dewey rescuing dewey essays in pragmatic naturalism


Upload: anne-fremaux

Post on 16-Apr-2017




3 download


Page 1: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 2: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Rescuing Dewey

Page 3: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 4: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Rescuing Dewey

Essays in Pragmatic Naturalism

Peter T. Manicas


A div i s ion o f


Lanham • Boulder • New York • Toronto • Plymouth, UK

Page 5: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


A division of Rowman & Littlefield Publishers, Inc.A wholly owned subsidiary of The Rowman & Littlefield Publishing Group, Inc.4501 Forbes Boulevard, Suite 200Lanham, MD 20706

Estover RoadPlymouth PL6 7PYUnited Kingdom

Copyright © 2008 by Lexington Books

All rights reserved. No part of this publication may be reproduced,stored in a retrieval system, or transmitted in any form or by anymeans, electronic, mechanical, photocopying, recording, or otherwise,without the prior permission of the publisher.

British Library Cataloguing in Publication Information Available

Library of Congress Cataloging-in-Publication Data

Manicas, Peter T.Rescuing Dewey : essays in pragmatic naturalism / Peter T. Manicas.

p. cm.Includes bibliographical references and index.ISBN-13: 978-0-7391-2515-1 (hardcover : alk. paper)ISBN-10: 0-7391-2515-X (hardcover : alk. paper)

1. Dewey, John, 1859-1952. I. Title. B945.D44M27 2008191—dc22 2008018082

eISBN-13: 978-0-7391-3020-9eISBN-10: 0-7391-3020-X

Printed in the United States of America

�™ The paper used in this publication meets the minimum requirements of AmericanNational Standard for Information Sciences—Permanence of Paper for Printed LibraryMaterials, ANSI/NISO Z39.48–1992.

Page 6: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


Acknowledgements vii

Introduction xi


1 Pragmatic Philosophy of Science and the Charge of Scientism 3

2 John Dewey and American Psychology 35

3 John Dewey and American Social Science 63

4 Culture and Nature 81


5 Naturalism and Subjectivism: Philosophy for the Future? 101

6 Naturalizing Epistemology: Reconstructing Philosophy 119


7 American Democracy: A New Spirit in the World 143

8 John Dewey: Anarchism and the Political State 187

9 Philosophy and Politics: A Historical Approach to Marx and Dewey 211

10 John Dewey and the Problem of Justice 237

11 Liberalism’s Discontent: America in Search of a Past That Never Was 251


Page 7: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


12 The Evasion of Philosophy 273

13 Democratic Hope 283

14 Analytic Pragmatism 287

15 Postmodern Pragmatism 295

Bibliography 305

Index 315

About the Author 323

vi Contents

Page 8: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


The original place and time of presentation and/or publication is indicated be-low. I am grateful to each of the editors and publishers for their permission toreprint.

Chapter 1, “Pragmatic Philosophy of Science and the Charge of Scientism,”was read as part of an invited panel, American Philosophical Association,Boston, December 1986, and appeared in The Transactions of the Charles S.Peirce Society, Vol. XXIV, No. 4 (Spring 1988), pp. 179–222.

Chapter 2, “John Dewey and American Psychology,” Journal for the Theoryof Social Behavior, Vol. 32, No. 3 (September 2002), pp. 267–94.

Chapter 3, “John Dewey and American Social Science,” in Larry Hickman (ed.),Reading Dewey (Bloomington: Indiana University Press, 1998), pp. 43–62.

Chapter 4, “Culture and Nature,” was the Patrick Romanell Lecture, Ameri-can Philosophical Association, Portland, March 27, 1992, published in Pro-ceedings of the American Philosophical Association, Vol. 66, No. 3 (1992),pp. 59–76.

Chapter 5, “Naturalism and Subjectivism,” read at the conference, The Futureof Realism in the American Tradition of Pragmatic Naturalism, University ofBuffalo, October 20–22, 2000. The essay appeared in John Shook, (ed.),Pragmatic Naturalism and Realism (Amherst, N.Y.: Prometheus Books,2003), pp. 79–106.


Page 9: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Chapter 6, “Naturalizing Epistemology: Reconstructing Philosophy,” in JohnJ. Stuhr (ed.), Philosophy and the Reconstruction of Culture: Pragmatic Es-says after John Dewey (Albany: SUNY Press, 1992), pp. 151–74.

Chapter 7, “American Democracy: A New Spirit in the World,” Chapter 13 ofWar and Democracy (Oxford: Basil Blackwell, 1989), pp. 338–78.

Chapter 8, “John Dewey: Anarchism and the Political State,” read at the an-nual meetings of the Society for the Advancement of Philosophy, VillanovaUniversity, Spring 1980, and appearing in Transactions of the Charles S.Peirce Society, Vol. 18 (Spring 1982), pp. 133–58. Reprinted in J. E. Tiles(ed.), John Dewey: Critical Assessments, Vol. II (London and New York:Routledge, 1992), pp. 407–29.

Chapter 9, “Philosophy and Politics: A Historical Approach to Marx andDewey,” originally, “Dewey’s Critique of Marxism,” Society for the Ad-vancement of American Philosophy, American Philosophical Association,Eastern Division, New York, December 28, l984, and appeared in W. J. Gavin(ed.), Context over Foundation: Dewey and Marx (Dordrecht: Reidel, 1988),pp. 147–75.

Chapter 10, “John Dewey and the Problem of Justice,” Journal of Value In-quiry, Vol. 15 (1981), pp. 279–91. Reprinted in J. E. Tiles (ed.), John Dewey:Critical Assessments, Vol. III (London and New York: Routledge, 1992), pp. 407–29.

Chapter 11, “Liberalism’s Discontent: America in Search for a Past ThatNever Was,” read at a panel discussion Midwest Political Science Associa-tion, Chicago, April 1998.

Chapter 12, “The Evasion of Philosophy,” Review of Cornel West, The Amer-ican Evasion of Philosophy, Transactions of the Charles S. Peirce Society,Vol. XXVl, No. 3 (Summer 1990), pp. 373–84.

Chapter 13, “Democratic Hope,” Review of Robert B. Westbrook, Demo-cratic Hope: Pragmatism and the Politics of Truth, in Perspectives on Poli-tics, American Political Science Association, Vol. 4, No. 2 (June 2006), pp. 373–75.

Chapter 14, “Analytic Pragmatism,” Review of Mathew Festenstein, Prag-matism and Political Theory: From Dewey to Rorty (Chicago: University of

viii Acknowledgements

Page 10: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Chicago Press, 1997), Transactions of the Charles Peirce Society, Vol.XXXV, No. 1 (Winter 1999), pp. 203–14.

Chapter 15, “Postmodern Pragmatism,” Review of Baert, Philosophy of theSocial Sciences: Towards Pragmatism (Cambridge: Polity Press, 2005), forth-coming in Journal of Critical Realism.

Acknowledgements ix

Page 11: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 12: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


There may be no philosopher who has provoked more books and articles thanJohn Dewey, who, of course, may also have set a record in producing booksand articles. (The Collected Works, now complete, number some thirty-sevenvolumes.) One wonders if anyone has read all of it. Indeed, one strategy, per-haps too frequently adopted, is to ignore all the critical literature and try tostick to his published texts—sometimes, not even to all of them! Nor, it maybe supposed, is any philosopher less in need of “rescuing.” He has been thesubject of several biographies, including two excellent recent biographies andhis name continues to reappear in all sorts of contexts. Moreover, Dewey’smany critics ranged pretty much across the philosophical spectrum and, to besure, there were plenty of sympathetic philosophers who responded to thesecritics. But the guiding idea in this volume is not to try to rescue Dewey fromhis critics (although that is sometimes also a consequence), but to rescueDewey from his friends.

The friends that he needs rescuing from fall into two main groups. On theone hand, there are those who either play down or ignore the implications ofDewey’s naturalism. For these philosophers, his version of pragmatism brokenew ground precisely because it overcame the fundamental impasses of tradi-tional metaphysics. Viewed from the perspective of academic philosophy,these philosophers have labored hard to preserve and extend Dewey’s prag-matism as an original and distinct American philosophy. While still marginalin the academy, they have made many important contributions. Prominent inthis group are philosophers who speak of Dewey’s “metaphysics of experi-ence.”1 For these philosophers, James’s radical empiricism is often taken asDewey’s point of departure. These philosophers see rightly that Dewey re-jected atomistic empiricist versions of experience and that, for him, experience


Page 13: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

was rich and informing, and included not only “relations” of all sorts, but both“doings” and “sufferings.” But as Tiles has recently remarked, while Deweyinsisted that in his Experience and Nature, he sought to provide an “empiricalnaturalism” or a “naturalistic empiricism,” he saw also that many philosopherswould find these expressions to be oxymoronic, like “talking of a roundsquare.”

These philosophers take seriously the problem set out by Kant and hold,not without reason, that Dewey is suspicious of metaphysics in Kant’s sense:claims about that which is not in experience. The problem here, as Ralph Bar-ton Perry saw, was the slip into philosophical idealism. He argued: “It wouldappear that while Dewey . . . rescues reality from dependence on intellect, heis satisfied to leave it in the grasp of more universal experience which is ‘amatter of functions and habits, of active adjustments and re-adjustments, ofcoordinations and activities,’ rather than of states of consciousness” (Perry,1955: 315). Some defenders of Dewey would, I think, also be satisfied. Perrywas not, of course, since he persisted that “a thoroughgoing realism must as-sert independence not only of thought, but any variety whatsoever of experi-ence, whether it be perception, feeling, or even the instinctive response of theorganism to its environment” (315).2

There was something radical and important about insisting on the rejectionof a “subject/object dichotomy” but on the usual terms, if existence is re-stricted to what can be experienced, it is hard to see how idealism is to beavoided. There is an alternative, a form of critical realism, which makesKant’s thing-in-itself knowable. That is, the causes of our experience cannotthemselves be in experience. There is a real tension in Dewey on this, a ten-sion examined in several of the papers in this volume. On the present view,the critical point is that, for Dewey, contrary to modern epistemology, theproblem of the external world was not a problem—for good reason; but evenso, there remained not only the question of the causal role of an independentlyexisting nature, but as part of this, the causal role of the theoretical entities ofscience. For the whole of Dewey’s long life, positivism was surely the un-challenged view on such matters (Manicas, 1989), and while recent pragmaticphilosophies of science were not particularly influenced by Dewey, they havehelped to promote the idea that Dewey could be enlisted in their cause. Thus,pragmatic philosophy of science rejects “realism” as an untenable and un-necessary metaphysical commitment. But this seems inconsistent with the ac-tual practices of the successful sciences (Manicas, 2006). Thus, while it mayseem obvious, we can only explain the rusting of iron if we have a theory thatpostulates the independent existence of Fe and which details the processcalled oxidation.

xii Introduction

Page 14: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Similarly, while it is clear that Dewey was a powerful advocate of “themethod of intelligence,” and that plainly, the practices of the sciences werepertinent to seeing what was involved in this, these philosophers have tendedto be uncritical of what might somewhat anachronistically be called Dewey’sphilosophy of science. Partly because Dewey wrote no explicit philosophy ofscience and partly because these philosophers have been rightly suspicious ofVienna-inspired Anglo-American philosophy of science, they have paid al-most no attention to Dewey’s original—but confusing—theory of science, in-cluding his vision of its role and relation in society. This is an important la-cunae from the present point of view. Dewey is rightly associated with“science” and “scientific method,” but neither idea can be taken for granted.For example, as Dewey made clear: the social sciences need to function in ademocratic society, but as he insisted, “the prime condition of a democrati-cally organized public is a kind of knowledge and insight which does not yetexist” (Dewey, 1954: 166; my emphasis). Quest for Certainty (1929) is hardlythe key text for getting a handle on Dewey’s theory of science. As it turns out,Experience and Nature (1934), Logic: The Theory of Inquiry (1938), andmany disparate essays are both far richer and much more clearly provide themain outlines of his distinctive views on the critical issues in philosophy ofscience.

These philosophers also see American pragmatism and especially Deweyas creating “an image of America” which made him a critical player in whathas been termed the “reformist left.” These writers may acknowledgeDewey’s vision of a democratic society, but hold that, for him, problems inAmerican society “could be corrected using the institutions of constitutionaldemocracy”: elect the right politicians and enact the right laws. Thus, he isseen as left-liberal who puts his trust in American exceptionalism: the historyof the United States was moving progressively toward a distinct American vi-sion. While it is true that Dewey rejected an insurrectionary politics and wasno “fire-eating leftist,” his analysis of the present was radical in the sense thatit went straight to the roots. This put his politics close to Marx’s in criticalways. The interpretation of Dewey as a left-reformist is best articulated by thesecond group of Dewey’s “friends.”

Rorty and those who follow him constitute this second group of friendsfrom whom Dewey needs rescuing. Again, speaking from the point of viewof academic philosophy, these philosophers are typically “Anglo-Americananalytic philosophers.” While their style of philosophy has somewhat waned,it is fair to say that they continue to dominate academic philosophy in theUnited States, if less so in other places. A good deal of recent Dewey schol-arship falls into this mode.

Introduction xiii

Page 15: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Rorty, of course, is key here. Himself a well-established analytic philoso-pher, his “discovery” of Dewey led him to a more general attack on the claimsof philosophy. He was correct to insist that epistemology is a modern sub-discipline of philosophy generated by the problem of legitimating the claimsof the new science (Rorty, 1979). And given this understanding of the historyof Western philosophy, he was correct also to see strong parallels between Nietzsche, James, Heidegger, Derrida, and Foucault. Thus, contemporary“textualism,” the idea that there is nothing but texts parallels the idealist no-tion that there is nothing but ideas (Rorty, 1982: 139). But there is, he insists,a critical difference between current textualism and classical idealism. In re-pudiating the tradition, textualists rejected the framework that allow for epis-temology and ontology. Thus, unlike idealists (or naturalists or materialists)so-called “postmodern” writers reject the idea that what is important is notwhether what we believe is true, but what “vocabulary we use.” Finally, then,for Rorty, pragmatism joins postmodern thinking in repudiating metaphysicalargument between idealist/naturalists and the epistemological idea of truth ascorrespondence with reality. But if Dewey was committed to naturalism,there would seem to be no escaping ontological commitments—including anaturalistic account of consciousness. Similarly, philosophical realists—asmost ordinary people, believe that true means “correspondence with reality.”But even if one assents that we can have no unmediated access to reality as itis itself, it does not follow that we cannot discriminate between true and false.Indeed, it is a scandal to think otherwise.

Rorty sees problems with postmodernist moves to escape traditional phi-losophy. But he sees also that Dewey does not exactly fit his larger picture. Inagreement with Santayana, Rorty insists that Dewey’s efforts at a “naturalis-tic metaphysics” betrays “a recurrent flaw in Dewey’s work: his habit of an-nouncing a bold new positive program when all he offers, and all he needs tooffer, is criticism of the tradition” (1982: 78). To be sure, Dewey does offer“a bold new positive program”—a naturalistic metaphysics with epistemol-ogy replaced by his version of “logic” (Sleeper, 1986). And he needed to dothis because he could not step out of history and argue, as Rorty does, thatknowledge and truth are pseudo problems that will go away once we abandonthe claims of philosophy. Indeed, it is quite one thing to try to convince usthat “warranted assertability” could replace “truth,” understood as certainty,and quite another to say that, for pragmatists, “there are no constraints on in-quiry save conversational ones—no wholesale constraints derived from thenature of objects, or the mind, or of language, but only those retail constraintsprovided by the remarks of our fellow-inquirers” (1982: 165). Worse, “theSocratic virtues—willingness to talk, to listen to other people, to weigh the consequences of actions on other people—are simply moral virtues. . . .

xiv Introduction

Page 16: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

The pragmatists tell us that the conversations which it is our moral duty to continue is merely our project, the European’s intellectual form of life” (1982:172).

As several of essays in both Part I and Part II try to make clear, there are,for Dewey, considerable constraints on inquiry, beginning with a taken-for-granted independently existing nature, our embeddedness in it, our historyand indeed, those ongoing institutional arrangements which often make im-possible the required “conversation.” Thus, the inquiry which produced mo-lecular chemistry as we now understand it very much depended both on thenature of the independently existing world and the practices which show that,as Peirce argued, there is a preferred mode of fixing belief about it. Similarly,our embeddedness in nature and history both enables and constrains us in ac-tion. Thus, for example, as Part III tries to show, there are good reasons to be-lieve that Dewey was fully aware that there were enormous obstacles to hav-ing the kind of knowledge that he thought was essential to a democraticsociety and a humane life.

The writer who is the inspiration for the main thrust of most of the essaysin this volume is Ralph W. Sleeper, my former colleague at Queens College.His wonderful The Necessity of Pragmatism (1986) provides a systematic ef-fort to respond to both sets of the friends of Dewey. The reader might noticehere that for many years four members of the Queens department had contin-uing conversation about Dewey and, more generally, about pragmatism.These include John J. McDermott, Jack B. Noone, and Eugene Fontinell. Ourconversations never lacked passion but never approached violence.

Because Dewey’s thought was both rich and provocative, it is hoped thatthe essays of this volume, written over a period of some twenty-five years,provide a contemporary refocusing of current problems, both philosophicaland political. The essays are easily organized under four main headings.


David Hollinger has rightly argued that the critical role played by the prag-matists was “to find and articulate” a “way of life consistent with what theyand contemporaries variously perceived as the implications of modern sci-ence” (Hollinger, 1985: 93). Part I finds a deep irony in this. It is widely held,by friends and enemies, that the pragmatists succeeded. On this interpreta-tion, the pragmatists adopted an “instrumentalist” view of science in whichsuccessful prediction and control-vindicated inquiry. By subordinating all in-quiry to “practical ends,” they could show that a belief was warranted only in-sofar as it was “scientific.” Finally, they could then vindicate a culture whose

Introduction xv

Page 17: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“social motor” was science. But there is a deep irony in this: As Chapter 1,“Pragmatic Philosophy of Science and the Charge of Scientism,” tries toshow, this “victory” was pyrrhic: In this chapter, I argue that the foregoing in-terpretation is a stunning distortion and that the pragmatists failed utterly intheir quest to set a new course for a “scientific” civilization. Not only werethe forces at work resistant to their criticisms, but their fundamental insights,in a paradoxical inversion, became absorbed in distorted forms. This was es-pecially critical as regards psychology and the social sciences—as Chapters2 and 3 try to show.

