limits in the necessary inexpressibility of whatever lies beyond the world suggests that we may draw from both philosophers the indication of a necessary complication, inherent to the problematic of sense, in the topology or limitology that is thereby invoked. In particular, that “being itself is not a being,” and that “whereof we cannot speak, there we must be silent” might then be the problematic claims through which, if it follows the suggestions made by both Heidegger and Wittgenstein, a future inquiry might pursue the question of the ultimate dispensation of sense as we meet with it in the forms of our lives. But is there a world, for us, today? If, as I have suggested, both Wittgenstein and Heidegger point to the question of the force of logos and logic in relation to the totality of the world, then the single question that is formulated by both may be put as the question of realism about the world as such a totality. This question bears close connections to the broader questions of realism and anti-realism that many contemporary and recent analytic and continental figures have discussed from different perspectives, and using various definitions of the terms. But because its topic is the world as a whole, it is distinct from the question of realism about any kind of entity or entities in any of their aspects (even “as they are in themselves”) or in general. It is also not primarily to be posed as the question of how or whether specific entities are conditioned by our epistemic “access” to them or (as we have seen) as that of the dependence of being or truth on our specifically human constitution or capacities (unless this just means the capacities that can be possessed by a being that can speak and understand language, and hence

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  239 grasp sense, at all). In particular, if the twentieth-century linguistic turn, in both its “analytic” and “continental” variants, already witnesses a decisive turn away from the idea of epistemology as “first” philosophy and toward a variety of practices of philosophical analysis and reflection grounded in the consideration of linguistic structure and sense, then it is plausibly no longer adequate in its wake to pose the question of realism about the world as one of the limits of our knowledge or epistemic access, or about the conditions under which a “subject” can have knowledge of “objects.” Rather, the question here to be formulated is, precisely, about sense: about if and how the world is meaningfully “given” as an intelligible whole to understanding or reasoning, or about whether the idea of such a whole, as what provides the broadest and most possible condition for the sense of things, is even so much as coherent.54 Posed this way, the question asks in “linguistic mode” whether there is a thinkable totality of all that is that is also, in some “logical” sense, a unity: a one whose composition or structure is itself intelligible on the basis of a formal consideration of it. In fact, this question was already central to discussion “between” recognizable ancestors of the analytic and continental traditions, long before the linguistic turn itself. The negative answer to it given by Kant in the argument of the “Antinomies” section of the Critique of Pure Reason, for instance, points to the broadly logical paradoxes that he took to be involved in any conception of the world as a completed totality of phenomena, while one of Hegel’s main goals in the Science of Logic was to overcome this Kantian antinomic reasoning by means of a speculative picture of the ultimate reconciliation of the paradoxes of the whole at the level of the Absolute.55 Whereas Husserl, in his later phenomenology, treated the world (in one of its senses) as the indeterminate but always-expansible horizon of possible experience, his analysis of meaning and sense also pointed to what he saw as its basis in the pre- or proto-rational foundation of the everyday “lifeworld.” In the early decades of what would become the “analytic” tradition, the question already posed by Kant received a more rigorous kind of formulation through Cantor’s set theory and the paradoxes of totality—most decisively, Russell’s paradox of the set of all sets not members of themselves—that appeared to show the impossibility of maintaining without contradiction the existence of a total set or set of all sets.56 Today, differing answers to the question of the world also situate and articulate key contemporary projects of both “analytic” and “continental” philosophers. On the one hand, recent prominent projects in “analytic” metaphysics, such as David Chalmers’s Constructing the World and Ted Sider’s Reading the Book of the World, have essentially presupposed or relied upon the coherence of the idea of the world in their strongly constructivist or naturalist projects of analyzing its structure.57 On the other hand, recent positions of a more directly continental lineage set out, by contrast, from a radical denial of the existence of a totality of all that is. For example,

240  Paul M. Livingston early in his 1988 Being and Event, Alain Badiou declares as a fundamental motivating basis for his project of set-theoretical ontology that “the one is not,” or more specifically that there is no total domain of entities, objects, facts or phenomena.58 This claim of radical non-totality is in one sense an axiomatic decision, but Badiou also seeks to motivate it on the level of ontological reflection by considering the implications of Russell’s paradox of the set of all sets that are not elements of themselves.59 Here, therefore, the world is understood as radically incomplete, or as fundamentally incapable of being gathered into a single totality, on pain of irreducible contradiction or paradox.60 From this perspective, the requirement to maintain the consistency of the language in which beings and phenomena are presented itself requires the non-existence of the world as a whole and in the singular; this position is extended, but not fundamentally modified, in Badiou’s 2006 sequel, Logics of Worlds, which posits an irreducible plurality of worlds as individually structured domains of presentation, each with its own (generally non-classical but consistent) logic. But another possibility, very much within the matrix of the same problems but suggesting a radically different answer to them, is the one explored by Graham Priest in a series of works: to explore the structure of the paradoxes that arise necessarily on the assumption of the existence of a totality of all that can be thought, or all that is linguistically expressible, or the “One” of all that is, accommodating them within a paraconsistent logic (one that does not assume the principle of ex falso quodlibet, according to which anything at all can be inferred from a contradiction) and maintaining the existence of actually true contradictions.61 These various contemporary projects are thus united in giving a presuppositional or foundational status to one or another assumption about the way in which formal and logical reasoning captures, or indicates the limits of, a consideration of the world as a whole. Accordingly, it appears possible to pose the question that divides them, that of the way in which formalism, at its limits, itself articulates the logical or meta-logical structure of the world as such. This is not, importantly, a question about the powers or limits of specific natural/historical languages, such as they might be thought to show up in the problems of mutual translation or in the question of the “incommensurability” of various languages with one another. It is, rather, essential to its posing—and, as I have argued, urgent in receiving the contemporary legacy of Wittgenstein and Heidegger—to understand it as a question about the structure and limits of language in the singular, or of language as such. It is true that twentieth-century and contemporary inheritors of the linguistic turn, even and perhaps especially in light of interpretations of Wittgenstein or Heidegger or both, have often denied the existence of any such thing, declaring rather the moral of an irreducible contingency and irremediable plurality. On this kind of position, there are indeed worlds in the plural, correspondent to the variety of cultures, languages, or practices, but there is no one world as the totality of all that is or all that can

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  241 be said or meant.62 The index and proof of this radical non-existence of the world-in-the-singular is often, here, the presumptive incommensurability of languages and cultures, the salutary differences that allow for the inscription in every communication and practical relationship the ineffable reserve of the “untranslatable” or the irreducibility of the individual, situated, or radically embodied. But if, as I have argued, the dynamics of the contemporary global regime of capital, technology, and information force us to consider in a renewed way the possibility of a critical reflection on the unity of the world, then it is relevant to consider the ways that both Heidegger and Wittgenstein (both early and late), through their considerations of language, meaning, and sense, themselves undertake this reflection and point toward its further development. What form might such a further development of the problematic of the relationship of the structure of language to the possibility of the world take, today? On the one hand, it would involve a renewed inquiry into the possible sense of the One of what is as it gives itself to be thought. This would be a contemporary repetition of the problem posed by Parmenides about the logical sense of being, to whose response Plato devoted the considerable resources of his late argumentation in dialogues such as the Parmenides, the Sophist, and the Philebus. On the other, and just as importantly, it would involve displacing this inquiry onto the transformed ground of contemporary logic and formal knowledge. To pursue the inquiry would then be (in the Heideggerian jargon) to grasp the very basis of linguistic sense as resting, not in factual institutions or given situations, but in the underlying or deeper structure of the granting of presence, the opening of the possibility of sense or the meaning of a life on the ultimate ground of what is. Or it would be to take up in a renewed way the inquiry into the essential and unitary dimension of form that arguably articulates, beyond or before empirical instances, the key concepts of “both” Wittgensteins, those of logical form and forms of life.63 And given both philosophers’ arguments, this would be something like the problem of the origin of sense in the abeyance of any source of sense in any subject, object, entity, fact, or situation. Because it is not possible to determine the sense of the world without contradiction from a position within it, and there is at the same time no position outside the world from which to determine it either, this inquiry would have to take up in a deepened way the critical question of what sets or establishes the limit of the world and its description, and how this setting itself complicates the easy topological distinction between a simple “inside” and a simple “outside.” The consideration of how possibilities and structures of sense arise, at or in the problematization of the boundaries of whatever is, and do so quite independently of any kind of human intentionality, activity, capacity, or purpose, might then also define a fitting topic for critical reflection within a philosophical praxis to come. Thus distinguished from the characteristic twentieth-century pathos of the human agentive subject of meaning and the

242  Paul M. Livingston constitutive anxiety of its nullification within a senselessly plural being, one task of twenty-first-century philosophy might then be to locate forms and structures of sense, in their original givenness, at the problematic limit of the world itself.64

NOTES   1 As Lee Braver has pointed out to me, in addition to the remark from Heidegger’s Le Thor seminar of 1969 that I will discuss below, Heidegger makes a brief mention of an analogy that he attributes to Wittgenstein in the seminar on Heraclitus (held jointly with Eugen Fink) of 1966–1967. See Heidegger and Fink (1967), p. 17. Gabriel Citron has pointed out to me one other remark attributed to Wittgenstein about Heidegger (in addition to the one discussed below): in a 1963 letter to Brian McGuinness, Rush Rhees relates that when he (Rhees) mentioned Heidegger to Wittgenstein, Wittgenstein “remarked that a man might be obscure and still have something important to say; but he added: ‘But I don’t trust Heidegger’.” See Wittgenstein, Rhees, and Citron (2015), p. 48.   2 Wittgenstein (1930), p. 68.   3 Murray (1974). The originally published text is Waismann (1965). Murray also discusses possible connections between Wittgenstein’s remarks and his “Lecture on Ethics,” as well as to Heidegger’s “What is Metaphysics?” lecture.   4 Heidegger (1927) (henceforth: S&Z), p. 186.   5 Wittgenstein (1921) (henceforth: TLP), 6.41–6.42.   6 Wittgenstein (1929), pp. 36–44.   7 Wittgenstein (1929), p. 39.   8 Wittgenstein (1929), p. 40.   9 Wittgenstein (1929), p. 40. 10 Wittgenstein (1929), p. 41. 11 Wittgenstein (1929), p. 44. 12 Wittgenstein (1929), p. 42. 13 Wittgenstein (1929), pp. 42–43. 14 Cf. TLP 6.41. 15 Heidegger (1929). As Murray (1974) suggests, it is certainly possible that Wittgenstein in late 1929 had some knowledge of the content of Heidegger’s lecture. 16 Heidegger (1929), pp. 99–100. 17 Heidegger (1929), p. 101. 18 Heidegger (1929), p. 101. 19 Heidegger (1929), p. 105. 20 S&Z, section 34. 21 S&Z, p. 160. 22 S&Z, p. 161. 23 S&Z, p. 161. 24 S&Z, p. 161. 25 S&Z, p. 161. 26 S&Z, p. 161. 27 S&Z, pp. 163–64. 28 S&Z, p. 161. 29 S&Z, p. 166. 30 “Keeping silent authentically is possible only in genuine discoursing. To be able to keep silent, Dasein must have something to say—that is, it must have at its disposal an authentic and rich disclosedness of itself. In that case one’s reticence [Verschwiegenheit] makes something manifest, and does away

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  243 with ‘idle talk’ [“Gerede”]. As a mode of discoursing, reticence Articulates the intelligibility of Dasein in so primordial a manner that it gives rise to a potentiality-for-hearing which is genuine, and to a Being-with-one-another which is transparent.” (S&Z, p. 165) 31 Carnap (1932). A number of recent commentators have explored the background of the famous controversy between Carnap and Heidegger, including the implications of Carnap’s attendance at the 1929 disputation in Davos between Heidegger and Cassirer: see, e.g., Friedman (2000) and Stone (2006). James Conant (2001) relates Heidegger’s position in “What is Metaphysics?” explicitly to Wittgenstein’s conception of sense and nonsense in the Tractatus, arguing that this provides a basis for resisting Carnap’s dismissal of it. 32 S&Z, p. 165. 33 TLP 6.421; 6.45. Partisans of the so-called “resolute” interpretation of the Tractatus typically oppose the claim that there is a substantive vision of the world that is mentioned or gestured at here, holding rather that the ethical sense of these remarks is exhausted by their use in exposing to critique our temptations to attempt to speak “metaphysically” of the world as a whole from beyond its boundaries. I do not take a position in the dispute between “resolute” and other more traditional interpretations here (but for some discussion of the issues, see e.g., Livingston (2008), chapter 3 and Livingston (2012), chapter 5). 34 Heidegger (1969). The original text is a protocol written in French which was read out in Heidegger’s presence at the seminar. 35 Kant (1763). 36 Heidegger (1969), pp. 35–36. 37 Cf., e.g., Heidegger’s opening discussion of the transition from the “guiding question” of the being of beings to the “grounding question” of the truth of being in the Beiträge zur Philosophie (Heidegger 1938, pp. 6–8). For the language of “beings as such and as a whole,” see, e.g., (Heidegger 1969b, p. 67.) 38 In the 1957 lecture “The Onto-Theo-Logical Constitution of Metaphysics,” Heidegger discusses the “historical stampings” [geschicklichen Prägungen] of being as “phusis, logos, hen, idea, energeia, substantiality [Substanzialität], objectivity, subjectivity, will, will to power, will to will.” (Heidegger 1969b, p. 134). For a helpful discussion, see also Thomson (2005), esp. chapter 1. 39 S&Z, p. 151. 40 S&Z, sect. 44. 41 E.g. Heidegger (1938), pp. 75–77. 42 The misquotation appears, in German, already in the original French protocol. In the French text, the gloss on it that immediately follows reads: “(est réel ce qui est le cas; ce qui veut dire: ce qui tombe sous une détermination, le fixable, le déterminable)” (p. 260). At TLP 2.063, Wittgenstein identifies the world with “Die gesamte Wirklichkeit” or the “sum-total of reality,” and at 2.06 he says that “The existence and non-existence of states of affairs is reality” (“Das Bestehen und Nichtbestehen von Sachverhalten ist die Wirklichkeit”); along similar lines, TLP 2.04 identifies the world with the “totality of existing states of affairs” (Die Gesamtheit der Bestehenden Sachverhalte). It is thus possible (though not seemingly likely, from the context) that Heidegger’s substitution of “Wirklich” for “Die Welt” in his attribution to Wittgenstein is intended as, at least in part, a gloss on these later remarks (or a combination of one or both of them with remarks 1 and/or 1.1). Even if this is so, however, there remains an important gap between the claim that the world, in the sense of all that is the case (or of the obtaining and non-obtaining of states of affairs, as in 2.06), is identifiable with “reality” (Wirklichkeit) as a whole and the different claim, which Heidegger effectively attributes to Wittgenstein, that the criterion for something’s being real (Wirklich) is its being determinable as a fact. I am indebted to Conrad Baetzel for pointing out the possible relevance of the

244  Paul M. Livingston remarks at TLP 2.04, 2.06, 2.063 to Heidegger’s reading of Wittgenstein here. The English translators of Heidegger’s seminar also note the misquotation and suggest the possibility of reading it in terms of TLP 2.063 (p. 107). 43 Heidegger (1969), p. 36 (translation slightly modified). 44 Heidegger (1969), pp. 36–37. 45 As Andrew Cutrofello has pointed out, the relationship that Heidegger draws between “Wirklichkeit” in the sense of the “real” and the sense of the rationally calculable “effective” can also be related to Hegel’s famous “identification” of the real and the rational in the Elements of the Philosophy of Right: ““Was vernünftig ist, das ist wirklich; und was wirklich ist, das ist vernünftig.” 46 See, e.g., Heidegger (1941, p. 445 (transl. slightly modified)): “The precedence of what is real [der Vorrang des Wirklichen] furthers the oblivion of Being [betreibt die Vergessenheit des Seins]. Through this precedence, the essential relation to Being which is to be sought in properly conceived thinking is buried. In being claimed by beings, man takes on the role of the authoritative [maβgebende] being. As the relation to beings, that knowing suffices which, according to the essential manner of beings [Wensensart des Seienden] in the sense of the planned and secured real [des planbar gesicherten Wirklichen] must issue into objectification and thus to calculation. The sign of the degradation of thinking [Herabsetzung des Denkens] is the elevation of logistics [Hinaufsetzung der Logistik] to the rank of the true logic. Logistics is the calculable [rechenhafte] organization of the unconditional lack of knowledge about the essence of thinking, provided that thinking, essentially thought, is that projecting knowledge which issues from Being in the preservation of truth’s essence [das in der Bewahrung des Wesens der Wahrheit aus dem Sein aufgeht].” 47 Heidegger (1969), p. 38. 48 TLP 2.02–2.0212; 3.142; 3.203–3.23. 49 TLP 2.021, 2.026. 50 Wittgenstein (1953), sect. 50, 81. 51 Cf. Livingston (2003). 52 See, e.g., Frege (1915) for the problem of the separation of assertoric force from content. 53 At TLP 6.41, Wittgenstein uses the phrase “the sense of the world,” saying specifically that it “must lie outside the world.” What does he mean by it? The paragraph goes on to deny that there is value (“Wert”) in the world, since “in the world everything is as it is, and everything happens as it does happen,” and moreover “all that happens and is the case is accidental.” In terms of the Tractatus’s structure of logical importance, the remark is moreover directly subordinate to 6.4, which states simply that “All propositions are of equal value.” [Alle Sätze sind gleichwertig.] Sense, value, and the non-accidental are, here, all closely aligned with one another, and jointly opposed to the totality of the world, understood as the totality of what can be described in (contingent) propositions. The “sense of the world” is thus, it is reasonable to assume, connected with the source of whatever value the world has, and it may not be amiss to understand this source as the “metaphysical subject” that “does not belong to the world” but is rather “a limit of the world” (5.632) and is thus to be sharply distinguished from the “subject that thinks or entertains ideas”[das denkende, Vorstellende, Subjekt] (which itself, according to 5.631, does not exist) (On the other hand, it may also not be incorrect to understand “the sense of the world” in terms of the characterization of “sense” suggested (e.g.) at 3.11–3.12, according to which a proposition is a sign “in its projective relation to the world” and the “method of projection” is “to

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  245 think of” the proposition’s sense. Synthesizing all of this, then, the “sense of the world” might be taken to be what is (as it were) projected in a thinking of the sense of the totality of propositions—or of propositions as such—from the “perspective” of the “metaphysical subject.” But this would again be sharply distinguished from anything that is or could be thought or represented by the (actually non-existent) innerworldly “subject that thinks” or represents. 54 It is here we can see the specific limitation, with respect to the current problematic, of Quentin Meillassoux’s recent realist challenge to what he calls “correlationism,” understood as a broad label for any position that holds that “we can only ever have access to the correlation between thinking and being, and never to either term considered apart from the other” and hence that (as he puts it) it is impossible “to consider the realms of subjectivity and objectivity independently of one another” (Meillassoux (2008), p. 5). This is reasonable as a characterization of some Kantian and post-Kantian idealist philosophy, but when Meillassoux goes on to attribute such a “correlationism” to both Heidegger and Wittgenstein (pp. 41–42), he misses the essential way in which the question of sense articulates both of their projects differently than any that accords a central role to questions of (epistemic) “access,” subjectivity, and objectivity. As a result, the critique fails to meet its target and many of the positions based on it largely miss the significance for the question of realism of the twentieth-century linguistic turn itself (see Livingston (2013) for a more detailed development of this point). 55 See Hegel (1812), especially pp. 157–65. 56 For connections and parallels between Kant’s cosmological antinomies and set-theoretical paradox, see, e.g., Hallett (1986), pp. 223–234; Moore (1990), pp. 149–51; and Priest (2002), pp. 87–102. 57 This is not to say, of course, that contemporary representatives of the “analytic” tradition are univocal in considering the idea of the totality of the world to be intelligible. An important strand of contemporary argumentation, inherited ultimately from the later Carnap’s idea of a plurality of language frameworks (see Carnap 1950) and perhaps passing through Putnam’s “internal realism,” challenges the idea of a unitary world by maintaining that any description of facts or phenomena is irreducibly situated within one or another theory, domain of discourse, or set of logical operators. This strand of argumentation is also by no means limited to projects that continue to see themselves as situated within the “linguistic turn” itself, but is just as often today articulated in terms of “metaphysical” or “metametaphysical” claims about the scope and structure of logic quite irrespective of considerations about natural languages (as recent discussions about “quantifier variance,” for example, witness.) 58 Badiou (1988), pp. 23–25. 59 Badiou (1988), p. 31; pp. 38–48. 60 For a closely related recent argument for the non-existence of the world in this sense, made from a partly Schellingian perspective, see Gabriel (2013). 61 See, e.g., Priest (2002, 2014, 2015). For a formally based framework that situates the differing positions outlined here in terms of the orientations of thought they represent, see Livingston (2012), chapter 1. 62 Cf. the conclusion that Richard Rorty draws from a consideration of the positions and development of the early and late Wittgenstein as well as the early and late Heidegger in his 1989 article “Wittgenstein, Heidegger and the Reification of Language.” In the article, Rorty opposes the tendency to “reify” or “hypostatize” language to his own “pragmatist” inclinations, which involve emphasizing instead the contingency of all languages and their thorough embedding in historically situated practices. This opposition produces a reading according to which Wittgenstein and Heidegger “passed each other

246  Paul M. Livingston in mid-career, going in opposite directions.” (Rorty 1989, p. 52), the former “advancing” in the direction of pragmatism and the rejection of the very idea of philosophy as positive theory, while the latter regressed from the social pragmatism formulated (on Rorty’s reading) in Being and Time to the later mysticism marked by the grand being-historical narrative of the totality and closure of “Western metaphysics,” a narrative of which Rorty himself is suspicious. Both Rorty’s characterization of the developments of the two philosophers and his critique of the tendency to “reify” language are, however, consequences of his own presumptive dismissal of the problematic of language and world that we have considered here: according to Rorty, for example, it is essential in avoiding the “reification” that he inveighs against “that we not think either of language in general or a particular language (say, English or German) as something which has edges, something which forms a bounded whole and can thus become a distinct object of study or of philosophical theorizing.” Insofar as Rorty cites here a basis for this dismissal, it is the moral he draws from Donald Davidson’s project of semantic analysis and its “naturalized” approach. However, it is certainly possible to read the implications of Davidson’s project differently, as suggesting both the positive possibility of a “transcendental” or quasi-transcendental analysis of the positive conditions of linguistic truth and meaning that itself develops the problematic of the structure of language and world considered here (cf., e.g., Wrathall, Okrent, Livingston 2014). 63 Although Wittgenstein himself uses “form of life” both in the plural or with an indefinite article (as in, e.g., PI sections  19 and 23) and in the singular (as, e.g., in PI 241, where he speaks of agreement “in language” as “agreement . . . in form of life”), it is more or less standard in many contemporary interpretations to read “forms of life” as irreducibly a plural concept of varying cultures or lifestyles. However, Giorgio Agamben has insightfully and partially independently developed “form-of-life” as an essentially unitary concept capturing the “linguistic being” or “belonging in language” that formally characterizes a linguistic life. See, e.g., Agamben (1996), p. 11: “Intellectuality and thought are not a form of life among others in which life and social production articulate themselves, but they are rather the unitary power that constitutes the multiple forms of life as form-of-life.” See also Agamben’s recent investigation of the relationship between form-of-life and rule in the development and practice of Western monasticism (Agamben 2011) and Livingston (2012), chapter 1. 64 I would like to thank Lee Braver and Iain Thomson for providing helpful comments on an earlier version of this article.

WORKS CITED Agamben, G. 1996 (2000). Form-of-life. In Vincenzo Binetti and Cesare Casarino, trans., Means Without End: Notes on Politics. Minneapolis, MN: University of Minnesota Press, 3–14. Agamben, G. 2011 (2013). The Highest Poverty: Monastic Rules and Form-of-Life. Trans. Adam Kotsko. Stanford, CA: Stanford University Press. Badiou, A. 1988 (2005). Being and Event. Trans. Oliver Feltham. London: Continuum. Carnap, R. 1932 (1959). The Elimination of Metaphysics Through Logical Analysis of Language. In A. J. Ayer, ed., Logical Positivism. New York: The Free Press, 60–81. Carnap, R. 1950. Empiricism, Semantics, and Ontology. Revue Internationale De Philosophie 4 (2): 20–40.

Wittgenstein Reads Heidegger, Heidegger Reads Wittgenstein  247 Conant, J. 2001. Two Conceptions of Die Überwindung der Metaphysik: Carnap and Early Wittgenstein. In T. McCarthy and S. C. Stidd, eds., Wittgenstein in America. New York: Oxford University Press, 13–61. Frege, G. 1915 (1997). My Basic Logical Insights. Trans. Peter Long and Roger White. In Michael Beaney, ed., The Frege Reader. Oxford: Blackwell, 322–324. Friedman, M. 2000. A Parting of the Ways: Carnap, Cassirer, and Heidegger. Chicago: Open Court. Gabriel, M. 2013. Warum es die Welt nicht Gibt. Berlin: Ullstein. Hallett, M. 1986. Cantorian Set Theory and Limitation of Size. Oxford: Clarendon. Hegel, G. W. F. 1812 (2010). The Science of Logic. Trans. and ed. George di Giovanni. Cambridge: Cambridge University Press. Heidegger, M. 1927 (2006). Sein und Zeit, 19te Auflage. Tübingen: Max Niemeyer Verlag. Translated as Being and Time: A Translation of Sein und Zeit, by John Macquarrie and Edward Robinson. San Francisco: Harper, 1962. Heidegger, M. 1929 (1993). What Is Metaphysics? Trans. D. F. Krell. In D. F. Krell, ed., Martin Heidegger: Basic Writings, revised and expanded edition. Harper San Francisco, 89–110. Heidegger, M. 1938 (2012). (GA 65) Beiträge zur Philosophie (Vom Ereignis), Herausg. von. Friedrich- Wilhelm von Herrmann (Frankfurt: Klostermann, 1989). Translated as Contributions to Philosophy: Of the Event. Trans. by Richard Rojcewicz and Daniela Vallega-Neu, Bloomington, IN: Indiana University Press. Heidegger, M. 1941 (1997). Recollection in Metaphysics. In GA 6, vol. 2: Nietzsche 2. Herausg. von Brigitte Schillbach. Frankfurt: Klostermann. Partially translated as Nietzsche, vol. 3: The Will to Power as Knowledge and as Metaphysics and Nietzsche, vol. 4: Nihilism, ed. by David F. Krell, Harper Collins, 1991. Heidegger, M. and Fink, E. 1967 (1993). Heraclitus Seminar. Trans. by Charles H. Seibert Evanston, IL: Northwestern University Press. Heidegger, M. 1969a (2006). Seminar in Le Thor 1969. In Andrew Mitchell and Francois Raffoul, trans., Four Seminars. Bloomington, IN: Indiana University Press. Translation of “Seminar in Le Thor 1969” in Heidegger, Seminare (GA 15) (Frankfurt am Main: Klostermann, 1986) and “Séminaire du Thor, 1969” in Heidegger, Questions IV (Paris: Gallimard, 1976). Kant, I. 1763 (1992). The Only Possible Argument in Support of a Demonstration of the Existence of God. In David Walford, ed. and trans., Theoretical Philosophy, 1755–1770. Cambridge: Cambridge University Press, 107–201. Livingston, P. M. 2003. Thinking and Being: Heidegger and Wittgenstein on Machination and Lived-Experience. Inquiry 46 (3): 324–345. Livingston, P. M. 2008. Philosophy and the Vision of Language. New York: Routledge. Livingston, P. M. 2012. The Politics of Logic: Badiou, Wittgenstein, and the Consequences of Formalism. New York: Routledge. Livingston, P. M. 2013. Realism and the Infinite. Speculations: A Journal of Speculative Realism 4: 99–117. Livingston, P. M. 2014. The Sense of Finitude and the Finitude of Sense. In Piotr Stalmaszczyk, ed., Semantics and Beyond: Philosophical and Linguistic Investigations. Berlin: Walter de Gruyter, 161–184. Meillassoux, Q. 2008. After Finitude: An Essay on the Necessity of Contingency. New York: Continuum. Moore, A. W. 1990 (2001). The Infinite. 2nd ed. New York: Routledge. Murray, M. 1974. A Note on Wittgenstein and Heidegger. The Philosophical Review 83 (4) (October): 501–503. Priest, G. 2002. Beyond the Limits of Thought. 2nd ed. New York: Oxford University Press. Priest, G. 2014. One: Being an Investigation into the Unity of Reality and of its Parts, including the Singular Object Which Is Nothingness. New York: Oxford University Press.