By looking at the views of Peirce as well as James, we can see more clearlyDewey’s distinctive and original response. Chapter 1 provides, as well, a gen-eral introduction to themes and issues taken up in subsequent chapters andparts of the volume. Thus, Chapter 2 turns to Dewey’s relation to the historyof American psychology and argues that contrary to much established opin-ion, not only did Dewey have no influence in the path taken, but that as earlyas 1896, he marked out a path which today offers considerable promise for agenuinely scientific psychology.

The key is a proper understanding of his much ignored and when noticed,misunderstood, ideas on logic, understood by him as the theory of inquiry. Itnot only points the way to a powerful conception of an “ecological psychol-ogy,” but as Part II argues, it is at the heart of Dewey’s rejection of traditionalepistemology.

It is striking here that his enemies, for example, Bertrand Russell, found itto be a confused mess and that most of his friends have paid no attention toit. Striking here also is the fact that the two most recent and otherwise veryuseful overall accounts of Dewey, Robert Westbrook’s John Dewey andAmerican Democracy (1991) and Alan Ryan’s John Dewey and the High Tideof American Liberalism (1995), almost entirely ignore it. Ryan, surely a com-petent philosopher of science, refers to the Logic as “vast and somewhat baf-fling” (309). Following on the excellent work of Tom Burke (1994), Chapter2 develops the central role and key ideas of the Logic as the critical part ofthe misunderstanding of Dewey’s relation to psychology as a science. I con-clude by arguing that, versus the dominant “Cartesian” varieties associatedwith a good deal of what is termed “cognitive science,” the Logic providesexcellent philosophical ground for an ecological psychology. Thus, as Burkewrites, “in contrast with a classical empiricist view of perception (involvingso-called sense data, sense impression, stimulations or nerve endings, irrita-tions of body surfaces, and so forth), ecological psychology emphasizes a dif-ferent array of theoretical concepts; one being the concept of ‘invariants’ andanother the concept of affordances . . .” (1994: 84). The pertinence of theseideas for a critical realist theory of science are picked up in Chapter 5.

xvi Introduction

Page 18: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Chapters 3 and 4 extend the argument to the social sciences. From the per-spective of naturalism, Dewey could easily respond to the fundamental prob-lems of the social sciences, but especially the series of invidious dualisms:subjectivism/objectivism, agency/structure, nature/culture, the idiographic,and the nomothetic. In this context, of special note, is his usually ignored ormisunderstood theory of meaning, a theory shared by his Chicago colleague,G. H. Mead. But while there is an independently existing external world, it isstriking that as social forms do not exist independently of the beliefs and ac-tions of situated agents, only a naturalism can escape the poles of subjec-tivism and materialism.


This part picks up another central theme raised in Chapter 1, the central po-sition in Dewey’s naturalism of his remarkable and much ignored Logic:The Theory of Inquiry (1938), which, as he insisted, was “not another epis-temology.” Chapter 5 considers the broader context of “naturalism” and“subjectivism,” and seeks to locate Dewey in this context. Chapter 6 as-sumes the main thrust of the Logic, and considers critically some compet-ing efforts at “naturalistic epistemology,” including the work of Quine,Rescher, and Laudan.

Two problems stand out. First, all of the many varieties of contemporaryanalytic epistemology share in what can be termed an “epistemological in-dividualism.” This is a legacy of ruggedly antiecological individualistic tra-ditional epistemology, a legacy which, it is critical to emphasize, also pro-foundly affects a great deal of work in current cognitive psychology(Chapter 2).

Quine, whose version of “naturalistic epistemology” has nothing in com-mon with Dewey’s, despite suggestions to the contrary, gives an exemplarycharacterization of “epistemological individualism”: Thus:

This human subject is accorded a certain experimentally controlled input—certainpatterns of irradiation in certain frequencies, for instance—and in the fullness oftime the subject delivers as output a description of the three-dimensional externalworld and its history (Quine, 1969: 77).

It is hardly clear how we get from “patterns of irradiation” to concepts, orwhether, finally, the “output” which is a “description of the three-dimensionalexternal world” can be established as true. The same problems arise for thosewho identify themselves as pursuing alternative epistemologies. “Internal-ists” and “reliabilists” like William Lycan and Alvin Goldman, but also

Introduction xvii

Page 19: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“externalists” and “naturalists” of various stripes, for example, Hilary Korn-blith and Philip Kitcher. On these views, the social is not denied; but it entersonly as regards either the explanation of false belief or “the social organiza-tion of knowledge”—with Robert K. Merton identified as having authored“the most important twentieth century work.” Remarkably, no attention ispaid to the important work in recent sociology of science, including herework by the so-called strong program, and work by Latour, Pickering, Hack-ing, and many others.

I said that there are two problems. Even we acknowledge the role of the so-cial in perception and understanding, it is hardly clear whether we can, fol-lowing Dewey, simply ignore the problem of Pyrhonian skepticism. Theproblem is not justifying the existence of an external world, or whether thatworld is structured in some fashion or other, but whether, given “the natura-listic equivalence of the knowledge of different cultures,” we can justifyclaims to even warranted belief while at the same time avoiding either circu-larity or dogmatism. Thus, can we say that when the Karam utters “I see a kobity now” he is wrong, that what he sees is really a cassowary?Quine’s version of naturalistic epistemology, as well as most traditional epis-temology, either assume that some privileged beliefs are true or they assumethat something vaguely identified as “science” yields truth. In Chapter 6, atleast in the spirit of Dewey, we consider three pragmatic approaches to theproblem of privileging the claims of science without circularity or dogma-tism. I suggest that instead of seeking to warrant pragmatically assertions ormethods, we take practices as our point of departure. This provides a far moreplausible, even if modest, outcome.


There is perhaps no term so badly abused as “democracy.” Originally, ofcourse, it meant (literally) that the people rule. While not all those living inAthens were citizens, citizens actually met and made decisions, which af-fected them all. When the U.S. Constitution convinced the world that the peo-ple could be sovereign and still be entirely excluded from participation in decision-making—exactly as Madison made clear, liberal republics became“democracies.”3 With this move, not only did capitalism become consistentwith democracy, it became the ideal arrangement! “People’s democracies”accepted Aristotle dictum—and Madison’s, that if the demos who are poorachieved power, they would abolish private property as contrary to their in-terests. The people’s democracies could be one-party states as long as theymade the effort to realize the democratic value of equality. After all, as Mar-

xviii Introduction

Page 20: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

shall Tito liked to point out, since the people did not rule in either the (nowgone) Yugoslavian state or in the United States, the difference was really onlyone party as against two!

The problem of democracy in the era of the modern nation-state was bril-liantly posed in the 1920s in a remarkable debate between Dewey and WalterLippmann who argued that in the United States, tweedle-dum and tweedle-dee best characterized our party system. The debate, considered in Chapter 7,remains of profound relevance especially given the nearly complete capitula-tion to the idea that “there is no alternative” to the liberal capitalist democra-cies of the advanced industrial societies. Dewey surely thought otherwise.

The occasion for this debate was World War I. But it is critical to noticethat the war changed the minds of both parties and that war remains the gen-erally unacknowledged problem for “democracy” in the modern world. Thatis, until the Great War, Dewey’s perception of American democracy waslargely uncritical and we cannot begin to understand him on the subject of de-mocracy until we locate his maturing ideas against the background of the warand of the writings of Walter Lippmann. Chapter 7 attempts an analytical/historical consideration of the conditions and content of this debate.

Briefly, Dewey fully grasped the power of Lippmann’s brilliant critique,but he could not accept Lippmann’s solution. Lippmann made two funda-mental moves. First, he insisted that the citizen cannot know what is happen-ing or what ought to happen, and even if they could, there is a structured in-capacity to constitute any sort of coherent “public opinion.” Only “mysticaldemocrats” could believe that the people had “a will” and that this was—evencould be—actually realized.

But Lippmann was not threatened by this outcome, since he also insistedthat the critical question of government was not whether citizens actually“participated,” or whether it sought and realized “the will of the people,” butwhether “it is producing a certain minimum of health, of decent housing, ofmaterial necessities, of education, of freedom, of pleasures, of beauty” (1954:196–97). Accordingly, “the essence of popular government,” notwithstandingtweedle-dum and tweedle-dee, is a choice between supporting “the Ins whenthings are going well” and supporting “the Outs when they seem to be goingbadly” (126).

The capacity to “throw the bums out” does give a minimum of accounta-bility and this should not be discounted,4 but Lippmann’s criteria are emptyof rational content. If the citizen is to appraise the success or failure of aregime in power, then she has to make impossible counterfactual judgmentsand to find some way for these to congeal into a coherent majority vote. Lipp-mann gave a host of reasons why, under present institutional arrangements,this is impossible. But if so, there is simply no rational ground for applying

Introduction xix

Page 21: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Lippmann’s criteria. What indeed is “minimum of health” or “freedom” andhow can one know if the Outs would have done better? Lippmann fell backon a thoroughly elitist version of his argument regarding science. “Gradually. . . the more enlightened directing minds have called in experts who weretrained, or had trained themselves, to make parts of this Great Society intelli-gible to those who manage it” (370). Though these “enlightened directingminds” knew that they needed help, they were “slow to call in the social sci-entist” (371). Lippmann hopes that the lesson has been learned. What isneeded, he opines, is presidential leadership responsive to the best of “socialscientific knowledge”!5

This is an unembarrassed technocratic solution to the problem of democ-racy. But, obviously, it assumes that “experts” can have the requisite knowl-edge, and it still confronts the problem of assessing counterfactuals. More-over, even assuming that “experts” make no mistakes, actions haveconsequences that generally cannot be undone. War is surely the most obvi-ous instance.

Perhaps the recent Bush dominance of American politics is the best andworst case for Lippmann. The capacity to “throw the bums out” is a test of“democracy,” but we need to see clearly what democracy thus amounts to.Put aside the fact that mechanisms of opinion formation allow those withhuge sums of money to manipulate the opinions of citizens. Put aside also thatit is relatively easy to disenfranchise voters, that there are serious problemswith the electoral system, including the electoral college and the U.S. systemof “representation.” Put aside also that, as we more recently have discovered,with electronic voting, there is no way to know if the results of voting usingthe new electronic technologies are even truthful!6

In the midterm election of 2006, it seems clear enough that American vot-ers did repudiate at least some of the policies of President Bush and the Re-publican controlled Congress. But not only is there no way to weigh the rel-ative importance of the many policies adopted (and rejected) and to identifysome coherent alternatives, but these decisions and their consequences haveinalterably reshaped the world. While this is always the case, decisions arenot all equally monumental. The most obvious case is the war in Iraq wherethe consequences include the death of thousands, the waste of billions of dol-lars, a civil war, and likely the promotion of global terrorism. By January2007, it seemed that the clear message of the voter on the war in Iraq couldbe ignored. Indeed, confirming Lippmann on tweedle-dum and tweedle-dee,the leaders of the Democratic Party comfortably took positions remarkablysimilar to those of the sitting President! Moreover, there is already an argu-ment about which of the President’s legacies will be more important: the war,the inattention to the environment, the attack on the division of powers and

xx Introduction

Page 22: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

civil liberties, the huge deficit or the stunning decline in America’s standingin the world.

Indeed, the main argument for self-determination is precisely that if inter-dependent persons must live with decisions that affect them all they musthave a hand in making them.7 But Dewey went further. For him, the problemof democracy and social science were intimately connected. Lippmann andthe technocrats failed to realize that “experts” could indeed provide “infor-mation,” but policy needs also a clear idea of the present situation of personsand of goals to be pursued. That is, generating coherent policy that affects thelives of interdependent individuals requires the direct participation of theseindividuals. Lippmann earlier had it right: “The scientific spirit is the disci-pline of democracy, the escape from drift, the outlook of the free man” (1961:151). But this requires not experts but a democratic social science. As Deweyinsisted, what is required is “the perfecting of the means and ways of com-municating meanings so that genuinely shared interest in the consequences ofinterdependent activities may inform desire and effort and thereby direct ac-tion” (1954: 155). A “public” which satisfied this goal would still make mis-takes and would still suffer the consequences of these; but they would be, atleast, their mistakes.

Dewey recognized full well that the conditions for realizing a democraticsocial science that could constitute a public required radical change in exist-ing institutions. And this was not merely a change in our electoral politics—important as these may be, but a change that acknowledged that in capital-ism, decisions of major social importance are legitimately made by personsentirely unaccountable to the electorate. If democracy means that personshave a say in determining the conditions of their everyday lives, then de-mocracy required some form of socialism. Dewey was never clear abouthow socialism was to be institutionalized, and there is evidence that he be-lieved, wrongly on my view of the matter, that “the difference and choice be-tween a socialism that is public and one that is capitalist” regarded a choicebetween markets versus planning.8 But however that may be, the problem ofrealizing democracy pushed him to what is easily read as an anarchist vi-sion!9 Chapter 8 considers this by looking carefully at what Dewey actuallysaid against the background of actual anarchist thought. Yet, at the sametime, Alan Ryan observes rightly: “He was not a fire-eating leftist, and neverbecame one” (1995: 117).

Unfortunately, not only was anarchism a dirty word by the time Deweywrote The Public and Its Problems—Sacco and Vanzetti were executed in1921 for their beliefs even though the charges against them were unproven, butthe leftism of the day was inevitably connected to Bolshevism and a version of Marx which was rooted in Engels (Manicas, 2000) and in the 2nd

Introduction xxi

Page 23: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

International, and, subsequently, with what became the standard, but also con-testable version of Lenin (Lewin, 2005). Indeed, it is absolutely crucial to no-tice that Dewey had read little of Marx and that his anticommunism wassquarely directed at the Stalinism that had solidified in the 1930s. Similarly, avery different reading of Marx became possible only in the 1930s with thepublication for the first time of the early writings of Marx (in German), thecomplete German Ideology and The Grundrisse. Chapter 9 offers an historicalreading of Marx and Dewey and shows that understanding Dewey requiresposing his thought against the Stalinist version of “Marxism/Leninism.”Dewey had no patience with the pseudoscience of a vulgar dialectical or his-torical materialism—mostly for good reasons, even if this leaves open thequestion of whether as Chapter 9 argues, some other version of Marx was eas-ily compatible with Dewey and whether some patent shortcomings in Dewey’sanalysis might be filled with some pertinent Marx. It is acknowledged by anumber of important writers that Dewey and Marx shared fundamental philo-sophical premises, but it is not always noticed that they also shared in their vi-sion of a good society and in thinking that a “gradualist” politics need not beantirevolutionary (Chapter 9).10 The point, more generally, is that Dewey triedto find a politics between liberals who insisted on parliamentary means butwho saw no need for radical change in the existing social structure, and the“scientistic,” eschatological and insurrectionary versions of the Marxists. Theproblem remains—assuring the continuing relevance of Dewey.

Finally, then, Dewey would insist that a politics without vision is merelyunprincipled opportunism. But Dewey’s vision of democracy is not merely anabstraction. As a practice and a process in which action is informed by arecognition of our inevitable interdependencies, it is a realizable ideal. AsRousseau, Marx, and Dewey saw, interdependency is inevitable, and interde-pendency does establish the conditions of injustice and tyranny. But democ-racy is its only solution. There are no assurances, to be sure, but as EmmaGoldman observed “The night cannot last forever.”

The final two chapters of Part III consider Dewey’s political theory in con-trast to three more recent and influential interventions. The publication ofJohn Rawls’s A Theory of Justice (1971) generated a veritable industry. Forthe first time in a very long time, here was a philosopher doing normative po-litical theory. Rawls articulated a liberal theory which had a New Deal lookabout it, progressive without being radical. It was quickly responded to byRawls’s Harvard colleague Robert Nozik. His book, Anarchy, State andUtopia (1974) was a criticism of Rawls from the Libertarian Right. It earnedhim a cover in the New York Times Magazine.

It is critical to notice that neither Rawls nor Nozick had much to say aboutdemocracy. Rawls assumed some form of “representative regime” and (with

xxii Introduction

Page 24: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

J. S. Mill) even defended plural voting. While democracy is not indexed inRawls’s book, Nozick surely went further. After acknowledging that democ-racy is the idea that “people have a right to a say in the decisions that impor-tantly affect their lives,” Nozick asserted, remarkably: “After we excludefrom consideration the decisions which others have a right to make and theactions which would aggress against me, steal from me, and so on, . . . it isnot clear that there are any decisions remaining about which even to raise thequestion” (1974: 270). For both, accordingly, justice was the key concept.

Dewey is quite the opposite. As argued in Chapter 10, if one surveys thevoluminous writings of Dewey, the first thing that one notices is the relativeinattention paid by Dewey to the problem of justice. Altogether, there are per-haps not more than a dozen pages of sustained discussions devoted explicitlyto the topic. These discussions are little gems, and they offer what are, I think,fatal criticisms of the liberal theories of both Rawls and Nozick. This, too, isgenerally ignored.

Dewey, always concrete and historical, recognized that what we call “lib-eral democracy” emerged at a specific time and place in world history, that itdid not have democracy as one of its goals, and that while it celebrated theautonomous individual—a prerequisite for the ideology of market capitalism,it offered a false picture of individuals and their relations. Not only were per-sons social beings, deeply interdependent and “encumbered” (to borrow aterm from Sandel), but “the control of the social environment which is fur-nished by the institution of property” makes the idea of equal freedom in lib-eral democracies “a pure absurdity” (Dewey, 1954: 271).

It is not that Dewey did not care about justice. Rather, he insisted that de-mocracy was the primary problem and that because the fundamental assump-tions of associated life were misconceived by liberal theories, it disvalued de-mocracy. Liberal theory thus shares with Lippmann the idea that“participation” was not an issue and that as long as the quality of everydaylife was as good as could be expected, all was well enough. For liberal the-ory, securing political and civil rights, some measure of opportunity for all,unimpeded markets and private property was all that “democracy” demanded.It is striking that when, in 1928, Dewey traveled to the New Soviet Union, he observed: “I was certainly not prepared for what I saw; it came as a shock” (LW, Vol. 3: 217). For him, the “experiment” had two goals. First,there was what had concerned Lippmann—security against want and illness,and for health, recreation and “a degree of material ease.” The other was the“familiar democratic ideals, familiar in words at least—of liberty, equalityand brotherhood.” The hope was that both will be “more completely realizedin a social regime based on voluntary cooperation, on conjoint worker’s con-trol and management of industry, with an accompanying abolition of private

Introduction xxiii

Page 25: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

property as a fixed institution” (LW, Vol. 3: 244). Dewey soon enough cameto see that “the experiment” which, under the prevailing conditions, could nothave succeeded, had turned to disaster (Manicas, 1989, Chapter 11). Sincethen we have been left with the idea, inherited from Wilson and currently pur-sued by Bush, that making the world “safe for democracy” really means mak-ing the world “safe for liberal capitalism.”

Apart from “Marxist” criticisms of liberal theory, there is a currently fash-ionable critique of “rights-based liberalism” often termed, “republican-communitarianism.” It is clear enough that Rawls and Nozick (along withFlathman, Dworkin, Feinberg, Gewirth, Sen, and many others) are, despitedifferences, “rights-based liberals.” The other side is a much less clear groupand might include any number of diverse writers who have criticized liberalphilosophy and promoted some version or other of “community,” includingDewey, Hannah Arendt, Robert Paul Wolff, Charles Taylor, Roberto Man-giabera Unger, Michael Walzer, Carol Gould, Hannah Pitkin, Amitai Etzioni,and some others. The relation to democracy of these writers is also very di-verse. But “republican-communitarism” is well represented by MichaelSandel’s, Democracy’s Discontent: American in Search of a Public Philoso-phy (1996), discussed in Chapter 11. Of particular interest here is Sandel’s ef-fort to link his views to those of Dewey.

There are two related features of Sandel’s position. First, he claims tohave a version of “self-rule” and second—the heart of his critique of liberalism—the real problem of government is not securing liberal justiceas defined by either Rawls or Nozick, but the cultivation of “civic virtue.”To be a citizen requires “a sense of belonging,” the existence of “a moralbond with the community whose fate is at stake.” On this view, govern-ments have legitimate concerns with “soulcraft,” what he elsewhere calls“the formative project.”

Sandel is quite right, of course, to say that Dewey was a critic of liberal in-dividualism, but Dewey called for radical version of the alternative, not as inSandel, a reactionary version of “encumbered selves” who, like Robert E.Lee, concluded that his obligation to Virginia (and to the institution of slav-ery) was not merely of sentimental import, but had “moral force” (Sandel,1996: 15). For Dewey, “community” was essential, but for him, in contrast toSandel, it was grounded on recognition of interdependence not on blood,habit, religion, or language. Similarly, Sandel seeks to capture the essentialingredient of democracy by speaking of “self-rule,” but in sharp contrast toDewey, there is simply no attention paid to what this means institutionally.One wishes that he had read Lippmann, or more lately, Robert Dahl. This absence is explained, in part at least, by his distorted view of American his-tory—a history that, as noted, was well understood by Dewey and Lippmann.

xxiv Introduction

Page 26: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Worse, perhaps, for Dewey, in contrast to Sandel, the problem of Americandemocracy was not “moral” but institutional and structural: In conditions ofalienation, “publics” cannot exist.


The problem of American democracy is well understood by Cornel West inhis important The American Evasion of Philosophy (1989), discussed inChapter 12. In the spirit of Rorty, he wishes that Dewey had been “a moreconsistent historicist pragmatist,” instead as I would have it, “a more consis-tent naturalist,” exactly in Marx’s sense. West, sympathetic to Marxian ideas,sees the radical and unfinished character of Dewey’s emancipatory project.While one may have some misgivings about both his account of Emerson andhis efforts at tapping American cultural materials, his critical reflections onthe failure of Dewey’s project are especially provocative and suggest the deepreasons for what remains an unsolved dilemma: How to be both radical andcommitted to democratic processes.

A useful comparison to West’s book is Robert Westbrook’s DemocraticHope: Pragmatism and the Politics of Truth (2005), discussed in Chapter 13.On his version, Dewey neither “evaded” nor transformed philosophy. Ratherhe holds that Dewey should be read as offering an epistemological defense ofdemocracy, something which he recognizes Dewey did not do. For West-brook, the “argument” has been filled in by the recent pragmatisms of Putnamand Misak. But the project is both alien to Dewey and, in contrast to his wellinformed book (1991), ignores the arguments that Dewey actually did make(see above).

Westbrook admits that he gave Dewey’s logic “short shrift” and he en-dorses Kloppenberg’s view that “Dewey was taking the challenge of ‘con-structing a democratic political culture on the quicksand of instrumentalistlogic’” (2005: 177). This is half-right: the Logic was the ground of his visionof a democratic culture, but one needs to overcome a good deal of the philo-sophical tradition before one can see that it can hardly be characterized as“quicksand.”

Similarly, Westbrook’s suggestion that Dewey never leaves the “populism”of late nineteenth-century “producer-republicanism” is a fairly typical criti-cism that fails to account for the changes in his views following the GreatWar. While Dewey’s socialism contradicted all “actually existing socialisms,”it was fully consistent with Marx’s idea that “producers” in capitalist societyare alienated, that neither wage workers nor “independent producers” on theJeffersonian model, are capable of what Marx termed, “free production.”

Introduction xxv

Page 27: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Again, one must see that Dewey’s critique was not nostalgic, but radical injust the ways that Marx would have endorsed.

The temptation to assist Dewey by constructing “arguments” which helacked is the central task of Festenstein’s Pragmatism and Political Theory(1997), discussed in Chapter 14. Like Westbrook, Festenstein writes solidlywithin the tradition of Anglo-American analytic philosophy and, viewedthrough these lens, he finds fatal problems in Dewey’s naturalistic ethicalframework (62, 99, 145). He concludes that Dewey had “a scientistic hope fora physics of problem-solving” (45) and that his “empirical theory of valuationseems to rest on the possibility of a prior science of problems and their reso-lution, which does not exist” (44). But inquiry, as Dewey understood it, wasnot some “prior” science of problem solving. It was the only defensible wayto address all problems, scientific and otherwise. Dewey did not, of course,embrace the prevailing “fact/value” dichotomy and he often spoke of “allegedscientific social inquiry.” The following text neatly sums up a theme whichhe pursued throughout his long life.

The sociologist, like the psychologist, often presents himself as a camp followerof genuine science and philosophy, picking up scraps here and there and piecingthem together in somewhat aimless fashion. . . . But social ethics is the changefrom inquiring into the nature of value in general to an inquiry of the particularvalues which ought to be realized in the life of everyone, and of the conditionswhich shall render possible this realization (Early Works, Vol. 5: 23).

Chapter 15, the final chapter, rejoins the question of Dewey and social sci-ence and argues that the currently fashonable postmodern reading of prag-matic social science will not do. So why not Dewey?


1. As Sleeper (1986) remarks, we need to resist “the almost universal habit oftaking for granted that experience is the subject matter of [Dewey’s] metaphysics”(1986: 6).

2. Perry’s version of direct realism is, to be sure, untenable. See Shook, Dewey’sEmpirical Theory of Knowledge and Reality. But see Roy Wood Sellar’s “Material-ism and Human Knowing,” in R. W. Sellars, V. J. McGill, and Marvin Farber (eds.),Philosophy for the Future (New York: Macmillan, 1949).

3. See my War and Democracy, especially Parts I and III. Lippmann and Deweywell understood the consequences of this shift in the meaning of democracy. Lipp-mann wrote: the fiction that the United States is a democracy owes “to the victory ofThomas Jefferson. . . . It is a fair guess that if everyone had always regarded the Con-stitution as did the authors of it, the Constitution would have been violently over-

xxvi Introduction

Page 28: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

thrown, because loyalty to the Constitution and loyalty to democracy would haveseemed incompatible” (Lippmann, 1954: 284). Dewey, who was never a “mysticaldemocrat,” offered similar sentiments.

4. Most Americans believe, it seems, that one has a democracy if there are “freeelections,” but free elections require political and civil liberties and it is these, not freeelections that mark the important difference between tyrannies and non-tyrannies. For this reason, as well, liberalism is often confused with democracy. SeePart IV.

5. Rorty remarks: “Even someone like myself, whose admiration for John Deweyis almost unlimited, cannot take seriously his defense of participatory democracyagainst Walter Lippman’s insistence on the need for expertise” (1998: 104). CompareWestbrook’s very useful Epilogue (1991) and his later account in Democratic Hope(Chapter 14 below) which seems, at least, to capitulate to a Rortyean problematic.

6. While they have been under attack by the Bush regime, political and civil lib-erties are not yet utterly compromised. People can still inquire, speak out, and orga-nize, even if this has little effect on policy. Accordingly, if democracy in the UnitedStates (as elsewhere) is profoundly constrained, the United States is not a tyranny.

7. There is, of course, absolutely no democracy as regards what is hypocriticallycalled “the community of nations.” See Dewey’s remarkable and much misunder-stood efforts in the campaign to outlaw war (Chapter 8).

8. It is easy enough to show that centralized planning, Soviet style, cannot be the-oretically sustained and is disastrous practically. As Mandel, for example, sees, onemust assume the whole of general equilibrium theory. But there are fatal objections tothis theory—as argued by Hayekians among many others. There are variant forms ofmarket socialism, but surely a vision to be pursued was laid out in the much ignoredessay by Diane Elson, “Market Socialism or Socialism of the Market” (1988).

9. Once we are clear about misconceptions of anarchist politics including the ideathat anarchists were terrorists and utopian in the worst sense, there is nothing prepos-terous about seeing Dewey as an anarchist. At the time of the Pullman strike, he wroteto Alice Dewey that he had realized “how ‘anarchistic’ (to use the current term here)our ideas and especially feelings are” (quoted by Westbrook, Democratic Hope: 86).See also Hook’s extremely useful account of the Marxian theory of “the state.” Hook(1933) distinguishes “society,” “government,” and “state,” and argues that “where thegovernment represents the needs and interests of the entire community, it does notneed [the state] special and coercive force behind it” (214). Hence, for Marxists, if de-mocracy is to prevail, the state must be “smashed.” Indeed, Lenin’s State and Revo-lution (1905) is an anarchist tract and a proper reading of “What Is to be Done” andof the period from the October Revolution to Lenin’s death, shows that while Leninmade many mistakes, including, for example, destroying the Constituent Assembly,he never wavered in his defense of the soviets, the most democratic of the institutionsin the evolving USSR (Lewin, 2005).

10. During the much-misunderstood period following the abdication of the Kaiser,the revolutionary goals of Social Democracy, as understood by Marx, Engels and the“revisionists” were betrayed by SPD leadership. After this betrayal, with the help ofBolshevik revolutionary practice, Social Democracy was redefined as consistent with

Introduction xxvii

Page 29: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

a capitalism “with a human face.” Similarly, the highly restricted choices of the Bolsheviks in Russia generated a very distorted view of socialism. See Manicas, Warand Democracy (1989), Chapters 11 and 12. In both the German and Soviet case, thecritical question is: “could it have been otherwise”? As Hook insisted, following both Marx and Engels, to be a socialist, one had to be a democrat. In our Orwellianworld, the meanings of “anarchism,” “socialism,” and “democracy” are thoroughlycorrupted.

xxviii Introduction

Page 30: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Part One


Page 31: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 32: Dewey Rescuing Dewey Essays in Pragmatic Naturalism



By the turn of the century, it was clear to William James, Thorstein Veblen,and John Dewey that science was giving “its tone to modern culture.” But forthem, the consequences were more than uncertain. The most well known ad-vocates of science, Spencer, Clifford, Huxley, and others, were not only de-fending agnosticism and positivism, but a view in which science was to be immunized from the biases and interests of human communities. James,Veblen, and Dewey were anything but enthusiastic about the situation as theysaw it. Indeed, James’s criticisms hinged on ideas about the foundations ofscience which were completely novel, and Veblen and Dewey were clearestin seeing that science was being shaped by changes “in industry and in theeconomic organization of society.” Science, pretender to transcendent au-thority, was becoming industrialized, technocratic.

David Hollinger has rightly argued that the critical role played by the prag-matists in American culture was “to find and articulate” a “way of life consistentwith what they and their contemporaries variously perceived as the implicationsof modern science” (Hollinger, 1985: 93).l It is widely held, by friends and ene-mies, that they succeeded. On this interpretation, the pragmatists adopted a viewof science in which successful prediction and control vindicated inquiry. By sub-ordinating all inquiry to “practical ends,” they could show that a belief was war-ranted only insofar as it was “scientific.” Finally, they could then vindicate a cul-ture whose “social motor” was science.2 In what follows, I suggest that theforegoing interpretation is a stunning distortion and that the pragmatists failedutterly in their quest to set a new course for a “scientific” civilization. Not onlywere the forces at work resistant to their criticisms, but their fundamental in-sights, in a paradoxical inversion, became absorbed in distorted forms.

Chapter One

Pragmatic Philosophy of Science and the Charge of Scientism

Page 33: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

But they were not entirely blameless. The problem is not merely that thelanguage they employed, including, of course, the term “pragmatism” itself,made it easy to misunderstand their views—although this was certainly true.3

Rather the problem was deeper. In particular, it regarded pragmatist views ofthe nature of philosophy and its relation to science. While the pragmatistswere correct in seeing themselves as innovators, and while, finally, they maywell have agreed on less than they disagreed, they did not wholly escape pre-vailing antimetaphysical attitudes. Indeed, the nature of their commitment toexperience made it easy to distort the nature of their commitment to practice.If I am correct, the deep problem in pragmatist thought is the turn away from“the epistemological question,” explicitly taken by Dewey. The troubles be-gin with Peirce’s verificationism and are enormously exacerbated by James’sshift from the philosophy of his Principles to his radical empiricism. The dif-ficulties in getting a handle on Dewey’s philosophy can hardly be overstated.My treatment, which relies heavily on R. W. Sleeper’s recent and importantstudy (1986), is but a sketch. In Sleeper’s useful terms, if I am correct,Dewey’s “logic of experience” and “metaphysics of existence” made his nat-uralism precarious and incomplete, a feature of his philosophy noticed, forexample, by Woodbridge and Santayana. For me, his logic of experienceneeded what his metaphysics of existence would not allow, an indirect real-ism which affirmed that there is a causally efficacious nonexperienceableworld sufficiently structured to be inferentially knowable. Peirce struggledwith the idea, and surely in Principles, James had held to such a view. But forreasons that I try to make clear, James and then Dewey supposed that it couldbe dispensed with. Spurred on by an emerging consensus over the nature ofscience, Peirce’s “doubt-belief” theory of inquiry allowed Dewey to dissolveboth “the problem of the external world” and “the mind/body problem.” Forhim, there was no problem of knowledge (überhaupt), even if there were par-ticular and concrete questions which cried for resolution, questions whichpersistent inquiry could answer. This meant, on Dewey’s view, how might themethods of science be turned to human use? Yet, if as a consequence of theturn away from the epistemological legacy of modern philosophy, the idea ofscience which he assumed had positivist elements, then it became easy to seepragmatism as a technocratic philosophy.4 The story I have suggested is com-plicated, of course, and this essay must be considered but a sketch.


As everyone knows, what came to be called “pragmatism” was first set outby Peirce in two remarkable essays published in Popular Science Monthly in

4 Chapter One

Page 34: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

1877–1878. In the first, “The Fixation of Belief,” he put forward his genuinelyoriginal “doubt-belief” theory of inquiry, what I shall take to be the core of“pragmatism.”5 Insisting that “that sole object of inquiry is the settlement ofopinion”—not as the tradition had held, the securing of truth, and that “beliefis of the nature of a habit,” he offered that of the possible modes of fixing be-lief, while all “do have their merits,” the method of science had, finally, to bethe one we must chose. Unlike “tenacity,” but like “authority,” it is consistentwith the fact that humans are social beings. On the other hand, while themethod of science can give us “a clear logical conscience,” as with “all that wecherish,” it “costs us dear.” The other methods, indeed, are psychologicallysatisfying and easy. The a priori method, e.g., allows “the action of naturalpreferences” to be “unimpeded” and “under their influences” lets people “con-versing together and regarding matters in different lights, gradually developbeliefs in harmony with natural causes” (Peirce, 1950: 105)—Peirce’s versionof philosophy as “conversation.” Similarly, the method of authority is not onlypropelled by our natural feelings of “sympathy and fellowship,” but, strikinga skeptical note, if for “the mass of mankind” it is “their highest impulse to beintellectual slaves,” then “slaves they ought to remain.”6 Nevertheless, as withthe mate one has selected, one must “work and fight” for the method of sci-ence, never complaining that “there are blows to take, hoping that there maybe as many and as hard to give . . .” (112). Why this effort? Exactly becausethe method of science alone “presents any distinction of a right and a wrongway” (108–9). Its “fundamental hypothesis” is that

there are real things, whose characters are entirely independent of our opinionsabout them; these realities affect our senses according to regular laws, and,though our sensations are as different as our relations to objects, yet, by takingadvantage of the laws of perception, we can ascertain by reasoning how thingsreally are . . . (l07–8).

I want to emphasize that (1) for Peirce, the doubt-belief theory rules out thequestion of the existence of an “external world “ as a skeptical question, evenif versions of “realism” and “idealism” remained open questions. That is, nei-ther solipsism nor the problem of other minds can be taken seriously. Peircecannot “prove” that there is something “which affects or might affect every[one],” but “upon which our thinking has no effect.” Yet there is no reasonthat a genuine doubt should arise in the practice of the method; indeed, “no-body . . . can really doubt there are realities, or, if he did, doubt would not bea source of dissatisfaction” (Peirce, 1950: 108). (2) We can know “how thingsreally are” even if the effects of reality on us “are necessarily as various asare individual conditions.” We can because we can assume that there are “reg-ular laws” involved in our transacting with “real things.” As Peirce later

Pragmatic Philosophy of Science and the Charge of Scientism 5

Page 35: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

makes clear, this guarantees that given infinite time, there will be agreementon the part of all inquirers. Finally (3), not only does “everybody use themethod,” hesitating only when “he does not know when to apply it,” but “sci-entific investigation has had the most wonderful triumphs in the way of set-tling opinion” (108). More generally, then, I am suggesting that the doubt-belief theory is a psychosocial—and thus scientific—theory of belief, but thatPeirce aims to provide a philosophical defense of a particular method of fix-ing belief, the scientific method, the foundations for which are in commonsense. To put the matter briefly (if cryptically), Peirce recasts the “epistemo-logical problem” by accepting the Kantian “insulation” against skepticism,but by rejecting Kant’s transcendental move.7

This enormously rich beginning was followed by “How to Make Our IdeasClear,” the essay which contains Peirce’s famous “pragmatic maxim.” Usu-ally read as the earliest—and clearest!—expression of the “operationalist”theory of meaning, the essay addressed a problem which a host of writers hadbegun to address: the distinctive character of the terms of science. Peirce, andof course, James and Dewey, began their inquiries at just the time that a hostof philosopher/physicists were producing books and articles in what wewould now call “the philosophy of science.” These included G. R. Kirchoff,Wilhelm Ostwald, Ludwig Boltzmann, Hermann Helmholtz, his pupil, Hein-rich Hertz, Ernst Mach, W. K. Clifford and his student, Karl Pearson, HenriPoincaré, and Pierre Duhem. These men all spoke with enormous authorityexactly because, by then, science was rapidly becoming an evident force inthe daily lives of people. Moreover, all of these men have been called “posi-tivists” in that, following Kant, they held, first, that scientific explanationmust eschew appeal to what in principle is beyond experience, that to do sotakes one into metaphysics, and second, following Berkeley and Hume, that“laws of nature,” are but empirical invariances.8 This thesis was related to thefirst, and, as we shall see, it was a critical one.