248  Paul M. Livingston Priest, G. 2015. The Answer to the Question of Being. This volume. Rorty, R. 1989 (1991). Wittgenstein, Heidegger, and the Reification of Language. In Essays on Heidegger and Others: Philosophical Papers, Volume 2. Cambridge: Cambridge University Press, 50–65. Stone, A. 2006. Heidegger and Carnap on the Overcoming of Metaphysics. In S. Mulhall, ed., Martin Heidegger. Aldershot: Ashgate, 217–244. Thomson, I. 2005. Heidegger on Ontotheology: Technology and the Politics of Education. Cambridge: Cambridge University Press. Waismann, F. 1965. Notes on Talks with Wittgenstein. The Philosophical Review 74 (1) (January): 12–16. Wittgenstein, L. 1921 (1974). Tractatus Logico-Philosophicus. Trans. D. F. Pears and B. F. McGuinness. London: Routledge. Wittgenstein, L. 1929 (1993). A Lecture on Ethics. In James Klagge and Alfred Nordmann, eds., Philosophical Occasions 1912–1951 Minneapolis, IN: Hackett. Wittgenstein, L. 1930 (2001). Ludwig Wittgenstein und der Wiener Kreis. Gespräche, aufgezeichnet von Friedrich Waismann (Suhrkamp, 2001), 68. (Translated as Ludwig Wittgenstein and the Vienna Circle: Conversations Recorded by Friedrich Waismann. Trans. by Brian McGuinness and Joachim Schulte (New York: Barnes and Noble Books, 1979)). Wittgenstein, L. 1953 (2009). Philosophische Untersuchungen. Trans. as Philosophical Investigations [in German and English.]. Rev. 4th ed. Trans. G. E. M. Anscombe, P. M. S. Hacker, and Joachim Schulte. Chichester. West Sussex, UK: Wiley-Blackwell. Wittgenstein, L., R. Rhees, and G. Citron. 2015. Wittgenstein’s Philosophical Conversations with Rush Rhees (1939–50): From the Notes of Rush Rhees. Mind 124 (493) (January): 1–71.

11 The Answer to the Question of Being Graham Priest

1. INTRODUCTION At the beginning of Sein und Zeit, Heidegger famously asked the question of being: what is it to be?1 The question was to exercise him, in one way or another, for the rest of his philosophical life. He came to the conclusion that, because of a certain aporia, the question could not, in fact, be answered. One cannot say what being is. Being shows itself, that is, beings show their being, and the job of the thinker is to open people’s eyes to see this showing.2 Heidegger notwithstanding, I think that the question of being can be answered, and the point of this essay is to do so. In the process we will also investigate the ground of Heidegger’s aporia and its implications. These matters will involve a certain construction concerning unity and its possibility, in the shape of gluon theory. In the next section, we will look at Heidegger and his predicament more closely. I will then explain enough of the theory of gluons to make what follows intelligible. After that, we will be in a position to answer the question of being. In the final section will see how the theory of gluons explains Heidegger’s aporia. 2. HEIDEGGER To start with, Heidegger.3 Heidegger poses the question of being as follows:4 What is asked about in the question to be elaborated is being, that which determines beings as beings, that in terms of which beings have always been understood no matter how they are discussed. The being of a being, then, is what it is which makes it be.5 And one thing that Heidegger is clear about from the very beginning is that being is not itself another being. As he puts it:6 The being of beings ‘is’ not itself a being. The first philosophical step in understanding the problem of being consists in avoiding telling the

250  Graham Priest mython tina diegeisthai, in not ‘telling a story’, that is, not determining beings as beings by tracing them back in their origins to another being—as if being had the character of a possible being. He takes this to be so obvious that he does not give an argument for it in what follows. Indeed, it is actually hard to find arguments for the claim in the Heideggerian corpus. But one might essay a couple: one metaphysical, one grammatical. The metaphysical one is Neo-Platonist. Being is the ground of beings. As such, it is not the kind of thing that can be a being. It can function as the ground only if it, itself, is beyond being, and so not a being.7 The grammatical argument is somewhat different. We say, for example: Heidegger is. ‘Heidegger’ refers to an object, a being. If ‘is’ referred to a being then the italicized phrase would simply be a list of two objects: Heidegger and being. But it is clearly not a simple list. In an obvious sense, it has a unity that a pair of objects lacks. One can hear both of these arguments in the following passage:8 If we painstakingly attend to the language in which we articulate what the principle of reason [Satz vom Grund] says as a principle of being, then it becomes clear we speak of being in an odd manner that is, in truth, inadmissible. We say: being and ground/reason [Grund] ‘are’ the same. Being ‘is’ the abyss [Ab-grund]. When we say something ‘is’ and ‘is such and so’, then that something is, in such an utterance, represented as a being. Only a being ‘is’; the ‘is’ itself—being—‘is’ not. The wall in front of you and behind me is. It immediately shows itself to us as something present. But where is its ‘is’? Where should we seek the presencing of the wall? Probably these questions already run awry. Whatever the reason, though, if being is not itself a being, we have a problem. If one is to answer the question of being one must say something of the form: being is such and such. This uses ‘being’ as a noun phrase, and so treats its referent as an object. It follows that one cannot answer the question of being. I take this insight to be what drives much of Heidegger’s later thought (though, again, I am aware of nowhere he says this explicitly).9 Being cannot be said; it can only be shown. If one has the eyes to see it, beings show their being. One’s eyes can be opened by art, poetry, language—before it becomes a dead metaphor. As he says in “On the Origins of the Work of Art”:10 The art work opens up in its own way the Being of beings. This opening up, i.e., this revealing, i.e., the truth of beings, happens in the work. In the art work, the truth of beings has set itself to work. Art is truth setting itself to work.

The Answer to the Question of Being  251 A solution to the question of being can, therefore, consist only in helping to open people’s eyes in this way. But discerning eyes will also perceive an aporia here. Never mind answering the question of being; if one cannot refer to being with a noun phrase, one cannot even ask it. (‘What is being?’) Indeed, for exactly this reason, one can say nothing at all of being. Yet, Heidegger’s own works are replete with statements about being. (Just look at the quotes above.) To bend a comment from Russell’s introduction to Wittgenstein’s Tractatus—which finds itself with a similar aporia:11 Everything involved in talking of being cannot, grammatically, be said. What may give some hesitation about this fact is that, despite his arguments to the contrary, Mr. Heidegger manages to say a good deal about what cannot be said. Heidegger was, of course, well aware of the problem, and he wrestled with various ways to avoid it. Thus, for example, he tried the technique of writing under erasure:12 a thoughtful glance ahead into the realm of ‘Being’ can only write it as Being. The crossed lines at first only repel, especially the almost ineradicable habit of conceiving ‘Being’ as something standing by itself . . . Nothingness would have to be written, and that means thought of, just like Being. The failure of this strategy is manifest, though. Heidegger has to talk about being in order to explain what it is that the erasure shows is not to be taken as standing by itself. As far as I know, Heidegger never solved this problem. Maybe, having no cure, he learned to live with the disease. Or maybe, once he figured out that the action was in opening people’s eyes in a certain way, it became irrelevant to explain what it is that is seen. As Zhangzi says (in a quite different context):13 A fish-trap is for catching fish; once you’ve caught the fish, you can forget about the trap. A rabbit-snare is for catching rabbits; once you’ve caught the rabbit, you can forget about the snare. Words are for catching ideas; once you’ve caught the idea, you can forget about the words. 3.  UNITY AND GLUON THEORY So much for the Heideggerian background. Before we turn to the answer to the question of being, I need to explain the background theory that informs it. This concerns unity and what makes it possible. I shall not attempt to justify the theory here, or to show its technical coherence. I do that elsewhere.14 I wish merely to explain the essence of the view in question.

252  Graham Priest Take an object with parts. What makes them into a single thing? There must be something in virtue of which they form a unity. Quite possibly, this thing depends on the unity in question. If the unity is a house, its parts are bricks, and maybe what makes them into a unity is their geometric configuration. If the unity is a symphony, its parts are notes, and maybe what makes them into a unity is their arrangement. But whatever the binding agent is, there must be one. Let us call it, neutrally, the gluon of the unity. It does not take a lot of thought to see that gluons are very strange things: they appear to have contradictory properties. A gluon is an object: we can think about it, quantify over it, refer to it. But it is not an object: if it were, the totality comprising it and the other parts would be just as much a congeries as the parts themselves, and we would want for an explanation of how the unity is achieved. Think of Bradley’s regress at this point. If the gluon were just another object, there would need to be a ‘hypergluon,’ holding the gluon and the other parts together. And so on . . . We are off on a vicious regress. So the gluon is an object and not an object. But how is it that it holds all the parts (including itself) together? If it were distinct from the other parts, Bradley’s regress would strike. It must, therefore, be identical with each of the other parts. How to make sense of this idea? We may define identity in the standard fashion, deploying Leibniz’ identity of indiscernibles: x = y is ∀Z(Zx ≡ Zy), where the quantifier is second-order, and the connective is the material biconditional. Note, however, that we are dealing with objects some of which have contradictory properties. Hence this must be the material biconditional of some paraconsistent logic (a logic that tolerates contradictions without blowing up).15 So suppose that we have a unity; let its gluon be g, and let a, b, c, and d be its (other) parts. Then g = a, g = b, g = c, and g = d:

=

d

=

a = g = c b

⊥

This will happen if g has all the properties of a, b, c, and d. Naturally, since the parts are liable to have a variety of disparate properties, g is liable to be an inconsistent object—but we knew that already. Note, also, that in virtually all paraconsistent logics, the material biconditional is not transitive: A ≡ B, B ≡ C, A ≡ C. So the fact that a = g and g = b does not entail that a = b. The various parts are not, generally speaking, identical. For a very simple illustration of how all this works, suppose that the parts are just a, b, and g, and that there is only one property at issue, P. Suppose that (consistently) Pa, ¬Pb, but that Pg ∧ ¬Pg. Then Pa≡ Pg and Pg≡

The Answer to the Question of Being  253 Pb, but it is not the case that Pa≡ Pb. Hence (since P is the only property at issue), a = g, g = b, but it is not the case that a = b. Note, also that since Pg ∧¬Pg, it follows that ¬(Pg ≡Pg), and so ∃X¬(Xg ≡ Xg), and ¬∀X(Xg ≡ Xg). That is, g ≠ g: g is not self-identical. Finally, and importantly for what follows, gluons are and are not objects/ beings.16 To say that something is an object is to say that it is something: ∃y y = x. This follows from the logical truth that x = x. So every x is an object. Everything is a being, gluons included. But as we have just seen, if g is a gluon, then g ≠ g. But if g ≠ g then, for any y, y ≠ g. For either y = g or y ≠ g. And in the first case, the result follows by the substitutivity of identicals. (Substitute y for g in the first occurrence of g ≠ g.) Hence, ∀y¬y = g. That is, ¬∃y y = g; that is, g is not an object. 4.  THE ANSWER TO THE QUESTION We are now in a position to answer the question of being.17 But first, we need to be clear about the sense of being involved. To be a being for Heidegger is simply to be an object; that is, to be anything that one can think of, predicate something of, refer to. As Heidegger says:18 Everything we talk about, mean, and are related to is in being in one way or another. What and how we ourselves are is also a being. Being is found in thatness and whatness, reality, the objective presence of things [Vorhandenheit], subsistence, validity, existence [Da-sein], and in the ‘there is’ [es gibt]. And:19 Being is used in all knowledge and all predicating, in every relation to beings and in every relation to oneself, and the expression is understandable without ‘further ado.’ Everyone understands ‘The sky is blue,’ ‘I am happy,’ and similar statements. That is, to be a being/object, is, in Meinongian terms, to have Sosein.20 Given this understanding of being, it is clear that being and unity come to the same thing. If something is an object, it is one thing; and if it is one thing, it is certainly an object. The thought is, in fact, as old as Aristotle. As he puts it (Met. 1003b 23–31):21 being and unity are the same and are one thing in the sense that they are implied by one another as principle and cause are . . . ; for one man and a man are the same thing and [man who is] and a man are the same thing, and the doubling of the words in ‘one man’ and ‘one [man who

254  Graham Priest is]’ does not give any new meaning . . . ; and similarly with ‘one’, so that it is obvious that the addition in these cases means the same thing, and unity is nothing other than being. And again (Met. 1054a13–19): And that in a sense unity means the same as being is clear from the fact that it follows the categories in as many ways, and is not comprised within any category, e.g. neither in substance nor in quality, but is related to them just as being is; and from the fact that in ‘one man’ nothing more is predicated than in ‘man’, just as being is nothing apart from substance or quality or quantity; and to be one is just to be a particular thing. The thought is expressed—in more pellucid terms—by Plotinus, at the beginning of Ennead VI, 9:22 It is in virtue of unity that beings are beings. This is equally true of things whose existence is primal and of all those that are in any degree to be numbered among beings. What could exist at all except as one thing? Deprived of unity, a thing ceases to be what it is called: no army unless a unity: a chorus, a flock, must be one thing. Even house and ship demand unity, one house, one ship; unity gone, neither remains; thus even continuous magnitudes could not exist without inherent unity; break them apart and their very being is altered in the measure of the breach of unity. Take plant and animal; the material form stands a unity; fallen from that into a litter of fragments, the things have lost their being; what was is no longer there; it is replaced by quite other things—as many others, precisely, as possess unity. To be, then, is exactly to be one. One might object. It would seem that we have plural forms of reference. Thus, we can say, for example, that Russell and Whitehead wrote Principia Mathematica, and that they were in Cambridge together at the time. The conjunctive noun phrase and pronoun appear to refer to objects that are inherently plural. Similarly, one can say that something is a square; but one also can say that some things have the same shape as each other. The italicized quantifier is plural, and refers to a plurality.23 There are, then, objects that are, but are not one, being a plurality. Russell and Whitehead, for example is (an object), but it is not one object. The reply is simple, however. The machinery does not allow us to refer to objects that are plural, but to a plurality of objects. Thus, when we say that Russell and Whitehead wrote Principia, we are not referring to some strange object, Russell and Whitehead; we are referring to Russell and to

The Answer to the Question of Being  255 Whitehead. (One cannot say: ‘Russell and Whitehead wrote Principia, and it was in Cambridge at the time.) Similarly, if we use plural pronouns and quantification we are referring to multiple objects. If something is, it is one, a unity; and if some things are, they are ones, unities. The machinery of plural reference does indeed enable one to refer to a plurality of objects, but each is one. So to be is still to be one. Let us now put all these thoughts together. The being of something is that in virtue of which it is. To be is to be one. So the being of something is that in virtue of which it is one. And what is it in virtue of which something is one? By definition, its gluon, g. The being of something is therefore its gluon. We have answered Heidegger’s question as to the nature of being.24 (To be clear about the exact form of this argument used here: We have established that the conditions ‘x is one thing’ and ‘x is’ are necessarily equivalent. Hence, the inference: ‘g makes it the case that x is one thing; hence g makes it the case that x is’ is an application of the substitutivity of such equivalents.) 5.  HEIDEGGER’S APORIA We are not finished yet. We still have to deal with Heidegger’s aporia. He is forced to talk about what cannot be talked about. How so? We are now in a position to see how. The being of any object/being, x, is its gluon. A gluon both is and is not an object/being. Since it is an object, it can have a name ‘α’ (‘the being of x’) Hence, one can talk about it by saying things like ‘α is not a being.’ But it is not an object/being as well. Since it is not an object, it cannot have a name. And if it cannot have a name, one cannot refer to it, and so one can say nothing about it. That, of course, is a contradiction. But that is exactly the terrain we are in.25 We can, in fact, make the matter quite precise. The intuitively correct principle governing truth is the T-schema. For every sentence, A: • ‘A’ is true iff (if and only if) A Thus, ‘Socrates was human’ is true iff Socrates was indeed human. Analogously, the intuitively correct principle governing denotation is the D-schema. For every name, ‘n’: • ‘n’ denotes x iff n = x Thus, ‘Socrates’ denotes Plato’s teacher iff Socrates is Plato’s teacher. Now, since is not α an object, ¬∃x x = α That is ∀x x ≠ α. So in particular, if ‘n’ is any name, n ≠ α. So by the D-schema (and contraposition), ‘n’ does not denote α. α has no name. Hence, one can say nothing about it, for to say something about it, one has to have some name to refer to it.

256  Graham Priest Heidegger was right, then, in his conclusion that one cannot say anything about the being of an object (even though one can)! And in case one thinks this is some peculiarity of Heideggerian philosophy, it is worth noting that the very same situation occurs in some paradoxes of self-reference. Thus, take König’s Paradox.26 This concerns ordinals. Ordinals are numbers that extend the familiar counting numbers, 0, 1, 2, . . . beyond the finite. Thus, after all the finite numbers there is a next, ω, and then a next, ω + 1, and so on. Crucially, ordinal numbers preserve the property of the counting numbers that any non-empty collection of them has a least member. How far, exactly, the ordinals go is a somewhat vexed question, both mathematically and philosophically, but it is not contentious that there are many more ordinals than can be referred to by names of a language with a finite vocabulary, such as English. This can be shown by a perfectly rigorous mathematical proof. Now, if there are ordinals that cannot be referred to in this way, then, by the properties of the ordinals, there must be a least. Consider the phrase ‘the least ordinal that cannot be referred to.’ This obviously refers to the number in question, and one can use it to say things about it, such as that it is the least ordinal that cannot be referred to. Yet, since it cannot be referred to, one can say nothing of it. In this regard, it is just like a gluon, and in particular, the being of an object. 6.  AND SO . . . We have answered Heidegger’s Seinsfrage. The being of a being is its gluon. Heidegger was just wrong to give up trying to answer the question. (Of course, the fact that being can be said does not—Wittgenstein ­notwithstanding— mean that it cannot be shown. I can tell you what a cricket bat is and show you one.) The answer, it must be admitted, uses the techniques of modern mathematical logic, and in particular paraconsistent logic, something that Heidegger could have known little about, since he died just about the time that this was being invented. The answer is none the worse for that. Nonetheless, Heidegger was right in insisting that the being of a being is not itself an object/being, and in concluding that one could therefore say nothing about it. The answer to the question of being requires us to talk of the ineffable. Being is indeed a strange beast.

NOTES   1 Some who translate Heidegger into English use a capital ‘B’ for being, and a lower case ‘b’ for beings. Though this has a certain point, it is entirely artifactual. In German, both words, being nouns, begin with capitals (Sein, Seindes). So I will not follow this practice, though I have not changed quotations that do so.   2 This is spelled out in a number of later essays, such as “The Origin of the Work of Art” and “What Calls for Thinking?” (Krell 1977, chs. 3 and 9).

The Answer to the Question of Being  257   3 For a more detailed discussion of some of the following matters, see Priest (2002a), ch. 15.   4 Stambaugh (1996), p. 4f.   5 Heidegger is clear that being is always the being of a being. He was, as it were, an Aristotelian about this universal. Thus we have (Stambaugh 2002, p. 61): ‘If we think of the matter just a bit more rigorously, if we take more heed of what is in contest in the matter, we see that Being means always and everywhere: the Being of beings.’   6 Stambaugh (1996), p. 5.   7 Heidegger’s discussion of being as ground can be found in “The Essence of Ground” (McNeill 1998, pp. 97–135) and The Principle of Reason (Lilly 1991). See also the relevant discussions in Caputo (1986) and Braver (2012).   8 Lilly (1991), p. 51f.   9 For some relevant discussion of the need for silence, see Sections 37 and 38 of Contributions to Philosophy (of the Event) (Rojcewicz and Vallega-Neu 2012). 10 Krell (1977), p. 166. 11 Pears and McGuiness (1961), p. xxi. 12 Kluback and Wilde (1959), p. 81. 13 Mair (1994), p. 277. 14 Priest (2014a), chs. 1, 2. 15 For a very non-technical introduction to paraconsistent logic, see Priest (1998); for a somewhat more technical account, see Priest (2008), ch. 7; for a very technical account, see Priest (2002b). 16 Strictly speaking, this is guaranteed only for objects with multiple parts. As far as Heidegger goes, I think we can simply assume that all objects have parts. What to say about objects with no parts, simpleces, is discussed in Priest (2014a), ch. 4. 17 The matter is discussed at greater length in Priest (2014a), ch. 4. 18 Stambaugh (1996), p. 5. 19 Stambaugh (1996), p. 5 20 See Priest (2014b). 21 The translation is from Barnes (1984), except that ‘man who is’ is my rendering of the text’s ‘existent man.’ This, I think, is more accurate. 22 Translation from MacKenna (1991), pp. 535–536. 23 See, e.g., Yi (2005). 24 And if a Heideggerian wants to object that the being of a being is not a unity, the reply is: Of course: it is not an object! But it is both an object and a one for all that. (You can, after all, speak of it.) That is the aporia. 25 Priest (2002a), ch. 15, argues that Heidegger should have been a dialetheist. In his forthcoming Ph.D. thesis (University of St. Andrews) Filippo Casati, drawing on some of Heidegger’s later writings that have been translated only recently into English, argues that he actually became one—at least in private. 26 See, e.g., Priest (2002a), pp. 131–134.

WORKS CITED Barnes, J., ed. 1984. The Complete Works of Aristotle. Princeton, NJ: Princeton University Press. Braver, L. 2012. Groundless Grounds: A Study of Heidegger and Wittgenstein. Cambridge, MA: Massachusetts Institute of Technology Press. Caputo, J. 1986. The Mystical Element in Heidegger’s Thought. New York, NY: Fordham University Press.

258  Graham Priest Kluback, W., and J. T. Wilde, trans. 1959. The Question of Being. Harrisonburg, VA: Vision. Krell, D. F., ed. 1977. Martin Heidegger: Basic Writings. New York, NY: Harper and Row. Lilly, R., trans. 1991. The Principle of Reason. Bloomington, IN: Indiana University Press. MacKenna, S., trans. 1991. The Enneads of Plotinus. London: Penguin Books. Mair, V., trans. 1994. Wandering on the Way: Early Taoist Tales and Parables of Chuang Tzu. New York, NY: Bantam Books. McNeill, W., trans. 1998. Martin Heidegger: Pathmarks. Cambridge: Cambridge University Press. Pears, D. F., and B. F. McGuinness, trans. 1961. Tractatus Logico-Philosophicus. London: Routledge and Kegan Paul. Priest, G. 1998. What’s So Bad about Contradictions? Journal of Philosophy 95: 410–426; reprinted as ch. 1 of G. Priest, J. C. Beall and B. Armour-Garb (eds.), The Law of Non-Contradiction: New Philosophical Essays. Oxford: Oxford University Press, 2004. Priest, G. 2002a. Beyond the Limits of Thought., 2nd ed. Oxford: Oxford University Press. Priest, G. 2002b. Paraconsistent Logic, vol. 6 of D. Gabbay and F. Guenthner, eds., Handbook of Philosophical Logic. 2nd ed. Dordrecht: Reidel, 287–393. Priest, G. 2008. Introduction to Non-Classical Logic: From If to Is. Cambridge: Cambridge University Press. Priest, G. 2014a. One. Oxford: Oxford University Press. Priest, G. 2014b. Sein Language. Monist 97: 430–442. Rojcewicz, R., and D. Vallega-Neu, trans. 2012. Martin Heidegger, Contributions to Philosophy (of the Event). Bloomington, IN: Indiana University Press. Stambaugh, J., trans. 1996. Being and Time. Albany, NY: State University of New York Press. Stambaugh, J., trans. 2002. Identity and Difference. Chicago, IL: University of Chicago Press. Yi, B. 2005. The Logic and Meaning of Plurals, Part I. Journal of Philosophical Logic 34: 459–506.

Part IV

Values, Personhood, and Agency

12 Relativism and Recognition Carol Rovane

INTRODUCTION I have a three-part agenda in this paper. All three parts are related, and their cumulative effect will, I hope, be to improve our philosophical understanding of what is most fundamentally, and interestingly, at issue in philosophical debates about relativism, and especially in debates about moral relativism. Before I present these three related pieces of the overall argument, I want to start by describing all three parts in somewhat greater detail than one would normally find in a mere introduction, so as to give some advance sense of the sort of improved understanding I hope to achieve. Relativism, Objectivity, and Universality. In the rhetoric of philosophers in the English-speaking world, the term “relativism” is often used as a term of opprobrium, something to be said of a philosophical position that falls short or afoul of a much lauded ideal of ‘objectivity’ that is supposed to follow upon a robust metaphysical realism. Richard Rorty responded to this rhetoric by urging that analytic philosophers should simply set aside the realist ideal of objectivity and embrace relativism. But of course his response did not unsettle the longstanding presumption that philosophical debates about relativism are centrally concerned with whether or not it is possible to retain that ideal.1 The presumption has fueled significant anxiety among analytic philosophers, who tend to worry that if relativism is true then there are no real normative constraints on belief—or, as the most well-worn, if imprecise and airy, cliché has it, anything goes. Although Rorty didn’t think such anxiety was warranted, he didn’t do very much to soothe it. He also didn’t do very much to clarify what relativism is exactly, except to emphasize that his embrace of it went hand in hand with a rejection of the realist ideal of objectivity. The same could equally be said of the other participants in the main twentieth-century debates about relativism within the analytic tradition—they too didn’t sufficiently clarify what relativism is. They did introduce a term through which they intended to capture a positive commitment of relativism over and above its presumed opposition to objectivity, namely, “alternativeness.” But the meaning of that term was left disturbingly vague. It shouldn’t

262  Carol Rovane be surprising, then, that when Donald Davidson set out to refute relativism, his first step was to offer a provisional definition of it—he defined what he called “alternative conceptual schemes” as languages that are true but not translatable.2 Since his aim was refutation, it also shouldn’t be surprising that he ended up concluding that there cannot be alternatives in the sense he defined. That was one main thrust of his argument against the very idea of a conceptual scheme. Rorty admired Davidson’s argument very much and pronounced it “the transcendental argument to end all transcendental arguments.”3 There is some irony there, because the argument not only left the relativist’s position insufficiently clarified, but also left an impression that there is no coherent way to formulate it. In a more recent literature, analytic philosophers have tried somewhat harder to formulate relativism by introducing the idea of a certain kind of irresoluble disagreement. While at first sight this may appear to be a departure from the older idea of alternativeness, closer examination reveals that the recent literature—almost unbeknownst to itself—is really a collective attempt to make sense of the older idea. As it happens, the attempt fails. Yet it is well worth thinking through why it fails, for doing so brings to light a significant point of convergence throughout the main analytic contributions to the topic of relativism, including the earlier ones from which Rorty’s own position descended as well as the more recent ones—a convergence that has not been sufficiently noticed due to the distracting and not entirely well placed concern about the realist ideal of objectivity. The convergence is on the matter of universality. All parties have agreed that relativists aim to challenge the idea that truth is universal—the idea that if something is true then it is true for everyone. My main thesis in the first section of the paper is that there is good reason to construe the debate about relativism directly and solely in terms of the question of whether truth is universal, while bracketing the issues of realism and objectivity. When we do so, the relativist’s position emerges as having a distinctive significance that is at once logical, metaphysical, and practical. To encapsulate the debate in two words—and I do not intend them metaphorically—relativists hold a “many-worlds” view, or Multimundialism, while their opponents hold a “one-world” view, or Unimundialism.4 The Analytic-Continental Divide. It will be useful to lay bare some ideological terms that have shaped many philosophers’ understanding of the significance of these issues. Rorty portrayed analytic philosophers as locked into a Cartesian model of philosophical reflection, which is an inside-out reflection that starts from direct knowledge of one’s self and one’s thoughts and then must confront skeptical problems about how we can know anything beyond them—skeptical problems that presuppose the ideal of objectivity that follows upon realism. Rorty portrayed continental philosophers, by contrast, as having thrown over the Cartesian problematic for a project that he wished to recommend to analytic philosophers, namely, politically engaged cultural criticism. He also portrayed them as appreciating that all of our concepts, including even our normative concepts, are historically situated. Since this point about the historical situatedness of our concepts is