In his Analytic Mechanics, then, Kirchoff had said that we understand theeffect of force, but do not understand what force is. Peirce found this self-contradictory: “the idea which the word ‘force’ excites in our minds has noother function than to affect our actions, and these actions can have no refer-ence to force otherwise than through its effects” (Peirce, 1950: 129). It surelyseems here that, as Ostwald and Mach argued, force is not some “mysteriouspower” but is nothing other than its “sensible effects.” Peirce illustrated hisfamous principle by asking if one could say of a diamond that had been crys-talized in the midst of a cushion of cotton and had remained there until it wasburned up, whether it was really hard? He responded confidently “the ques-tion of what would occur under circumstances which do not actually arise isnot a question of fact, but only of the most perspicuous arrangement of them.”

6 Chapter One

Page 36: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

We know that Peirce came to see that his initial attempt to reconcile realismand phenomenalism, the characteristic drift of positivist philosophies of sci-ence, founded on the assumption that nothing is possible which is not actualor will not become actual. The issue was not merely whether unscratched dia-monds are hard, but more generally, there was the question of that Realitywhich he had posited as so essential to the method of science. When Peirce ap-plied his principle to the meaning of “the real,” he was led, as everyone knows,to assert that “the opinion which is fated to be ultimately agreed to by all whoinvestigate is what we mean by the truth, and the object represented in thisopinion is the real” (133). Against himself, he asked whether this was consis-tent with the definition given in his fixation essay? Did it not, in idealist fash-ion, make “the characters of the real depend upon what is ultimately thoughtabout them.” He answered that “reality is independent, not necessarily ofthought in general, but only of what you or I or any finite number of men maythink about it” (133). Murray Murphey was surely correct to conclude:

In Peirce’s system, the infinite future plays the part of the philosopher’s stone;it transforms possibility into actuality without compromising either the inex-haustibility of the possible or the limitations of the actual. On the one hand, thereal must be a permanent and inexhaustible possibility of sensation; on the other,it must be wholly cognized (1961: 169–70).

But if “the real” is to provide a constraint on current belief adequate for epis-temic purposes, will this do?

In The Monist of 1905, he returned to these problems. In the first of two es-says, he made clear that “instead of merely jeering at metaphysics, like otherprope-positivists, the pragmaticist extracts from it a precious essence, whichwill serve to give life to cosmology and physics” (Peirce, 1950: 192). This“precious essence” was his “scholastic realism”—and precious it indeed was.

But if pragmatism was “prope-positivist,” what did this mean? He beganthe essay with an anecdote about an “experimentalist.” He wrote:

If you talk to him as Mr. Balfour talked not long ago to the British Associationsaying that “the physicist . . . seeks something deeper than the laws connectingpossible objects of experience,” that “his object is physical reality” unrevealedin experiments, and that the existence of such non-experienceable reality is “theunalterable faith of science” . . . you will find the experimentalist mind to becolor-blind (182).

Although the phrase, “unrevealed in experiments,” is more problematic thanmay appear, this could have been written by Ostwald or Mach, Poincaré orPearson: Science aims at discovering laws which connect possible objects of

Pragmatic Philosophy of Science and the Charge of Scientism 7

Page 37: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

experience, where this means, laws of invariance between phenomena. But inthat same essay, Peirce was as emphatic about his scholastic realism as he wasemphatic about what for him was the real novelty of the new pragmatic the-ory: “its recognition of an inseparable connection between rational cognitionand rational purpose,” the connection which James will be so pleased to develop.

Whatever Peirce intended by his scholastic realism, it is clear enough to see that it is inconsistent with all the positivisms, Comte’s, Mill’s, Mach’s, or later Vienna varieties. While for Peirce, there was no “nonexperienceable reality”—in this he agreed with Kant and the positivists, “there are real ob-jects that are general, among the number being the modes of determination ofexistent singulars.” The article of 1878 had either glossed over this point as“unsuited” to the public there addressed or, he noted, “perhaps the author wavered in his own mind” (215). In that essay, he had written: “it would bemerely a question of nomenclature whether that diamond should be said tohave been hard or not.” This is, he now writes, no doubt true, “except for theabominable falsehood in the word ‘merely,’ implying that symbols are un-real.” “Nomenclature involves classification,” he continued, “and classifica-tion is true or false.” Thus, “the generals to which it refers are either reals . . . or figments” (215). In this case, the “generals” are real: There are dia-monds and anything which is really a diamond is really hard because beinghard is an inseparable property of at least some of those other propertieswhich make a diamond what it really is. It must be hard. He wrote:

Being a diamond, it was a mass of pure carbon, in the form of a more or lesstransparent crystal . . . it could be found to be insoluble, very highly refractive,showing under radium rays . . . a peculiar bluish phosphorescence, having ashigh a specific gravity as realgar . . . and giving off during its combustion lessheat than any other form of carbon would have done. From some of these prop-erties hardness is believed to be inseparable. For like it they bespeak the highpolemerization of the molecule (219).

The point must not be missed. On positivist versions, laws of nature areconstrued as universal conditionals of the form (x) (Fx → Gx) where “→”is “suitably” interpreted. That is, a law is construed as a contingent rela-tionship between the extensions of its terms, “all F’s are G’s.” But onPeirce’s view of the matter, a law expresses a nomic relationship betweenproperties, between F-ness and G-ness, properties to which we refer withcorresponding abstract terms.9 The reality of the diamond is expressed inthe truth of “general conditional propositions,” but these are not construedin a Humean fashion, for as Peirce saw (and Kant before him), on such aview, science is not possible.

8 Chapter One

Page 38: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

In an unpublished manuscript, “Laws of Nature and Hume” (1901),Peirce’s criticism of Hume (and the Humeans) is decisive. He writes:

. . . we do not say that the alternation of day and night is necessary, because itdepends upon the circumstance that the earth continually rotates. But we do saythat by virtue of gravity every body near the surface of the earth must be con-tinually receiving a component downward acceleration. . . . Nor do Hume or hisfollowers dream of denying that. But what they mean when they say there is not“necessity” in gravitation is that every “event” which gravitation formulates isin reality totally independent of every other; just as Hume supposes the differ-ent instances of induction to be independent “evidences.” One stone’s fallinghas no real connection with another’s fall. . . . The objection to Hume’s concep-tion of a Law of Nature is that it supposes the universe to be utterly unintelligi-ble, while, in truth, the only warrant for any hypothesis must be that it rendersphenomena intelligible (Peirce, 1950: 310).

Abduction leads us to conclude that gravity is a structural property of all bod-ies; hence, d � 1/2 at2 is a law. It is a contingent fact that the world is con-stituted such that gravity is true of all bodies and it is a contingent fact thatsome particular body be near the earth, but if the theory is true, then that body“must be continually receiving a component downward acceleration”; in free-fall, it must fall as 1/2 at.2 Science needs real connectedness; but such con-nectedness is not the product of constitutive features of the mind, as Kant hadit. Connectedness is in the mind-independent world. It is thus that for Peirce,there are “objective possibilities,” unactualized, but real.

Plainly, I cannot here do any sort of justice to Peirce’s complicated andingenious philosophy of science. Murray Murphey has, I think, caught itsmost fundamental premises in his account of the “material aspects” ofFirstness, Secondness and Thirdness. Briefly, Firstness and Secondness in-volve, critically, a psychological theory of perception. The phenomenalsuchness of a percept, Firstness is a product of unconscious inferences ofneural stimuli—as Helmholtz had argued. Secondness is the stubbornnessor compulsory character of sensation. Thirdness, then, is lawfulness. AsPeirce wrote:

Whatever is subject to law is capable of representation by a sign of which thatlaw is the meaning, and whatever is subject to law is itself a sign of the law towhich it is subject. It is in this sense that Thirdness is at once the category of lawand of rationality and intelligibility (Quoted by Flower and Murphey, 1977:604).

Since the pragmatic theory of meaning holds that “what a thing means is sim-ply what habits it involves,” and since “habits” are themselves analyzable as

Pragmatic Philosophy of Science and the Charge of Scientism 9

Page 39: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

conditionals supporting expectations—realistically understood—we can seehow critical is Peirce’s anti-Humean concept of law.

There are all sorts of questions that might be asked here, not leastwhether, in whole or in part, Peirce’s philosophy of science can be sus-tained? I forego the temptation to engage this question. Instead, I merelyemphasize that Peirce packed a great deal of empirical science into his the-ory of knowledge and that because all of the usual categories, positivist,realist, idealist, fit his thought, none of them did. He did deny with the pos-itivist that there was a “nonexperienceable reality,” a consequence of hisingenious, if unsuccessful, effort to combine realism and phenomenalism.No doubt this fostered confusion. Moreover, like them, he offered a verifi-cationist theory of meaning, but unlike them, he tied this to a psycho-social theory of belief and a strong version of lawfulness, a version whichwas undergirded by his scholastic realism. His conditionals were not ex-hausted by the material conditionals of later verificationist theories. ForPeirce, Thirdness was a critical ontological category guaranteeing hissemiotic. The ideologists of “scientific method” liked the “operationist”part of the story and at least in part because so much of Peirce’s work wasunpublished, the very complicated metaphysics that sustained it was ig-nored. Aided and abetted by the loose language of James’s Pragmatism,the former came to fit neatly into the Weltanschauung of the times, initiat-ing the myth of Peirce, the seminal American pragmatist cum positivist.James and Dewey each found different things in Peirce, but it is fair to saythat they both liked what Peirce saw to be the genuine novelty of “prag-matism,” namely, the “inseparable connection between rational cognitionand rational purpose.” Although James had arrived at a similar notion atabout the same time as Peirce, it found its most developed expression inJames’s Principles of Psychology (1890).10


James’s Principles is perhaps one of the two or three greatest books in the his-tory of psychology. Yet contrary to the conventional wisdom, it had practi-cally no influence on the development of American psychology. It would takeanother essay to begin to show this, but since the problem is germane to thepresent argument, something needs to be said. (For a fuller account, seeChapter 2.)

We can notice, first, that Principles was published when the subject matterand method of psychology as a science were still very much unsettled (Man-icas, 1987, Chapter 9). For example, was psychology concerned with the

10 Chapter One

Page 40: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“laws of the mind” (as in Mill, Bain, the later Wundt)? Or with giving a neu-rophysiological account of the phenomenon of mind, including the physiol-ogy of sensation and the genesis of “reasoning” (as for example, in Helmholtzor Spencer)? What was its relation to “logic” (as in Venn or Lotze)? Or per-haps the task of psychology was “practical,” in “behavior”? Was its concern“the generalized, normal human adult mind” or the psychology of individualdifferences? And finally, what of its methods? Was introspection part of “ex-perimental psychology,” independent of it, or to be completely rejected?

James offered, modestly and misleadingly, that the originality of his Prin-ciples consisted in its “strictly positivist point of view” (James, 1981, Vol. 1:6). It is important first to see what James did not mean by this.

He was clear that the results of scientific inquiry were in no way the “imme-diate results of experience” nor were “scientific objects” restricted to what isfound in experience. Thus, “the essence of things for science is not to be whatthey seem, but to be atoms and molecules moving to and from each other ac-cording to strange laws. . . . What we experience, what comes before us, is achaos of fragmentary impressions interrupting each other; what we think is anabstract system of hypothetical data and laws” (1981, Vol. 2: 1230–31).

Plainly, we do something with “what comes before us.” But there are twoaspects of this. There is first what we all do if we are to have coherent expe-rience, if we are to convert “the chaos of fragmentary impressions” to a graspof the “habitudes of concrete things.” The grasp of these, the “proximate lawsof nature,” for example, that heat melts ice and salt preserves meat, form “anenormous part of human wisdom.” These “empirical truths” are practical.Indeed, they are indispensable to the continued reproduction of human com-munities. In James’s view, getting an understanding of how we come to havesuch knowledge was the first problem for a scientific psychology. But thereis, as well, what as scientists we do: The effort to explain these proximatelaws by means of theories that, for example, speak of polemerization or grav-itation. For James, such theories have an entirely different aim and ground.“The popular notion that ‘Science’ is forced on the mind ab extra, and thatour interests have nothing to do with its constructions, is utterly absurd.” ButJames emphatically denied that the “interest,” which generates science is“practical.” Picking up a theme he had advanced in “The Sentiment of Ratio-nality,” he insisted that “the craving to believe that the things of the world be-long to kinds which are related by an inward rationality together, is the par-ent of Science as well as of sentimental philosophy. Moreover, “the originalinvestigator always preserves a healthy sense of how plastic the materials arein his hands” (1260).

Scientific inquiry might yield technologies, but James was clear that thiswas neither its motivation nor its vindication—a point that Veblen put to such

Pragmatic Philosophy of Science and the Charge of Scientism 11

Page 41: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

good work in his “The Place of Science in Modern Civilization.” In contrastto the proximate laws of nature, scientific theories are “abstract systems oflaws.” They “have to harmonize” with the proximate laws of nature, yet theyare tested not in the course of everyday experience, but in “artificial experi-ments in the laboratory.” James seems to see that, in order to set up an ex-periment, we need to “conjecture” that there is some unobservable mecha-nism whose processes have predicted effects. We contrive the experiment,then, so as to eliminate conditions that, in uncontrolled common experience,would interfere with its uncomplicated operation. That is, “experience,” inBaconian fashion, does not “engender” the “inner relations.” Rather, in ex-perimentation, we generate experiences that give us evidence of the realitypostulated by the theory.11

Accordingly, what is pertinent to defining success will differ as well. Prac-tical purposes offer practical tests; the interests of theoretic rationality, the“constructions” which bring “a strong feeling of ease, peace, rest,” the “livelyrelief” which comes with “rational comprehension,” answer to “the aestheticPrinciple of Ease” (James, 1978: 35), what Veblen termed, “the test of dra-matic consistency.”

In Principles, James’s selected example is a long text from Helmholtz’sDie Efhaltung der Kraft. Helmholtz had it right: Theoretical science “tries todiscover the unknown causes of processes from their visible effects; tries tounderstand them by the law of causality. . . . The ultimate goal of theoreticalphysics is to find the last unchanging causes of the processes of nature”(1981, Vol. 2: 1261).12

To be sure, James gave this a novel twist: “What makes the assumption [ofunchanging causes] ‘scientific’ and not merely poetic, what makes aHelmholtz and his kin discoverers, is that the things of Nature turn out to actas if they were of the kind assumed” (1261). Over metaphysics, aesthetics,and moral philosophy, science has an advantage:

Though nature’s materials lend themselves slowly and discouragingly to ourtranslation of them into ethical forms; but more readily into aesthetic forms; totranslation into scientific forms they lend themselves with relative ease andcompleteness. The translation, it is true, will probably never be ended. The per-ceptive order does not give way, nor the right conceptive substitute for it ariseat our bare word of command. It is often a deadly fight (1981, Vol. 2: 1236).

This is perhaps the basis of James’s most profound ethical claim, repeated inmany different formulations, that “the inmost nature of . . . reality is congenialto powers which [we] possess.” Moreover, saying that “the translation . . . willprobably never be ended” suggests that James would reject, as I think he should,the Peircean notion that in the end, there will be some one true “description” that

12 Chapter One

Page 42: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

is the product of persistent inquiry. Indeed, this would seem to be the case, aswell, as regards ethical and aesthetic matters. Yet, the belief that there are “atomsand molecules moving to and from each other according to strange laws” is a be-lief about the nature of a hidden reality. Indeed, in his notes for the 1879 “TheSentiment of Rationality,” there is a brilliant argument for the pragmatic perti-nence of the idea of a nonexperienceable reality. James says:

The principle of “pragmatism” which allows for all assumptions to be of iden-tical value so long as they equally save the appearances will of course be satis-fied by this empiricist explanation . . . [viz., as according to Mill, that no mys-terious “outness” needs to be postulated]. But common sense is not assuaged.She says, yes, I get all the particulars, am cheated out of none of my expecta-tions. And yet the principle of intelligibility is gone. Real outness makes every-thing simple as the day, but the troops of ideas marching and falling perpetuallyinto order, which you now ask me to adopt, have no reason in them—theirwhole existence is de facto and not de jure (James, 1978: 374).

Nevertheless, if British phenomenalism did not suffice, neither could he ac-cept a “more” beyond the actual as it functioned in Spencer and Kant. Ap-pealing to Peirce’s arguments, he first notes that “most scientific readers ofSpencer wholly fail to catch the destructive import of his theory. . . . They arewilling to believe with the Master that the deepest reality is the absolutely ir-rational, because that reality is unknowable, but few of them ultimately real-ize that the knowable of their philosophy forms a world of Chance pure andsimple” (1978: 369). Spencer’s “unknowable” cannot function to give order,since to do this it must be known to have properties which could explain theorderliness of experience. It was thus that the “plus ultra in many philoso-phies—in Mr. Spencer’s and in Kant’s, e.g., the noumenon is a dog in themanger, it does nothing for us itself but merely stands and blasts with itsbreath the actual” (371).

James was haunted by the apparent intractability of making sense of a re-lation between “outer” and “inner,” between mental facts and facts in theworld independent of mind. At this point at least, none of the inherited formsof phenomenalism would suffice. Moreover, so as to be clear, they did notsuffice not because, or only because, of flaws in the associationist treatmentof the connectedness of experience, but because “the troops of ideas . . . haveno reason in them.” James agreed here with Peirce that the real could not bereduced to the actual: “There are still other forces at work in the mind whichlead it to suppose something over and above the mere actuality of things.”These include “the sense of futurity, the power of expectation” and our moraljudgments, which “also involve [. . .] the notion of something related to theinstant representation and yet lying beyond its mere actuality” (369–70).