Relativism and Recognition  263 what drove his own commitment to relativism, he naturally viewed continental philosophers as fellow travellers on the path to relativism. However, in taking this view he overlooked various tendencies of thought within the continental tradition that pointed away from relativism. These tendencies were not animated by a desire to retain the realist ideal of objectivity that Rorty thought analytic philosophers should give up, but rather by more directly moral and political concerns. For example, many continental philosophers agreed with Hegel and Marx that history will ultimately issue in a kind of moral and political progress that is not compatible with relativism; and Jürgen Habermas explicitly argued that public debate within democratic societies requires a framework of universal rationality that is likewise incompatible with relativism.5 One of my larger aims in this paper is to show that we can better understand the moral sources of resistance to relativism on both sides of Rorty’s analytic-continental divide if we formulate it as I propose to do, as a challenge to the universality of truth, while letting go of his presumption that relativists must also be aiming to challenge the realist ideal of objectivity. Another more specific aim of the paper, which will occupy me in Section 2, is to show that the most important considerations that speak in favor of moral relativism have been perfectly visible for a long while from the analytic side of the divide, most notably (though not exclusively) in the work of Bernard Williams.6 Williams, too, made the usual mistake of presuming that relativism is something we are left with if we give up the realist ideal of objectivity; and he thought we must give it up in the moral domain. Yet if we look closely at the terms in which he tried to clarify the moral relativist’s position—such as his idea of a “merely notional confrontation,” and his distinction between “thick” and “thin” moral concepts, and his claim that “convergence” means something different in morals than in science—then we shall see that his concerns about objectivity recede to the background and cede to other more directly moral concerns—concerns that turn on the question that I claim really divides moral relativists from their opponents, concerning whether moral truths are universal. I will close the section by identifying what I see as a moral division over moral relativism, which I doubt can be resolved by appealing to underlying metaphysical issues. When we construe relativism as Multimundialism, it follows from the moral relativist’s position that certain forms of engagement are foreclosed, because to encounter others whose moral beliefs are alternatives to one’s own is to encounter a normative boundary across which moral argument cannot reach. Unimundialist opponents of moral relativism condemn this Multimundial stance of non-engagement on the ground that it would lead to alienation from others and a loss of potential community with them, and in its stead they adopt a Unimundial stance, which recognizes a moral imperative to universal engagement with all others. But in their turn, Multimundialists see some serious moral mistakes in the Unimundialists’ determination to universal engagement—namely, the mistake supposing that the moral values by which we live have a normative force that reaches

264  Carol Rovane beyond the particular circumstances in which we are able to meaningfully live by them, and the correlative mistake of supposing that others cannot be right to live by their moral values if we are not willing to adopt them for ourselves. Moral Relativism and Recognition. In general, philosophers do not attach very much importance to the distinction between universality and objectivity, and in the context of moral philosophy the distinction itself nearly vanishes from view. For it is commonly assumed by moral philosophers that if we can show that moral truths are universal then we will thereby have shown them to be objective and, conversely, that if we can show that moral truths are objective then we will thereby have shown them to be universal. One influential method by which moral philosophers have tried to establish such a conflated objectivity-cum-universality is by invoking a conception of the person as a reflective rational being who is a being capable of conceiving itself as a member of a kind that includes all and only reflective rational beings. The conception seems to guarantee the possibility of a universal and mutual recognition among all persons of their shared status as persons. It is hard to see how this could fail to be a shared moral status; and insofar as that is so, the conception would appear to be in some tension with moral relativism. For this reason, when Axel Honneth proposed to make recognition the fundamental relation in morals and politics, he identified it as a serious challenge to reconcile his position with his commitment to relativism—a commitment that he, along with virtually every moral philosopher, assumed would simultaneously challenge both the objectivity and the universality of morality.7 In Section 3 I will explain that relativism actually presupposes this universal conception of the person, because relativism involves adopting a stance that only a person can adopt, and only towards another person. So moral relativism would not involve a failure of mutual recognition among persons; it would involve rather a positive recognition among persons that we do not all face the same moral problems, and as a result, we do not all require the same moral truths to live by. I will close Section 3 with a proposal about how we should proceed in the face of doubts about how to resolve the debate about moral relativism, once it is construed as a debate about whether to adopt the Multimundial or the Unimundial stance in the moral domain. We shall see that such doubts are natural and reasonable because, on the one hand, we are not in a position to settle this debate on purely metaphysical grounds, and, on the other, we can see that there are important moral considerations that speak for and against both sides. 1.  RELATIVISM, OBJECTIVITY, AND UNIVERSALITY Within the analytic tradition, there has been a presumption not only that relativism and realism are mutually opposed doctrines, but also that

Relativism and Recognition  265 arguments for relativism should proceed in a roughly neo-Kantian way. Kant argued that we must give up a certain ideal of objectivity that is associated with what he called “transcendental realism,” on which we would aspire to know things as they are in themselves, and that we must put in its stead a “transcendental idealist” metaphysical orientation on which we cannot know things as they are in themselves, but only as they can appear to us given our particular forms of sensibility and understanding. Insofar as what we then come to know is a world, it seems to follow that there can in principle be more than one world, indeed as many worlds as there are kinds of minds with different cognitive constitutions. All of the various projects and movements in early twentieth-century analytic philosophy that seemed to have relativist implications followed roughly the same Kantian pattern of argument, and the main differences among them concerned how they proposed to conceive the subjective conditions in which things can be known. Whether the context was cultural anthropology, logical positivism, philosophy of language, philosophy of science, or pragmatism, arguments for relativism tended to proceed by first casting aside a realist aspiration to know things as they are in themselves and then taking on something like Kant’s transcendental idealism—a vision of the world as mind-dependent— and then concluding that there can be as many worlds as there can be kinds of minds to know a world. The main candidates for subjective conditions on which worlds might thus depend included linguistic frameworks, theoretical paradigms, and social practices, all of which Davidson lumped together under the heading “conceptual schemes.” During that period, there was an oft-repeated knee-jerk reaction against relativism with three main components: The first was to insist upon the realist ideal of objectivity; the second was to protest that there is no coherent way to formulate the relativist’s position anyway, and the third was to back up that protest with the following series of queries, which I have often personally witnessed in philosophical discussion: What makes these different conceptual schemes competitors? Are they inconsistent? If so, isn’t that just a case of ordinary disagreement? If not, how can the schemes possibly exclude one another? I claimed in my introductory remarks that the relativists’ position was never given an adequate formulation in the early analytic debates. This was mainly due to the fact that the one idea that was most closely associated with the position—alternativeness—was never elucidated in a fully satisfactory way. Intuitively, alternatives were supposed to be truths that cannot be embraced together. And one thing that convinces me that the idea of alternativeness lay at the heart of the main twentieth-century debates about relativism within the analytic tradition is that the protest I just rehearsed was so often made. For the protest is really directed against the very idea of alternativeness in this intuitive sense. Let me now give that knee-jerk resistance some philosophical clarity, in the form of a Dilemma for Alternativeness. According to the dilemma, every pair of truth-value-bearers is either

266  Carol Rovane consistent or inconsistent; if they are consistent then they can both be true but they can also be embraced together; if they are inconsistent then they cannot be embraced together but they also cannot both be true; in neither case do we have what is intuitively required for alternativeness, which is that both are true and yet they cannot be embraced together. I also noted in my introductory remarks that the more recent analytic literature on relativism has focused on another idea that appears to be quite different from the idea of alternativeness, namely, the idea of a certain kind of disagreement that is alleged to be relativism-inducing because it is irresoluble.8 Minimally, we can say that disagreements are irresoluble when neither party can convince the other to her side. But the question arises, why can’t they convince one another? In ordinary cases the reason why is that the parties do not agree on background issues like standards of evidence, or some basic beliefs about the world, and as a result, the reasons that move each to adopt her own view fail to move the other. But in such ordinary cases, the parties would still agree that one of them must be in error—and so the irresolubility of their disagreement poses an epistemological difficulty concerning which, if either of them, is really right. If this is a kind of relativism, it is mere epistemic relativism. In contrast, when it is alleged in the recent literature that some disagreements are relativism-inducing, they are claimed to be irresoluble in a different and, I would say, metaphysical sense. The claim is that there may be domains in which anti-realism holds, and in these domains there is no more to the truth than what an ideally informed subject would be warranted in asserting, and, furthermore, there is no logical bar to there being two sets of standards for warranted assertibility such that a given claim might count as true according to one while its negation counts as true according to the other. On the assumption of anti-realism, if two parties were to embrace such contradictory claims, and thereby disagree with each other, there wouldn’t be any further facts in the light of which either of them could be mistaken—and it is for this reason that I say that their disagreement would be metaphysically irresoluble. One much discussed example of such a relativism-inducing disagreement is: I say, truly, that those snails are delicious while Crispin Wright says, equally truly, that those snails are not delicious.9 Analytic philosophers who propose to formulate relativism in terms of the idea of a metaphysically irresoluble disagreement do not present their proposal as an effort to make sense of the idea of alternativeness. All the same, their proposal does provide for a kind of alternativeness and, moreover, in a way that actually provides a direct response to the dilemma that seems to stand in the way of making sense of it. For these philosophers are claiming that some truths cannot be embraced together—and such truths would qualify as alternatives in the intuitive sense I have identified; and these philosophers are claiming that these truths cannot be embraced together for the specific reason that they are contradictory—and this is a flat rejection of the dilemma’s second horn, which insists that inconsistent claims cannot

Relativism and Recognition  267 both be true. However, we should not attempt to portray alternativeness along these lines, because doing so would saddle the relativist with having to allow that there can be violations of what is arguably the most fundamental law of logic, the law of non-contradiction. If this is the best we can do in our efforts to formulate relativism in a coherent way, then I think we should side with the dilemma and conclude that it cannot be done. One main goal of the recent analytic literature has been to show that this can be done, by portraying the truths at issue in metaphysically irresoluble disagreements as relative to different standards of warranted assertibility. This formal move promises two advantages: it removes the appearance of any outright violation of the law of non-contradiction, and it affords an intuitive sense of why metaphysically irresoluble disagreements are situations of relativism—it is because these situations now emerge as ones in which truth is relative. Going back to the example above, the proposal is: if I say snails are delicious, while Wright says they are not delicious, the reason we can both be right is that our claims are true relative to different standards for warranted assertions about deliciousness. But note that there really is no need to portray me and Wright as affirming and denying the same proposition. We can just as easily suppose that he and I don’t really mean the same thing by the term “delicious”—that what I am claiming is that snails are delicious-according-to-Rovane’s-gustatorystandards, while what he is claiming that they are not delicious-accordingto-Wright’s-gustatory-standards. The reason why this latter interpretation tends not to be favored is that it fails to preserve an appearance of disagreement between me and Wright. But we would do well to bear in mind that the usual normative point of registering a disagreement is missing in this sort of case anyway. The usual normative point of portraying two parties as being in disagreement to draw attention to the fact that there is something to be resolved between them, namely, which (if either) of them is right. But if the parties are both right, then there is no longer anything to be resolved, and so it is highly misleading to describe their situation as an irresoluble disagreement or, for that matter, as a disagreement at all. It would be far less misleading to interpret the parties’ respective claims so that there is no logical obstacle to their both being right, which we can always do by disambiguating terms in the way I have just done, to clarify that they are not making contradictory claims. I am not suggesting that the sort of situation that contemporary relativists have in mind should not be thought of as relativism-inducing. I am suggesting that we should try to understand what is going on in them differently. We should not posit a disagreement when the normative point of positing one is missing, but instead, we should explore how and why it might be the case that the claims made in such situations cannot be embraced together even though they are not in strict contradiction. If we can work this out, then we shall have worked out the very notion of alternativeness that was introduced in the main analytic debates about relativism in the

268  Carol Rovane twentieth century. It was never claimed then—or at any rate, it was rarely claimed, and never coherently claimed—that alternative conceptual schemes are in conflict. What was claimed was that they do not share meanings at all—and obviously, if this is true then it is impossible for parties who hold different schemes to affirm and deny the same thing. My suggestion still leaves us with the task of finding a way out of the original Dilemma for Alternativeness. The burden of my discussion so far is that we cannot hope to do this by denying either horn: consistent claims can always be embraced together when they are both true, and inconsistent claims cannot both be true. Clearly, the only way out is to find a third possibility that the dilemma overlooks, and there is exactly one possibility to consider, namely, that some claims are neither consistent nor inconsistent. I cannot reproduce here the many detailed arguments that I have given elsewhere for construing the idea of alternativeness along these lines.10 All I  can do is explain very briefly what alternativeness amounts to on this construal. To say that claims are neither consistent nor inconsistent is to say that they do not stand in any logical relations—they are, as I shall put it, normatively insulated from one another. Thus, to oppose relativism is to say that there are no alternatives in this logical sense; equivalently, it is to say that logical relations run everywhere among all truth-value-bearers, and that all of the true ones can therefore be embraced together (at least in principle). Metaphysically speaking, it is to say, with the early Wittgenstein, that the world is all that is the case; equivalently, it is to say that there is just one world, i.e., Unimundialism. Practically speaking, it is to adopt a stance of epistemic engagement towards all others. Whenever I encounter someone who believes something I do not believe, as a Unimundialist I must accept that their belief is either consistent or inconsistent with my beliefs, and that if it is consistent then it is a candidate for belief by me to be embraced together with the rest of what I believe, and that if it is inconsistent then either I must reject it as false, or I must give up my own beliefs that are inconsistent with it (this may involve actually rejecting certain of my own beliefs as false in order to consistently take on the other party’s belief, or it may involve going into a state of doubt concerning which of us is really right). To affirm relativism is to say that there are alternatives in this logical sense, which means that logical relations do not run everywhere among all truth-value-bearers but only among some of them (those that figure within a single conceptual scheme). Metaphysically speaking, it is to say that there are many incomplete bodies of truths that cannot be embraced together, and that they constitute separate worlds, i.e., Multimundialism. Practically speaking, it is to adopt a stance of epistemic indifference towards others who do not occupy the same world. This may seem like a very unappealing stance to take. But it should seem less unappealing once we see that it would be the only appropriate way to respond to normative insularity. If I were to encounter someone whose beliefs are normatively insulated from mine, her beliefs would not be candidates for belief by me, and yet it would not be

Relativism and Recognition  269 because her beliefs were inconsistent with mine and hence ruled out as false, but rather because her beliefs were not consistent with mine and so not conjoinable with them. Thus to encounter normative insularity would be to encounter an epistemic barrier—the truths on the other side of it would not be truths-for-me even though they would be truths-for-others. I need not be indifferent in every way towards those who live on the other side of that barrier—for example, I can want to learn about their beliefs. But in learning about them I would not be embracing them myself—that, at any rate, is what it means to suppose that they are alternatives in the sense that was at issue in the main twentieth-century debates about relativism. So the epistemic indifference amounts to a recognition that if they are truths, they are not truths-for-me. This account of how we may formulate relativism, as Multimundialism, has been relentlessly abstract, and it may not yet be clear how it would apply to the one example I have introduced so far. Before clarifying that, let me issue a caveat about the use of illustrative examples in connection with relativism: it is impossible to establish any interesting claim about relativism just by appealing to an example. Even if it is possible to interpret what is going on in a given example in such a way as to accord with a given formulation of relativism, it is also always possible to interpret it in such a way as to accord with a different formulation as well, and for that matter, in such a way that it does not illustrate any recognizably relativist position. Thus we might say in connection with the example about gustatory attitudes towards snails, that Wright and I are just having an ordinary disagreement about whether snails are delicious in which we are making contradictory claims that cannot both be right. We might also say, as Wright himself suggests, that there is no more to the truth about whether snails are delicious than what a subject would be warranted in asserting on the basis of having tasted them, and so, if he and I, both having tasted them, go on to make contradictory claims, then we are both right. I have recommended against this interpretation on the ground that it asks us to accept that there can be violations of the law of non-contradiction. I have also argued that if contemporary relativists try to avoid outright contradiction by portraying the truth of Wright’s and my claims as relative to different contexts, in which different standards of warranted assertibility apply, they will remove the appearance of disagreement as well as contradiction. After all, it would allow Wright and me to agree on two related points: first, we can both be right, and, so, second, there is nothing to be resolved between us; and once we get this far in our understanding of our situation, we may as well remove the appearance of disagreement altogether by reinterpreting our claims in such a way that we are not actually affirming and denying the same proposition, in the way I already suggested above—so that what I am affirming is that snails are delicious-accordingto-Rovane’s-gustatory-standards, while what Wright is denying is that snails delicious-according-to-Wright’s-gustatory-standards. The question then is, how can this situation be of the sort that the relativist intuitively has in

270  Carol Rovane mind, once the appearance of disagreement has been entirely removed? My answer is that we must look for a reason why Wright’s and my claims cannot be embraced together, even though they are not inconsistent. The reason I have offered is that they are not consistent either, but are normatively insulated from one another—that is, they are alternatives. Logically speaking, this means that, while Wright’s and my claims do not logically rule each other out in the way that they would if they were inconsistent, they also cannot be logically conjoined in the way they could if they were consistent. Metaphysically speaking, it means that he and I inhabit separate gustatory worlds. Practically speaking, it means that he and I should adopt an attitude of epistemic indifference toward one another, in the following sense: neither of us can regard the other as possessing a truth about the gustatory properties of snails that we can take on together with the rest of our beliefs, and yet neither of us has any normative basis on which to correct the other’s belief either—in short, we have nothing to learn from each another concerning the gustatory properties of snails. It may well seem to readers who have anti-relativist instincts that it is of course possible for Wright and me to embrace both of our beliefs together—each of us can perfectly well claim both that snails are delicious-according-to-Rovane’s-gustatory-standards and that they are not delicious-according-to-Wright’s-gustatory-standards. But all the same, if we want to try to understand the relativist’s position as sympathetically as we can, at least for the sake of formulating the issue that seems to be in dispute between relativists and their opponents, then this is how we should try to envisage the situation: although Wright and I can learn about one another’s gustatory attitudes—that is, he and I can come to understand one another’s gustatory attitudes and appreciate the sense in which we are both right—there is nevertheless a sense in which neither of us can adopt the other’s attitudes together with our own, which has to do with the fact that neither of us can adopt the other’s attitudes as a basis for our own gustatory judgments and responses.11 In the next sections, I will turn to the more serious and important case of morality, and I will try to say more about what is involved in regarding another’s moral attitudes in the way that would embody the Multimundialist’s stance of epistemic indifference—and why it is not necessarily the case that the Unimundialist’s stance of universal epistemic engagement is always to be preferred. But I want to conclude this section with some initial remarks concerning the distinction between universality and objectivity, and why it makes sense to see the relativist as directly challenging the former rather than the latter. The universality of truth is the social meaning of the Unimundialist’s commitment to the oneness of the world, on which there is just one comprehensive body of truths. In a way, I have already explained this social meaning of Unimundialism under the heading of its practical meaning. I noted that Unimundialism brings in train a stance of universal epistemic engagement. As a Unimundialist, I must suppose that the beliefs of others are

Relativism and Recognition  271 always either consistent or inconsistent with mine and that, if their beliefs are consistent with mine, they can be embraced together with mine, and if they are inconsistent with mine, then either I must reject them as false or I must revise my beliefs in order to accommodate their truth (or perhaps suspend belief on the matters at issue). This means that we can always learn from one another—we can always acquire the truths that others possess, and vice versa. To suppose that the same truths are thus available for us all is the same thing as to suppose that they are universal—that if they are true for anyone then they are true for everyone. It does not follow that we must all be interested in the same truths; what follows is that all of the truths are candidates for belief by all of us, should we become interested in them. This is precisely what the Multimundialist denies when she denies that logical relations run everywhere among all truth-value-bearers. The barriers of normative insularity that arise when logical relations are missing effectively separate inquirers into separate worlds; those who inhabit different worlds are not in a position to inquire into the same truths, but only into the truths that hold within their own worlds; the truths they learn are not true for everyone but only for those who inhabit the same world; thus, according to the Multimundialist, truth is not universal but local. Now that I  have clarified why the logical issue that divides Unimundialists and Multimundialists—concerning whether there is one world or many—has to do with the universality of truth, let us consider how and why it is different from the issue that divides realists from their opponents, concerning whether reality is mind-independent. Realism does seem to provide for a sense in which reality is one, for the realist may say that there is a single totality of all that there is. This single totality might be conceived as something like a giant set that includes every single mind-independent thing that there is within it. But note that there are no normative or logical constraints on the relations that can hold among the members of this all-inclusive set (except, of course, that it cannot contain itself). In other words, the members of a set can be consistent, they can be inconsistent, they can be normatively insulated from one another, and they can be without logical significance at all (if it is possible that anything can be without logical significance). Thus it seems to me that Kant was right: when we define things as radically mind-independent, we thereby deprive ourselves of the right to infer anything about their nature from the conditions of our own subjectivity—that is, we cannot suppose that they must have the sort of logical significance that they would have to have in order to be intelligible to us. This shows that the oneness of reality that follows upon realism is not the same thing at all as the oneness that the Unimundialist posits, which is an expressly logical conception of oneness. Another understanding of the difference between these two conceptions of oneness is afforded by Thomas Nagel’s characterization of the ideal of objectivity that goes together with realism. This ideal says that we should aspire to know things as they really are—as Kant would say in themselves; and Nagel pointed out that this aspiration drives us to try

272  Carol Rovane to transcend the limits of our own subjective viewpoints, so as to achieve a View from Nowhere.12 In contrast, the Unimundialist holds that all of the truths can in principle be embraced together; and what this vision of truth requires is not a View from Nowhere but a View from Everywhere—that is, a point of view from which every truth that can be apprehended from every other point of view can be embraced together. I hope it is now clear how and why there is at least a conceptual distinction between the sort of objectivity to which realists characteristically aspire and the universality of truth that follows upon Unimundialism. The history of debate about relativism might seem to indicate that this conceptual distinction holds little or no significance. For even if I am right that analytic arguments for relativism in the twentieth century generally did deliver a Multimundial conclusion—that people who employ alternative conceptual schemes inhabit different worlds—I have also made clear that they generally employed a neo-Kantian strategy whose first step was to deny the mind-independence of reality, and so whatever challenge they posed for the universality of truth went together with a challenge for objectivity too. Indeed, it may seem hard to see how there could be any other way to challenge the universality of truth than by challenging its objectivity; and if there is no other way, then perhaps moral philosophers are not really mistaken after all when they assume that universality and objectivity must go hand in hand. Nevertheless, I do want to insist that the conceptual distinction between universality and objectivity is important, and, furthermore, it is not just an abstract point that deserves registering for the sake of conceptual clarity. It is a point that can and should affect the task of assessing what reasons there may be to affirm Multimundialism. On the one hand, it reminds us that not every challenge to the realist ideal of objectivity will carry Multimundial implications—for such a challenge might lead to a variety of idealism or skepticism that does not call into question whether logical relations run everywhere among all truth-value-bearers. On the other hand, it should also make us wonder why every argument for Multimundialism should have to start by challenging the realist ideal of objectivity—especially since, as I just explained above, the realist conception of mind-independence does not so much as speak to the logical issue that divides Multimundialists from Unimundialists. 2.  THE ANALYTIC-CONTINENTAL DIVIDE Rorty gave us a picture of an analytic-continental divide in which he and continental philosophers stand united against analytic philosophy and in favor of relativism. He also gave us a picture of relativism that in a way lines up with the main twentieth-century analytic debates about it, insofar as he shared the presumption that I have just called into question but which

Relativism and Recognition  273 was never questioned in those debates, that relativism and realism are mutually opposed doctrines—from which it would appear to follow that any argument for relativism would have to employ a neo-Kantian strategy that calls the mind-independence of reality into question. There was, however, a real difference between Rorty and the other participants in those debates, which he regarded as all-important. When he rejected the realist ideal of objectivity, he did not did not take up a position within metaphysics that stands opposed to realism—such as Berkeley’s idealism, Kant’s transcendental idealism, the logical positivists’ radical empiricism, or contemporary anti-realism. Rather, he proposed to set aside the issue of realism. He was perfectly right in thinking that his own position came much closer to that of the American pragmatists, who tended to agree with him that there is no philosophical problem about the mind-world relation that needs to be solved either by vindicating the realist ideal of objectivity, or by retreating to some variety of idealism. Yet all the same, it is still fair to say that his position was a direct response to an alleged problem about objectivity that had continued to be central for analytic philosophers, and, moreover, he did a great deal to help entrench their presumption that giving up on the realist ideal of objectivity—for whatever reason—is a first step on the path to relativism. In contrast, the continental turn away from the ‘problem’ of objectivity was not a turn towards relativism, but, on the contrary, was accompanied by significant resistance to it, especially in the domains of morals and politics. Rorty’s rejection of the realist ideal of objectivity was part of a wider polemic against the Cartesian model of philosophical reflection, which is an inside-out reflection that takes for granted our knowledge of our own states of consciousness and then finds difficulties standing in the way of knowing anything beyond them. Rorty responded to the Cartesian skeptical problematic by invoking a social account of normativity that is very close to the account that Kripke arrived at through his engagement with Wittgenstein on rule following. Here are the bones of that account: we cannot think without concepts; we cannot apply concepts without bearing in mind that there is a normative distinction between the correct vs. incorrect application of a concept; the only way to ground this distinction is by equating the correct application of a concept with what a community agrees is the correct application. Obviously, if this social account of normativity is correct, then the Cartesian skeptical problematic simply cannot get going—a reflective subject can never be locked inside her own consciousness, because it is a condition of her very ability to think at all that she be in epistemic touch with a whole social world. The sort of relativism that Rorty took to follow is an across-the-board relativism that would hold in any and all domains—for it was his view that, in general, all there is to truth is what is warranted by the lights of a given community agreement. Predictably, many analytic philosophers responded to his arguments with the clichéd knee-jerk response that I  identified at the outset—they saw in his brand of across-the-board

274  Carol Rovane relativism a threat that anything goes. Rorty’s counter-response was simply to enjoy their discomfort. My own complaint is somewhat different—not that his relativism poses a threat to objectivity, but that it is not clear what he took relativism to be. If this seems an unfair complaint, just consider what we learned in the last section about how much more precise it is possible to be in our formulations of it. Continental philosophers had found a route out of the Cartesian problematic well before Rorty’s post-Wittgensteinian social account of normativity came on to the philosophical scene, through a line of argument that was pioneered by Rousseau and later developed by Hegel (and by George Herbert Mead in North America). According to this line of argument, first person reflection on my own thoughts cannot proceed in the inside-out way that Descartes supposed, but must rather proceed from the outside-in. The reason why is that my attention is always naturally directed outwards on the objects of my consciousness, and the only occasions when my attention is directed inwards on myself and my own subjectivity are occasions when that is required in order to apprehend and cope with other objects of my consciousness. These would be social occasions in which I encounter another subject for whom I am an object, where I cannot know this object-who-is-asocial-subject without coming to know what it knows about me. It should be clear that this account of self-consciousness as emerging through social interaction rules out the Cartesian skeptical problematic just as effectively as Rorty’s social account of normativity. Both accounts guarantee that a self-conscious subject will not be locked inside her own consciousness, cut off from epistemic relations to things outside it. Yet the social interaction account of self-consciousness does not speak directly to the issue of relativism, and this is something that it is easy to miss  in the wake of Davidson’s influential discussion of relativism. At the end of his argument against it in “On the Very Idea of a Conceptual Scheme,” he declared that the so-called scheme-content distinction is a third dogma of empiricism that we should give up along with the other two dogmas that Quine had identified, and once we give it up we can “reestablish unmediated epistemic touch with the familiar objects whose antics make our sentences and opinions true and false.”13 Davidson didn’t make fully clear what he meant by those last words about “unmediated epistemic touch” until much later, when he offered his own version of the social interaction account of self-consciousness. This was his “triangulation argument,” in which he aimed to show that thought itself is a social achievement.14 Since he took thought to be essentially self-conscious, the triangulation argument was about both thought and self-consciousness at once. For the purposes of the argument, he construed thought as involving an epistemic relation to objects, and his main premise was that we cannot be conscious of our thoughts as thoughts without seeing ourselves as standing in epistemic relations to objects. He then reasoned that the only way in which we can come to see ourselves as standing in epistemic relations to objects is by apprehending

Relativism and Recognition  275 another subject’s point of view on those same objects, where that point of view is also a point of view from which our own epistemic relations to those objects are visible. So when that other point of view comes into view for us, we finally gain through it a vantage point from which to see our own epistemic relations to objects—where, to see ourselves from their vantage point is precisely to see ourselves as enjoying unmediated touch with objects. Davidson was much more explicit about the anti-Cartesian thrust of his argument than was generally so in the parallel continental discussions of the social nature of self-consciousness, for he underscored that the philosophical mistake that Descartes and others have made when they took skepticism seriously was precisely to see our epistemic relations to objects as mediated by something subjective. In Descartes’ case, that something was consciousness, and Davidson claimed to see a similar mistake at work in the relativist’s position, insofar as the relativist sees our epistemic relations to things as mediated by some sort of conceptual scheme. If Davidson himself connected his two arguments so closely—for the social character of thought and against the scheme-content distinction—why do I insist that the former argument does not really speak to the issue of relativism? The reason is similar to the one I offered earlier when I explained that the realist’s conception of reality as mind-independent does not really speak to the issue of relativism—at least not when we formulate relativism as Multimundialism and frame the issue that divides relativists from their opponents in terms of whether logical relations run everywhere, among all truth-valuebearers. The social interaction account of self-consciousness does not really speak to that logical issue either, because the claim that self-consciousness is achieved through social interaction with others does not settle whether all social subjects must inhabit a single world. I think even Davidson himself could have been brought to see this point had it been put to him, and I say this in spite of the fact that he clearly did believe that skepticism and relativism rest on a common mistake, which is to see our knowledge of objects as mediated by something subjective. For in his anti-conceptual schemes argument he offered quite specific considerations against relativism that did not enter into his triangulation argument at all—considerations that had to do with how, in his view, all languages are inter-translatable and therefore have the same expressive power.15 We should not expect that the specific line of resistance to relativism that Davidson offered in his anti-conceptual schemes argument would be widely shared among twentieth-century continental philosophers. On the whole, they were much more historically minded, and therefore much more willing to concede the starting point that set Rorty on his path to relativism, which is the claim that concepts are always historically situated, and so not all of them are available to all people in all times and places—and it is that very starting point that Davidson aimed to cut off with his argument against the very idea of a conceptual scheme. So the question arises, what form did resistance to relativism take among the more historically oriented

276  Carol Rovane philosophers of the continent? I claimed in the introduction that their resistance was largely moral. Now, this is hardly the place, and I am hardly the person, to attempt a serious historical defense of this general claim. But I do want to make a few observations about why Hegel and Marx should be counted as Unimundialists, which I think should apply virtually anywhere in the continental tradition where their influence has reached. Then I’ll briefly note the reasons why it is obvious that Habermas should also be counted as a Unimundialist. Although Hegel and Marx both held that there are truths concerning what is the best form of life for human beings—a life in the right kind of community with others—it is not immediately obvious that this suffices to make them Unimundialists, given that they also held that it will not be possible for human beings to live by those truths until the right historical conditions obtain and that throughout the rest of human history, before sufficient progress has occurred, it has not been possible for human beings to live by those truths. For it seems to follow from this historical point that it would not be wrong for human beings who live in different times and places to live in different ways, and it is tempting to infer a Multimundial conclusion—that when they live in these different ways they are living by different truths, ones that hold in their different historical conditions. But this Multimundial conclusion would follow only if there was a complete absence of logical relations with respect to moral matters across different historical contexts, and such an absence would rule out a kind of comparative evaluation to which Hegel and Marx were both clearly committed, according to which the best form of life that will ultimately be available to human beings will also be better than the forms of life that had gone before. To be committed to making such comparative judgments is to be committed to a kind of universal engagement with all others concerning moral matters. There are two further signs that progressivist interpretations of history conceive moral truths as universal. First, if history is a moral progress, then whatever human beings have taken to be true in their actual historical conditions will either survive the process of history or not—that is, it will either emerge at the end of history as part of the truth about what the best form of life is or it won’t—and if it does, then it will emerge as universally true, and if it does not, then it will emerge as having in some real sense been false, even though it was not optional for human beings to avoid taking it to be true. Second, when we arrive at the time when it is possible to live the best form of life and look back, we must view the human beings who lived in times when that life was not available as lacking, both in moral development and in moral knowledge—and the very idea of such moral deprivation incorporates the sense in which the moral truths that will be known at the end of history really are universal, for they provide the critical standard in the light of which such moral deprivation can be conceived. This leaves us only to consider how human beings can or should engage with one another in the midst of history while is still unfolding. The Hegelian view definitely

Relativism and Recognition  277 requires the Unimundial stance—for on the Hegelian view, it is precisely by engaging with the ideas of others that the mistakes of history will get identified and corrected and left behind. The case of Marx is somewhat less clear, because he took history to be a material process rather than a development of Spirit. Yet the Unimundial stance is appropriate on the material interpretation so long as the following holds: if, at times of historical change when human beings find themselves in new historical conditions, it is appropriate for them to regard their new conditions as an improvement over the old—if, in other words, it is appropriate for them to adopt a critical attitude towards earlier times in history. I simply do not see how this could fail to be appropriate if history is truly a progress. Habermas’s view is, as I’ve said, explicitly committed to Unimundialism. He recommended that in the public sphere of democratic debate, and in communication more broadly, we should be maximally inclusive in our conception of who our audience is—our aim should be to make claims that could be accepted by a universal audience, provided of course that they are reasonable and rational. This recommendation would be utterly incoherent if he didn’t conceive truths as truths for everyone. It would be equally incoherent to embrace that conception while denying the Unimundial corollary that there is a single, comprehensive body of truths that could in principle be embraced together—for if there were any truths that could not be embraced together from the same point of view, they would be truths for some and not others, which is to say, they would not be universal after all. I want to reiterate that the resistance to relativism that comes with these three views—the Hegelian, the Marxian, and the Habermasian—is a moral resistance, by which I mean, it is not driven by purely metaphysical considerations. I do see, of course, that there are such things as Hegelian, Marxian, and Habermasian metaphysics, and that these metaphysics have been offered in support of correlative moral points of view—or at least as harmonizing accompaniments. But I also see in these views a direct moral imperative to moral engagement, the full force of which will be easier to appreciate by contrast with the avowedly relativist attitude of Bernard Williams. Williams did share the analytic philosopher’s characteristic preoccupation with the ‘problem’ of objectivity, and, like many, he regarded scientific realism as an adequate response to that ‘problem’ in the domain of natural facts that are the appropriate targets of scientific investigation. So when he suggested that we should take moral relativism seriously, his official reason was that there is a ‘problem’ about moral objectivity that cannot be solved on the model of scientific realism. At the same time, he did much more than other analytic philosophers to try to formulate relativism in such a way as to bring out its practical and normative and ultimately moral significance. I won’t be able to give a complete account of the many strands that made up his overall view, but only a quick summary of the main strands that matter for the contrast I want to draw here between his sympathy for moral relativism and continental resistance to it.

278  Carol Rovane Williams proposed to formulate relativism in terms of the idea of a merely notional confrontation between two systems of belief, which he distinguished from a real confrontation. The latter is really an ordinary disagreement in the sense I described in the last section: the holders of one system find a logical conflict between their system of beliefs and another system held by others, and all parties agree that they cannot all be right; so it is in principle possible for the holders of the first system to go over to the second, if they come to believe that they are the mistaken parties and that they can correct their mistake by adopting the second system in place of their own. In contrast, in a merely notional confrontation, the holders of one system find that they cannot go over to the other. Williams didn’t say nearly enough about why they cannot go over, and the main reason he gave sounds singularly unphilosophical: he said that they would lose their grip on reality. One slightly more philosophical sounding thing he said was that any effort to rationally appraise the other system would be pointless—but again, he didn’t say enough about why such appraisal would be pointless. I have argued at length elsewhere that, with his notion of a merely notional confrontation, Williams was struggling to make sense of the notion of alternativeness that had figured centrally in all of the main twentieth century-debates about relativism, and that the shortcomings of his account of that notion can easily be rectified by bringing in the idea of normative insularity. If we construe a notional confrontation as an counter with normative insularity, then it makes literal sense to say, as Williams did, that one could not go over to the other side of it without losing one’s grip on reality, for one would literally be leaving one world for another; and we can also see why it would be pointless to try to rationally appraise the other system of beliefs—it is because those beliefs would not stand in any logical relations to one’s own beliefs. I turn next to Williams’s reasons for taking relativism—now construed as Multimundialism—seriously in the domain of morals. Because he was much better informed than the average analytic philosopher about the variety of historical circumstances that human beings have faced, he was highly critical of the universalist tendency in Western moral theorizing, which has led philosophers to focus on what he called thin moral values like ‘right’ and ‘wrong,’ as opposed to thick moral values like ‘honor’ and ‘modesty.’ The former are too underdescribed to give real guidance to us in any real choice situation, and it is only in conjunction with the latter that their action-guiding content can be determined. Take, for example, the value we all attach to well being, in the light of which it seems we can all agree that it is wrong to deliberately harm someone. What we have here is a thin moral value, which seems to support a universal moral principle. Yet we cannot possibly take this allegedly universal moral principle as a guide in our moral deliberations and actions unless we know how, concretely, our actions may harm others, and this is something we cannot generally know without bringing in thick moral values that speak to the specific ways in which life around us is organized—the specific relationships and roles and institutions

Relativism and Recognition  279 we find ourselves in, where all of these are precisely not a universal condition. In historical conditions in which honor and modesty count for a great deal, impugning them will count as a great harm, even though similar actions would not be harmful everywhere; likewise, in historical conditions in which individual self-determination and privacy matter a great deal, intruding in the lives of others in certain ways may count as great harms, even though they wouldn’t qualify as harms elsewhere. To portray us all as being bound by a single and universal principle “To harm is wrong” would therefore be misleading—the portrayal would fail to register the many more specific prohibitions by which people actually live in their different historical conditions—prohibitions against insulting honor, offending modesty, engaging in paternalistic interference, violating someone’s right to privacy, etc. But if the thick moral values by which we actually live have only local application, then we ought to adopt the Multimundial stance towards others who live in different contexts from our own and allow that we are separated by barriers of normative insularity. Having recognized those barriers, we may hold our own moral beliefs to be true without having to infer that others who do not embrace them must be either mistaken or ignorant; and we may find ourselves unable to adopt others’ moral beliefs together with our own without having to infer that those others must be mistaken to hold them. In short, we may find it appropriate to retreat from epistemic engagement about putatively common moral matters to a stance of epistemic indifference about moral matters that arise in circumstances not our own. This doesn’t mean that we can’t be curious about others’ moral outlooks, or the circumstances in which they arise. It means that all we can expect to gain from our curiosity is sociological insight, as opposed to moral insight that we might carry into our own lives—so the epistemic indifference reflects an apprehension that we can neither give nor receive specifically moral instruction across a barrier of normative insularity in the moral domain. Having argued that we do not and cannot all live by the same thick moral values, Williams went on to consider: What should we make of the fact, if it emerges to be a fact, that our varied moral outlooks end up converging over the course of history? He did not take this question lightly, because he thought that there has been a similar convergence in the history of science, which invites a scientific realist explanation. According to this explanation, scientific methods are designed in such a way as to let the mind-independent facts tell us when our theories are failing to get them right, and so when our theories converge, this can be explained by the hypothesis that we are succeeding in getting them right. But Williams did not see scope for a parallel realist explanation of convergence in morals. He thought that if we find that moral beliefs that incorporate certain thick moral values tend to disappear over the course of history while others remain, and, moreover, if we find that those that remain are ever more widely shared, this will not signify that we have finally discovered certain moral truths that had somehow been there all along, like natural facts out there waiting to be discovered; it would signify rather that history had produced a homogeneous culture in which

280  Carol Rovane different people no longer face substantially different moral problems, but rather similar ones—ones that can appropriately be navigated by employing the very same thick moral values. When we see the significance of moral convergence in this way, we must turn our backs on the progressivist interpretations of history given by Hegel and Marx. We will not view the form of life that ultimately emerges from history as the best form of life; nor will we view earlier forms of life as somehow less good. Rather than engaging in such comparative evaluation, we will retreat to the disengaged stance of the Multimundialist, who is prepared to countenance many distinct moral worlds in which human beings have lived by moral truths that hold only there, where they live. We can accept Williams’s non-progressivist vision of history without going so far as to agree with him that the moral relativist must relinquish the realist ideal of objectivity in the moral domain. Let us suppose that it is not possible for human beings to navigate the actual options they face except by recourse to thick moral values that actually speak to those options. The question then arises, is there any sense in which any of the courses of action that are open to those human beings can actually be correct? I think the answer Williams should give is yes. What would be correct is for them to live in the best way that is actually feasible, given the actual options they face, and given the best guidance they may draw from the thick moral values that speak specifically to those options. Insofar as this is the correct way for them to act, then they will be living by moral beliefs that are actually true. And it does not matter that such moral truths are not universal—all that matters is that such truths are a correct basis for living a life where they apply. Thus, there is no loss of, or threat to, objectivity on this relativist view, but only a loss of universality. Williams made a regrettable move when he reasoned that the relativist must give up on moral objectivity. The move led him to suppose that moral relativism lands us with a new kind of uncomfortable and distinctly modern moral challenge, which concerns how to maintain confidence in our own moral outlook once we have given up on its objectivity. The reason why Williams saw it as a new kind of challenge was that he thought the rise of science has placed a new kind of pressure on moral philosophers, to try to make sense of moral objectivity on the model of scientific objectivity. If he was right, then the only way in which we could retain confidence in our moral outlook is by scientizing it. He of course would never have recommended that we do that, since he understood very well that the enterprise of morality is wholly different from the enterprise of science. In this section, I have been recommending that we can and should view morality as capable of a kind of objectivity that is completely unlike the objectivity of science. If I am right about that, then we shouldn’t have to suffer any crisis of confidence in our own moral outlooks, even if we accept a historically based argument for moral relativism. All that such an argument would show is that the normative force of moral values cannot be universal, given that they must speak to the options that human beings actually face, and given that

Relativism and Recognition  281 these options are not the same in all times and places. When the normative force of moral values is thus local rather than universal, they hold only in some moral worlds and not others. But this does not mean that there are no objective truths concerning how it is best for human beings to live; it only means that these truths are not timeless and universal—unlike in the case of science. It is important not to confuse the Multimundial view of moral relativism with the “When in Rome” view of morality. The Multimundialist does not think we are free to move from world to world and take on whatever moral outlook prevails there. We can, however, encounter worlds not our own, and really experience them in that way—as not our own. (This is why Williams’s phrase “notional confrontation” is apt.) Such an experience is bound to be somewhat alienating, and this brings me, finally, to why I  see the main resistance to moral relativism in the continental tradition as a moral resistance. Hegel and Marx did view certain social conditions as unavoidably alienating. But they thought the right response to alienation is always to strive to overcome it, through the right kind of community with others. And of course, when we explicitly embrace an ideal of universal community, as Habermas did, we are recognizing a normative demand to overcome alienation from others wherever and whenever it arises. So a common view emerges from these parts of the continental tradition that we should never permit ourselves to respond to others in the disengaged manner of the Multimundialist, because doing so would close off a possibility of community, which would be an essentially alienating thing to do and, as such, it would work against human flourishing.16 This is a perfectly understandable moral concern about what it would mean to live moral relativism. But there is a perfectly reasonable response on behalf of moral relativism, which is equally moral in its orientation. While the Unimundialist may put up moral resistance against adopting the Multimundial stance on the ground that it accepts the possibility of a kind of alienation from others that would be insuperable, the Multimundialist may in turn put up moral resistance against adopting the Unimundial stance on the ground that it refuses to acknowledge the genuine differences that there are in human situations and the implication that one’s own moral values may have only local normative force—the moral concern here is that refusing to recognize this can be a moral mistake, a mistake that has arguably been repeated throughout the long human history of conquest and colonization, in which conquerors and colonizers wrongly assumed that any moral differences they encountered must be instances of ignorance and error. Many philosophers would automatically suppose that whatever moral divisions there may be over moral relativism, we should still seek to find a purely metaphysical basis on which to resolve them. If we are persuaded by Williams that people require thick moral values by which to live, then we will not find this metaphysical basis in the standard moral theories that employ only thin moral concepts. And if we are persuaded by arguments that I have given in this section, then we will not find this metaphysical basis

282  Carol Rovane in arguments for and against moral objectivity—for it is the universality of moral truths that is at issue in debates about moral relativism, which is a quite different thing. As I see it, then, the only remaining metaphysical issue that could possibly be relevant is whether history is indeed a progress in the way that Hegel, Marx, and some others have claimed—for that would settle the issue about moral relativism in favor of Unimundialism. I am frankly skeptical that we can settle the question of history in a value-neutral way. So I do not find myself in step with philosophers who expect to resolve the issue of moral relativism on purely metaphysical grounds. I think moral grounds may well have to be brought in. But note that, in any case, the moral considerations I have raised here, both for and against moral relativism, would still remain to be addressed even if we were to eschew metaphysics altogether, as Rorty recommended. He may have thought that we can go ahead and embrace moral relativism without addressing the specifically moral sources of resistance to it, on the basis of his social account of normativity and the across-the-board relativism that he took to follow. But those of us who think there may be more to normativity than communal consensus cannot settle the issue in Rorty’s way. 3.  MORAL RELATIVISM AND RECOGNITION All that has gone before may be regarded as set-up for what can now be a relatively short discussion of how we can reconcile moral relativism with Honneth’s proposal to base morality in recognition. I want to begin by canvassing some powerful considerations that seem to speak against such a reconciliation. If we accept the social interaction account of self-consciousness, then to exist at all as a self-conscious being is to stand in social relations with others with whom one stands in relations of mutual recognition. If we have rational capacities, then we will be capable of specifically moral reflection, through which we can recognize a certain kind of parity among all points of view—we can recognize that, just as things matter from our own points of view, they likewise matter from other points of view as well. And then we can also recognize that it matters to us all to be recognized by others. For not only is it a fact that being so recognized is a condition of our very existence as self-conscious beings, but it is also a fact that when such beings undertake moral reflection, and consider how they should act given that there are other points of view from which things matter, it will matter to them that others who are also capable of moral reflection will duly take into account what matters to them—which is to say, it will matter to them to be recognized as objects of moral consideration as well as subjects who can extend such consideration. Kant wrought out of these facts a highly unified account of what mutual recognition among self-conscious rational beings might deliver by way of moral insight: such beings will recognize a moral law whose normative force is universal, in the sense that it is binding on all of them; they will recognize

Relativism and Recognition  283 that this universally binding law is also universal in scope, in the sense that it takes them all equally into account; they will recognize their common rational nature; they will recognize themselves as forming a single moral kind that includes all and only rational beings like them, who are all capable of mutual recognition; thus, they will all mutually recognize one another as members of a single kind who all recognize that they all fall under and are bound by a single moral law. Even if we agree that differently situated people will face different morally significant options and choices, which they cannot navigate without the aid of thick moral values that are not universal, we must also concede that the forms of mutual recognition that Kant exploited in his account of morality are available to them all. In conceding this, we are really conceding that there is a universal concept of a person, where to be a person is to stand in relations of mutual recognition to all other persons. When we frame the Kantian view in terms of personhood, we neatly sum up one way in which it matters to us all to be recognized by others: we want to be recognized as persons by all other persons, and if any person should withhold this recognition from us we shall experience that as a wrong of some kind. This way in which recognition clearly matters to all persons has obviously been central in political movements for civil rights, in which various human beings have claimed that it was unjust that they (and other members of their ‘groups’) were not recognized as meriting equal status as persons under the law. But to a historically minded philosopher, this rights-based way of thinking about the importance of personhood does not track timeless moral truths to be framed in terms of thin moral values; it is a way of thinking that makes sense only when certain political institutions and supporting social practices are in place, which afford life in accordance with a whole slate of liberal values thickly construed. The progressivist interpretation of the moral significance of the emergence of these values, or other values that a Hegel or a Marx would expect to come later, seems optional, at least to those of us who suspect that Williams’s suggestion may have been right—that any convergence that now seems to be emerging in our moral outlooks may simply reflect a process of cultural homogenization, which is not guaranteed to issue in better forms of life, but simply new ones. I think the argument that Williams offered against the progressivist view of history has even deeper relativist implications than I have so far managed to trace. So far I have merely emphasized that history isn’t necessarily issuing in ever better forms of life. But let me now add a related point about history that strengthens the case for moral relativism, which is that history is not a process that can be brought directly under our intentional control. Insofar as that is so, then it simply isn’t appropriate to attempt any comparative moral evaluation of different forms of life. Why should this not be appropriate? Because moral evaluation is pointless if it cannot be action-guiding. Thus, comparative moral evaluation of whole forms of life would be pointful only if we could make and unmake them at will—and to suppose that we could would be to suppose that history is indeed under our intentional control.

284  Carol Rovane However, although this feature of history—that it cannot be brought under our direct intentional control—does strengthen the case for moral relativism, it doesn’t decisively settle it. For we can still raise the question, should we not always strive to overcome alienation from others and, in doing so, strive to realize whatever potential there is for life in community with them? Should we not always morally prefer that way of relating to others? This is another way of getting at why Honneth may feel that there is philosophical work to do in order to reconcile a morality of recognition with moral relativism. Can we really be giving others due recognition when we adopt the Multimundial stance of the relativist—that is, when we do not strive to overcome the form of alienation that goes together with that stance? Here I think it is crucial to acknowledge that we cannot so much as consider the possibility of adopting the Multimundial stance towards others unless we already recognize them as persons in the sense I just described above, that we associate so closely with Kantian morality and theories of liberal rights. This recognition is sometimes only implicit, because it is not always accompanied by the language of personhood. But all the same, there is a clear distinction between those whom we might engage because they are the sorts of beings who can be engaged by someone and those, such as our pets, whom we can’t engage because they are not the sorts of things that can be engaged by anyone. That is the distinction between persons and non-persons. So I’m saying that if we could ever appropriately adopt the Multimundial stance, it would have to be towards a person; indeed, it is precisely because persons are the sorts of beings who can be engaged that it is an issue of such significance whether to approach our dealings with them in the Unimundial way that presumes that we should and must always attempt engagement, or in the Multimundial way that allows that it may be appropriate to refrain from such attempts in certain conditions. Like Williams and virtually all others, Honneth presumes that to give up on the idea that morals are necessarily universal is to give up on their objectivity. I’ve already said what I have to say about why that is a regrettable philosophical error. It muddies the waters as we take up the important moral issue about whether universal engagement with all others is or is not appropriate, by fanning the flames of anxiety that talk of relativism so often produces in philosophers, who worry that if there is no objectivity then anything goes. But to recognize that there has been a diversity of historical conditions with which people have had to cope is to recognize perfectly objective reasons why they might appropriately bring to bear different thick moral values that speak to the moral options and choices that they confront. The crux of the matter concerning the issue of moral relativism, then, is to figure out whether that is the human situation. It would be nothing like a failure of recognition to learn that it is. On the contrary, it would be to fully recognize the persons we encounter, in a way that fully appreciates the concrete historical conditions in which they must live their moral lives.

Relativism and Recognition  285 How should we proceed if we are not sure who is really right in the debate about moral relativism—if we are not sure whether history really is a progress, or if there will ever be any meaningful scope for ideals of universal human community? Some moralists might think the best way to respond to such doubts is by acting as if the Unimundial view is true—for then we can be sure that if there is any real prospect for historical progress towards universal ideals, we will not have missed the opportunity to realize them. But note that if we take that tack, then if we ever find that we are unable to take on the moral beliefs of others together with our own, we shall be forced to reject them as false; and in being so forced, we may fail to see that their beliefs speak to moral options and choices that we do not ourselves face and that they are good and true guides for those who face those options and choices. It may be the better tack, therefore, when we encounter others whose moral beliefs differ from our own, to investigate with some care whether we and they really do face common moral problems or different ones. As we investigate this, we leave open the possibility that we may have nothing to learn from them or to teach them about the proper moral bases of our respective moral choices—not because there are no moral truths at stake, but because we need different moral truths to live by. There is no danger that such an open-minded investigation would foreclose possibilities of engagement and community with others or stop the progress of history, if that is what history is. If we do face common moral problems, for which we need recourse to the same moral truths, this will surely emerge from the investigation. But we will not have arrived at this outcome while refusing to recognize the specificity of human circumstances and the possibility that they have thrown up many moral worlds and not just one. If that possibility is our actual situation, I reiterate that we shall not fail in any morally important form of recognition when we recognize that it is so. NOTES   1 I will be discussing Rorty’s overall position in a very general way throughout this paper, but not discussing any of his texts in close detail. My discussion will be guided by my understanding of his most relevant works, including Philosophy and the Mirror of Nature (Princeton University Press 1979), Consequences of Pragmatism (University of Minnesota Press 1981), Contingency, Irony and Solidarity (Cambridge University Press 1989), and Objectivity, Relativism and Truth (Cambridge University Press 1991).  2 See his “On the Very Idea of a Conceptual Scheme” (Proceedings and Addresses of the American Philosophical Association, Vol. 47, 1973–1974).  3 See his “Transcendental Arguments, Self-Reference, and Pragmatism,” in Transcendental Arguments and Science, ed. Peter Beri, Rolf-P. Horstmann, and Lorenz Kruger (Reidel Publishing, 1979).   4 See my The Metaphysics and Ethics of Relativism (Harvard University Press 2014) for a full defense of this formulation of the relativist’s position, as well as some initial attempts to evaluate what reasons there are for and against it once it is formulated along these lines.

286  Carol Rovane   5 For a mature working out of his position, see his Between Facts and Norms (MIT Press 1995) and Moral Consciousness and Communicative Action (MIT Press 1999).  6 Among his many works on this topic, the most important are “The Truth in Relativism” (Proceedings of the Aristotelian Society, New Series Vol. 75, 1974–1975) and Ethics and the Limits of Philosophy (Fontana Press 1985).   7 Honneth’s book, The Struggle for Recognition: The Moral Grammar of Social Conflicts (Polity Press 1995) contains his basic account of recognition; for a brief statement of his position on relativism and recognition, see “Social Critique between Anthropology and Reconstruction: An Interview with Axel Honneth” (Norsk Filosofisk Tiddskrift 2010 No. 3).  8 The Blackwell Companion to Relativism, ed. S. Hales (Blackwell 2011) gives a good sense of the current state of that literature.   9 Wright argued for this alleged connection between relativism and anti-realism in a series of papers, of which perhaps the most representative is “Intuition, Relativism, Realism and Rhubarb” in Truth and Realism, ed. P. Greenough and M. Lynch (Oxford University Press 2006). 10 See The Metaphysics and Ethics of Relativism, op. cit. 11 There are some common confusions that arise in connection with what it means to “share” attitudes, especially in the case of evaluative attitudes. It does not suffice for two people to share the very same evaluative attitudes that they agree about what their respective evaluative attitudes are—they must each hold that attitude, and what is required in order to hold an evaluative attitude is that one be committed to taking that attitude as a basis for one’s own deliberations and actions. In the example under discussion, even though Wright and I come to agree about what our respective evaluative attitudes are toward the gustatory properties of snails, neither of us thereby comes to hold the other’s attitude, in the sense of taking it as a basis for our own deliberations and actions—thus in particular, the fact that snails are delicious-according-to-Wright’s-gustatory-standards is not supposed to be a basis on which I should try to work out whether they are worth eating. 12 See his book of that title, The View From Nowhere (Oxford University Press 1986). 13 See Davidson, op. cit. 14 He pursues this line of argument in a series of papers that are all reprinted in the third volume of his collective papers, Subjective, Intersubjective, Objective (Oxford University Press 2001)—see in particular, “Rational Animals,” “The Second Person,” and “The Emergence of Thought.” 15 See my “The Philosophical Significance of Holism” in Companion to Davidson, ed. E. Lepore and K. Ludwig (Wiley-Blackwell 2013) for an account of how, unlike arguments against relativism from realism, the overall account of meaning and belief that drives Davidson’s anti-conceptual schemes argument most definitely does address the logical issue that divides Multimundialists from their Unimundialist opponents—in favor of the Unimunidal conclusion that logical relations do indeed run everywhere among all truth-value-bearers. 16 For a recent argument for this conclusion by a philosopher of analytic training, see Akeel Bilgrami, “Secularism, Liberalism and Relativism” in A Companion to Relativism, ed. S. Hales (Blackwell 2011).