Pragmatic Philosophy of Science and the Charge of Scientism 13

Page 43: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

The Sentiment of Rationality is important in another way. In holdingthat “conceptions, ‘kinds’ are teleological instruments,” serving the needsof “theoretic rationality,” he hinted at an utterly novel solution to someage-old problems, problems given a full-blown naturalistic treatment inPrinciples in the chapters on conception (7), reasoning (22), and necessarytruths (28).

On this view, classification, judging, and predicating presuppose “a ratherintricate system of necessary and immutable ideal truths of comparison.” The“empiricists” are wrong in supposing that necessary truths are merely the re-sult of “experience” or as Spencer had it, of “mere paths of ‘frequent’ associ-ation which outer stimuli . . . ploughed” into the brain. But the a priorists arealso wrong since the “eternal verities” which “our mind lays hold of do notnecessarily themselves lay hold on extra-mental being, nor have they, as Kantpretended later, a legislating character even for all possible experience.”Rooted “in the inner forces which make the brain grow,” and therefore nottranscendental, they can be given a wholly naturalistic explanation. More-over, psychology shows that classification is functional in the sense that es-sential attributes are nothing more than abstracted properties which serve in-ference. While “universals” need not be grounded in “reality,” if we are tothink at all, they are nevertheless indispensable.

It is not surprising, accordingly, that if Mill et al. “begin with a clear nom-inalist note, they are sure to end with a grating rattle which sounds very likeuniversalia in re, if not ante rem” (James, 1978: 49). As Peirce had alreadyinsisted, if “particulars” are wholly independent, inference is impossible. Buton James’s view, Peirce’s “generals” did not need to have ontological status,either ante rem or in re. “The only meaning of essence is teleological . . . clas-sification and conception are purely teleological weapons of the mind” (1981,II: 961).

Yet it is critical to see also that James’s “pragmatic” account presupposes—as he sees—that there are relatively enduring “things,” that “the world” whichis independent of mind is not Heraclitean: “This world might be a world inwhich all things differed, and in which what properties there were were ulti-mate and had not farther predicates.” Fortunately, our world “plays right intologic’s hands. Some of the things . . . are of the same kind as other things;some of them remain always of the kind which they once were; and some ofthe properties of them cohere indissolubly and are always found together”(1981, II: 1246–47). That is, as Peirce had insisted, the “objects” of the ex-ternal world have some “character” or other, even though they need not beself-identifying to be cognized. If they are not self-identifying, however, theway they got identified can be largely a function of human purposes, generi-cally understood. H2O is not “more deeply and truly” the essence of water

14 Chapter One

Page 44: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

than it is “a solvent of sugar or a slaker of thirst”—since “it is all of thesethings with equal reality” (II, p. 961, note). Still, for scientific purposes, H2Ois primary, exactly because the scientific interest in “the interest of theoreticrationality.” The foregoing shows, I believe, that James offered a powerfulphilosophy of science that was not vulnerable to the difficulties in Peirce’smore speculative view. And it was not a positivism: “All ages have their in-tellectual populace. That of our own day prides itself particularly on its loveof Science and Facts and its contempt for all metaphysics (1978: 56). Posi-tivists fool themselves if they suppose that they dispense with metaphysics.Indeed, “Metaphysics of some sort there must be. The only alternative is be-tween the good Metaphysics of clear-headed Philosophy and the trashy Meta-physics of vulgar Positivism” (57).13 James’s philosophy of science was no“trashy metaphysics of vulgar Positivism.” Yet, James prefaced Principles byinsisting that in writing it, he had adopted “a strictly positivist point of view.”If I am correct, the nature of James’s “positivism” is centrally connected tothe problem acknowledged by James, that Principles was confusingly a psy-chology and an epistemology.


James insisted that a psychology which takes “the natural science point ofview,” must assume as data (a) mental states of humans (experience), (b)physical things and states in a spatio-temporal environment, and (c) knowl-edge by humans of things of type (b) (1981, Vol. 1: 6, 184). But he also as-serted that “the relation of knowing is the most mysterious thing in theworld,” that “if we ask how one thing can know another we are led into theheart of Erkenntnisstheorie and metaphysics” (1981, I: 212). A moment’sconsideration will show, however, that if the latter is true, there are some se-rious problems for psychology as a science. Indeed, it was the fear of meta-physics which had led Mill, for example, to restrict psychology to investiga-tion of mental states,14 just as it led the behaviorists to redefine the goals of ascientific psychology. James nevertheless insisted that while psychology was“the Science of Mental Life,” this necessarily committed the psychologist toinvestigation of not just its “phenomena,” e.g., consciousness and the streamof thought, but to investigation of its “conditions,” physiologically and in the“outer world.”15

Similarly, while he seems to deny that psychology, approached from “thenatural science point of view,” needs to solve the mind/body problem, he at-tacked materialism, spiritualism, parallelism, and epiphenomenalism. Forhim, “mental phenomena are not only conditioned a parte ante by bodily

Pragmatic Philosophy of Science and the Charge of Scientism 15

Page 45: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

process, but they lead to them a parte post” (I: 18). That is, although brain ac-tivity is a necessary causal condition for mental phenomena, for James, itseemed obvious that mental phenomena were themselves irreducibly real andhad causal efficacy. Not only do they lead to acts—“of course the most fa-miliar of truths”—but mental states “occasion . . . changes in the calibre ofblood vessels . . . or processes more subtle still, in glands and viscera.” Hisemergent naturalism recurs throughout Principles. The following brilliantlysuggests his notion of this:

What happens in the brain after experience has done its utmost is what happensin every material mass which has been fashioned by an outward force. . . . Thefashioning from without brings the elements into collocations, which set new in-ternal forces free to exert their effects in turn. And the random irradiations andresettlements of our ideas, which supervene upon experience, and constitute ourfree mental play, are due entirely to those secondary internal processes, whichvary enormously from brain to brain. . . . The higher thought-processes owetheir being to causes which correspond far more to the sourings and fermenta-tions of dough, the setting of mortar, or the subsidence of sediments in mixturesthan to the manipulations by which these physical aggregates came to be com-pounded (II: 1234–35).

James was not, however, clear on how such a nonreductive naturalistic re-sponse to the mind/body problem had to be worked out.16 Indeed, one mightsay that the Cartesian (ontically dualist) formulations which also occur inPrinciples suggest more than unclarity, that they reveal tensions not resolvedin James’s own mind. James’s 1894 Presidential address to the American Psy-chological Association, “The Knowing of Things Together” (1978: 71–89)was a clear step in the resolution of the tensions of the Principles. In that es-say, he considered a host of theories in response to the psychological problemof “the nature of the synthetic unity of consciousness” and concluded that forvarious reasons, none of the theories can be accepted. But what of his ownview in the Principles? He there had proposed “to simply eliminate from psy-chology ‘considered as a natural science’ the whole business of ascertaininghow we come to know things together or to know them at all” (1978: 87, myemphasis). “That we do know things, sometimes singly and sometimes to-gether, is a fact. That states of consciousness are the vehicle of knowledge,and depend on brain states, are two other facts.” At that time he supposed that“a natural science of psychology might legitimately confine itself to tracingthe functional variations of these three sorts of fact” (1978: 87).

It was precisely, then, in his claim that a science had to restrict itself to“functional variations” between “facts” that James was a positivist. For de-spite texts that suggest the opposite, for example, his willingness to counte-

16 Chapter One

Page 46: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

nance Helmholtz on the search for hidden causes, he more generally seems toagree with the positivists that all talk about causes as productive powers ismetaphysical. Science aims at “functional variations.”

The point is critical. The clearest statement is in Volume I, of Principles, inhis criticism of “the automation theory.” James insisted “the whole questionof interaction and influence between things is a metaphysical question. . . . Itis truly hard enough to imagine the ‘idea of a beefsteak binding two mole-cules together’; but since Hume’s time it has been equally hard to imagineanything binding them together” (1981, I: 140). The problem is not merelymind and matter as different “stuffs,” but of the causal interactions betweenmolecules constituting a beefsteak! The whole idea of “binding,” he wrote,“is a mystery, the first step towards the solution of which is to clear scholas-tic rubbish out of the way.” It was true that “popular science talks of ‘forces,’‘attractions’ or ‘affinities’ as binding the molecules,” but while such wordsmay be used “to abbreviate discourse,” “clear science [Mach, Pearson, Ost-wald!] . . . has no use for the conceptions, and is satisfied when she can ex-press in simple ‘laws’ the bare space-relations of the molecules as functionsof each other and of time” (1981, I: 140).

The automatists “pull the pall over the psychic half of the subject only . . .and say that that causation is unintelligible, whilst in the same breath onedogmatizes about material causation as if Hume, Kant, and Lotze had neverbeen born” (I: 140). James insists that one must be “either impartially naif orimpartially critical,” either “pull the pall” over the whole business or admitboth physical and psychic causation: If the latter, the reconstruction must bethorough-going or “metaphysical,” and will probably preserve the common-sense view that ideas are forces, in some translatable form. But Psychology isa mere natural science, accepting certain terms uncritically as her data, andstopping short of metaphysical construction (I: 141).

He had defined psychology as “the Science of Mental Life, both of itsphenomena and their conditions,” and as he says many, many times, he isinterested in ascertaining all sorts of “conditions”—of, e.g., memory, I: 17;of discrimination, I: 494–98; of thinking that what we believe is real, II:917–35, etc. But not only is he never bashful about employing causal lan-guage and in implying that it is explanatory—no scientist is!—but one mayreasonably wonder what are the conditions for something existing or hap-pening if, taken together, they are not its causes and do not explain? More-over, it was clear to James, even if amazingly missed by so many, that onecannot experience the causes of experience. Indeed, there had to be com-plicated causal relations between the three sorts of “facts” involved inknowing. Even the “spiritualist” and the “associationist” are “cerebralists”since they must admit, “certain peculiarities in the way of working of their

Pragmatic Philosophy of Science and the Charge of Scientism 17

Page 47: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

own favorite principles are explicable only by the fact that the brain lawsare a codeterminant of the result” (I: 18, my emphasis).

Nevertheless, for good historical reasons, James, like Spencer, remainedtrapped in the idea that scientific causation was merely empirical invari-ance. But if so, then if he were consistent, psychology, as James had de-fined it, was not likely to achieve its goals. It is true that one begins“naively,” by taking for granted that people have minds, that there is aphysical world and that people have knowledge of it, but if one then re-fuses to consider how, one is surely not going to explain perception, con-ception, reasoning, learning, memory, or anything else. On the positivistview, of course, explanation is either subsumption (deduction, inclusion in“a class already known”) or it is “metaphysical”—appealing to “occultforces,” “substances,” “powers.”17 James offered no deductions and heproduced no empirical invariances between the three sorts of “facts,” be-tween, for example, the experience of red tomato, some discharge of neu-rons and some “thing” in the external world. Indeed, exactly because anenormous number of very different kinds of causal mechanisms are in-volved in my experiencing red tomato—are “codeterminant of the re-sult”—it is hard to see how this could be possible! On the other hand, whenin terms of “conditions,” he did propose an explanation, for example, ofmemory, he was then well on his way to that “thorough-going,” “meta-physical” reconstruction which would have been at least part of naturalis-tic epistemology. An excellent example of this tension is James’s discus-sion of memory. The reader has been led to believe that he is getting someexplanations, but just before James concludes his brilliant account, he asserts:

A word, in closing, about the metaphysics involved in remembering. Accordingto the assumptions of this book, thoughts accompany the brain’s workings, andthose thoughts are cognitive of realities. The whole creation is one, which wecan only write down empirically, confessing that no glimmer of explanation ofit is yet in sight. That brains should give rise to knowing consciousness at all,this is the one mystery, which always returns, no matter of what sort the con-sciousness or of what sort the knowledge may be (1981, I: 647).

It is a remarkable fact, but nonetheless a fact, that positivist assumptionsabout causality and lawfulness are adequate to a scientific psychology, whichdenies that it is “the Science of Mental life, both of its phenomena and theirconditions.” By taking “prediction and control” as its “theoretical” (sic) goal,such a psychology avoids the troublesome “metaphysics” of mind/body, andof knowledge and reality. Or better, as James had it, it assumes “the trashyMetaphysics of vulgar Positivism.”

18 Chapter One

Page 48: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


By the 1894 “Knowing of Things Together,” James “became convinced . . .that no conventional restrictions can keep metaphysical and so-called episte-mological inquiries out of the psychology books” (1978: 88). But this did notlead him in the direction of a metaphysical reconstruction, which would haveallowed a full-blown nonreductive, physiological psychology replete withphysical and psychic causation as a replacement for some of the central prob-lems of traditional epistemology. It led him, instead, to “radical empiricism,”as John McDermott rightly says, to a novel metaphysics of experience.

This then is the heart of the problem. Was it possible to have both a phys-iological psychology, which displaced traditional epistemology and a meta-physics of experience, or indeed, are the two ideas at the bottom inconsistent?

From the beginning, James had sought to transform empiricism. AsHollinger says, his previous efforts had been in work which James himselfcalled “practical and psychological.” Thus, the idea that relations as well asparticulars come to us as part of a single “stream” was offered in response toboth “associationist” psychology and to neo-Kantian alternatives. As I havesuggested, this attack was fully consistent with the indirect realism of Princi-ples. Still, the problem of the relation of knower to known and of mind tobody haunted James. He made the break with “radical empiricism.” It seemsthat, originally, “radical empiricism” was, for James, merely a name of “anattitude.” But increasingly he came to think of it as the name for a technicalposition in epistemology and metaphysics, a “doctrine” that allowed him toovercome the limitations of conventional empiricisms and to dispense withthe “inclusive mind” of the idealists. Nothing could be clearer, I believe, thanthat James overcame with the new doctrine of pure experience what, for him,were ultimately invidious dualisms in his own psychology. What, in Princi-ples, he had called “the most mysterious thing in the world” was now, re-markably, fully intelligible.18

If we start with the supposition that there was only one primal stuff or materialof the world, a stuff of which everything is composed, and if we call that stuff“pure experience,” then knowing can easily be explained as a particular sort ofrelation towards one another into which portions of pure experience may enter(1978: 4, my emphasis).

Yet I do not see how to square radical empiricism with any vision of an indi-rect realism. Thus the well-known “postulate” of radical empiricism asserts:

. . . the only things that shall be debateable among philosophers shall be thingsdefinable in terms drawn from experience. Things of an unexperienceable

Pragmatic Philosophy of Science and the Charge of Scientism 19

Page 49: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

nature may exist ad libitum, but they form no part of the material for philosophicdebate.

If what is not definable “in terms drawn from experience” cannot be a matterof rational debate, it is hard to see how this is not a phenomenalism.19

One might hold here that this takes James too strictly or that it reads Jamesin the light of developments in epistemology and the philosophy of sciencewith which he had little concern. Why, for example, hold James to the logi-cian’s sense of “definable in terms drawn from experience”?

There is, I believe, an important case to be made here. James as a philoso-pher was a “popular” writer whose concerns far outran the concerns of theprofessional community (including the present writer!), who now struggle toget straight his “epistemology” and “metaphysics.” James could applaudMach and influence Russell, but the James who left the psychological labo-ratory was ambivalently an “epistemologist” and a trenchant cultural critic.As Hollinger has argued, “especially did James fear that his contemporary in-tellectuals were forming a culturally destructive idea of what it meant to be‘scientific’”(Hollinger, 1985: 5).20 Thus, the tragic irony: Because his scien-tific psychology seemed to him to involve an invidious dualism, James optedfor an innovative metaphysics of experience. This theory, he hoped, gave log-ical rights to those “too tender to give up religion, but too tough to give upscience.” But unnoticed was the fact that this theory also undercut the theoryof science he so brilliantly sketched in Principles. Because it has been diffi-cult not to read radical empiricism as a phenomenalism, James’s pragmatismwas plagued by an incipient subjectivism, and by the collapse of realism intoactualism, the problem which haunted Peirce, and if I am correct, hauntedDewey, as well. The issue is not, so as to be clear here, the status of“essences” or universals, for James’s own psychological account of these isfully consistent with his own earlier indirect realism. Nor does it regardJames’s lifelong criticism of the correspondence theory of truth, for rejectionof it is also consistent with forms of indirect realism. James’s most penetrat-ing text on this score is his very early “Remarks on Spencer’s Definition ofMind as Correspondence.” James there wrote:

the knower is not simply a mirror floating with no foot-hold anywhere, and pas-sively reflecting an order that he comes upon and finds simply existing. Theknower is an actor, and co-efficient of truth on one side, whilst on the other he reg-isters the truth, which he helps to create. Mental interests, hypotheses, postulates,so far as they are bases for human action—action which to great extent transformsthe world—help to make the truth which they declare. . . . The only objective cri-terion of reality is coerciveness, in the long run, over thought. Objective facts,Spencer’s outward relations, are real only because they coerce sensation (1978: 21).

20 Chapter One

Page 50: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

But if this is true, as I think it is, then “outward relations” are real enough,even though all we can experience are the effects of a world which is neverexperienced as it is in itself. This means that no sense can be made of test-ing truth by “correspondence”; but it means as well that as James had as-sented, the “world” contains relatively enduring “things” which exist in-dependently of us, and that these are the “objects” at the object end of the“subject/object” dichotomy. From the point of view of science, these are thetheoretized objects of physical science, just as so much in Principles sug-gested. Accordingly, foundationalist epistemology can give way to a natu-ralized epistemology in which a critical part of the story will be showinghow, for example, theorized photons, being emitted from “things” not in ex-perience affect our retina, and how, through some very complicated causalprocess, “things” in the “outer world” become the “things” of ordinary ex-perience.21 That is, either we admit the existence and causal powers of pho-tons which are in principle not “definable in terms drawn from experience,”or as radical empiricists, we merely accept, unexplained, the de facto relat-edness of experience.22


This is, if I am correct, another way of saying, as John Smith said of Dewey,that the essence of his well-known rejection of the epistemological problemwas his unwillingness to countenance that there was a theoretical subject/object problem, that “in effect all attempts at making knowledge itself intel-ligible are greeted by pointing out that science is a fact and that is the end ofthe matter” (Smith, 1970: 2). Dewey was right in rejecting “the spectator the-ory of knowledge”—the kernal of pragmatic cognition, and right also in in-sisting that inquiry had a biological and a social “matrix”—as Peirce hadseen. His theory of inquiry was a naturalist epistemology, but it was incom-plete because he finally rejected “the problem of an external world” and “themind/body problem” as nonproblems. (See Chapter 5.)