13 Revolutionary Actions and Events Andrew Cutrofello

For most analytic philosophers, events are ubiquitous. Whether identified with property exemplifications (Kim 1996), causes and effects (Davidson 1969), repeatable states of affairs (Chisholm 1970), or entities of some other sort, “analytic” events pervade worlds. By contrast, for many continental philosophers, events are comparatively rare. Instead of pervading worlds, “continental” events unexpectedly disclose (Heidegger 1989), actualize (Deleuze 1969), or radically transform (Badiou 1988) worlds. Generally speaking, analytic events are ordinary or normal, continental events extraordinary or revolutionary. Likewise with actions. For most analytic philosophers, an action is a normal event with the right sort of intentional structure. For many continental philosophers, an action is a revolutionary event with a correspondingly more demanding intentional structure. Generalizing again, “analytic” actions are ordinary or normal, “continental” actions extraordinary or revolutionary. In this paper, I will discuss revolutionary actions and events using a theoretical framework more commonly used to think about normal actions and events, namely, John McTaggart’s distinction between A series and B series representations of time. I will focus on Hannah Arendt’s conception of revolutionary actions and Alain Badiou’s conception of revolutionary events. I will represent Arendt as a kind of A series theorist and Badiou as a kind of B series theorist. After contrasting their positions, including their respective interpretations of the French Revolution, I will briefly consider the significance of revolutionary actions and events for theories of possible worlds. Drawing on David Lewis’s distinction between modal realism and the ersatz form of modal realism he calls linguistic ersatzism, I will distinguish four different types of possible worlds, two of which permit, and two of which preclude, revolutionary actions and events. In The Human Condition, Hannah Arendt distinguishes three different types of intentional activity: labor, work, and action.1 Each is characterized by a distinctive type of temporality. Labor consists in the maintenance and reproduction of life. Its temporality is cyclical, like that of the natural needs it seeks to satisfy. Work (poiesis) interrupts cyclical nature by setting, and aiming at, a future end. Its temporality is linear and historical. Finally,

288  Andrew Cutrofello action (praxis) interrupts history by beginning something new in the present. Its temporality is “natal,” or emergent. All three—cyclical labor, teleological work, and emergent action—are essential to human existence, but action is the noblest and most distinctively human of the three. Each of these activities has its own proper sphere: family life (labor), civil society (work), and the polis (action). The polis, or public sphere, is the proper space for action because action is inherently political. It is also inherently revolutionary. If the polis were the fruit of labor, it would be something to be consumed. If it were a finished product, such as a table or a work of art, it would be something to be used or admired, occasionally repaired but not continually transformed. Because it is constituted through action, it must be reconstituted, again and again. All action has this reconstituting, or revolutionary, dimension: it gives birth not only to itself, but to the political space within which it appears. At the same time, it has a conservative dimension, for it must sustain the conditions necessary for continual political renewal. One of the most fundamental problems of politics is to reconcile these two aspects of revolutionary action. Arendt discusses this problem in On Revolution, her comparative analysis of the American and French revolutions. In her view, both revolutions failed in the long run, but for completely different reasons. The French Revolution began with the Third Estate’s declaration of itself as the National Assembly. Its first order of business was to give France a revolutionary constitution, one that could sustain the revolution itself. Despite numerous efforts, this task was never satisfactorily completed. The first republican constitution (the second of three constitutions adopted between 1791 and 1795) was suspended by the dictatorial government that made terror the order of the day. Arendt identifies several causes of the revolutionaries’ failure to make action the order of the day. These included their lack of prior participatory political experience, their Rousseau-inspired belief in the non-divisibility of political power, and the overwhelming needs of the poor, whose appearance on the revolutionary stage made it impossible to disentangle social problems from political problems (Arendt 1963, pp. 50, 66ff., 157). The American revolutionaries faced none of these obstacles. They had prior participatory political experience; they rejected Rousseau’s ideal of an undivided general will in favor of Montesquieu’s model of a division of power; and, in the absence of either a strong abolitionist movement or a slave rebellion, they faced no comparable problem of economic misery (Arendt 1963, pp. 58–62, 141, 167–170). For these reasons, they were able to adopt a republican constitution with comparative ease. However, despite making long-term provisions for future constitutional amendments, the American revolutionaries failed to incorporate into their constitution the conditions necessary to sustain revolutionary action. They took it for granted that the qualitative superiority of “public happiness” over “private welfare” would not be forgotten (Arendt 1963, pp. 117–119). Jefferson soon perceived the danger of growing political quietism. At first he believed that

Revolutionary Actions and Events  289 only periodic rebellions could keep the American revolutionary spirit alive (Arendt 1963, p. 225). Eventually, he advocated a more stable institutional remedy, proposing that state counties be divided into wards so that every citizen would have the opportunity to directly participate in the day-to-day business of government (Arendt 1963, p. 241). Without such an opportunity, he warned, Americans would gradually cease to think of themselves as citizens, i.e., as political agents. Arendt argues that this is exactly what happened. Within a generation, Americans were by and large more interested in a negative right to be protected from government interference than in a positive right to participate in politics. They had forgotten the pleasures of revolutionary action and had come to think of themselves primarily as private laborers and workers. The revolutionary ideal of action in the presence of others had been reduced to a limited right to cast an occasional ballot in a semi-private voting booth. Jefferson’s idealized wards hearkened back not only to the town hall meetings of the American Revolution, but also to the clubs and societies that flourished during the first years of the French Revolution. When Robespierre was just another member of the Jacobin Club, he defended the rights of citizens to directly participate in the political life of the nation. Later, as a member of the ruling Committee of Public Safety, he criticized the clubs and societies for fragmenting the general will (Arendt 1963, p. 232). It was enough, he now maintained, for citizens to be represented by the revolutionary government instead of presenting themselves in public forums. Former members of peremptorily disbanded clubs and societies suddenly found themselves disempowered. Ironically, so did the members of the Committee of Public Safety (and the Committee of General Security), for despite possessing a (precarious) monopoly on the means of violence, they lacked genuine political power, which for Arendt can only be generated through the joint actions of citizens. Stripped of power, even enterprises of as great pith and moment as the French Revolution lose the name of action. From the end of the eighteenth century to the middle of the twentieth, comparable instances of sustained revolutionary action were relatively few and far between. Writing in 1963, Arendt’s primary examples include the Paris Commune of 1871 and the workers’ and soldiers’ councils that spontaneously appeared in Russia in 1905, in Germany in 1918, and in Budapest during the Hungarian Revolution of 1956 (Arendt 1963,pp. 249, 254).2 Unfortunately, each of these revolutionary forms of self-government was quickly destroyed or incorporated into a ruling party. By 1963, the party system had become nearly universal. Modern systems of government differ from one another insofar as they involve single, dual, or multiple party rule (Arendt 1963, pp. 259–261). Some governments are more stable than others, and some guarantee a wider range of civil and even political rights. But even the best of representing parties is no substitute for self-presentation in public.

290  Andrew Cutrofello The difference between the party system and the council system can be modeled set-theoretically. Parties, whatever their number, partition a set into non-overlapping subsets. Party members are represented by the party itself. Councils, by contrast, are vehicles for self-presentation. Spontaneously generated local councils group themselves together to form a larger set of which they become subsets. Members of first-order subsets are represented in higher-order subsets by having one of their members belong to the latter. Thus the council system includes a representational dimension. But unlike representation in a party, which relieves the majority of citizens of any self-presentation, representation in a higher-order council presupposes widespread, if not universal, self-presentation at the first-order level. In principle, a council system could allow for greater complexity. Citizens could belong to more than one first-order subset—for instance, by being a member of a neighborhood council, a workplace council, a chatroom council, etc. Likewise, there could be multiple membership and overlapping subsets at higher levels. At the ideal, if somewhat fanciful and certainly impractical, limit, a complete Arendtian council system would have the structure of a power set, i.e., the set of all subsets of the body politic (including, therefore, singleton sets, the empty set, and the set consisting of the body politic as a whole). The term “power set” felicitously captures Arendt’s idea that power is generated through acts of collective constitution. The revolutionary power of the power set, or republic, would be conserved by a generic political subset (a revolutionary “élite”) whose authority would rest not on any non-political property shared by its members (such as intelligence, wealth, etc.), but on its fidelity to the constitutional conditions necessary for sustained revolutionary action.3 Representing Arendt’s council system in set theoretic terms makes it easier to compare her conception of revolutionary actions to Badiou’s conception of revolutionary events.4 Like Arendt, Badiou advocates a “politics without party,” but he doesn’t support her alternative model of a council system (Badiou 2005 [1998], p. 120). Instead of identifying revolutionary action with the self-presentation of citizens, Badiou focuses on the self-presentation of an “event.” An event, in his technical sense of the term, is a member of a set, or “situation,” that is indiscernible in that situation prior to its self-presentation. For an event to appear, it must be borne witness to by a collective “militant” subject that functions at the margins of the state, or power set, rather than as a governing “elite” (Badiou 1988, p. xvi). A Badiouian event requires an “evental site,” a presented subset of a situation none of whose members are presented in that situation (Badiou 1988, p. 183). An event takes place by presenting some of the site’s elements while simultaneously self-presenting, or naming, itself. Whether or not an event takes place is always a contestable matter. Badiou’s primary example of a possible event is the French Revolution, one of whose (contested) elements is or was the name “the French Revolution” (Badiou 1988, p. 189). It is impossible to prove that the French Revolution, or any other supposed

Revolutionary Actions and Events  291 event, is or was a genuine event. (I will explain in a minute why I keep hesitating between the tenseless “is” and the tensed “was.”) Because Badiou identifies ZFC set theory with ontology, and because ZFC set theory prohibits self-membership, events are ontologically impossible or “illegal” (Badiou 1988, p. 216). According to Badiou, however, they are trans-ontologically possible, for reasons having to do with the independence of the continuum hypothesis, and thus with an entirely different interpretation of the political significance of power sets than the one I have somewhat speculatively attributed to Arendt. For Badiou, situations are infinite in a specifically Cantorian sense. If the cardinality of a situation is transfinite, then the cardinality of its power set must be of a higher order of infinity. Were the continuum hypothesis provable for ZFC set theory, the cardinality of the power set of an infinite set would necessarily be that of the next infinity in the hierarchy of transfinite numbers. In such a case, it would be impossible to produce the type of “generic” set that Badiou identifies with the infinite truth of an event. In the absence of such an infinite truth, there could be no violation of the ontological prohibition of events. There could only be veridical statements about ordinary happenings, whether these be construed as property exemplifications, causes and effects, repeatable states of affairs, etc. To avoid this implication, Badiou appeals to the independence of the continuum hypothesis. Kurt Gödel demonstrated that the continuum hypothesis is compatible with ZFC set theory, but Paul Cohen later showed that it isn’t provable in ZFC theory. Specifically, Cohen showed that from any countably infinite set it is possible to extract, or “force,” a “generic” subset that isn’t a member of the original set. Adding this generic subset to the original set makes it possible to identify the cardinality of the power set of the original set with the cardinality of any of an indefinite number of higher order sets. Badiou’s interpretive strategy is to identify the generic subset with the infinite truth of an event.5 Armed with this interpretation of Cohen’s “generic procedure,” Badiou can justify his own fidelity to the French Revolution, that is, his subjective conviction that the French Revolution counts as a genuine event. Fidelity to an event can only be a matter of subjective conviction because it is always “too soon to tell” if the purported event will have taken place.6 This temporal constraint is not merely epistemic but ontological, because the infinite procedure by which it will have been determined whether or not an event has taken place is itself temporal. Only retroactively will an event have taken place by having been connected in the right way to subsequent event-generating “interventions” and event-discerning acts of fidelity. Thus truths of events are “produced” in time. Yet despite his evident commitment to the openness of the future, Badiou characterizes truths of events as eternal or timelessly true. This paradoxical position has A series and B series aspects. For McTaggart, A series events belong, successively, to the future, the present, and the

292  Andrew Cutrofello past. Statements about their temporal relations to one another are irreducibly tensed. By contrast, B series events are spread out on a temporal continuum. They are ordered by relations of succession and simultaneity, but they are not divided into future, present, and past. Hence statements about their temporal relations to one another are essentially tenseless. Badiouian events exhibit both sets of characteristics. From the standpoint of faithful subjects, events can be divided into those that have occurred, those that are occurring, and those that may occur in the future. But from the standpoint of the infinite truths associated with them, events form an eternal series. In Logics of Worlds, as we will see in a minute, Badiou attempts to do justice to the A series aspects of this two-fold picture. In Being and Event, he emphasizes the B series aspects. Faithful subjects project themselves into the standpoint of eternity by dating events, that is, by locating them on revolutionary calendars whose temporal relations are tenseless (hence Badiou’s recourse to phrases such as “five years after the October Revolution,” “France between 1789 and, let’s say, 1794,” etc. [BE 114, 189]). Without an A series indexical, calendrical relations between events can be specified only in the B series terms of “before e,” “after e,” and “simultaneous with e,” for various values of e. A faithful subject may say that an event has taken place, is taking place, or will take place, but such a statement can have no determinate truth value until it can be dated. (Nor can it have a “veridicality” value, for in Badiou’s sense of the term veridicalities are finite rather than infinite.) Unlike Badiou, Arendt has an A series conception of revolutionary actions. Earlier I called attention to her view that revolutionary actions create the space in which they appear. They also create the time of their appearance, namely, the present. Arendt conceives the present as a “hiatus,” or “gap,” “between past and future” (Arendt 1968). Revolutionary action interrupts the flow of historical time, the time measured by clocks and calendars. Hence revolutions cannot be dated in the way that ordinary activities and events can: If one dated the revolution, it was as though one had done the impossible, namely, one had dated the hiatus in time in terms of chronology, that is, of historical time. It is in the very nature of a beginning to carry with itself a measure of complete arbitrariness. . . . it is as though it came out of nowhere in either time or space. For a moment, the moment of beginning, it is as though the beginner had abolished the sequence of temporality itself, or as though the actors were thrown out of the temporal order and its continuity. (Arendt 1963, p. 198) Slavoj Žižek has characterized Arendt’s “suspension of temporality” as fundamentally “Badiouian” in character, but there is a fundamental difference between Arendt’s and Badiou’s conceptions of the way in which an action or event suspends temporality (Žižek 2009, p. 122). For Badiou, an event

Revolutionary Actions and Events  293 suspends temporality by producing an eternal truth.7 For Arendt, an action suspends temporality by disclosing the present. The apparent impossibility of dating a revolution is due to the fact that there is no absolute B series sequence in which the present can be inserted. We can situate present actions with respect to the past and the future, but these are relative rather than absolute designations. Relative to the present, we characterize past actions as past, present actions as present, and future actions as future. These are the tenses, respectively, of historical, political, and speculative discourse. History tells us what has happened, politics what is happening, and speculation what might happen. As a resolute A series theorist, Arendt emphasizes the non-reducibility of these three tensed discourses. To date a revolution is to reduce political discourse about the present to historical and/or speculative discourse about the past and future. Arendt thinks that Hegel’s and Marx’s dialectical discourse reduces politics to history and/or speculation in just this way. Dialectical statements treat the present as if it fulfilled a past future and/or anticipated a future past (Arendt 1982, pp. 4–5, 57). By reinserting the undatable present into an absolute B series calendar, they reduce revolutionary actions to objects of historical and speculative reflection. Leaving aside the merits of Arendt’s interpretation and critique of Hegel and Marx, Badiou’s discourse is not reductively dialectical in this sense. While retaining the B series standpoint of eternity, he situates it with respect to the A series perspective of faithful subjects.8 In Logics of Worlds, he distinguishes three subjective attitudes toward events: those of “faithful,” “reactive,” and “obscure” subjects (Badiou 2006, p. 47ff.). Faithful subjects affirm the present reality of an event, reactive subjects deny the present reality of an event, and obscure subjects occlude the present reality of an event. These intentional stances are essentially tensed. However, despite this new emphasis on the present, Badiou’s conception of evental truth retains a tenseless dimension. An event still has the character of a participation-like procedure through which an eternal truth is produced. Instead of distinguishing between historical, political, and speculative statements about an event, Badiou distinguishes the discourses of the reactive, obscure, and faithful subjects. Reactive subjects deny that the French Revolution will ever take place, obscure subjects mask its taking place, and faithful subjects affirm its having taken place. Yet despite the tensed language that I have just used, and that Badiou’s quarreling subjects may use as well, it is in fact only superficially tensed. Its deeper metaphysical structure is that of a B series debate in which the reactive subject insists on beforeness (nothing succeeds the Ancien Régime), the faithful subject on afterness (the French Revolution succeeds the Ancien Régime), and the obscure subject on non-simultaneity (the French Revolution doesn’t co-occur with the ordinary happenings that occur between 1789 and 1794). What Logics of Worlds has added to Being and Event is a phenomenological description of the ways in which these intentional attitudes appear to time-bound subjects. They appear in the guise of subjective convictions for which Badiou’s faithful, reactive, and

294  Andrew Cutrofello obscure subjects can only purport universality. This gives the discursive conflict between them the appearance of a Kantian quarrel between competing judgments of taste. Up until now I have ignored the fact that Arendt treats revolutions as series of actions, while Badiou treats them as series of events. According to Arendt, there is a fundamental difference between the engaged standpoint of an agent and the disengaged standpoint of a spectator who observes and judges events. Given Badiou’s characterization of revolutions as events rather than actions, his theory of generic truth procedures might seem to have a fundamentally spectatorial character. This impression is reinforced by the resemblance of the act of a faithful subject—the bearing witness to an event—to a Kantian judgment of taste. However, Badiou’s faithful subjects don’t simply judge events from afar; on the contrary, they intervene in situations by bearing witness to events. According to Badiou, moreover, it is this active dimension of his generic truth procedures that is missing from Arendt’s political philosophy. As he rightly points out, it is she rather than he who explicitly models political discourse on Kantian judgments of taste. Such judgments, he contends, are anything but truth procedures (Badiou 1998, pp. 12–13; cf. Arendt 1982, passim). Likewise, he accuses Arendt of sharing Kant’s moral horror at the actual revolutionary deeds performed by the French revolutionaries. In “An Old Question Raised Again: Is the Human Race Constantly Progressing?” Kant argues that human history is inexorably progressing toward the republican ideal that the French revolutionaries sought to actualize. He bases this prediction not on the actions of the revolutionaries, but on the enthusiasm felt by those who witnessed them (Kant 1798, p. 302). Instead of endorsing a militant politics, Kant advocates a merely discursive politics (as in his famous formula “Argue as much as you will and about whatever you will, but obey!” [Kant 1784, p. 18]). Badiou suggests that the same holds for Arendt. On her conception of the council system, political action consists in nothing more than the voicing of opinions. Moreover, the “power set” model that I have speculatively attributed to her seems to preclude the possibility of revolutionary events by limiting the “excess” of the power set over the situation it governs. In the language of Being and Event, Arendt seems to give us a “constructivist” rather than a “generic” politics, one that, again, can yield no truth procedure. There are, however, several problems with this critique. First, Arendt doesn’t actually condemn revolutionary action. On the contrary, she fully embraces its self-justifying character. Her primary objection to the deeds of the Jacobin populists is that they conflated political freedom with liberation from economic necessity. In this respect, Badiou is right to characterize her politics as anti-Jacobin. Arendt believed, as did Condorcet and the Brissotins, that there could be no successful revolution, in France or anywhere else, without the appearance of laboring (or simply starving) citizens in the public sphere. Anyone restricted to a life of labor is, as it were, “not yet”

Revolutionary Actions and Events  295 in the present insofar as she hasn’t appeared in the polis. Conversely, anyone lucky enough to work toward self-conceived ends is directed toward the future and so, in this sense, “no longer” in the present. Revolutionary action requires collective self-constitution in the present. What Arendt objected to in Jacobin populism was its subversion of this process of collective self-constitution. In claiming a monopoly of power, the militant Jacobins acted as if they wanted “a revolution without a revolution”—to turn Robespierre’s famous phrase against Robespierre himself (Robespierre 2007 [1792], p. 43). A second problem with the Badiouian critique is that the function of Arendt’s council system isn’t to limit excess by reducing presentation to representation—that, as we have seen, is the function of political parties—but to enable newcomers to appear and act. Likewise, Arendt doesn’t restrict political intervention to speech acts in constituted assemblies. Instead, she endorses acts of civil disobedience that are themselves inherently revolutionary (Arendt 1972). Finally, the reason she compares political opinions to Kantian judgments of taste isn’t to distinguish them from truth procedures per se, but to contrast them with dogmatic truth procedures, a point to which I will return. Together, these aspects of her picture of revolutionary action suggest that her alternative to Badiou’s generic politics isn’t a constructivist politics that would preclude the possibility of revolutionary events. Rather, it is a political application of an orientation of thought that Badiou overlooks: what Paul Livingston has characterized as the “paradoxico-critical” orientation (Livingston 2012, pp. 11, 56ff.). As Livingston points out, Badiou’s generic orientation rests on a choice that ZFC set theory leaves undetermined. The choice, first articulated by Gödel, is between completeness and consistency for formal systems. Badiou’s generic orientation sacrifices completeness for the sake of preserving consistency. The paradoxico-critical orientation preserves completeness at the cost of accepting inconsistencies. These two “post-Cantorian” orientations can both be opposed to two others: the “constructivist” and the “onto-theological.”9 The constructivist orientation limits excess (and so prohibits revolutionary events) by restricting the hierarchy of transfinite sets to those that can be constructed on the basis of predicates available in lower-order sets. The onto-theological orientation (which Badiou calls “transcendent”) circumvents Gödel’s dilemma by positing a complete and consistent totality that transcends the axiomatic constraints of formal systems. Each of these four orientations can be associated with a different type of politics. Badiou’s generic politics is based on an affirmation of multiplicity and repudiation of every claim to totality. Arendt, despite or because of her critical engagement with totalitarianism, seems to accept the ideal of a political totality on condition that its inherent inconsistencies be acknowledged and embraced (Arendt 1951). Inconsistency in politics takes the form of incompatible truth procedures, a discursive predicament that

296  Andrew Cutrofello she represents as the clash of political opinions.10 Political opinions aren’t ordinary beliefs about veridicalities. If they were, then in the language of Kant they would be determining judgments subject to the principle of non-contradiction (Kant 1790, p. 67). Neither are they non-cognitive expressions of preference. If they were, they would be equivalent to Kantian judgments about the agreeable and so wouldn’t logically conflict at all (Kant 1790, p. 91). Instead, they are cognitively reflective judgments about singular cases (e.g., “This is beautiful,” or, for Arendt [though not for Kant], “This is good” or “This is just”). Unlike logically reflective judgments that seek universals under which to subsume particulars, political opinions resemble aesthetically reflective judgments in that they concern the singular, i.e., unique, forms of particulars. Such judgments purport subjective universality, and as such they can and do contradict each other (e.g., “This is just” vs. “This is not just”). For Kant, aesthetic arguments aim at a consensus the possibility of which is grounded in the idea of a transcendent object (what he calls, in the case of judgments of taste, the “supersensible substratum of humanity” [Kant 2000, p. 216]). For Arendt, who rejects Kant’s transcendent orientation in favor of what I am characterizing, with Livingston, as a paradoxico-critical orientation, there is nothing transcendent to guarantee that political consensus will ever be reached, strive for it though we may. The reason she compares political opinions to Kantian judgments of taste is not to privilege judgment over action (pace Badiou), but to model the discursive inconsistencies that a paradoxico-critical politics must affirm. Badiou’s generic orientation yields a politics of consistent truth procedures. It doesn’t preclude political contestation, but it does preclude inconsistent truths. Badiou’s faithful, reactive, and obscure subjects may each purport subjective universality for their respective points of view, but there is only a single, infinite truth about which they disagree. That truth, as we have seen, transcends the finite A series order of political contestation. It is, as it were, a date on a calendar, giving the series of revolutionary events its B series character. By contrast, Arendt’s paradoxico-critical orientation precludes any reference to eternal truths that could reconcile inconsistent truth procedures. This explains, from another point of view, the sense in which, for her, it is impossible to date revolutions. Each of the other two orientations of thought—the constructivist and onto-theological—can also be associated with a particular type of politics. For our purposes, what matters is that they both preclude the possibility of revolutionary actions and events. The constructivist orientation does so in the way indicated by Badiou, namely, by restricting the universe of infinite sets to those that can be constructed on the basis of predicates available in lower-order sets. So restricted, a constructed set or world can contain only ordinary actions and events. Within each set or world, the series of actions and events has the form of a consistent A series for subjects situated within it. (We will see how in a minute.)

Revolutionary Actions and Events  297 The onto-theological orientation precludes the possibility of revolutionary actions and events in a different way, namely, by positing a complete and consistent totality that isn’t subject to the axiomatic constraints of set theory. Once again we are left with only ordinary actions and events, conceived now not in situ but sub specie aeternitas. In transcending discourse, the onto-theological orientation has transcended time as well. Hence its series of actions and events has the character of a B series—or, rather, a C series. This is McTaggart’s original position. McTaggart began with the paradox that all events (including actions) have the seemingly incompatible properties of being future, present, and past. Clearly, he thought, no event can be “at once” future, present, and past. However, we cannot avoid the paradox by dropping the “at once” in favor of “successively,” for that simply pushes the problem to a meta-level and thence to an infinite regress. It does so because we must explain in A series terms what it means for events to have the A series properties “successively.” To avoid this paradox, McTaggart’s B series theorist posits mere ordering relations. But the B series doesn’t allow for change, and so doesn’t count as a temporal series at all. McTaggart concludes that time is unreal. However, he doesn’t deny the reality of the elements of the B series or the order that they constitute. Instead he identifies the elements—what we misperceive as temporal events—with sets, and he identifies the order that these sets constitute with that of an “Inclusion Series” (McTaggart 1927, p. 247). On this picture, “earlier” events are simply proper parts of “later” events. Because we misperceive the Inclusion Series—or “C series”—as a temporal series, it seems meaningful to us to characterize inclusion relations as B series relations. But this is an illusion. Were there such a thing as temporal change, only the A series could adequately model it. B series thinking arises from inserting the atemporal C series into the A series. To arrive at an adequate picture of reality, we must subtract the C series from the A series setting that gives it the appearance of a B series. Like McTaggart, Badiou models the series of ordinary happenings as an inclusion series (Badiou 1988, p. 143). Unlike McTaggart, Badiou goes on to distinguish the series of ordinary happenings (“nature”) from the series of genuine (revolutionary) events (“history”) (Badiou 1988, pp. 197–198). Since natural sequences of ordinary happenings are ahistorical, they can be regarded as comprising a fully atemporal C series. If nothing else happened but what happens in nature there would be no such thing as a true event, for events involve genuine change (on this point Badiou is in complete agreement with McTaggart). Badiou’s theory of generic truth procedures can be understood as an account of what would happen if a genuinely temporal entity appeared within McTaggart’s fully atemporal C series. An event doesn’t just interrupt the sequence of ordinary happenings; it introduces temporality itself into an otherwise atemporal reality. By allowing for the possibility of change, Badiou justifies his representation of the series of events as a B series.

298  Andrew Cutrofello Arendt’s paradoxico-critical orientation offers us another way of responding to McTaggart’s paradox. In a word, we can accept it. When we say that an event will take place, we are speculating. When we say that it is taking place, we express a political opinion. When we say that it has taken place, we are making an historical judgment. Thus to say that events are successively future, present, and past is to say that they are the objects, successively, of speculative, political, and historical discourse. The challenge raised by McTaggart is to say whether “successively” should be construed as a speculative, political, or historical term, or, instead, as a neutral B series term. As a paradoxico-critic, Arendt could reply that her use of “successively” cannot be reduced to speculative, political, or historical terms. It is, in fact, a B series term. She could say this, however, without abandoning her commitment to the A series picture, for her concession would simply amount to accepting the paradox: an option uniquely available to a paradoxico-critic. Perhaps we can make sense of this picture by reconsidering the way in which the present is situated between past and future. In The Life of the Mind, Arendt suggests that, in the absence of thinking subjects, “there would be no difference between past and future, but only everlasting change” (Arendt 1978, p. 208). I take this to mean that, in the absence of thinking subjects, time would in fact have a B series structure. When we observe ourselves from this asubjective point of view, we are not so much adopting the standpoint of eternity as counterfactually representing the world we inhabit without us in it. When we do this, we lose our temporal bearings because in subtracting ourselves we subtract the present with respect to which past and future can be distinguished. There is something inherently paradoxical in the B series picture that emerges when we perform this thought experiment, for we can make sense of change only from our usual A series vantage point, namely, the present. There is something similarly disorienting about McTaggart’s invitation to consider what we mean when we say that events are successively future, present, and past. The paradox is that we both can and cannot transcend our A series perspective. For Arendt, the present isn’t simply the junction between future and past, the place through which future events must pass to become past events. For thinking subjects, the present is the gap where two opposing vectors meet: the vector of an infinite past and the vector of an infinite future. Situated in this gap, we are doubly burdened: with “the dead weight of the past” and dread of the unborn future. We seek refuge from the past in hope for the future, and refuge from the future in “nostalgia” for the past (Arendt 1978, p. 205). One way of dealing with this unstable oscillation is to seek to transcend time itself, to take refuge in the thought of eternity. Another is, simply, to think. Arendt suggests that thinking is a way of responding to the fact that time past and time future are both bearing down on (without being quite present in) time present.11 When we think, we move from the present “toward an undetermined end,” but without transcending time (Arendt

Revolutionary Actions and Events  299 1978, p. 209). Likewise, when we act we respond to the past and future by opening up the gap between them. In this respect, revolutionary actions take regard of ancestors and descendants. But it is always with respect to a privileged moment in time—now—that the present past is past and the present future is future. Analytic responses to McTaggart’s paradox have typically taken one of two lines: either to defend the A series picture of time against the alleged paradox or else to defend the B series picture of time. Many, if not all, A series theorists are constructivists. On their view, time genuinely passes and there are only ordinary actions and events. Conversely, most B series theorists adopt a modified version of McTaggart’s onto-theological or transcendent orientation. Instead of interpreting the B series as a C series—i.e., as an inclusion series of sets—they characterize it as a temporal series, albeit one that doesn’t allow for genuine change. In support of their view, B series theorists often appeal to the “space-like” representation of time in general relativity. They are comfortable with the idea that four-dimensional objects have “temporal parts” as well as “spatial parts.” Just as the four orientations can be associated with four different conceptions of time, so they can be associated with four different conceptions of possible worlds. Two of the four are instanced in David Lewis’s modal realism and the constructivist position that he calls “linguistic ersatzism” (Lewis 1986a, p. 142ff.). Lewis’s modal realism is onto-theological—or it would be if he retained the theistic ground of Leibniz’s modal realism; as is, we can simply call it transcendent. On this view, logical space is complete. It is divided into distinct possible worlds, each of which is fully determinate. Each world is united by its own spatiotemporal relations or some analogue thereof (Lewis 1986a, pp. 74–75). Within each world, everything that exists is compossible, so there are no ontological inconsistencies in the sense of impossible worlds (Lewis 1986a, p. 7n.). All worlds have the same ontological status, so the one we call actual has no special privilege (Lewis 1986a, pp. 92–93). Likewise for every part of each world’s spacetime: whatever portion of ours we call the present has no special privilege. Lewis is a B series theorist. He regards events as properties of regions of spacetime (Lewis 1986b). Collectively, events comprise an entire world’s history, but it is a history not in Badiou’s sense, for Lewisian events are all normal. In effect, all time is eternally present, making all time “unredeemable.”12 Linguistic ersatzers, or constructivists, deny the trans-linguistic reality of Lewis’s logical space. They take possible worlds to be set theoretic representations of the actual world, the only world there is. Only one such representation is accurate. The others are false. Constructivists, then, are actualists.13 They are typically also presentists, for just as they take the actual world to be the only real world, so they take the present to be the only real part of time. Worldly objects aren’t spread out in time in the way that they are spread out in space; if they exist for more than an instant they “endure” rather than “perdure” (Lewis 1986a, p. 202). Constructivists can,

300  Andrew Cutrofello though they need not, accept the openness of the future, but their theoretical framework precludes the possibility of revolutionary events. Even if it isn’t determined now whether or not a sea battle will occur tomorrow, its eventual occurring or not occurring won’t involve a radical transformation of the existing ontological order. In Lewis’s terminology, it won’t involve the instantiation of a hitherto alien property, a property not previously present in the world.14 Whatever can happen in the constructivist’s world is already present as possible; things change, but possibilities don’t. Constructivists are A series thinkers, but not paradoxico-critical A series thinkers. Instead of accepting McTaggart’s paradox, they must respond to it in some way. One extreme way is to deny that past and future have any reality at all: if events are only real in the present, they don’t successively have the properties of being future, present, and past. Finally, Arendt and Badiou offer paradoxico-critical and generic representations of possible worlds. Arendtian worlds—or perhaps we should say Arendt’s world, for she may be an actualist of some sort—are (or is) complete but inconsistent. Conversely, Badiou’s worlds—here we must speak of an incomplete infinity—are consistent but incomplete. A crucial feature of each of these models is the way in which it incorporates irreducibly de se attitudes. For Badiou, every event requires a faithful subject for whom it is an event. This requirement is connected to the A series setting that makes his series of events a B series rather than a C series. Likewise, for Arendt, revolutionary actions and events are objects of irreducibly de se attitudes. As we have seen, these attitudes may conflict, whether as clashing intentions (inconsistent actions) or as clashing opinions (inconsistent truth procedures).15 A paradoxico-critic need not (and probably should not) posit inconsistencies everywhere. She need only posit them at the limits of the world.16 That way, the worlds in question will still be possible and not impossible. Badiou’s generic orientation requires that events be essentially rare: an event must stand out against a normal background; that is, it must appear to faithful subjects as an exception, albeit one that attests to a universal truth about an entire situation. Nevertheless, we might wonder whether there are worlds in which every happening is revolutionary for some subject. All we would need, it seems, is enough faith or fidelity to go around. There couldn’t be a plenum of events from any (let alone every) available perspective, but we could theoretically have enough perspectives (those of monads, for example) to fill out the whole. In such a world, analytic debates about the nature of ordinary actions and events would take on new significance. If ordinary events are property exemplifications, then every event would exemplify a hitherto alien property for some subject. If, instead, ordinary events are causes and effects, then every event would interrupt the causal nexus for some subject. If events are repeatable states of affairs, then every event would repeat something unrepeatable.17 Finally, and most importantly, we may ask whether our world contains revolutionary actions and events, and how we should think about designated revolutions such as the French Revolution. If the four types of

Revolutionary Actions and Events  301 worlds—transcendent, constructivist, generic, and paradoxico-critical— partition logical space into regions, then the answer to this question will depend on where our world lies in logical space. But if the four types of worlds are taken instead to represent alternative models of worlds in general, then the answer will depend on the nature of logical space itself: an issue that may, in the end, be determinable not by ontological prose but by a kind of metaphysical poetry.18 NOTES   1 Together, these three activities comprise the “vita activa,” in contrast to the “vita contemplativa.” See Arendt (1958), especially Chapters III, IV, and V.   2 Arendt (1963), pp 249, 254. Schell (2006) argues that the Hungarian Revolution was the first of a “wave” of subsequent “democratic revolutions” of which Arendt would have approved (Arendt 1963, p. xxi). This may be so, but it should be noted that Arendt’s revolutionary ideal was far more republican (in the eighteenth-century sense) than democratic (in the modern sense).   3 Arendt makes clear her ambivalence about the word “élite,” which she herself always puts it in scare-quotes: “My quarrel with the ‘élite’ is that the term implies an oligarchic form of government, the domination of the many by the rule of a few. . . . The ‘élite sprung from the people’ [a phrase used by Maurice Duverger] has replaced the pre-modern élites of birth and wealth; it has nowhere enabled the people qua people to make their entrance into political life and to become participators in public affairs. . . . Of course the men who sat in the councils were also an élite, they were even the only political élite, of the people and sprung from the people, the modern world has ever seen, but they were not nominated from above and not supported from below. With respect to the elementary councils that sprang up wherever people lived or worked together, one is tempted to say that they had selected themselves; those who organized themselves were those who cared and those who took the initiative; they were the political élite of the people brought into the open by the revolution” (Arendt 1963, pp. 268–270).  4 I do not mean to give the impression that all Badiouian events are revolutionary in the conventional political sense. Besides political events, Badiou recognizes amatory, artistic, and scientific events, speaking, for example, of “the Galilean revolution” (Badiou 1988, p. 76), “Eudoxas’ [sic] geometrizing revolution” (Badiou 1988, p. 167), “the conceptual revolution of Evariste Galois” (Badiou 1988, p. 214), etc.   5 Badiou’s Cantorian account of the trans-ontological possibility of ontologically prohibited events resembles Kant’s account of the noumenal possibility of phenomenally impossible acts of freedom (acts of freedom being phenomenally “illegal” in the sense of violating deterministic laws of nature). Badiouian events are exceptions in the sequence of ordinary happenings, just as Kantian acts of freedom are exceptions in the sequence of heteronomous performances. To every Badiouian event there corresponds a transcendental subject whose faithful action attests to an infinite truth associated with it, just as for Kant a dutiful subject acts as if it were giving a new law to nature.   6 Here I allude to the famous (and famously inscrutable) remark of Zhou Enlai, the first Premier of the People’s Republic of China, who is reported to have told Henry Kissinger in 1972 that it was “too soon to tell” what the significance of the French Revolution was. For context and contrasting interpretations see William Doyle’s and Krishan Kumar’s successive letters to the editor in The

302  Andrew Cutrofello Times Literary Supplement, 8 November 2013 (“Tocqueville in China”) and 15 November 2013 (“Which Revolution?”). For Badiou, I am suggesting, it is always “too soon to tell” whether a purported event has taken place, despite the confident asseverations of faithful subjects.   7 This is why it can seem, as it does to James Williams, that “Badiou denies time” (Williams 2012, p. 115).   8 This is arguably true of Hegel and Marx as well. Their shared appreciation of the distinctive character of the present as the proper time (and place) for action can be seen in their respective appropriations of Aesop’s fable of the braggart who claimed to have jumped further than anyone else had—but elsewhere (at Rhodes), and in the past. To this he was challenged “Hic Rhodus, hic saltus” (“Here’s Rhodes; the jump should take place here [and now]”). Hegel draws the moral that “each individual is . . . a child of his time” (Hegel 1820, p. 21). Marx ups the ante by changing the noun “saltus” (jump) to the imperative “salta” (dance!) (Marx 1852, p. 35). Badiou repeats Marx’s version just before introducing his conception of an event (Badiou 1988, p. 188).   9 Livingston (2012), pp. 54, 67ff. Cf. Badiou (1988), p. 298. 10 Or what Jean-François Lyotard (1983) calls differends, unresolvable conflicts between incompatible idioms. 11 Here I allude to the language of T.S. Eliot’s “Burnt Norton” in Eliot (1944), p. 3. 12 Once again I allude to Eliot: “If all time is eternally present / All time is ­unredeemable” (Eliot 1944, p. 3). Badiou finds a similar point expressed in Mallarmé’s “Un coup de dès”: “nothing will have taken place but [the] place [rien n’aura eu lieu que le lieu].” Cited in Badiou (1988), p. 204. Or in the words of Pete Townsend: “There’s nothing in the street / Looks any different to me.” 13 For Lewis’s characterization of this position, see Lewis (1986a), pp. 142–165. Pictorial ersatzism may be another instance of a constructivist approach to modality. Magical ersatzism is another instance of the ontological (or onto-theological) approach. 14 Lewis (1986a), p. 159. I do not mean to imply that for Lewis it would make sense for an alien property to be instantiated at any point of a world’s history, for then it wouldn’t be alien to that world. The point I am making about “hitherto alien” properties applies only within the A series framework that the constructivist shares with the paradoxico-critic. 15 Carol Rovane’s multimundialism, which posits a plurality of actual worlds, could perhaps be construed as another paradoxico-critical model. See Rovane (2013) and “Relativism and Recognition,” Chapter 12 of this volume. 16 See Livingston (2012, p. 36) and (cited there) Priest (2002). 17 This is how Deleuze characterizes “festive” events: “This is the apparent paradox of festivals: they repeat an ‘unrepeatable’ ” Deleuze (1968), p. 1. 18 For helpful input on this chapter I thank my co-editors, Jeffrey Bell and Paul Livingston, as well as Sara Beardsworth and Maria Alexandra Keller.

WORKS CITED Arendt, H. 1951 (1976). The Origins of Totalitarianism. New York: Harcourt. Arendt, H. 1958 (1998). The Human Condition. 2nd ed. Chicago: The University of Chicago Press.

Revolutionary Actions and Events  303 Arendt, H. 1963 (2006). On Revolution. New York: Penguin. Arendt, H. 1968 (2006). Between Past and Future. New York: Penguin. Arendt, H. 1972. Civil Disobedience. Crises of the Republic. New York: Harcourt, Brace, Jovanovich, 49–102. Arendt, H. 1978. The Life of the Mind: One. Thinking. New York: Harcourt, Brace, Jovanovich. Arendt, H. 1982. Lectures on Kant’s Political Philosophy. Ed. Ronald Beiner. Chicago: The University of Chicago Press. Badiou, A. 1988 (2013). Being and Event. Trans. Oliver Feltham. New York: Bloomsbury. Badiou, A. 1998 (2005). Metapolitics. Trans. Jason Barker. New York: Verso. Badiou, A. 2006 (2009). Logics of Worlds: Being and Event II. Trans. Alberto Toscano. New York: Continuum. Chisholm, R. 1970. Events and Propositions. Noûs 4 (1):15–24. Davidson, D. 1969 (2001). The Individuation of Events. In Essays on Actions and Events. 2nd ed. New York: Oxford University Press, 163–180. Deleuze, G. 1968 (1994). Difference and Repetition. Trans. Paul Patton. New York: Columbia University Press. Deleuze, G. 1969 (1990). The Logic of Sense. Trans. Mark Lester with Charles Stivale. New York: Columbia University Press. Eliot, T. S. 1944 (2001). Four Quartets. London: Faber & Faber. Hegel, G. W. F. 1820 (1991). Elements of the Philosophy of Right. Ed. Allen W. Wood, trans. H. B. Nisbet. New York: Cambridge University Press. Heidegger, M. 1989 (2012). Contributions to Philosophy (Of the Event). Trans. Richard Rojcewicz and Daniela Vallega-Neu. Bloomington: Indiana University Press. Kant, I. 1784 (1996). An Answer to the Question: What Is Enlightenment? In Mary J. Gregor, Trans. and ed., Practical Philosophy. New York: Cambridge University Press, 11–22. Kant, I. 1790 (2000). Critique of the Power of Judgment. Trans. Paul Guyer and Eric Matthews. Cambridge: Cambridge University Press. Kant, I. 1798 (1996). The Conflict of the Faculties. Trans. Mary J. Gregor and Robert Anchor. In Allen W. Wood and George di Giovanni, eds., Religion and Rational Theology. New York: Cambridge University Press, 233–328. Kim, J. 1966. On the Psycho-Physical Identity Theory. American Philosophical Quarterly, 3 (3) (July): 227–235. Lewis, D. 1986a (2001). On the Plurality of Worlds. Malden, MA: Blackwell. Lewis, D. 1986b. Events. In Philosophical Papers Volume II. New York: Oxford University Press, 241–269. Livingston, P. M. 2012. The Politics of Logic: Badiou, Wittgenstein, and the Consequences of Formalism. New York: Routledge. Lyotard, J.-F. 1983 (1988). The Differend: Phrases in Dispute. Trans. Georges Van Den Abbeele. Minneapolis, MN: University of Minnesota Press. Marx, K. 1852 (1996). The Eighteenth Brumaire of Louis Bonaparte. In Terrell Carver, ed., Later Political Writings. New York: Cambridge University Press, 31–127. McTaggart, J. 1927 (1988). The Nature of Existence Volume II. New York: Cambridge University Press. Priest, G. 2002. Beyond the Limits of Thought. 2nd ed. New York: Oxford University Press. Robespierre, M. 1792 (2007). Extracts from ‘Answer to Louvet’s Accusation.’ In Jean Ducange, ed., and John Howe, trans., Virtue and Terror. New York: Verso, 39–48.

304  Andrew Cutrofello Rovane, C. 2013. The Metaphysics and Ethics of Relativism. Cambridge, MA: Harvard University Press. Schell, J. 2006. Introduction: The Arendtian Revolutions. In H. Arendt, ed., On Revolution. New York: Penguin Classics, xi–xxix. Williams, J. 2012. A Critique of Alain Badiou’s Denial of Time in his Philosophy of Events. In Sean Bowden and Simon Duffy, eds., Badiou and Philosophy. Edinburgh: Edinburgh University Press, 113–131. Žižek, S. 2009. In Defense of Lost Causes. New York: Verso.

14 Varieties of Shared Intentionality Tomasello and Classical Phenomenology Dan Zahavi and Glenda Satne

1. INTRODUCTION There are many things we do together: we enjoy listening to Beethoven’s Ninth Symphony, we paint a house, we discuss a proof of Fermat’s last theorem together. We also share a tradition with our ancestors, the world of nature with people from distant countries we have never met and responsibility with our co-workers for the outcome of a given task. There are many ways in which we can be together, have joint goals, share intentions, emotions and experiences. However, current accounts tend to explain these different ways of being-together in terms of one single form of collective intentionality. In doing so, they endorse three main assumptions: First, most theories that purport to give an account of collective intentionality assume, even if many times implicitly, a “uniformity thesis,” i.e., they are committed to the idea that all forms of collective intentionality ultimately reduce to one. Some understand it as a form of mutual commitment (Gilbert 2013), others as a state entertained by an irreducible plural subject (Schmid 2009). Some have thought it to be construed out of the individual’s mental states, either as states with an irreducible we-mode (Searle 1995) or as states that, suitably combined, provide the basis for a shared state (Bratman 1992). Second, the literature on collective intentionality has focused primarily on intentional action, asking questions such as when an action can be said to be collective, when an intention can be said to be shared, or when we are assuming collective responsibility for a given action or the practical reasoning behind it. Third, most accounts of collective intentionality analyze it in abstraction from the various specific forms that social relationships assume in actual and historical contexts. In this way, they ignore the possibility that collective intentionality might assume different forms according to specific social types and relations, with respect to the dimensions of past and future, and in connection with different communal practices and institutional settings. In this contribution, we want to explore the topic of shared intentionality and ultimately criticize all three assumptions. We shall (i) investigate

306  Dan Zahavi and Glenda Satne how the debate on the nature of collective intentionality might be enriched by not accepting the uniformity thesis. We will take our point of departure in the idea that we can participate in collective endeavors in different ways and defend the idea that we need to operate with a richer framework, if we are to understand the features of these different forms of collective intentionality. Furthermore, we will (ii) discuss different concrete forms of collective intentionality, widening the spectrum of the analysis, and include not only joint or collective action, but also shared emotions, shared norms, and forms of intentionality that underlie communal life as exhibited in sharing institutions, social types, a common history, and our capacity to think and reason about states of affairs. In pursuing (i) and (ii) we will show how classical phenomenology offers important resources for these tasks. We will conclude by showing how collective intentionality can be a topic around which some strands in analytic philosophy of mind and language and classical phenomenology can profitably work together. 2.  JOINT AND COLLECTIVE INTENTIONALITY Most accounts of collective intentionality assume, though often tacitly, that in thinking of collective intentionality we need to draw a line between individual forms of intentionality, including thinking, feeling, and acting, and collective forms, in which a number of different subjects come together in the performance of such activities. What has been less explored, and often neglected, is the possibility that shared intentionality may come in many forms and, importantly, in forms that in significant ways differ from each other. In the recent literature about the topic, there is nevertheless a noteworthy exception: Michael Tomasello’s most recent book A Natural History of Human Thinking (Tomasello 2014). This book draws on previous work that Tomasello and others have been carrying out over the last decade or so and defends the idea that in order to make sense of both the ontogeny and phylogeny of human thinking, we need to distinguish two different forms of shared intentionality (Tomasello 2014, p. 31). Even if Tomasello’s investigation of shared intentionality is indebted to the contribution of various philosophers of action, including Bratman, Searle, and Gilbert, his distinction between these two forms of shared intentionality is original and not to be found in the work of these authors. The two forms in question are the following: 1.  A form of shared intentionality that involves small-scale collaboration and joint attention. Tomasello calls this joint intentionality since it involves a short-lived relation between ad hoc pairs of individuals in the here and now. He also speaks of it in terms of a second-personal joint intentionality, or as a we-intentionality with a particular other (2014, p. 33). This second-personal social engagement between ‘I’ and ‘you’

Varieties of Shared Intentionality  307 has two minimal characteristics: (i) rather than observing the social interaction from the outside, the individual is directly participating in it; and (ii) the interaction does not take place with some general indefinite group, but with a specific individual with whom there is a dyadic relationship (2014, p. 48). 2. A form of shared intentionality that refers to large-scale forms of collaboration (and competition), that is, forms that go beyond the here and now and involve the construction of a common cultural ground (involving conventions, norms, and institutions) as well as in-group/ out-group differentiations (2014, p. 5–6). This is what Tomasello calls collective intentionality.1 Some further specifications concerning Tomasello’s account of shared intentionality are necessary. First, in his discussion of joint intentionality, Tomasello is adamant that it requires both sharedness and individuality. These requirements are not unlike those that must be met in the case of joint attention. Just as joint attention doesn’t involve a fusion of perspectives, but an awareness of the fact that different perspectives are brought to bear on the same target—in the case of a young child, the whole point of proto-declaratives is to bring someone else’s focus of attention in line with her own—joint intentionality requires that each partner plays her own interdependent role. Second, Tomasello links the distinction between joint and collective intentionality to issues of validity and objectivity. When two people are aware that they are both attending to the same thing simultaneously, an understanding of different perspectives might arise (2014, p. 45). In the case of joint intentionality, the individual could encounter another distinct perspective that he or she could contrast with her own. However, although we already can have several perspectives in play at this stage—when dealing with small-scale I-thou relations—we are still far from having attained real objectivity in the sense of universal validity. A significant further step takes place the moment individuals become group-directed or group-minded, i.e., the moment they move from a restricted second-person dyadic interaction and adopt a group identity, i.e., start to identify not only with coexisting strangers belonging to the same in-group, but also with ancestors and descendants. From being dyadic, the we transforms into something more akin to an enduring culture, where we do things in a certain way, and where we identify with others who share this common ground, even if we don’t know them personally as individuals (2014, pp. 83–84). For an individual born into a culture with an already existing structured inventory of conceptualizations and norms, those standards will seem natural and readymade. This process of group-identification allows for a new level of generality. It is not simply a question of adding more perspectives to the two already in play, but a move to the level of any perspective whatsoever, i.e., a view from everywhere (and nowhere) (2014, p. 122).

308  Dan Zahavi and Glenda Satne Third, Tomasello introduces a corresponding distinction between second-person and group-level normative pressure. In the former case, I am sensitive to how another individual evaluates me and might regulate my actions accordingly. In the latter case, a more objective or agent-neutral standard is at play. Even if the group norm is enforced by a particular individual, she is, so to speak, doing so as an emissary of the group as a whole (2014, p. 88). Tomasello rightly, in our view, rejects the idea that language is the alpha and omega of sociality. Obviously, language does play a crucial role for the development of human thinking, but it plays that role only at a relatively late stage in the process. It is, as Tomasello puts it, “the capstone of uniquely human cognition and thinking, not its foundation” (2014, p. 127). In fact, language could not get off the ground, nor make its decisive contribution, if it was not because of an already existing social infrastructure (2014, p. 128). However, in explaining how this process unfolds and what this awareness of perspective differences present in joint intentionality amounts to, Tomasello takes on board without further ado a number of assumptions found in the traditional theory-of-mind literature on social cognition. He highlights the need for recursive inferences, simulations, and ongoing self-monitoring and metarepresentations (2014, pp. 5, 9, 143) and speaks of how communication requires me to simulate my partner’s abductive inferences ahead of time, in order to anticipate how I might be comprehended or interpreted by him (2014, p. 94).2 Tomasello is in part following Bratman here (1992), whose conception of joint intentionality also requires the satisfaction of a common knowledge clause that involves metarepresentations of other people’s mental states. Despite Bratman’s influence, Tomasello must still be recognized for enriching the standard proposal by emphasizing the importance of integrating an account of social cognition and an understanding of face-to-face encounters into a proper model of joint action. Having said that, much of what Tomasello writes about social cognition and the dyadic encounter comes across as somewhat problematic.3 Tomasello simply assumes that an account of social cognition must start from the observation of mere behaviour and then explain how we come to infer the existence of an underlying inner mind. Whereas Tomasello has earlier spoken out in favor of a simulation theory of mind (1999, p. 70), according to which individuals understand others in analogy with themselves, he now appears to opt for a model that integrates elements from both simulation theory (ST) and theory-theory (TT) of mind. He is thereby following the consensus view, which has shifted from considering ST and TT as mutually exclusive theoretical paradigms to recognizing the need for a hybrid account that draws on both theories (cf. Nichols and Stich 2003, p. 132). What Tomasello overlooks in this move, however, is the possibility that some of the assumptions shared by both ST and TT are mistaken and ultimately impede a correct understanding of central forms of

Varieties of Shared Intentionality  309 social cognition, including basic face-to-face engagement. This possibility, which remains unexplored by Tomasello, is one that has been energetically pursued in recent years. Many of the arguments heralded in support of what might amount to an alternative third option have a distinct phenomenological heritage (cf. Gallagher and Zahavi 2012; Zahavi 2014). What is the main objection to the ST and TT framework? Although there are noticeable exceptions (e.g., Currie 2011; Lavelle 2012), many prominent defenders of both ST and TT have routinely assumed that other minds are characterized by a fundamental invisibility. They remain concealed and hidden, and it is precisely because we lack a direct experiential access to the mental states of others—which are frequently described as “inherently unobservable constructs” (Mitchell 2008)—that we need to rely on and employ either theoretical inferences or internal simulations. Thus TT and ST often share a fundamental background assumption, namely the view that the minds of others are hidden, which is why they have considered one of the main challenges facing a theory of social cognition to be the question of how and why we start ascribing such hidden mental entities or processes to certain publicly observable bodies. But is this assumption really sound? Is it really true that mental states (be it all of them or only those belonging to others) are invisible constructs and that our engagement with others as minded creatures is initially (or even exclusively) a question of attributing such hidden states to them? Is it really the case that when faced with a weeping person, I first perceive drops of liquid rolling from her eyes, distortions of her facial muscles, and broken sounds, and only then in a subsequent move come to realize that the person is grieving? This is a view that many phenomenologists have called into question. Consider, for instance, the following quote by Merleau-Ponty: Imagine that I am in the presence of someone who, for one reason or another, is extremely annoyed with me. My interlocutor gets angry and I notice that he is expressing his anger by speaking aggressively, by gesticulating and shouting. But where is this anger? People will say that it is in the mind of my interlocutor. What this means is not entirely clear. For I could not imagine the malice and cruelty which I discern in my opponent’s looks separated from his gestures, speech and body. None of this takes place in some otherworldly realm, in some shrine located beyond the body of the angry man. It really is here, in this room and in this part of the room, that the anger breaks forth. It is in the space between him and me that it unfolds. I would accept that the sense in which the place of my opponent’s anger is on his face is not the same as that in which, in a moment, tears may come streaming from his eyes or a grimace may harden on his mouth. Yet anger inhabits him and it blossoms on the surface of his pale or purple cheeks, his blood-shot eyes and wheezing voice. (Merleau-Ponty 2004, pp. 83–84)

310  Dan Zahavi and Glenda Satne As argued by many phenomenologists, rather than attempting to bridge the gap between a visible but mindless behavior and an invisible but disembodied mentality, it might be more promising to simply take the embodied and embedded mind seriously and to locate the foundation of interpersonal understanding in a fundamental sensitivity to animacy, agency, and emotional expressivity rather than in a capacity for detached belief-ascriptions. As Merleau-Ponty also puts it, the problem of knowing how I can come to understand the other is far less difficult to solve, if the other is understood primarily as a way of intending and grasping the world that surrounds us, than if she is understood as a radically alien psyche (1964, p. 117). More generally speaking, and this an idea found in Husserl, Stein, Merleau-Ponty, and others, the face-to-face encounter might allow for a form of quasi-perceptual empathic understanding of others that is more direct and immediate than any mindreading involving theoretical inference or imaginative perspective taking (cf. Zahavi 2014). Furthermore, this sort of social understanding is one than can primarily (though not exclusively) be found in second personal exchanges, in which both subjects are aware of being directed to one another in a relationship of mutual address. It is not hard to connect such ideas to findings in developmental psychology. There is currently ample evidence that children have some understanding of other people’s perceptions, attentions, desires, intentions, and emotions long before they are able to understand (false) beliefs. Indeed, long before the child starts to wonder about your specific belief-content, it has interacted with and treated you as a social partner. As Rochat and Striano conclude in an article surveying and summarizing research on the social-cognitive development in infancy: (1) infants manifest an essentially innate sensitivity to social stimuli, (2) there is already an early form of intersubjectivity at play from around two month of age, where the infant has a sense of shared experience and reciprocity with others, and (3) the “echoing of affects, feelings and emotions that takes place in reciprocal interaction between young infants and their caretakers” is a “necessary element to the development of more advanced social cognition, including theory of mind” (Rochat & Striano 1999, p. 8). Excepting some forms of pathology, there is no initial phase in which the infant is confronted with meaningless behavior; rather the infant comes equipped with an innate, automatic, and prereflective ability to tune in and respond to the expressive behavior of others. The ideas just outlined suggest that there are other ways of characterizing our basic social cognitive abilities than the proposals offered by Theory-Theory and Simulation Theory and that it is possible to problematize Tomasello’s claim that dyadic interaction and basic joint intentionality presuppose a number of fairly sophisticated processes (including recursive inferences, self-monitoring, simulations, and metarepresentations). Furthermore, even if by distinguishing two forms of shared intentionality, Tomasello has gone beyond the uniformity thesis, he still remains