Dewey did, of course, struggle with these problems, from his explicitly ide-alist beginnings until perhaps the exhaustion of the heated realist controver-sies at the end of the second decade of the last century. As Sleeper has con-vincingly demonstrated, Dewey generated an entirely novel metaphysics,rejecting both classical realism, even as that had lingered in the thought ofPeirce, and the Jamesian metaphysics of experience. Neverthless, in my view,Dewey’s rejection of traditional epistemology was insufficiently radical; in-deed, it was conservative insofar as his “metaphysics of existence” satisfiedKantian strictures about metaphysics.

Pragmatic Philosophy of Science and the Charge of Scientism 21

Page 51: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

The issue of realism was particularly bothersome since critics of pragma-tism persistently said that it was an idealism. Consider but these few transi-tional texts: In “Reality as Experience” (1906), he offered, cautiously, that“early reality”—reality which lacked “conscious organisms,” is “at any andevery point on its way to experience” (MW, Vol. 3: 102). The answer couldhardly be satisfactory. One might admit that if minded beings had not arrivedon the scene, there could be no knowledge of early reality, but surely this re-ality would still have existed independently? In “The Realism of Pragma-tism,” (1905) he wrote that “ideas, sensations, mental states, are, in their cog-nitive significance, media of so adjusting things to one another, that theybecome representative of one another. When this is accomplished, they dropout; and things are present to the agent in the most naively realistic fashion”(MW, Vol. 3: 153). His defense of a kind of operational naive realism recursin “The Postulate of Immediate Empiricism” (1905), where he asserts: “Im-mediate empiricism postulates that things—anything, in the ordinary or non-technical use of the term ‘thing’—are what they are experienced as” (MW,Vol. 3: 158). In “Experience and Objective Idealism” (1906), he defended theempiricism of (absolute) idealism but rejected its rationalism, the idea that“thought or reason” provides objectivity to sensory data. But he left openwhat did. It was thus that R. B. Perry was happy to accept Dewey’s rescue ofreality “from dependence on intellect,” but was not happy that Dewey was“satisfied to leave it in the grasp of that more universal experience which is‘a matter of functions and habit, of active adjustments and readjustments, ofcoordinations and activities, rather than states of consciousness.’” For Perry,“a thoroughgoing realism must assert independence not only of thought, butof any variety whatsoever of experiencing, whether it be perception, feeling,or even the instinctive response of the organism to its environment” (Perry,1912: 323–24).23

Dewey’s earliest writings on mind and body are patently dualist and, as inhis 1884 “The New Psychology,” he is at pains to deny that physiology cangive any sort of explanation of “psychical events.” Following Wundt, he wasat that time insisting that “of itself,” physiology “has no value for psychol-ogy” (EW, Vol. 1). The justly famous “Reflex Arc Concept in Psychology”(1896) has often been taken as Dewey’s solution to the mind/body problem.But though it does mark a decisive break from his earlier dualism, its key fea-ture is the way that Dewey preserves, in a biological setting, the teleology ofhis earlier idealism. He firmly and rightly rejects mechanistic biology as in-adequate to the facts, then generalizes this to include the psychical. To besure, the mechanistic view did assume that “the sensation is an ambiguousdweller on the borderland of soul and body, the idea (or central process) ispurely psychical, and the act (or movement) is purely physical,” but it was no

22 Chapter One

Page 52: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

answer to assert merely that “the reflex arc formulation is neither physical (orphysiological) nor psychological: it is a mixed materialistic-spiritualist as-sumption” (EW, Vol. 5: 104).

Had Dewey been more critical of the new “behaviorists” in American psy-chology, he would have seen—and might have taught them, that his brilliantcritique of mechanism utterly undermined their research program. But in fact,his easy “functionalist” solution to the ontological issue surely contributed tothe view, shared by all the behaviorists, here including Skinner, that mind/body was a nonissue. (See Chapter 2.)

Finally, while there are texts that support the view that Dewey assumed thatpsychology (along with sociology) had displaced traditional epistemology, heseems not to have noticed that there were matters left unfinished, matterswhich rightly puzzled his critics. Here is one example: “. . . when a writer en-deavors to take a frankly naturalistic, biological and moral attitude, and to ac-count for knowledge on the basis of the place it occupies in such a reality, heis treated as if his philosophy were only, after all, just another kind of episte-mology.” (See Part II of this volume.) For Dewey, of course, the root fallacyof all epistemology was “the failure to recognize that what is doubtful is notthe existence of the world but the validity of certain customary yet inferentialbeliefs about things in it” (MW, Vol. 8). Presumably, once this is admitted,one ought to get on with the real questions of which inferred beliefs werevalid—a wholly scientific problem.

But it hardly satisfied his critics to treat them as if they failed to understandthat their problems were nonproblems and to assert that “[pragmatism] occu-pies a position of an emancipated empiricism or thoroughgoing naive real-ism,” that “[it] is content to take its stand with science . . . [and] daily life”(MW, Vol. 10: 39). By doing this, Dewey simply took for granted both com-mon sense and science and thus refused to acknowledge with Hume andKant, that there was a problem of knowledge, not merely of certain knowl-edge, but of how we can at all connect “subjectivity” to “the world.”

Nevertheless, innocence once lost cannot be regained. Even if Dewey didshow that the skeptical objections of modern philosophy foundered on a mis-construal of “experience,” there remained the problem of “reconciling the re-ality of the physical object of science with the richly qualitative object of or-dinary experience.” Dewey, who I quote here, called this problem “factitious”(1960: 131). Yet however much the problem was “unnatural,” an “artifact” ofsome accidental developments in the history of Western civilization, it was, ifunfortunately, a genuine theoretical problem—as James surely would haveacknowledged. In the present view, the much heralded overcoming of subjectand object, the duality bequeathed by modern philosophy, is “overcome” onlyby eliminating one or the other: reductive materialism (the “naturalism” of

Pragmatic Philosophy of Science and the Charge of Scientism 23

Page 53: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Rorty), or (“basement”) idealism, the metaphysics so frequently imputed toDewey.24 The other alternative, recommended here, is to insist that even ifthere is no duality between knower and cognized object, the famous “ego-centric predicament,” there must be a duality between knower and that“world” which exists independently of knowers. Indeed, it is just this thatmakes possible that naturalized epistemology which James was so reluctantto pursue, but which, if I am correct, Dewey took for granted.


James and Dewey had rich doctrines of experience. There is no doubt of that.Yet if I am correct, they became convinced, with Ostwald and Mach, andlater, as regards Dewey, with Bridgman, Schlick, and Carnap, that scientificknowledge could dispense not only with “essentially metaphysical” causaltalk, but with talk of any nonexperienceable reality. But if so, both “radical”and “emancipated empiricism,” can get only to positivist versions of science.That is, James and Dewey had to give a thoroughgoing instrumentalist read-ing to the theoretical terms of science and thus to accept the idea that an ex-planation was “subsumption,” explanation and prediction were symmetrical.

Peirce’s “How To Make Our Ideas Clear,” now shorn of his realism, was,of course, the original inspiration for the pragmatic treatment of theoreticalterms. Thus Dewey writes “the resolution of objects and nature as a wholeinto facts stated exclusively in terms of quantities which may be handled incalculation . . . is a declaration that this is the effective way to think things . . . to formulate their meanings.” Conceptions are either to be defined “oper-ationally” or they are “purely dialectical inventions” (1960: 118). Dewey as-serts that most of Newton’s analytical work “would remain unchanged, if hisphysical objects were dropped out and geometrical points were substituted”(118–19). But this could only be so if, to quote Duhem, “a physical theory isnot an explanation. It is a system of mathematical propositions, deduced froma small number of mathematical principles, which aim to represent as simply,as completely, and as exactly as possible a set of experimental laws.”

Once Dewey assented to this, it was easy to waffle over the goals and vin-dication of scientific theory. Did science aim at understanding or at predictionand control? Was a theory “valid” if and only if it “predicted?” Dewey wasnot, to be sure, alone in not seeing how critical these questions were. In fact,he seems to have anticipated Ernest Nagel’s influential view that the differ-ences between realist and instrumentalist construals of theory reduce to but “aconflict over preferred modes of speech” (Nagel, 1961: 153). Yet, the differ-ences are fundamental—as Hempel’s “theoretician’s dilemma” shows

24 Chapter One

Page 54: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

(Hempel, 1965). Thus, if theoretical terms “serve their purpose” and “estab-lish definite connections among observable phenomena,” then they are un-necessary. If, on the other hand, they lack such connections, then as but “di-alectical inventions,” to use Dewey’s terms, they are surely unnecessary. AsHempel came to see, however, the dilemma depends upon holding that “thesole purpose of a theory is to establish deductive connections among obser-vation sentences.” If this were the case, then theoretical terms would indeedbe unnecessary. But theory has other purposes. As James’s Principles had in-sisted, theory satisfies “the interest of theoretic rationality—just as had “sen-timental philosophy.” Prediction and technical control is one thing, the satis-faction that comes with understanding quite another. On the positivist view,since explanation and prediction are symmetrical, this distinction is col-lapsed; and if so, prediction and control can be taken to define science—athoroughly technocratic view.25


Pragmatism had the burden that Americans could never accept that prakticheand pragmatisch were, as Peirce said, ‘‘as far apart as the two poles.” When thisconfusion is joined to some confusions over science, one easily produces thecharacteristic misunderstanding of Dewey’s profound analysis of “the logic ofjudgments.” By developing in an original way the pragmatic insight which hadlinked rational cognition with rational purpose, Dewey tried to show that practi-cal judgments answered to norms and conditions which made them as war-rantable as “theoretic” judgments; but not because the justification of theoreticaljudgments is that they have some “practical” use, but because all inquiry is con-strained by similar conditions, the indeterminate situation in which inquiry be-gins, the inherited materials with which it works, a “reality” which imposes itsown limits, and consequences which are produced by acting on “hypotheses,”consequences whose pertinence will be a function of the rational purpose of theinquiry. There could still be differences in practices aimed at satisfying theoreticrationality and those aimed at solving “practical” problems. Fully followingPeirce and James, Dewey held that the “peculiarity of scientific abstraction liesin the degree of its freedom from particular existential adhesions” and that “inscientific inquiry . . . meanings are related to one another on the ground of theircharacter as meanings, free from direct reference to the concern of a limitedgroup” (1938: 119). It was not, that is, that scientific inquiry was freed of its ex-istential conditions and purposes, for it was exactly the pragmatist’s point that noinquiry was or could be. Rather, the words to emphasize in the foregoing are“particular” and “limited,” just as Peirce would have insisted.

Pragmatic Philosophy of Science and the Charge of Scientism 25

Page 55: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


Finally, there was Dewey’s well-known emphasis on science and scientificmethod. It is a persistent theme of all the pragmatists that although knowledgeis fallible, inquiry has an integrity. Likewise, James and Dewey were em-phatic that the world was “responsive to human purposes,” that inquiry couldchange the world. But it was profoundly easy to miss the real force of this,especially if, as I have been arguing, there was an incipient subjectivism andidealism in pragmatist ontology, and in consequence, an inevitable instumen-talism in their conceptions of science. But adding to this, as Hollinger pointsout, it is a “striking feature of the history of pragmatism” that the detailed de-velopment of Dewey’s pragmatic theory of inquiry, in his 1938 Logic, “ap-peared long after his more vague and question-begging pronouncements hadhelped win for his reconstructionist vision a following greater than it has en-joyed during the more than forty years since he did his best to justify it philo-sophically” (Hollinger, 1985: 98).26 (See Chapter 5.)

Nevertheless, Dewey was sufficiently clear on other matters that are es-sential. First, like Veblen, he believed that science was a critical part of theproblem now being faced by inhabitants of “modern civilization.” UnlikeComte, Spencer and a host of nineteenth-century writers, Dewey did not holdthat as science gave us new knowledge, there would be continuous improve-ment in human life. In an 1893 assessment of Renan’s The Future of Science,Dewey endorsed Renan’s view that “the definition of science . . . is to knowfrom the standpoint of humanity; its goal is such a sense of life as will enableman to direct his conduct in relation to his fellows by intelligence and not bychance.” But he was sympathetic to Renan’s “loss of faith in science,” ac-knowledging that “the forty years since Renan wrote have not done much toadd to the human spirit and the human interpretation to the results of science;they have gone to increase its technical and remote character.” To be sure, heaffirmed that “any lasting denial of dogmatic authority is impossible save asscience itself advances to that comprehensive synthesis which will allow it tobecome a guide of conduct, a social motor” (EW, Vol. 4: 12, 17). In this es-say, he did not say what needed to happen if science was to become a “socialmotor” for human progress. He returned to this theme in 1900, pointing outthat “the anomaly of our present social life is obvious enough.” He went on:

With tremendous increase in control of nature, in ability to utilize nature for theindefinite expansion and multiplication of commodities for human use and sat-isfaction, we find the actual realization of ends, the enjoyment of values grow-ing unassured and precarious. At times it seems as if we were caught in a con-tradiction; the more we multiply means, the less certain and general is the usewe are able to make of them (MW, Vol. 1: 76).

26 Chapter One

Page 56: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

If science were to become a “social motor” for human progress, there was, asVeblen was insisting, the need to gain “knowledge of the conditions throughwhich possible values become actual in life.” Lacking an understanding of thecauses of outcomes, we are “at the mercy of habit, of haphazard, and henceof force.” Physical science had outdistanced human science. We have applied“intelligence” to “the control of nature” but we have not put it to use to makea human world, to structure society so that our technologies are used for therealization of human values. (See Chapters 3 and 4)

Secondly, Dewey was also clear that the problem was essentially political.When he wrote the foregoing words, the social sciences were just then be-coming institutionalized as specialized “disciplines,” each struggling toachieve that “expertise” which would make them authoritative in a civiliza-tion being shaped by industrialized science (Manicas, 1987). Dewey and Veblen agreed both on what was happening and why. Veblen was pessimisticthat much could be done. In his view, American social science could not helpbut succumb to the temptations and atmosphere of the times. As he saw, it waseasy to reject the idea that social science was “inquiry into the nature andcauses, the working and outcome of [the] institutional apparatus.” Such in-quiry was dangerous, since even if it “should bear no colour of iconoclasm,”its outcome “will disturb the habitual convictions and pre-conceptions onwhich they rest.” In the spirit of middle-class reform, social science couldconcentrate “on what ought to be done to improve conditions and to conservethose usages and conventions that have by habit become imbedded in the re-ceived scheme of use and wont, and so have been found to be good and right”(Veblen, 1957: 132). So as Veblen saw it, “habit, the haphazard and force”were being reinforced by social science.

Dewey, always hopeful, believed that the new social sciences could be partof the solution. Yet, in that early review of Renan, he had himself suggestedan analysis that could have been endorsed by a Marxist, just as it became thepoint of departure of Veblen’s later trenchant analysis. Dewey wrote: “Renandoes not seem to have realized sufficiently the dead weight of intrenchedclass interests which resists all attempts of science to take practical form andbecome a ‘social motor’” (EW, Vol. 4: 17). (See Chapter 10.)

His most significant treatment of the critical issues is perhaps his 1927 ThePublic and Its Problems. Dewey there offered a radical critique of the prob-lem of democracy and concluded his account with a chapter entitled “TheProblem of Method.” He affirmed some characteristic Deweyan themes, e.g.,the “absolutist” character of political philosophies and the diversion ofthought away from fruitful questions. In this context, he affirmed Veblen’spoint, that while we willingly spent money responding to “results of bad conditions,” we needed to identify the causes of our problems. Forgetting his

Pragmatic Philosophy of Science and the Charge of Scientism 27

Page 57: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

earlier insight, he argued that the reason for this “anomaly” was “clearenough”: “There is no conviction that the sciences of human nature are farenough advanced to make public support of such activities worthwhile.” Yet,if on this issue he had been nearer to the truth earlier, there was a deeper prob-lem. It was the problem of democracy, “the problem of the public,” the inca-pacity of citizens to communicate freely, to overcome “secrecy, prejudice,bias, misrepresentation, and propaganda as well as sheer ignorance,” to re-place these “by inquiry and publicity,” and thence to act so as to overcomepresent problems. This was the deep problem, since for Dewey, if social sci-entific knowledge of the causes of our problems was the work of “experts,”the problem of putting such knowledge to work so as to make possible valuesactual in life, was the problem of the Democratic Community. If people wereunable to act “intelligently,” it was because they lacked the means. If peoplewere now incapable, it was because they were incapacitated by the conditionsof life, economic, political, and pedagogic.27 Nor did people need to be “ex-perts:” “what is required is that they have the ability to judge of the bearingof the knowledge supplied by others upon common concerns.” It was in thissense, indeed, that “scientific method” had to be the method of all, and it wasin this sense that it was part of the democratic mode of life. (See Chapter 8.)Dewey tried desperately to convince us that we must “apply science to life,”but even given the difficulties in his conception of science and even given hisfrequent use of “control” metaphors, his commitment to democracy kept hisvision from being technocratic. (See Chapters 14 and 15.)

In this paper I have suggested that pragmatism was a novel and liberatingphilosophy but that its fundamental insights became distorted as they becameabsorbed. The epigones and enemies of American pragmatism have beenpleased to make it into the American philosophy of technocracy, celebratingpositivist science and bourgoise society. This is profoundly ironical. Al-though, of course, there are very large differences in their philosophies, itwould be fairer to say that the project of the pragmatists, like Marx’s, was toassist us in de-alienating our increasing alienated world. (See Chapter 9.)