Varieties of Shared Intentionality  311 committed to the second assumption insofar as he continues to analyze joint intentionality exclusively in terms of joint action. What we want to explore in the following sections is whether a consideration of other forms of shared intentionality might provide us with an enriched understanding of we-intentionality. We have already said that Tomasello sees the evolution of culture as the milestone of the human species. But what does it mean to belong to a culture? What is it to be part of a temporally and spatially extended we, a we that spans both time and space? 3.  VARIETIES OF WE The question concerning how we become part of a we has been a crucial worry for classical phenomenology. Paradigmatically, one of the central questions raised by Gerda Walther in her dissertation Zur Ontologie der Sozialen Gemeinschaften from 1919 concerns the nature of a social community. What must be in place in order for a plurality of individuals to form a community? What is required in order for an individual to employ the first-person plural and articulate an experience, an action, or a belief as ours, as something we experience, do, or believe? It is neither sufficient nor necessary for the individuals in question to have the same kind of intentional states with the same kind of object. Not only could such similarity obtain by mere happenstance in situations in which the individuals had no awareness or knowledge of each other, but there are obviously also communities whose members pursue very different goals while nevertheless remaining part of one and the same community. To make headway, Walther introduces a more nuanced distinction between different kinds of communities than the one we find in Tomasello. If we start with a case where a number of individuals are aware of and interacting with each other with the same goal in mind, would this be enough to constitute a community? Walther’s reply is negative. Consider, for instance, a situation in which a number of workers who otherwise know nothing about each other are brought together to finish a construction. They interact in order to accomplish the same goal. To some extent they work together, but in the case in question they all consider each other with suspicion or at best indifference (Walther 1923, p. 31). Seen from without, they might be indistinguishable from a community, but they only constitute an instrumental partnership. For the former to obtain, something remains missing. What more is needed? According to Walther, the missing element is an inner bond or connection (innere Verbundenheit), a feeling of togetherness (Gefühl der Zusammengehörigkeit). Only when the latter is present, does a social formation, on her account, become a community (1923, p. 33): We are standing here on the same ground of those theorists [. . .] that consider the essential element of the community to be a ‘feeling of

312  Dan Zahavi and Glenda Satne togetherness’, or an inner unification [innere Einigung]. Every social configuration that exhibits such an inner unification, and only those configurations are, in our opinion, communities. Only in communities can one strictly speak about communal experiences, actions, goals, aspirations, desires, etc. (in contrast to experiences, actions, etc. that may be the same or similar, and that can be present in societal relations). (Walther 1923, p. 33). According to Walther, the direct awareness of and interaction with others allows for a special kind of community, one that Walther labels purely personal communities or life communities (rein personale Gemeinschaften or Lebensgemeinschaften). Some of these communities are not regulated by a shared external object or goal. This holds true, for instance, for friendships, families, and marriages. Even in these cases, the common life is penetrated by a shared meaning or goal, but the goal is the flourishing of the community itself. Walther calls these forms of communities “reflexive communities” (1923, p. 67). But communal life is not restricted to these forms since it cannot be a necessary feature of every community that all its members engage in reciprocal interaction (1923, pp. 66, 68). In fact, people can experience themselves as members of a community, can identify with other members of the same community, and can have group experiences even if they do not live temporally and spatially together. In such cases, shared objects, goals, rituals, conventions, norms, etc. play a crucial role. It is as a result of being attached to such external objects, factors, and features that individuals might come to identify with other individuals whom they might never have met in person, but who are also members of the same community (1923, pp. 49–50). The more the unification of the members is conditioned by the unification with external objects (rather than directly bound to immediate interpersonal contact) the greater the spatiotemporal separation of the members can be (1923, p. 82), and the more replaceable the individual members will be. Walther labels such (institutionalized) communities “objectual communities” (gegenständliche Gemeinschaften) (1923, p. 50). Walther next explores the question of whether members of a community must necessarily realize or recognize this membership. As she points out, there is a difference between identifying with and being united with certain other people, and knowing that one belongs to a particular community. It often happens that one lives together with others in a reciprocal unity without reflecting on this relation. But this can change, for instance, because of intergroup conflict. Thus, as Walther remarks, war can often make people aware of themselves as members of a special community (1923, p. 96). From that moment on, one’s interaction with out-group members might also be explicitly framed and influenced by one’s own group identity. In other cases, however, knowledge about the group precedes knowledge about the individual members. Consider, for instance, someone who converts to a new religion or acquires a new citizenship. Here one might identify with the new

Varieties of Shared Intentionality  313 community before one starts to identify with particular individuals, and one’s relation to these individuals might at first be only as representative group-members rather than as unique individuals (1923, p. 99–100).4 Finally, Walther also discusses the case in which a community is recognized as such not only by its own members, but also by members belonging to another community. When that happens, and especially if the out-group members in question are representatives of, and act in the name of, their own community, one might talk of a higher-order interaction between communities. Walther suggests that the community through this kind of external recognition acquires a new and more objective status (1923, p. 121). So far we have identified a strand in classical phenomenology that, while keeping in mind the important distinction between face-to-face interaction and communal intentionality, goes beyond a simple bipartition of shared intentionality into two kinds and offers a more refined discrimination of the different ways in which individuals can relate to each other. This analysis provides useful tools for thinking about some constitutive features of collective intentionality, namely, inner unification and the feeling of togetherness, as well as about mediated forms of shared intentionality that allow communities to extend in time and space. In order to enrich the analysis even further, let us next consider how others are given to me in the social world, as present or absent, as past or contemporary, as personal or anonymous, and mediated by social functions and roles. We find such an account in the work of Alfred Schutz. In The Phenomenology of the Social World, Schutz faults Weber for failing to offer a proper account of the constitution of social meaning and, more generally, for being indifferent to more fundamental questions in epistemology and theory of meaning. It is this lacuna that Schutz then seeks to overcome by combining Weber’s interpretive sociology with reflections drawn from Husserl’s phenomenology. According to Schutz, one of the more specific shortcomings of Weber’s theory is that it fails to acknowledge the heterogeneity of the social world. He then directly targets what we previously identified as the third weakness in current debates on collective intentionality, namely the tendency to discuss social relations at only a very high degree of abstraction. As Schutz writes, “Far from being homogeneous, the social world is structured in a complex way, and the other subject is given to the social agent (and each of them to an external observer) in different degrees of anonymity, experiential immediacy, and fulfilment” (1967, p. 8; modified translation). More specifically, Schutz (1967, p. 14) distinguishes four different spheres within the social world: the sphere of the “directly experienced social reality” (soziale Umwelt) (1967, p. 142), the “social world of contemporaries” (soziale Mitwelt) (1967, p. 142), the “social world of predecessors” (soziale Vorwelt) (1967, p. 143), and the “social world of successors” (soziale Folgewelt) (1967, p. 143). The realm of directly experienced social reality or, to put it differently, the social surrounding world, is the one in which the social world is open

314  Dan Zahavi and Glenda Satne for direct experience, and within which others are presented as fellow men (Mitmenschen). It would be wrong, however, to restrict social reality to this specific dimension. According to Schutz, we must recognize that there is also a social world of contemporaries (Nebenmenschen) that coexists with the subject and is simultaneous with his duration, although the lack of spatial proximity prevents other subjects’ experiences from being grasped as originally and directly as is possible in the social surrounding world. Furthermore, a subject can also be directed to a world of predecessors (Vorfahren) that existed at some point but does not exist anymore and to a forthcoming world of successors (Nachfahren) that can be apprehended only in a vague and indeterminate manner. According to Schutz, the face-to-face encounter characteristic of the social surrounding world provides for the most fundamental type of interpersonal understanding (Schutz 1967, p. 162). It is at the basis of what he terms the “we-relationship” or “living social relationship,” which is the central concept in his account of experiential sharing (cf. Zahavi 2015). But despite its central importance, we shouldn’t forget that I am also able to understand those whom I have previously encountered face-to-face, but who now live abroad, or those of whose existence I know, not as concrete individuals, but as points in the social space defined by certain roles and functions, say, tax officials or train conductors. In contrast to the situation in which I am directly aware of the living presence of the other’s consciousness as it is manifested in expressive movements or expressive acts, my understanding of my contemporaries is always general in form, is always shaped and framed by structures of typicality (Schutz 1967, pp. 181, 184). When I understand a contemporary, I do not consider him as a unique person. Rather, I conceive of him as an instantiation of a type and leave individual characteristics and changes out of my consideration. Consider, for instance, the kind of social understanding that occurs when I mail a letter. When doing so, my action is guided by assumptions I make regarding some of my contemporaries, namely the mail carriers. I assume that they will read the address and send the letter to its recipient. I do not know them personally and I do not think of them as particular individuals, but by behaving the way I do, I relate to them as ideal types, as bearers of certain functions. As Schutz puts it, when I am they-oriented I have ‘types’ for partners (Schutz 1967, p. 185). And of course, for this social process to work, the mail carriers have to relate to me as well, not as a particular individual, but as a typical customer (Schutz 1967, p.  202). Typification (and stereotyping) consequently facilitates everyday predictability. In ordinary life, we move between Umwelt and Mitwelt constantly and, as Schutz points out, the change from one to the other presents no problem. This is so because we always interpret our own behavior and that of the other within contexts of meaning that transcend the here and now. In that sense, a narrow concern with the question of whether our relationship is

Varieties of Shared Intentionality  315 direct or indirect is a somewhat academic exercise (Schutz 1967, p. 178). This is even more so given that the use of ideal types is not limited to the world of contemporaries (or the world of our predecessors or successors). The ideal types we acquire become part of our stock of knowledge and start to influence our face-to-face interactions as well, that is, they come to serve as interpretive schemes even in the world of direct social experience (Schutz 1967, p. 185). 4.  UNIVERSALITY AND OBJECTIVITY Schutz complements Walther’s analysis of we-intentionality by ascribing a central role to the process of typification.5 Nevertheless, the description given so far still seems incomplete insofar as it fails to account for a more universal conception of membership, one that spans temporal, spatial, and cultural differences. Apart from being able to relate to others distant in time and place or pertaining to different cultures, we are also able to engage with one another as members of a universal class: the class of rational beings. This possibility is exhibited in some of our joint activities, such as theoretical endeavors. More general speaking, we have a capacity to take judgments to be true from a view from nowhere, i.e., according to how things are independently of what anyone thinks of them (McDowell 1994; Satne 2015). As we explained above, one of Tomasello’s central claims is the idea that the human capacity to go beyond the here and now of face-to-face interactions and acquire a conception of ourselves as members of a universal class is at the basis of the uniquely human capacity for objective thinking. This is an idea that in turn can be traced back to Donald Davidson’s and Robert Brandom’s accounts of the social origin of objective meaning. Let us in the following pursue this philosophical link further. Doing so will allow us to unearth a nexus of convergence between classical phenomenology and contemporary analytic philosophy. Davidson and Brandom both think that having a notion of perspective goes hand in hand with the very possibility of objective thinking and that it ultimately amounts to the same capacity. For Davidson, objective thinking is tied to the capacity to conceive of a different perspective that refers to the same world. The notion of objective truth is consequently available only to beings that can grasp that others, similar to themselves, are perceiving the same world from a different perspective. For Davidson, objectivity might be said to be constituted in a triangle, where one apex is oneself, the second apex another creature similar to oneself, and the third an object located in what becomes a common space (2001, p. 121). Being part of such a triangle allows one, at least implicitly, to grasp the difference between belief and objective truth, i.e., to acquire the concept of belief. The notion of belief entails that my take on things can be incorrect and thus false with respect to how things stand in the world. Importantly,

316  Dan Zahavi and Glenda Satne The only way of knowing that the second apex of the triangle—the second creature or person—is reacting to the same object as oneself is to know that the other person has the same object in mind. But then the second person must also know that the first person constitutes an apex of the same triangle another apex of which the second person occupies. For two people to know of each other that they are so related, that their thoughts are so related, requires that they be in communication. Each of them must speak to the other and be understood by the other. They don’t [. . .] have to mean the same things by the same words, but they must be interpreters of each other (2001, p. 121). Only in the context of communication between two interpreters and a common world are language and objective thought possible. Importantly, understanding the other and grasping the world as an objective common realm go hand in hand. It is worth noticing that triangulation does not purport to provide a history of the emergence of social cognition or of the evolution of the capacity for objective thinking, but that it is a transcendental picture of the mutual dependence between interaction, meaning and objectivity.6 Understanding, according to Davidson, is not dependent on “the existence of any form of preexistent, determined, ‘internalised’ agreement” (Malpas 2011, p. 261) whether in the form of a common language, shared conventions, or implicit norms. Rather, the possibility of objective thought presupposes a relation of mutual address in which each participant is an interpreter of the other and both know this to be so. Hence, the very capacity to think depends on the capacity to engage in an interpretative relation with someone else.7 Interpretation is to be understood as a hermeneutical holistic process by means of which an utterance or action is located within a wider context. Such a context includes the situation in which interpreter and interpretee are mutually engaged, as well as an interpretative background constituted by an attributed set of other interrelated intentional states (including beliefs, desires, intentions, and so on) as well as rational principles that govern the relations among them and serve to explain and predict the other’s behavior. Interpretation is thus not mainly directed to linguistic utterances but rather to the behavior of another as a whole. If interpretation does not depend on any shared common ground, there is no difference between understanding a fellow of the same community and understanding an alien community (see Brandom 2010, pp. 33–34): they both employ the same fundamental tools. This is why Davidson speaks of “radical interpretation,” echoing Quine’s “radical translation” (Quine 1960). This means that interpretation works across cultures and is not dependent on the interpreter and the interpretee being acculturated or belonging to the same linguistic and cultural group. Rather, interpretation requires an understanding of the other as a member of the community of rational beings. It is only because I take the other to be as rational as I am that I can make sense of her conduct in the same terms I would make sense

Varieties of Shared Intentionality  317 of mine and anybody else’s under the same conditions. Hence, the very possibility of interpretation entails an understanding of the other and of myself as members of a universal class, the class of rational beings.8 Brandom, following Davidson, claims that providing an elucidation of objective thinking requires adopting the point of view of an interpreter (Brandom 2010, pp. 33–34): we cannot make sense of objective meanings without acknowledging the perspectival character of meaning-attribution within the social practices of human beings. This is what Brandom (1994, pp. 37–42, esp. 39) calls the I-Thou relation, a dyadic relation of interpretation that holds between communicating subjects. In fleshing out the idea of interpretation, Brandom gives a further characterization of what it means to be a member of a particular community. When understanding others as members of particular communities, say Hindus, Chinese, or Europeans, we are adopting a particular interpretational stance. Any concrete cultural or linguistic community is thus to be understood through the lens of the interpretation we can provide of its members’ attitudes and behavior. Whether one sees oneself as a member or an outsider of these particular communities is also a matter of the interpretations we can provide of ourselves and others. Furthermore, it is only within the context of these interpretations that the notion of objective truth can get a grip on our thinking. Against this interpretative background, the very notion of correction is historically relativized. Whether or not an interpretation is correct is not a matter of a simple correspondence between the interpretation and the world, but depends on how the interpretation fits the historical context and the wider interpretational framework. The question of whether a certain claim is correct then becomes a question of whether it coheres with other interpretations given by other interpreters throughout the history of the use of a given concept. This can be illustrated by concepts like “weight,” the meaning of which has progressively changed, from the Aristotelian theory of natural loci to contemporary physics. This, however, is not supposed to imply a relapse into some form of genealogical reduction of norms to purely factual and hence contingent descriptions of particular customs, dissolving the normative force of true claims to non-normative social facts about how people interpret meanings. The objective meaning of a concept is not just a matter of what the community takes it to be. Precisely to overcome this peril, Brandom (2013) invokes what he calls, following Hegel, a hermeneutics of magnanimity—a hermeneutic based on trust for what previous interpretations have put forward as our tradition and forgiveness for what they may have missed. The hermeneutics of magnanimity conceives of the contingent historical practices as the way in which a community progressively determines and shapes the content of rational norms (Brandom, 2013, p. 15). In our prior example, for instance, the contingent history in which different theories of the physics of the bodies developed is understood as the way in which the quest for truth is carried on. Interpretation is then the indefinite

318  Dan Zahavi and Glenda Satne historical process of conceiving the content of norms in terms of more and more determined and integrated wholes in the pursuit of truth. While difference and relativism amongst diverse communities is acknowledged, the very concept of interpretation pushes our thinking in the direction of an always more inclusive and encompassing community of interpreters. The need to reach an encompassing and widely acknowledged interpretation is the way in which objectivity governs the process. While Davidson’s account of interpretation goes hand in hand with a notion of universal community (Davidson 1984, p. 183ff; 2001, p. 137ff), this concept, much in line with the third assumption listed at the outset, remains very abstract and Davidson offers no analysis of its concrete emergence or development. Brandom complements this analysis by conceiving of the universal we as a historical and concrete community of interpreters that evolves and develops through time. What seems to be missing from his account, however, is an analysis of how one comes to be conscious of oneself as a member of such a universal we. How does one come to see oneself as one among others, as a mere somebody, i.e., as part of and included in the notion of everybody? Such an analysis, however, is at the very center of Husserl’s account of the link between intersubjectivity and objectivity.9 Husserl agrees with Brandom and Davidson that objectivity in the sense of ‘valid for everybody’ requires an analysis of intersubjectivity (Husserl 1974, p. 243). More specifically, Husserl argues that my apprehension of objects as real and objective is mediated by and depends upon my encounter with other world-related subjects. Much like Davidson, Husserl claims that it is by experiencing that others experience the same objects as I do that I come to experience objects as being more than merely objects-for-me. Ultimately, Husserl isn’t simply claiming that the sense and validity of the categories objective and real are intersubjectively constituted. On his account, the same holds true for the categories subjective and appearance.10 When I realize that the intentional object of my apprehension can also be experienced by others, I come to understand that the same object can appear for different subjects (Husserl 1973a, p. 9) and that there is a difference between the object as such and the way it appears to me. It consequently only makes sense to designate something as a mere appearance, as merely subjective, the moment I have experienced other subjects and through them acquired the concept of intersubjective validity (Husserl 1962, p. 453, 1973a, pp. 382, 388–389, 420–421). Davidson’s and Brandom’s accounts of triangulation conceive of it as the very condition of possibility of objective thought. Husserl agrees, but complements such analysis by exploring in greater detail the self-other relation that is at stake in the process. In his description of the different ways in which I relate to others, Husserl highlights the importance of a special kind of other-experience, namely the one where I experience the other as experiencing myself. My experience of being an object-for-another, my other-mediated self-apprehension, is on his view of decisive importance for the constitution of objectivity. When I realize that I can be an alter ego for

Varieties of Shared Intentionality  319 the other, just as she can be one for me, when I realize that the other conceives of me as an other, just as I conceive of her as a self, The difference between oneself and the foreign I vanishes; the other apprehends me as foreign, just as I grasp him as foreign for me, and he himself is a “self,” etc. Parity thus ensues: a multiplicity of feeling, willing I’s that are alike in kind and each independent in the same sense. (Husserl 1973a, pp. 243–244; cf. Husserl 1973c, p. 635) Some differences between Davidson and Brandom’s views, on the one hand, and Husserl’s on the other, remain. Both Davidson and Brandom conceive of triangulation as a process involving communication and hence linguistic capacities. Husserl, by contrast, focuses on pre-linguistic forms of intersubjectivity. As we pointed out in section 2, this is also the approach that Tomasello (along with most developmental psychologists) adopts.11 Furthermore, for Davidson and Brandom, it is the universal applicability of the notion of interpretation that allows us to move from a particular encounter with another to a universal concept of truth. This notion remains somewhat abstract even in Brandom’s account, since no description of the concrete experience of being part of a we is given. Husserl’s account, by contrast, can be seen as providing precisely such an analysis. According to Husserl, it is through the encounter with another that I come to realize that I am only one among many, and that my perspective on the world is only one among several (Husserl 1974, p. 245, 1973c, p. 645, 1950, p. 157). It is when I discover myself to be one among others that I can come to apprehend myself as “a member of a community of I’s in the sense of a ‘we’community” (Husserl 1973b, p. 468). As Husserl formulates it in Ideen II: In my own case, I arrive at the apprehension of man (in the Sense of spirit) by way of a comprehension of others, i.e., insofar as I comprehend them as centers not only for the rest of the surrounding world but also for my Body, which is for them an Object of the surrounding world. It is precisely thereby that I comprehend them as apprehending me similar to the way I apprehend them, thus as apprehending me as a social man [. . .]. The comprehensive representation others have, or can have, of me is of service to me as regards the apprehension of myself as a social ‘man,’ hence the apprehension of myself totally different from the way I grasp myself in direct inspection. By means of this apprehension, with its complicated structure, I fit myself into the family of man, or, rather, I create the constitutive possibility for the unity of this ‘family.’ It is only now that I am, in the proper sense, an Ego over against an other and can then say ‘we.’ (Husserl 1952, p. 242) Another striking similarity between Husserl’s, Brandom’s, and Davidson’s views is the emphasis on the importance of disagreement between world‑experiencing subjects in the grounding of objectivity. As we saw, the

320  Dan Zahavi and Glenda Satne notion of disagreement and perspective difference plays a key role in both Davidson’s and Brandom’s analyses of the conditions of emergence of objective thought. Nevertheless, as with the universal we, their analyses remain abstract: they emphasize the need for perspective differences but do not provide any account of how such differences are experienced by individuals nor of how the notion of disagreement itself is affected by such perspective differences. Husserl takes an extra step towards concreteness in his analysis of the role of normality. Already in Ideen II, Husserl pointed to the fact that there, next to the tendencies originating from other persons, also exist indeterminate general demands made by custom and tradition: ‘One’ judges thus, ‘one’ holds the fork in such and such a way, etc. (Husserl 1952, p. 269). I learn what counts as normal from others—and indeed, initially and for the most part from those closest to me, hence from those I grow up with, those who teach me, and those belonging to the most intimate sphere of my life (Husserl 1973c, pp. 428–429, 569, 602–604). And it is in this way that I participate in a communal tradition. In this vein, Husserl stresses, as Brandom does, the fact that each new generation inherits what was constituted through the labor of previous generations, and then by reshaping this heritage makes its own contribution to the constitution of the communal unity of the tradition. Husserl has a term for this: he refers to normal life as generative life and states that every (normal) human being is historical by virtue of being constituted as a member of a historically enduring community (Husserl 1973c, pp. 138–139, 431). Community and communalization inherently point beyond themselves to the open endlessness of the chain of generations, and this in turn points to the historicity of human existence itself. Ultimately, Husserl would consider the subject’s birth into a living tradition to have constitutive implications. It is not merely the case that I live in a world, which as a correlate of normality is permeated by references to others, and which others have already furnished with meaning, or that I understand the world (and myself) through a traditional, handed down, linguistic conventionality. The very meaning that the world has for me is such that it has its origin outside of me, in a historical past. Furthermore, Husserl offers important tools to qualify the notion of disagreement that plays a key role in the understanding of objectivity. He distinguishes different levels of normality. First of all, we speak of normality when we are dealing with a mature, healthy, and rational person. Here the abnormal will be the infant, the blind, or the mentally ill. Secondly, we speak of normality when it concerns our own (sub)culture or homeworld, whereas abnormality is attributed to the foreigner, who, however, if certain conditions are fulfilled, can be apprehended as a member of a foreign normality. It is precisely in the latter context that the disagreement gains a vital constitutive significance, since the experience of discrepancy between normal subjects (including the experience of a plurality of normalities, each of which has its own notion of what counts as true) can motivate us

Varieties of Shared Intentionality  321 to search for a truth that is valid for us all (Husserl 1954, p. 324). That is to say, “if we set up the goal of a truth . . . which is unconditionally valid for all subjects, beginning with that on which normal Europeans, normal Hindus, Chinese, etc., agree in spite of all relativity,” we are on the way to an objective science (Husserl 1954, p. 142). It consequently turns out to be necessary to differentiate between (1) ‘normal’ objectivity, which is correlated with a limited intersubjectivity (a community of normal subjects) and (2) ‘rigorous’ objectivity, which is correlated with the unlimited totality of all rational subjects (Husserl 1973b, p. 111). Husserl accordingly believed a correlation to exist between different levels of normality and different levels of objectivity (Husserl 1973c, p. 155). Even absolute objective being and truth is correlated with a subject‑dependent normality: the normality of rational subjects (Husserl 1973c, pp. 35–36). Ever new generations cooperate in transcendentally building up the structures of validity pertaining to the objective world, which is precisely a world handed down in tradition (Husserl 1973c, p. 463). There is, as Husserl writes, no fixed world; rather, it is what it is for us only in the relativity of normalities and abnormalities (Husserl 1973c, pp. 212, 381, Husserl 1954, p. 270). 5.  CONCLUDING REMARKS At the beginning of this contribution, we referred to three underlying assumptions shared—even if only implicitly—by most contemporary accounts of collective intentionality. The assumptions were the following. First, a commitment to a uniformity claim about collective intentionality that conceives of all its forms as ultimately reducible to one. Second, a privileging of joint action vis-à-vis other collective forms of intentionality, such as shared emotions or other distinct forms of relating to others proper to communal and institutional forms of life. Third, the dominance of abstract models of collectivity, leaving aside the historical, temporal, and spatial contexts of such social formations. Throughout this contribution, we have shown how the debate can be enriched by including considerations from a number of different sources, including developmental and evolutionary psychology, social approaches to mind and language, and classical phenomenological accounts of shared intentionality. Furthermore, we hope to have made a good case for promoting further dialogue between the analytic-minded literature (represented by the work of Tomasello, Brandom, and Davidson) and authors that belong to the classical phenomenological tradition, including Walther, Schutz, and Husserl. In analyzing all these different contributions to the topic of collective intentionality, it becomes apparent not only that there are important overlaps between their approaches, but also that there are obvious ways in which they can complement each other.

322  Dan Zahavi and Glenda Satne When it comes to the relation between analytic philosophy and phenomenology, there is consequently no reason to accept either Dummett’s claim that “we have reached a point at which it’s as if we’re working in different subjects” and that “it is no use now shouting across the gulf” (1996, p. 193), or Searle’s assertion that phenomenologists “do not have much to contribute to the topics of the logical structure of intentionality or the logical structure of social and institutional reality” (Searle 1999, p. 11).

NOTES  1 Thus, according to Tomasello, what made the difference for the evolution of human-specific forms of intentionality was the need for cooperation that resulted in more and more complex forms of cooperative sociality: “Human beings, in contrast, are all about (or mostly about) cooperation. Human social life is much more cooperatively organized than that of other primates, and so, in the current hypothesis, it was these more complex forms of cooperative sociality that acted as the selective pressures that transformed great ape individual intentionality and thinking into human shared intentionality and thinking” (Tomasello 2014, p. 31).   2 More specifically, Tomasello lists the following three key components: “(I) the ability to cognitively represent experiences to oneself ‘off-line’; (2) the ability to simulate or make inferences transforming these representations causally, intentionally, and/or logically; and (3) the ability to self-monitor and evaluate how these simulated experiences might lead to specific behavioral outcomes” (Tomasello 2014, p. 4).  3 Another problem for such an account is an ‘essential tension’ found in its account of the objective content of thought and language (Hutto and Satne 2015). Such tension consists in postulating a role for social interaction in an account of thinking and at the same time assuming that thinking is required for social interaction to happen. The problem is due to the fact that the theory assumes that in order to engage in joint intentionality one needs to already entertain representations of other people’s mental states as well as of external states of affairs. For a more in-depth treatment of this issue, which also discusses how the essential tension affects the approaches of Tomasello and others, see Hutto and Satne 2015.   4 Walther’s distinction bears a clear affinity with what Hogg and Abrams much later have termed personal and social attraction. Whereas personal attraction is an interindividual attraction that is firmly rooted in close personal relationships and whose object is a unique individual, social attraction is a form of intragroup attraction that is grounded in group membership: “It is an attraction to an ingroup stereotype and hence to any and all individuals who are perceived to be stereotypic or prototypic of the group” (Hogg and Abrams 1998, p. 94).   5 For a comparison of other aspects of Walther’s and Schutz’s theories, see León and Zahavi 2015.   6 For a thorough defence of this reading of Davidson see Malpas (1997, 1992). Empirical conditions on how to occupy an apex in the triangle have been explored by different authors (e.g., Gomila 2010; for criticism, Andrews 2002), thereby complementing Davidson’s remarks on the conditions of possibility of

Varieties of Shared Intentionality  323 objective thinking with theories of development and evolution of such capacities. See fn. 11 below.   7 Tomasello (2014, pp. 149–151) differentiates his view from Davidson’s precisely on this point, since he claims that language capacities are not required in order to jointly intend with another. One should make clear, however, that Davidson, as said, does not think that a common language is required or that the interactants already need to know a language. What is required is a capacity, the capacity for interpretation, and it is this capacity that is also required for language.   8 From a technical point of view, Davidson formulates this requirement in terms of a set of principles that govern and make possible interpretation: the principle of charity (I take most of the beliefs of the other to be true), coherence (I take others not to have contradictory beliefs), etc. These details, however, are not important for our present purposes.   9 For an in-depth discussion, see Zahavi (2001). 10 In strikingly similar terms, Davidson claims that it is only by grasping the very notion of perspective that the distinction between how things appear to me and how things are emerges (Davidson 2001, p. 105). 11 It must be emphasized that Davidson and Brandom are not out to offer either a psychological account or an evolutionary or developmental story of how thinking emerges (for further discussion, see Weiss and Wanderer 2010). Rather, they are providing the necessary conditions for the obtaining of objective thinking. Objectivity is possible only in the context of an intersubjective engagement with another similar to oneself with whom one is in communication. Their account of triangulation can, however, be complemented both by Tomasello’s two-step theory of shared intentionality that targets the evolution of the capacity for objective thinking and by a phenomenological account of the pre-predicative and pre-linguistic strata of intersubjectivity.