1. The many revisions of my essay have profited from Hollinger’s work.2. It is also customary to hold that Veblen’s essay was a direct critique of Dewey

and of pragmatic philosophy. This is symptomatic. Since Henry Waldgrave Stuart’sreview of Veblen’s remarkable “The Place of Science in Modern Civilization” (1906),Veblen’s “idle curiosity” has been widely interpreted as offered, as Diggins writes, inopposition to “both the Deweyite determination to make all knowledge expedient andthe Jamesian desire to allow man to believe what he wills to believe.” But it is clear

28 Chapter One

Page 58: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

enough that “idle curiosity” is a direct and acknowledged appropriation from James.Moreover, Veblen put to work Dewey’s important reflex arc paper in his argument,and as I argue below, Veblen and Dewey fully agreed that as “science” was part of theproblem, science was also part of the solution. As Diggins sees, Veblen “is careful toexpress great respect for John Dewey and William James,” but he fails to see, oddly,that this is exactly because Veblen is not attacking their pragmatism, but the alreadypersuasive technologism of American culture. For an example of the misunderstand-ings on these critical points, see John Diggins, The Bard of Savagery: Thorstein Veblen and Modern Social Theory (New York: Seabury Press, 1978): 30, 82f, and 182.Perry Miller is more cautious. See his “Introduction,” American Thought, Civil Warto World War I (New York: Reinhart, 1954): xlviii. More recently, see Dorothy Ross,“American Social Science and the Idea of Progress,” in Thomas L. Haskell (ed.), TheAuthority of Experts (Bloomington: Indiana University Press, 1984): 165 and DorothyRoss, The Origins of American Social Science (Cambridge: Cambridge UniversityPress, 1991).

3. James’s 1907 Pragmatism contributed greatly to confusion here. It is, of course,an exciting book which squarely faced “the present dilemma of philosophy”: “Ourchildren . . . are almost born scientific. But our esteem for facts has not neutralized inus all religiousness. It is itself almost religious.” Yet, James let his language carry himaway, as Dewey was quick to point out. See Dewey’s review, “What PragmatismMeans by Practical” (1908), The Middle Works, Vol. 4 (Carbondale: Southern IllinoisPress, 1977).

4. While it is not part of the present argument, if I am correct, Richard Rorty’s in-terpretation of Dewey falls into the family of misinterpretations which stem from theproblem discussed in what follows. Versus Rorty, Sleeper is correct in seeing that (1)Dewey did not try to overcome the tradition, but to transform it, and (2) Dewey didnot try to make everything scientific if by “science,” one means what Rorty means.Dewey’s conception, as I argue, was faulty, but not quite as faulty as the one promotedby Rorty and imputed to Dewey by him. Sleeper, John McDermott, and AbrahamEdel have considered Rorty’s views of pragmatism in the Winter 1985 issue of TheTransactions of the Charles S. Peirce Society, Vol. XXI, No. 1. See also James Camp-bell’s excellent treatment, “Rorty’s Use of Dewey,” Southern Journal of Philosophy,XXII, No. Vol. 2 (Summer 1984).

5. More generally, the idea that there is an “inseperable connection between ra-tional cognition and rational purpose,” is held by all the pragmatists. Remaining cita-tions from Peirce are from P. P. Weiner (ed.), Values in a Universe of Chance (NewYork: Doubleday, 1950).

6. The text looks elitist, of course, but this is doubtful. Not only must we acceptPeirce’s mix of irony and seriousness, but he was no doubt correct in judging both theeffectiveness of the method of authority and its “psychology.” Yet he may well be of-fering here that, if we chose, we can all always use the scientific method. As I note be-low, he says “everybody uses the scientific method about a great many things.” Forsome interesting discussion, see Thomas L. Haskell, “Professionalism versus Capi-talism: R. H. Tawney, Emile Durkheim, and C. S. Peirce on the Disinterestedness ofProfessional Communities,” in Haskell (ed.), The Authority of Experts.

Pragmatic Philosophy of Science and the Charge of Scientism 29

Page 59: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

7. See M. F. Burnyeat, “The Skeptic in his Place and Time,” in R. Rorty et al.(eds.), Philosophy in History (Cambridge: Cambridge University Press, 1984). On thepresent view, if Peirce recast the epistemological problem and refused the transcen-dental move, Dewey, and James, in his radical empiricism, refused the transcenden-tal move and the epistemological problem.

8. For discussion of these figures, see John Passmore, A Hundred Years of Phi-losophy (Harmondsworth: Penguin, 1966), Chapter 14. My definition of “positivism,”it should be noted, has two components. The label was not true of Helmholtz, norlikely of Hertz, nor, as I shall argue, of Peirce or of James in Principles. An enor-mously useful account of Helmholtz, pertinent to the present essay, is to be found inM. Mandelbaum, History, Man and Reason (Baltimore: Johns Hopkins Press, 1971).

9. See Fred I. Dretske’s “Laws of Nature,” Philosophy of Science, Vol. 44 (1977).If the properties are real and abstract, that is, sui generis real, we have a Platonism.On the other hand, as Peirce also suggests, if they are but abstracted real properties ofcomplexes, we can have a far more modest realism.

10. See Elizabeth Flower, “The Unity of Knowledge and Purpose in James’s Viewof Action,” in W. R. Corti (ed.), The Philosophy of William James (Hamburg: FelixMeiner, 1976).

11. Although it cannot be pursued here, the point is more fundamental than mayappear. Experiment, properly understood, implies that causal laws cannot be se-quences of events and that causal laws continue to operate under “open” conditions,when there are no empirical invariances. See Roy Bhaskar, A Realist Theory of Sci-ence. 2nd Edition (Atlantic Highlands, N.J: Humanities Press, 1978): 33–36.

12. There can be no doubt that Helmholtz had a realist view of causality. For ex-ample, he wrote, “the word Ursache (which I use here precisely and literally) meansthat existent something (Bestehende) which lies hidden behind the changes we per-ceive. It is the hidden but continuously existent basis of phenomena,” (Selected Writ-ings, Russell Kahl (ed.), (Middleton, Conn.: Wesleyan University Press, 1971). Seeespecially 521–26. Veblen concurred. Singling out Karl Pearson, he wrote: “Thoseeminent authorities who speak for a colorless mathematical formulation invariablyand necessarily fall back on the (essentially metaphysical) preconception of causationas soon as they go into the actual work of scientific inquiry (The Place of Science inModern Civilization and Other Essays [New York: Russell and Russell, 1961]: 15).And in a distinctly Jamesian formulation, he noted that “the concept of causation isrecognized to be a metaphysical postulate, a matter of imputation, not of observation;whereas it is claimed that scientific inquiry neither does not legitimately, nor, indeed,currently make use a postulate more metaphysical than the concept of idle concomi-tance of variation . . . ” (35). Dewey is much less clear, offering very positivist sound-ing utterances. See Chapter 4.

13. Compare Joseph Margolis, Pragmatism without Foundations (Oxford and NewYork: Basil Blackwell, 1986), esp. 101–4; 284–89.

14. Mill followed Comte in noting that “states of mind are caused either by otherstates of mind or by a state of the body” (Mill, Logic, Bk. IV, ch. 4). This last wasphysiology, since “sensation” always has “for its proximate cause some affection ofthe portion of our frame called the nervous system.” “The laws of the mind,” then,

30 Chapter One

Page 60: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

were 1aws regarding “the succession of states of mind”—British associationism, as itcame to be called.

15. In the radically rewritten second edition of his Principles of Psychology(1870), Spencer defended an indirect realism, which he called “transfigured realism.”Spencer distinguished physiology, “aestho-physiology,” that is the discovery of the connections between the data of consciousness and physiology and psychology. “. . . That which distinguishes Psychology from the sciences on which it rests [i.e.,physiology and aestho-physiology] is, that each of its propositions takes account ofthe connected internal phenomena (James’s “inner”] and of the connected externalphenomena to which they refer (James’s “outer”] [A psychological proposition] is theconnection between these two connections (Spencer, Principles of Psychology, NewYork: Appleton-Century-Crofts, 1950, I: 132). In his “Remarks on Spencer’s Defini-tion of Mind as Correspondence,” James gave a devastating critique of Spencer’sprinciple, the “adjustment of inner to outer relations,” a principle offered to explainboth life and “the entire process of mental evolution.” James argued that if “the as-certainment of outward fact” is supposed to be the evolutionary task of organisms,then the principle cannot be true: “‘Mind,’ as we actually find it, contains all sorts oflaws—those of logic, of fancy, of wit, of taste, decorum, beauty, morals, and so forth,as well as perception of fact” (8). But if “correspondence” is loosened to avoid ab-surdity, the principle is quickly seen to be vacuous: “Everything corresponds in someway with everything else that co-exists in the same world with it” (10). What Jamesseems not to have seen is that the flaw he correctly diagnosed was itself a result ofSpencer’s commitment to Humean causality, “causes” as empirical invariances.Mach’s influential physiological psychology, it might here be mentioned, made thesame assumption. For discussion of Spencer and Mach on these critical themes, seeManicas, History and Philosophy of the Social Sciences, Chapter 9.

16. Owen Flanagan Jr. has provided a reading of Principles along the lines of annonreductive neuropsychology in his The Science of the Mind (Cambridge: MITPress, 1984). See also my “Whither Psychology?” in J. Margolis, R. Harré, P. T. Man-icas, and P. F. Secord (eds), Psychology: Designing the Discipline (Oxford: BasilBlackwel1, 1986); Manicas, A History and Philosophy of the Social Sciences, Chap-ters 2 and 14.

17. For evidence that James was sensitive to the problems about explanation whichhad emerged at just this time, see his note for “the Sentiment of Rationality,” in Es-says in Philosophy: 340–41. A brilliant summary of twenty-five years of debate on thequestion is Pierre Duhem’s 1906 La Theories Physique; Son Object, Sa Structure(translated as The Aim and Structure of Physical Theory [Princeton, N.J.: PrincetonUniversity Press, 1954]). “To explain,” Duhem said, is “to strip reality of the appear-ances covering it like a veil.” Explanations, accordingly, are always metaphysical. Inagreement with Ostwald and Mach, then, “a physical theory is not an explanation. Itis a system of mathematical propositions, deduced from a small number of mathe-matical principles, which aim to represent as simply, as completely, and as exactly aspossible a set of experimental laws” (19).

18. See John J. McDermott, “Introduction,” Essays in Radical Empiricism,The Works of William James (Cambridge, Mass.: Harvard University Press, 1976),

Pragmatic Philosophy of Science and the Charge of Scientism 31

Page 61: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

and Gerald Myers’s introduction to James’s Principles of Psychology, Works: xi–xl.Myers offers that James “had set out to write a book that introduced psychology as anatural science. . . . But the project forced him, he confessed, to operate with an as-sumption that the philosopher in him seriously mistrusted” (xiii).

19. See Gerald Myers, William James: His Life and Thought (New Haven, Conn.:Yale University Press, 1986), Chapter 11. Myers comments, “the view that physicalthings are merely collections of sensations or sensible qualities (sometimes calledphenomenalism, sometimes Berkeleyan idealism) might be hard to defend on the du-alist premise of Principles, but it fits neatly into the scheme of radical empiricism,where everything is made of sensations or sensible qualities” (319). Myers notes also“if we examine the eight essays that represent James’s radical empiricism, we cannotdetect any effort to work out in technical detail the vague blueprint of pure experi-ence” (316). Myers suggests an explanation of this in citing James’s view that“philosophies are only pictures of the world which have grown up in the minds of dif-ferent individuals” (317). Do “pictures” need philosophical argument? Or perhaps,was James satisfied to leave this to “epistemologists”?

It is true that James denied, as per most phenomenalisms, that physical objects are“constructions” from private particulars. See Edward H. Madden and Peter H. Hare,“James’s View of Causality,” in Corti (ed.), 1976: 113ff. It was thus that as Jamessaid, “radical empiricism has more affinities with natural realism than with the viewsof Berkeley or Mill. . . . ” Yet, as Madden and Hare argue, James’s commitment to theprivacy of the particulars of direct experience was fatal, exactly because he then couldnot provide an adequate account of potentiality.

20. Dewey shared Hollinger’s view. In his own “Development of American Prag-matism,” Dewey wrote, “Peirce was above all a logician; whereas James was an ed-ucator and a humanist and wished to force the general public to realize that certainproblems, certain philosophical debates, have a real importance for mankind, becausethe beliefs which they bring into play lead to very different modes of conduct.”

The premature narrowing of context, audience, “discipline,” and definition of prob-lems in reading James has hermeneutic implications which are far more importantthan one is likely to think.

21. The psychological story will be but part of the story because it will need to besupplemented by a sociology of knowledge and a philosophical argument. See Chap-ter 6. For the pertinence of sociology of knowledge to Dewey’s naturalistic episte-mology, see Thelma Lavine, “Naturalism and the Sociological Analysis of Knowl-edge,” in Y. H. Krikorian (ed.), Naturalism and the Human Spirit (New York:Columbia University Press, 1944).

22. W. Donald Oliver remarks that “in the world of pure experience there are nomysteries.” Once the distinction between mental and physical is expressed in terms oftwo set of relations, “no explanation need be sought for the occurrence of an item ofexperience in the one or the other set of relations. Indeed, the very notion of expla-nation is deprived of meaning, hence to seek one is to fail to understand the import ofJames’s radical empiricism” (“James’s Cerebral Dichotomy,” in Corti, 1976: 36).

One might hold, equally, that James wanted to keep the basis of the relatedness of experience mysterious so as to make possible a source of order, which is non-

32 Chapter One

Page 62: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

naturalistic. See Eugene Fontinell, Self, God and Immortality: A Jamesian Investiga-tion (Philadelphia: Temple University Press, 1986).

23. A useful volume in this regard is S. Morgenbesser (ed.), Dewey and His Crit-ics (New York: Journal of Philosophy, 1977), esp. Sections II and III.

Sleeper argues that Dewey held to what he calls “transactional realism,” but seesthat Dewey “seems unable to recognize that what bothers Woodbridge is the conclu-sion that . . . real objects are not merely antecedent to being known, but antecedent tobeing had in experience, antecedent to experience altogether” (1986: 115).

24. The term “basement idealist” in reference to Dewey is T. V. Smith’s, myteacher at Syracuse.

25. Morgenbesser notes “Dewey need not be interpreted to argue against a realistview of the nature of theoretical entities, to argue against their reality. His argumentwas rather that we can understand the significance and point of postulating theoreti-cal entities or understand the role a theoretical term plays in a theory only if we un-derstand the use to which the theory will be put and the specific problems to which itis addressed” (1977: xvi). This is substantially the interpretation I have offered ofJames. While Dewey can be read in this way, Dewey sometimes seemed willing toaccept a straightforward instrumentalist view of theoretical terms.

We should note also here that to say this is not to say that Dewey’s philosophy oflogic or of language has much in common with recent empiricism. See Chapters 2and 4.

26. Sleeper rightly puts the Logic in the center of his interpretation of Dewey andshows that its ideas had antecedents as early as Dewey’s 1892 syllabus for “Course5.” But, of course, Sleeper had the advantage of reading the Logic while Dewey’searly followers did not. As Ernest Nagel’s near dismissal of the doctrines of the Logicshows, by the time it was published, the philosophers could no longer welcome itscentral message—when it was understood. See Nagel’s “Introduction” to the Carbon-dale edition of Logic.

27. Although Dewey had considerable influence on public education in the UnitedStates, the result was, for the most part, a caricature of his views. The appropriationof Dewey by interests antithetic to his was possible, at least partly, because he failedto communicate that his pragmatism was not, despite his protests, a scientism. Oneexample must suffice: When educationists began to seek professional status withinthe university, the question arose, Is there a science of education? James, Royce, andDewey were clear in insisting that the answer was in the negative. But as Silbermanargues, “most educationists ignored Dewey’s insistence that the study of education berooted in philosophy and the social sciences.” Instead, teacher education alienated it-self within the university and, in an orgy of empiricism, “the survey became the foun-dation on which the entire study of education and training of teachers was built”([Charles E. Silberman, Crisis in the Classroom, New York: Random House, 1970:428–29]). Thus, just as Dewey could be associated with Watson’s behaviorism, edu-cationists could absorb Dewey into their Taylorism and scientism.

Pragmatic Philosophy of Science and the Charge of Scientism 33

Page 63: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 64: Dewey Rescuing Dewey Essays in Pragmatic Naturalism



John Dewey is always included as a critical player in the development of psy-chology in America. But his relationship to this development is complicatedand, for this reason—and others, too, nearly always misunderstood. More-over, there is a double irony in this.

First, it is often assumed that he played a significant role in the develop-ment of behaviorist psychology, which quickly came to dominate Americanpsychology. In part, this depends upon the view, that as Boring put it, Amer-ican psychology “got its mind from Darwin” and dealt with “a mind in use,”and, as “an experimental science,” was to be pursued in terms already wellestablished in the biological sciences (1950: 553). It also depends on the as-sumption, related to this, that Dewey gave Progressivism its voice that “asthe twentieth century went on, psychologists would fulfill Dewey’s hopes.Psychologists would increasingly move out into society, remaking its mis-fits, its children, its schools, its governments, its businesses, its very psy-che” (Leahy, 1992: 277). Finally, the misunderstanding depends upon a dis-tinctly positivist reading of pragmatism, the genuinely distinctive Americanfamily of philosophies. Thus Hilgard writes of “a conception of scientificmethod that was continuous with the pragmatism of James and the instru-mentalism of Dewey” that “continued to influence the nature of psycholog-ical investigations in America even after Watson had dropped conscious-ness in favor of behavioral measurement” (Hilgard, 1987: 778). Dewey, the functionalists, and the behaviorists resisted mind/body dualism and focused in “adaptive behavior.” This allowed an easy collapse of the

Chapter Two

John Dewey and American Psychology

Page 65: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“functionalism” of James, and then of Dewey, into later versions of func-tionalism, the “characteristic American psychology” and from there to later“operationalist” versions of behaviorism.

Indeed, Dewey was powerfully influenced by Darwin and was a strongadvocate of “experiment,” and he often spoke not only of “the mind in use”but also of “control” (Hickman, 1990). But while Dewey was a leadingvoice in Progressive thought, it is easy to demonstrate that he was a stren-uous critic of the sort of scientism so well expressed in the text cited fromLeahy. (And see Chapter 1.) And, it was easy but wrong to think that thedifferences between positivism and variant forms of pragmatism were of lit-tle importance to the idea of “a scientific psychology.” Dewey was com-mitted to an empirical approach to mind and did reject dualist psychology,but his “instrumentalism” was no positivism. Indeed, it is not difficult toshow that he played nearly no role in the development of mainstream Amer-ican psychology. Part I of this essay addresses these confusions. It requiresproviding a reconstruction of the development of American psychology, ifonly in sketchy fashion, assaying what Dewey actually did and said, andclarifying some key concepts.