REFERENCES Andrews, K. 2002. Interpreting Autism: A Critique of Davidson on Thought and Language. Philosophical Psychology 15 (3): 317–332. Brandom, R. 1994. Making It Explicit. Cambridge: Harvard University Press. Brandom, R. 2013. Reason, Genealogy, and the Hermeneutics of Magnanimity (Online). Available http://www.pitt.edu/~brandom/downloads/RGHM%20%201211-21%20a.docx. Accessed January 23, 2015. Bratman, M. 1992. Shared Cooperative Activity. Philosophical Review, 101 (2): 327–341. Currie, G. 2011. Empathy for Objects. In A. Coplan and P. Goldie, eds., Empathy: Philosophical and Psychological Perspectives. Oxford: Oxford University Press, 82–95. Davidson, D. 2001. Subjective, Intersubjective, Objective; Philosophical Essays. Oxford: Oxford University Press. Davidson, D. 2005. Truth, Language, and History: Philosophical Essays. Oxford: Oxford University Press. Dummett, M. 1996. Origins of Analytical Philosophy. Cambridge, Harvard University Press. Gallagher, S. and D. Zahavi. 2012. The Phenomenological Mind. 2nd ed. London: Routledge. Gilbert, M. 2013. Joint Commitment: How We Make the Social World. New York: Oxford University Press.

324  Dan Zahavi and Glenda Satne Hogg, M. A. and D. Abrams. 1998. Social identifications. London: Routledge. Husserl, E. 1950. Cartesianische Meditationen und Pariser Vorträge. Husserliana 1. Den Haag: Martinus Nijhoff. Husserl, E. 1952. Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie. Zweites Buch. Phänomenologische Untersuchungen zur Konstitution. Husserliana 4. Den Haag: Martinus Nijhoff. Husserl, E. 1954. Die Krisis der europäischen Wissenschaften und die transzendentale Phänomenologie. Eine Einleitung in die phänomenologische Philosophie. Husserliana 6. Den Haag: Martinus Nijhoff. Husserl, E. 1962. Phänomenologische Psychologie. Husserliana 9. Den Haag: Martinus Nijhoff. Husserl, E. 1973a. Zur Phänomenologie der Intersubjektivität. Texte aus dem Nachlass I: 1905–1920. Husserliana 13. Den Haag: Martinus Nijhoff. Husserl, E. 1973b. Zur Phänomenologie der Intersubjektivität. Texte aus dem Nachlass II: 1921–1928. Husserliana 14. Den Haag: Martinus Nijhoff. Husserl, E. 1973c. Zur Phänomenologie der Intersubjektivität. Texte aus dem Nachlass III: 1929–1935. Husserliana 15. Den Haag: Martinus Nijhoff. Husserl, E. 1974. Formale und Transzendentale Logik. Husserliana 17. Den Haag: Martinus Nijhoff. Hutto, D. and Satne, G. Forthcoming. The Natural Origins of Content. Philosophia. Philosophical Quarterly of Israel. Lavelle, S. 2012. Theory-theory and the Direct Perception of Mental States. Review of Philosophy and Psychology 3: 213–230. León, F. and D. Zahavi. in press. Phenomenology of Experiential Sharing: The Contribution of Schutz and Walther. In A. Salice and H.-B. Schmid eds., Social Reality: The Phenomenological Approach. Springer. Malpas, J. 1992. Donald Davidson and the Mirror of Meaning Holism, Truth, Interpretation. Cambridge: Cambridge University Press. Malpas, J. 1997. The Transcendental Circle. Australasian Journal of Philosophy 75 (1): 1–20. Malpas, J. 2011. What Is Common to All: Davidson on Agreement and Understanding. In Jeff Malpas, ed., Dialogues with Davidson: Acting, Interpreting, Understanding, Cambridge, MA: MIT Press, 259–280. McDowell, J. 1994. Mind and World. Harvard: Harvard University Press. Merleau-Ponty, M. 1964. The Primacy of Perception. Trans. W. Cobb. Evanston: Northwestern University Press. Mitchell, J. P. 2008. Contributions of Functional Neuroimaging to the Study of Social Cognition. Current directions in Psychological Science 17/2: 142–146. Nichols, S. and S. Stich. 2003. Mindreading: An Integrated Account of Pretence, Self-Awareness, and Understanding Other Minds. Oxford: Oxford University Press. Quine, W. V. O. 1960. Word and Object. Cambridge: MIT Press. Rochat, P. and T. Striano. 1999. Social Cognitive Development in the First Year. In P. Rochat, ed., Early Social Cognition. Mahwah, NJ: Lawrence Erlbaum Associates, 3–35. Satne, G. 2014. Brandom and McDowell: Hermeneutics and Normativity. In H.-H. Gander and J. Malpas, eds., The Routledge Companion to Hermeneutics. London: Routledge, 236–238. Satne, G. in press. What Binds Us Together. Normativity and the Second Person. Philosophical Topics. Schmid, H. B. 2009. Plural Action: Essays in Philosophy and Social Science. Dordrecht: Springer. Schutz, A. 1967. The Phenomenology of the Social World. Evanston, IL: Northwestern University Press. Searle, J. 1995. The Construction of Social Reality. Cambridge: The Free Press.

Varieties of Shared Intentionality  325 Searle, J. 1999. Neither Phenomenological Description nor Rational Reconstruction: Reply to Dreyfus. http://socrates.berkeley.edu/~jsearle/articles.html. Accessed July 2014. Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge: Harvard University Press. Tomasello, M. 2014. A Natural History of Human Thinking. Cambridge: Harvard University Press. Walther, G. 1923. Zur Ontologie der sozialen Gemeinschaften. In E. Husserl, ed., Jahrbuch für Philosophie und phänomenologische Forschung VI. Halle: Niemeyer, 1–158. Weiss, B. and J. Wanderer. 2010. Reading Brandom. On Making it Explicit. London: Routledge. Zahavi, D. 2001. Husserl and Transcendental Intersubjectivity. Athens: Ohio University Press. Zahavi, D. 2014. Self and Other: Exploring Subjectivity, Empathy, and Shame. Oxford: Oxford University Press. Zahavi, D. 2015. You, Me and We: The Sharing of Emotional Experiences. Journal of Consciousness Studies 22(1–2): 84–101.

This page intentionally left blank

Contributors

Jeffrey A. Bell is Professor of Philosophy at Southeastern Louisiana University, USA. Lee Braver is Associate Professor of Philosophy at the University of South Florida, USA. James Conant is Chester B. Tripp Professor of Humanities, Professor of Philosophy, and Professor in the College at the University of Chicago, USA. Andrew Cutrofello is Professor of Philosophy, Loyola University Chicago, USA. Richard Eldridge is Charles and Harriett Cox McDowell Professor of Philosophy at Swarthmore College, USA. Paul M. Livingston is Associate Professor of Philosophy at the University of New Mexico, USA. Tamsin Lorraine is Professor of Philosophy at Swarthmore College, USA. John McCumber is Professor of Germanic Languages at the University of California, Los Angeles, USA. Catarina Dutilh Novaes is Assistant Professor and Rosalind Franklin Fellow in Theoretical Philosophy at the University of Groningen, The Netherlands. Graham Priest is Boyce Gibson Professor of Philosophy at the University of Melbourne, Australia, and Distinguished Professor of Philosophy at CUNY Graduate Center, USA. Carol Rovane is Professor of Philosophy at Columbia University, USA. Glenda Satne is Research Fellow at the Center for Subjectivity Research (2012–2015) and Assistant Professor at Alberto Hurtado University, Chile, 2015 onwards. David W. Smith is Professor of Philosophy at the University of California at Irvine, USA. Samuel C. Wheeler III is Professor of Philosophy at the University of Connecticut, USA. Dan Zahavi is Professor of Philosophy and Director of the Center for Subjectivity Research at the University of Copenhagen, Denmark.

This page intentionally left blank

Index

Abelard, P. 83 action(s) 10, 29, 68, 125, 127, 163 – 164, 182, 195, 202 – 203, 207, 212, 235 – 236, 238, 251, 278 – 280, 283, 286; collective 305 – 306, 308, 311 – 316, 321 – 322; ordinary vs. revolutionary 287 – 290, 292 – 297, 299 – 302 Agamben, G. 246 alienation 263, 281, 284 Al-Saji, A. 68 alternativeness 261 – 262, 265 – 268, 278 analysis 4 – 6, 9 – 10, 11 – 12, 19, 43, 75 – 77, 83, 85, 89 – 91, 99 – 101, 102, 125, 225; conceptual 1, 22, 59 – 62, 64, 70 – 71, 104; genealogical 77, 79, 83, 92 – 94, 96, 99 – 101; historical 75, 77 – 78, 80, 82, 89, 91, 95, 101, 161, 164; linguistic 4 – 5, 22, 239, 246; logical 21, 41, 215; paradox of 59 – 60; phenomenological 5, 111 – 112, 175, 318 – 319; transcendental 246; truth-conditional 117 analytic philosophy 1, 4, 8 – 9, 17 – 20, 22 – 55, 59, 75 – 79, 82, 91, 101 – 102, 130, 132, 135, 151, 161 – 162, 164, 193, 213 – 215, 217, 261 – 263, 265 – 266, 272 – 273, 277 – 278, 287, 306, 315, 322 Angst/anxiety 222 – 223, 226, 228, 242 Anscombe, G.E.M. 41, 48, 51 – 52 anti-realism 6 – 7, 130, 151, 161, 163 – 164, 238, 266, 273 Aquinas, T. 38, 83

archaeology 75, 84, 89 – 92,  104; of knowledge 89 – 90; of the subject 84 Arendt, H. 10, 287 – 291, 296, 298, 301; relation to Badiou 292 – 295, 300 Aristophanes 93 Aristotle 24, 38 – 39, 41 – 42, 72, 83, 85, 87 – 88, 98 – 99, 105, 178 – 179, 187 – 188, 196, 211, 230 – 231, 253 Armstrong, D. 131, 135 – 138 attention 63, 65, 67, 70, 124, 203, 214, 216, 218, 267, 274, 292, 306, 307, 310 aufhebung 174, 178, 193 Augustine, St. 83, 223 – 225 Austin, J.L. 5, 9, 22, 24, 59, 62 – 66, 70 – 72, 163 Bachelard, G. 82 Bacon, F. 69, 73 Badiou, A. 11, 12, 240, 287, 290 – 295, 297, 301 – 302 Bates, S. 65 Beiser, F. 103 Beistegui, M. de 73 Benardete, J. 144 – 145 Bergmann, G. 4, 11 Bergson, H. 31 – 32, 68, 73 Berkeley, G. 20, 273 boredom 226 Bradley, F.H. 252 Brandom, R. 12, 26, 103, 315 – 321, 323 Bratman, M. 305 – 306, 308 Broad, C.D. 195, 217 Buridan, J. 99

330  Index Cameron, R. 131, 138, 140 – 142, 145 Canguilhem, G. 82 – 85, 90, 93, 99, 101, 103 – 104 Cantor, G. 125, 239, 291, 295, 301 Carnap, R. 19, 21, 24, 27, 103, 115, 228, 243 Casey, E.S. 197 – 203, 205 – 206, 211 – 212 Cavell, S. 31, 65, 72 Chalmers, D. 11, 239 Chisholm, R. 287 Christianity 80 – 81, 103 cinema 66, 68, 73 Cohen, P. 291 Cohen, T. 65 community 10, 31, 33 – 36, 87, 153, 212, 263, 273, 276, 281, 284 – 285, 311 – 313, 316 – 321 Conant, J. 151, 243 conceptual schemes 24, 152 – 154, 156, 162, 164, 166, 262, 265, 268, 272, 274 – 275, 286 Condorcet, Marquis de 294 continental philosophy (or tradition) 1, 4, 9, 17 – 18, 56, 102, 129 – 130, 146, 214, 222, 263, 281 contingency 81, 84, 86 – 87, 90, 92 – 93, 95, 98, 105, 240, 245 Craig, E. 75 – 77, 89, 92 – 97, 105–106 Darwin, C. 89, 94, 105 Davidson, D. 6, 8, 27, 112, 172 – 173, 175, 188, 246, 262, 265, 274 – 275, 286 – 287, 315 – 323; on metaphor 178, 181 – 183; relation to Derrida 183 – 187 deconstruction 77, 185 Deleuze, G. 61, 72 – 73, 130 – 131, 149, 218, 287, 302; on concepts 66 – 71; on differencemakers 142 – 146; Difference and Repetition 143 – 144 on multiplicity 143 – 144; Francis Bacon: The Logic of Sensation 68 – 69; The Logic of Sense 5 Derrida, J. 8, 130, 146, 172 – 181, 188 – 189, 211; relation to Davidson 183 – 187; “White Mythology” 172 – 181 Descartes, R. 83, 85, 181, 233, 237, 274 – 275 Dewey, J. 26 dialectics 88, 233, 293

Drummond, J. 124 Dummett, M.A.E. 6, 11, 56, 163 – 165, 167 – 168, 322 emotion 70, 200, 206, 305, 306, 310, 321 empathy 310 empiricism 19 – 20, 41, 52, 274 essentialism 61 – 62 ethics 7, 30, 93 – 94, 163, 223 – 225 events 66 – 67, 69 – 70, 72 – 73, 80, 86 – 87, 90, 156, 159, 162, 168, 207, 215 – 216, 218, 236; event of Ereignis 230; ordinary vs. revolutionary 10, 287, 290 – 302 existentialism 1, 223 facts 50, 60, 64, 92, 93, 95, 112, 115, 118, 134, 137 – 139, 141 – 142, 162, 172, 182 – 183, 224 – 225, 227 – 228, 232, 234 – 235, 240, 245, 266, 277, 279; institutional 92; normative 141; social 317 Fodor, J. 27, 95 Føllesdal, D. 180 forms of life 10, 68, 89, 91, 97, 104, 241, 246, 276, 280, 283, 321 Foucault, M. 7, 9, 12, 75 – 77, 82 – 83, 89 – 94, 96 – 98, 101, 104 – 106, 151 – 152, 155 – 169 Frege, G. 1, 9, 36 – 37, 40, 43 – 44, 60, 71, 112 – 114, 180, 236, 244; Begriffsschrift 112, 126; conception of sense [Sinn] 60, 112 – 114; “On Sense and Reference” 44 Geach, P. 41, 88 genealogy 7, 75 – 106, 152, 185, 189; Derrida’s 185; Nietzschean 77, 79 – 81, 87, 90 – 92, 97, 101, 103, 104, 106; Foucault’s 90, 152 Geuss, R. 79 – 81, 87, 103 Gilbert, M. 305 – 306 gluons 249, 251 – 253, 255 – 256 Gödel, K. 3, 115, 291, 295 Greene, G. 205 – 209; The Honorary Consul 205 – 206 Grice, P. 24, 27, 30, 37 Griffin, J. 51 – 52 Habermas, J. 40, 102, 159, 163, 263, 276 – 277, 281 Hacking, I. 75, 77, 82, 103, 105

Index  331 Harman, G. 214 Hegel, G.W.F. 1, 6, 12, 25 – 26, 31 – 32, 37 42, 57, 80 – 81, 102 – 103, 152, 154, 168, 173 – 174, 178 – 180, 183 – 84, 215, 218 – 219, 233, 239, 244 – 245, 263, 274, 276, 280 – 283, 293, 302, 317 Hegelian/Hegelianism 6, 25, 31 – 32, 41 – 42, 57, 86, 215 – 216, 276 – 277 Heidegger, M. 9, 103, 121, 126, 130, 146, 158, 197, 209 – 210, 218, 238 – 246, 256 – 257, 287, 295; question of being 226, 229 – 230, 249 – 251, 253; on Wittgenstein 229 – 237; Wittgenstein on 223 – 228 hermeneutics 316 – 317 Hilbert, D. 88, 115, 126 Hintikka, J. 116, 125 – 126 historicism 61, 80 – 82, 124 history of philosophy 25 – 26, 36 – 40, 46, 50, 52 – 54, 56, 75 – 76, 83 – 84, 102, 223 Hobbes, T. 83, 93 – 94 Honneth, A. 264, 282, 284, 286 Horwich, P. 194, 212 – 213, 216 Hume, D. 40, 93 – 94, 136, 148 Husserl, E. 12, 40, 114 – 115, 119 – 122, 126 – 128, 180, 197, 201 – 203, 211, 218, 239, 310, 318 – 321; Cartesian Meditations 203; epoché 111, 113, 123 – 125, 129 – 130; Ideas I 113, 117, 123; Ideas II 319 – 320; Logical Investigations 112 – 113, 116, 122 – 123 Hylton, P. 51 – 54, 77 – 78, 168 idealism 5, 6, 25 – 26, 42, 123 – 124, 152, 163, 265, 272 – 273; see also Russell, rejection of idealism intentionality 111 – 125, 127, 130, 186, 241, 283 – 284, 287, 293; collective (or “communal,” “joint” or “shared”) 10, 305 – 308, 310 – 311, 313, 315, 318, 321 – 323 intuition 32, 59, 85, 99, 119 – 121, 155 – 56, 166, 181, 194; visual 121 intuitionism 6

James, W. 26 Jefferson, T. 288 – 289 Kant, I. 38, 56, 105, 127, 151 – 156, 166, 208, 210 – 211, 218, 229, 233, 265, 282 – 283, 294, 301; cosmological antinomies 239; on judgments of beauty 65, 72, 296; on time and space 194, 196 Kierkegaard, S. 223 – 225 Kim, J. 287 Koopman, C. 90 – 91, 94, 96 – 98, 100, 102,  105 Kripke, S. 61, 71, 188, 273 Kuhn, T. 12, 82, 154 – 155, 162, 164 Kusch, M. 78, 82, 104–105, 126 language 4 – 9, 11, 21 – 22, 24, 29, 44, 62, 64 – 66, 68, 70, 72, 102, 106, 112 – 113, 115 – 116, 120, 122, 125 – 127, 132, 134, 155, 157, 165, 172, 174, 177, 181 – 188, 205, 211, 222 – 223, 225 – 228, 233, 235 – 238, 240 – 241, 245 – 246, 250, 256, 262, 265, 275, 284, 293, 306, 308, 316, 321 – 323 Leibniz. G.W. 38 – 39, 83, 130, 144, 252, 299 – 300, 302 Leiter, B. 11, 81, 87, 89, 103, 105 Lewis, C.I. 26 Lewis, D.K. 8, 10, 130, 137, 146 – 148, 215, 287, 299, 302 Libera, A. de 83 – 84, 103 – 104 linguistic turn 4 – 6, 11, 102, 239 – 240, 245 Livingston, P. 121, 295 – 296 logic 1, 19, 21, 24, 42 – 43, 100, 112, 125 – 127, 226, 228, 237 – 238, 244 – 245, 267; Aristotelian 87 – 88; critique of 95; force of 235 – 236; paraconsistent 240, 252, 256 – 257 logical atomism 19, 50 – 51 logical form 87, 98 – 99, 132 – 134, 234 – 235, 241 McCumber, J. 161 McDowell, J. 11, 26, 31, 33 – 34, 133 – 134, 145, 315 McTaggart, J.M.E. 213, 215 – 217, 219, 287, 291, 297 – 300 Marx, K. 42, 103, 168, 263, 276 – 277, 280 – 283, 293, 302

332  Index Marxism 1, 41, 87 – 88, 97 Mead, G.H. 274 Meillassoux, Q. 130, 214, 245 Meinong, A. 50, 253 Merleau-Ponty, M. 197, 205, 309 – 310 metaphysics 157, 173 – 177, 181, 187, 228 – 30, 243, 277, 282; “analytic” 239; critique of 7 – 8; descriptive 22; neo-Aristotelian 138; return to 7 – 8, 130 – 131; “Western” 246; see also ontology Méthot, P.-O. 83, 103 – 104 Meyerhoff, H. 213 Miller, I. 213 Montesquieu, C. 288 Moore, G.E. 12, 19 – 20, 22 – 26, 50, 59, 71, 223 – 225 Morris, M. 131, 133 – 136, 139 Multimundialism 262 – 264, 268 – 272, 275 – 276, 278 – 281, 284, 286, 302 Murdoch, I. 31 mysticism 32, 246 Nagel, E. 100 Nagel, T. 271 naturalism 8, 19, 22, 24, 61, 124 natural kinds 60, 71, 78 natural sciences 21; relation to philosophy 23 – 24, 29, 52 Nietzsche, F. 75, 77, 79 – 92, 94, 97, 101, 103 – 106, 151, 161, 174, 234; see also genealogy, Nietzschean nonsense 20, 62, 64, 154, 223 – 225, 227, 243 normality 83 – 84, 125, 156, 163 – 164, 205, 210, 287, 299 – 300, 320 – 321 normative insularity 268 – 269, 271, 278 – 279 normativity 273 – 274, 282 nothing/nothingness 226, 228, 251 objectivity 213 – 214, 228, 233, 243, 245, 261 – 265, 270 – 274, 277, 280, 282, 284, 307, 315 – 316, 318 – 321, 323 Ockham, William of 83, 95 ontology 8, 115, 118, 124, 131, 136, 158; historical 157; of meaning 118; monistic 215 – 16; phenomenological 158; settheoretical 240, 291

Parmenides 241 persons/personhood 24, 66 – 67, 79, 87, 94, 160, 182, 184, 186, 198, 207 – 208, 214, 217, 219, 264, 283 – 284, 312 – 314, 320, 322; first-person 113, 117, 119 – 122, 224, 274, 311, 316; secondperson 306 – 310, 316; thirdperson 120 phenomenology/phenomenological 4, 5, 7, 8, 10, 11, 90, 111 – 13, 118, 120, 122 – 127, 175, 197 – 98, 204, 213 – 15, 293, 306, 309 – 13, 315, 321 – 22; linguistic 22; ontology 158; of time 213; transcendental 123 – 125; see also analysis, phenomenological Pippin, R. 42, 57 Plato 24, 38 – 39, 41 – 42, 50, 52, 64, 83, 88, 93, 159, 163, 181, 188, 196, 241; Symposium (dialogue) 93; Theaetetus (dialogue) 50 Plotinus 254 possible worlds 59, 111 – 112, 115 – 117, 126, 130, 215, 287, 299 – 300 pragmatism 6, 25 – 26, 41, 151, 245 – 246, 265, 273 Priest, G. 8, 240, 245, 257 Prior, A. 194, 213, 215, 219 Proust, M. 68 – 69, 73 psychologism 62, 71, 180; Chomskyan 61 Putnam, H. 7, 26 – 27, 30, 56, 129 – 131, 139, 145 – 147, 151 – 156, 158, 160 – 166, 168, 245 Quine, W.V. 1, 8, 22, 24, 27, 36, 61, 126, 132, 136, 147, 151, 154, 162, 166, 175, 183, 185, 188, 274, 316 rationality 10, 154, 161 – 162, 164 – 165, 263 realism 6 – 7, 19, 155 – 156, 163 – 164, 168, 245, 262, 265, 271, 273; about world 238 – 239; internal 139, 152 – 153, 165 – 166, 245; metaphysical 129, 131, 139, 145, 261; modal 287, 299; moral 168, 279; scientific 166, 277, 279 recognition 264, 282 – 286, 313 Reichenbach, H. 26 – 27

Index  333 relativism 7, 154, 160, 166, 185, 261 – 270, 272 – 275, 277 – 278, 280 – 286, 318 revolutions 85, 155, 157; American 288 – 289; French 287 – 289, 291, 293 – 295, 300 – 301; in physics 46; philosophical 51 – 52 Robespierre, M. 289, 295 Rorty, R. 4, 11, 12, 103, 147, 168, 245 – 246, 261 – 263, 272 – 275, 285 Rousseau, J.-J. 76, 93, 188, 274, 288 Russell, B. 20, 22 – 23, 41 – 43, 51 – 54, 115, 126, 251; Lectures on Logical Atomism 50; “On Denoting” 44; Principia Mathematica 254; rejection of idealism 6, 12, 19, 25; Russell’s paradox 239 – 240; “The Philosophy of Bergson” 31 – 32; The Philosophy of Leibniz 39, 56 Ryle, G. 21, 23 – 24, 39 – 40, 42, 48, 50, 56, 61, 201 Sartre, J.-P. 237 Schaffer, J. 131, 138 – 139, 141 – 142, 145, 148 Schlick, M. 21, 23, 222 Schutz, A. 313 – 315, 321 – 322 Searle, J. 305 – 306, 322 Sellars, W. 25, 41 – 42, 103, 133, 147, 188 set theory 11, 140, 239, 291, 295, 297 Sider, T. 239 simulation theory of mind 308 – 310 Socrates 34, 50 space 8, 67, 83, 93, 149, 193 – 198, 202 – 205, 209 – 213, 216 – 218, 299, 309, 311, 313, 315; logical 232, 299, 301; socio-political 288, 292, 314; of theoretical possibilities 99 Stein, E. 310 Strawson, P.F. 22 – 25, 41 Tarski, A. 5, 111 – 113, 115 – 120, 122, 125 – 127, 131 – 132, 147, 181, 183, 188 theory theory of mind 308 – 310 time 8, 9, 66 – 69, 73, 78, 80 – 81, 83 – 90, 93, 95, 97 – 99, 104, 154 – 155, 160 – 162, 182, 186,

193 – 198, 202, 205, 208 – 219, 275 – 277, 281, 302, 311, 313, 315, 318; A series vs. B series 194, 212, 215 – 217, 219, 287, 291 – 293, 296 – 300 Tomasello, M. 306 – 311, 315, 319, 321 – 323 Treisman, M. 198 triangulation 274 – 275, 315 – 316, 318 – 319, 323 truth(s) 5 – 9, 12, 19, 24, 44, 60, 68, 70 – 71, 76, 78, 82, 94 – 97, 106, 111 – 128, 130 – 132, 135 – 136, 138 – 141, 143, 146 – 147, 152 – 156, 158 – 163, 168, 172, 174, 180 – 188, 214 – 215, 218, 222, 227, 230 – 233, 235, 237 – 238, 243 – 244, 246, 250, 253, 255, 262 – 263, 265 – 273, 275 – 277, 281, 291 – 297, 300 – 301, 315, 317 – 319, 321; moral 263 – 264, 276, 279 – 280, 282 – 283, 285 truthmakers 6, 129 – 133, 135 – 143, 145 – 147 Unimundialism 262 – 264, 268, 270 – 272, 276 – 277, 281 – 282, 284 – 286 universality 125 – 127, 214, 219, 262 – 264, 270 – 272, 276 – 285, 294, 296, 300, 307, 315, 317 – 320 Urmson, J.O. 51 value(s) 6 – 7, 10, 76 – 77, 79 – 81, 83, 94, 96 – 97, 163, 178, 224 – 225, 244; moral 263 – 264, 278 – 284; truthvalues 78, 265, 268, 271 – 272, 275, 286, 292 van Inwagen, P. 56 Waismann, F. 222 Walther, G. 311 – 313, 315, 321 – 322 Weber, M. 313 Whitehead, A.N. 201, 254 – 255 Williams, B. 31, 35, 38 – 39, 56, 75 – 77, 93 – 94, 97, 263, 277 – 281, 283 – 284 Williams, C. 213 – 214 Wilson, J. 131, 141 – 142, 146, 148 Wittgenstein, L. 21, 23 – 24, 104, 126, 165, 166, 211, 228, 230,

334  Index 232, 234, 238, 240, 242, 246, 256, 268, 273; on Heidegger 222 – 223, 242; “Lecture on Ethics” 224 – 226; Tractatus Logico-Philosophicus 132 – 133, 228, 243, 244

wonder 60, 63, 224 – 226, 310 Wright, C. 33 – 34, 266 – 267, 269 – 270, 286 Zhangzi 251 Žižek, S. 8, 292