What, then of the second irony? Although as I shall argue, Dewey hadgreat hopes for “the new psychology,” and early on in his career, he identi-fied himself as a psychologist, but he subsequently abandoned psychologyand came to believe that it ill served what became his primary intellectualgoal, that philosophy must address not the problems of philosophy, but theproblems of humankind. While he continued to argue that “the nature of allobjects of philosophical inquiry is to be fixed by finding out what experi-ence has to say about them,” instead of getting answers from “scientificpsychology,” problems he was interested in addressing would respond to anew conception of inquiry, work which culminated in his 1938 Logic: TheTheory of Inquiry. This profound shift is missed primarily because Dewey’stheory of inquiry is so fundamentally in opposition to the dominating logi-cal empiricist theory of science, which had by then captured psychology,that it was misunderstood and then ignored. The second irony then is this:Having abandoned even the more refined forms of behaviorism, the cuttingedge of current work in psychology is so-called cognitive psychology. But,remarkably, not only does Dewey’s Logic, misunderstood when it is not ig-nored, give us prophetic insights into the most fruitful of these approaches,an ecologically oriented, biologically grounded cognitive science, butshows us decisively why “symbolic” AI (artificial intelligence) modelsmust fail. One wonders whether Dewey will be ill served once again? PartII of this essay addresses these complicated issues.

36 Chapter Two

Page 66: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


“The New Psychology”

One of Dewey’s earliest essays was an examination of “The New Psychology”(1884). Historians agree that “the new psychology” derived from WilhelmWundt who founded the first psychological laboratory in 1879 and whoseGründzüge der physiologishe Psychologie (1st Edition, 1873) was an enormoussuccess. But Wundt’s psychology was complicated and offered many not alwaysclearly consistent strands. Drawing on all the recent advances in physiology,neurology, and psycho-physics, Wundt seems to have put it all together: brain lo-calization, sensory psychology, will, memory, and cognition. Moreover, in con-trast to the reigning British tradition, mind was conceived as both active and uni-tive. Thus “apperception,” a central idea, was a psychic mechanism, which gaveus the power of “selective attention” and “discriminative judgment.” “Intro-spection,” which for Wundt was the “foundation of psychology,” licensed thisview. Third, for Wundt, mind also was social, an idea thoroughly developed inhis massive Volkerpsychologie (1900–1920). Finally, psychology earned its cre-dential as a science (and not a branch of philosophy) because it was “experi-mental” and because it disclaimed irrelevant metaphysical issues, e.g., the rela-tion of mind and body. To be sure, this made problematic the question of therelation of physiology to psychology, a question that still haunts inquiry.1

The entirely new American universities provided ample opportunity to in-stitutionalize the “new,” scientific psychology (Manicas, 1987: Chapter 10).In short order, American psychology would pre-empt all others. By 1900,some 42 psychology laboratories were established in American colleges anduniversities; by 1926, there were 117. Of the first of these, 13 of the foundershad taken degrees with Wundt.

We get a clear picture of Dewey’s assessment of the high importance of the“new psychology,” by looking at his essay, so entitled. Dewey first notes thatthe new psychology is part of the new Zeitgeist of the new sciences, includ-ing the advances in physiology, but Dewey sees a confusion. The commonview is that “some or all the events of our mental life physically conditionedupon certain nerve structures, and thereby explains these events.” But, he in-sists, “nothing could be further from the truth” (Early Writings, Vol. 1: 52).According to Dewey, “all the leading investigators clearly realize that expla-nations of psychical events, in order to explain, must be psychical and notphysiological” (52). Professor George Ladd was an exception. In Dewey’s re-view of Ladd’s Elements of Physiological Psychology (1887), Ladd is

John Dewey and American Psychology 37

Page 67: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

charged with holding to both “the older opinion” that physiological psychol-ogy is the science of the relations of mind and body; of the correlations of thephysical and the psychical” (EW, 1: 200) and to the new view, promoted byWundt, of “physiological psychology as a method, whose end in not the par-allelisms between the brain and consciousness, but the investigation of con-sciousness itself by physical methods” (200). This very well characterizesnearly all the Wundtian inspired work then going on in America, for example,as represented by the influential work of Dewey’s teacher at Johns Hopkins,G. Stanley Hall, and of E. B. Titchener, fresh from his PhD under Wundt. Itwas “experimental” but it relied heavily on introspection and its primary goalwas investigation of “consciousness.”

But, according to Dewey, advances in the biological sciences have had an-other direct effect on the new psychology: “To biology is due the conceptionof organism. . . . In psychology this conception has led to the recognition ofmental life as an organic unitary process developing according to the laws ofall life, and not a theater for the exhibition of independent autonomous facul-ties, or a rendezvous in which isolated, atomic sensations and ideas maygather, hold external converse and the forever part” (EW, I: 56).

This, of course, is directed at British-style associationist psychology, but aspart of this, Dewey, still the Hegelian, endorses an ecological conception andthe Wundtian premise that mind is social. Thus, “the idea of environment is anecessity to the idea of organism, and with the conception of environmentcomes the impossibility of considering psychical life as an individual, iso-lated thing, developing in vacuum” (56).

But there is more. As a “movement” the new psychology has certaingeneral features: “The chief characteristic distinguishing it from the oldpsychology is undoubtedly rejection of a formal logic as a method and test.The old psychologists almost without exception held to a nominalist logic”(EW, I: 58), a pronounced tendency especially among those “who pro-claimed that ‘experience’ was the sole source of all knowledge” (EW, I:59). Hume destroyed all relations except as “accidents” and “denied alluniversality.” But he did this on the basis of “purely logical models,” “ab-stract principles of difference and identity . . . put in the guise of psycho-logical expression.” The reaction to this, as in Kant, was to “fall back oncertain ultimate, indecomposable, necessary first truths immediatelyknown through some mysterious faculty of mind. . . . Such intuitions arenot psychological; they are conceptions bodily imported from the logicalsphere” (59). These ideas are, of course, familiar Deweyan themes, to berearticulated and developed in the remainder of his long life—even if, aswe shall see, they will have negligible impact on the psychology, whichcomes after “the new psychology.”

38 Chapter Two

Page 68: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Dewey’s hopes for “the new psychology” were not restricted to psychol-ogy only; he believed that it held enormous promise for philosophy itself. Inthree essays published in 1884 and 1886, Dewey offered a redefinition of phi-losophy and a new role for psychology: “in the ordinary way of putting it, thenature of all objects of philosophical inquiry is to be fixed by finding out whatexperience has to say about them. And Psychology is the scientific and systematic account of this experience” (EW, I: 123). To be sure, not any psychology will do. Dewey’s first effort was his 1886 Psychology, publishedfour years before James’s Principles of Psychology.

Dewey intended the book to be an introductory text in psychology, butwanted also that it be an introduction to philosophy. In his Preface, he re-marks: “How shall we make our psychology scientific and up to the times,free from metaphysics—which, however good in its place, is out of place ina psychology—and at the same time make it an introduction to philosophy ingeneral?” (EW, 2: 4). In good Wundtian fashion, the book covers all the char-acteristic topics, but despite the disclaimer of the preface, the argument iscouched thoroughly within a Hegelian frame.

The ground for this was set in the three programmatic essays just men-tioned. In “Kant and Philosophic Method” (1884), Dewey challenged the“method of ‘intellectualism’ begun by Descartes” (EW, I: 34) and he argues,as before, that Humeans, avowedly empirical, distorted experience. Kant’s at-tempted repair failed: “Though the categories make experience, they make itout of foreign material. . . . They constitute objects, but these objects are notsuch in universal reference, but only to beings of like capacities of receptiv-ity as ourselves. They respect not existence in itself, but ourselves as affectedby that existence” (EW, I: 39). Again, “the only conception adequate to ex-perience as a whole is organism,” a conception which Dewey found inHegel’s Logic (EW, 1: 42, 43).

The rationale for Absolute Idealism emerges more clearly in “The Psycho-logical Standpoint” (1886). “The fact that sensations exist before knowledgeand that knowledge come about by their organic registration and integrationis undisputed” (EW, I: 127). If one accounts for this by something not in con-sciousness, then it is not known and we have abandoned “the psychologicalstandpoint” for “an ontological standpoint” (EW, 1: 128–29). But “either thismatter is unknown, is a thing-in-itself . . . or is known, and then becomes oneset of the relations which in their completeness constitute mind—when to ac-count for mind from it is to assume as ultimate reality that which has exis-tence only as substantiated by mind” (EW, 1: 129).

The argument is straightforward: “From the psychological standpoint therelation of subject and object is one which exists within consciousness”(EW, 1: 131). Hence, materialism and all forms of dualism, as for example,

John Dewey and American Psychology 39

Page 69: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

the “Transfigured Realism” of Spencer, all fail. But so too does SubjectiveIdealism.

The essence of Subjective Idealism is that the subject, consciousness or mind,which remains after the “subject world has been subtracted,” is that for whichafter all this object world exists. Were this not so—were it admitted that this sub-ject, mind, and the object, matter, are both but elements within, and both existfor, consciousness—we would be in the sphere of an eternal absolute con-sciousness, who partial realization both the individual “subject” and the “exter-nal world” are (EW, I: 135).

Similarly, in “Psychology as Philosophic Method,” Dewey writes: “The rela-tion of Psychology to Philosophy now stands, I suppose, something like this:There is an absolute self-consciousness. The science of this is philosophy.This absolute self-consciousness manifests itself in the knowing and acting ofindividual men. The science of this manifestation, a phenomenology, is psy-chology” (EW, I: 136).

Before the century had ended, Dewey did become uncomfortable with hisHegelianism, abandoning it for the variety of naturalism, which now sostrongly identifies him, but it is essential to see that while he abandonedHegel and psychology, he never did abandon the seminal psychological in-sights that his Hegelianism afforded.2

William James and the Problem of a Scientific Psychology

But before we pursue the divergence between Dewey and the development ofacademic psychology, we need here to bring in the work of William James.Along with Wundt, James is often credited as being one of the founders. Aswith Wundt, James’s psychology was complex, freely drawing from a widevariety of sources and orientations. And like Wundtian psychology, it wasalso to be rejected.3

James himself played a role in this. Despite his extremely rich andprovocative beginning,4 there were some genuine tensions in his account, ten-sions which led him, ultimately to despair of the very idea of a scientific psy-chology. The center of this was precisely the problem that had led Wundt andDewey to abandon physiology. In James’s terms (in criticism of Spencer) ifthere was no “correspondence” between “inner” and “outer” relations, thenthe problem of knowledge was insoluble!5

In his 1894 Presidential Address to the American Psychological Associa-tion, “The Knowing of Things Together,” James returned to the problem of“the nature of synthetic unity of consciousness” and concluded that none ofthe available theories could be accepted. What of his own account in the Prin-

40 Chapter Two

Page 70: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ciples? He there had proposed “to simply eliminate from psychology ‘con-sidered as a natural science’ the whole business of ascertaining how we cometo know things together or to know them at all” (James, 1978: 87). As Deweyhad earlier also noted, “That we do know things, sometimes singly and some-times together, is a fact. That states of consciousness are the vehicle of knowl-edge, and depend on brain states, are two other facts.” At the time of the writ-ing of the Principles, he believed that “a natural science of psychology mightlegitimately confine itself to tracing the functional variations of these threesorts of fact.” But he now believed that this was a dead-end. Remarkably, af-ter struggling for some twelve years to write his great book, James concludedthat it was “a loathsome, distended, tumified, bloated dropsical mass, testify-ing to nothing but two facts: 1st, that there is no such thing as a science ofpsychology, and 2nd, that W. J. is an incapable.”6

One must dismiss James’s self-deprecation. But could he be right: There isno such thing as a science of psychology? Well, at least not quite in the waythat either Dewey or James conceived it. The last piece of our account regardsthe prior question: What would count for psychology to be a science?

Two things may immediately be said: It had to be experimental and it hadto be free of metaphysics. But what were to be the data of experiment? In-deed, this could not be answered without first deciding what was to be in-cluded as not metaphysical. Some experimental psychologists were interestedin establishing “functional relations” between physiological events and men-tal events. As noted, this was essentially the position of Ladd and James, butalso of the earlier psycho-physicists, Fechner, for example, and of some of the“old psychologists,” (e.g., Helmholtz and Spencer). Moreover, there was cur-rently available a powerful argument that such inquiry was not metaphysical.Ernst Mach, the eminent philosopher/physicist whose enormously influentialAnalysis of Sensations (1883) provided a clear and forceful statement of a“positivist”—antimetaphysical—theory of science which did this. He heldthat “sensations” are the data of all science. Versus the metaphysicians, in-cluding here Kant, and realist and dualist versions of Kant, Mach asserted:

For us, the world does not consist of mysterious entities which by their interac-tion with another, equally mysterious entity the ego, produce sensations, whichalone are accessible. For us, colors, sounds, spaces, times, . . . . are provision-ally the ultimate elements, whose connexion (sic) it is our business to investi-gate” (Mach, 1959: 29f.).

Since following Hume, “causes” are not “productive powers,” on this view ofscience, scientific laws are merely “connexions.” As Comte (who both intro-duced the term positivism and ably defined it) had argued, they are merely“invariant associations,” taking the form, “if this, then that.” This idea was

John Dewey and American Psychology 41

Page 71: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

powerfully propelled by Mach and his followers, including Karl Pearson and,at just about the same time, by Francis Galton, “the pioneer of a ‘new’ psy-chology in Great Britain” (Boring, 1950: 482). Galton accepted the idea thatquantitative measurement is the mark of a science and following AdolphQuetelet, applied the “law of error” to biological and psychological data, in-cluding, of course, “mental inheritance” (Boring, 1950: 476). At this point,“experimental psychology” and “the psychology of individual differences,”which employed the new statistical methods, were developing independentlyof each other. In the following decades, fully legitimated by a positivist the-ory of science, these research programs would converge.7

Or one could deny the explanatory relevance of physiology, as Dewey hadsuggested, and argue, with Titchener, for an experimental science of con-sciousness which included data gained by “trained” introspective experi-menters. Titchener also was deeply influenced by Mach. Indeed, as Hilgardsays, “because sensation was the unit out of which mental processes werebuilt, Titchener’s position was called structuralism, with its emphasis on thewhat of consciousness with somewhat less concern for the how or why (Hil-gard, 1987: 73–74). Attention to “function” would be the response to Titch-ener’s “structuralism.”

Pragmatism and Functionalism

In all the standard histories, “functionalism,” “the distinctive American psychol-ogy” figures heavily in the subsequent development. Moreover, the “pragma-tists” figure heavily in the development of fumctionalism from its beginnings inChicago to its variant forms elsewhere (Boring, 1950: Chapter 22; Hilgard andBower, 1966: Chapter 10; Hilgard, 1987: 73–103; Leahy, 1992: 285–90; Ben-jamin, 1988; Murray, 1983; Schultz, 1987).8 The so-called functionalist schoolincluded James Mark Baldwin, James McKeen Cattel, both trained by Wundt,and James Rowland Angell, who did an MA under Dewey and took on James ashis mentor at Harvard. In these histories, the “pragmatists” figure heavily in thedevelopment of “functionalism.” While Hilgard regards James as the guidingspirit, Boring remarks, “it was the philosopher, John Dewey . . . who was the or-ganizing principle behind the Chicago school of functional psychology” (Bor-ing, 1950: 539). Leahy holds that Dewey’s “important but tediously written”mid-1890s papers “furnished the central conceptions of America’s native psy-chology, functionalism” (Leahy, 1992: 281).

In these histories, the path from “functionalism” to “behaviorism is less clear,even if Dewey, as the mentor of Angell, Watson’s dissertation supervisor andthe premier functionalist, remains in the fuzzy background. Boring could note

42 Chapter Two

Page 72: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

that “Watson was a functionalist (with a small f) but he could not tolerate forlong the requirement of the Chicago school that even the animal psychologistmust take time to translate (sic) positively observed behavior into the vagueterms of inferred consciousness” (1950: 641). Indeed, with this not subtle era-sure of differences between pragmatism and positivism, Boring could concludethat “[operationism] was there all along,” recognized well before Percy Bridge-man and Vienna positivism offered their putative “advances” over the views ofAmerican psychologists. Similarly, Willard Day Jr. notes, in his otherwise veryhelpful account (1998) of the antecedents of behaviorism, that “James was . . .concretely influential in the thinking of certain individuals who are importantin the history of behaviorism” and goes on to name John Dewey and WilliamMcDougall. But he provides no argument as to how these two figure in this his-tory. Hilgard offers that “it would be a mistake . . . to think of Dewey’s influ-ence upon the psychology of learning as limited to the work of those who de-veloped functional psychology within the laboratories (1966: 299). He seesthree additional lines: learning as the “product of schoolroom practices, “soci-ological social psychology” “by way of Mead,” and Dewey’s influence as aphilosopher “carrying on the pragmatic tradition (300). In his later history, Hil-gard notes that Dewey-inspired Chicago functionalism became, with E. L.Thorndyke’s “connectionism,” a “stimulus-response psychology.” Thorndyke,whose later work in educational psychology would cross paths with Dewey,had begun his animal studies in William James’s Cambridge basement and,contestably, was already a behaviorist. His influential “law of effect” is alsotermed “functionalist.” As Hilgard sees matters, although functionalism “de-clined as a recognizable school, it was destroyed by its own success, and in partby the success of its intellectual progeny,—behaviorism” (Hilgard, 1987:87–88). Leahy sees “functionalist” controversy, especially as regards the “mo-tor theory” of consciousness playing a key role. He offers that Thaddeus Bolton(1902) “integrated theoretical development since James, including the motortheory of consciousness and Dewey’s account of the reflex arc, into a theory ofperception embodying the coming behaviorist psychology” (Leahy, 1992: 286).For Leahy, Angell’s APA Presidential address (1906) was a “milestone on theroad to behavioralism” (288). As Angell later concluded:

There is unquestionably a movement on foot in which interest is centered in theresults of conscious, rather than the processes themselves. This is particularlytrue in animal psychology; it is only less true in human psychology. In thosecases interests in what may for lack of a better term be called “behavior”; andthe analysis of consciousness is primarily justified by the light it throws on be-havior, rather than vice-versa (cited by Leahy, 1992: 305).

John Dewey and American Psychology 43

Page 73: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Boring was probably correct: “Watson touched a match to this mixture, therewas an explosion, and behaviorism was left” (Boring, 1950: 506). “Watsonfounded behaviorism because everything was ready for the founding. Other-wise, it could not have been done” (ibid.). With Watson’s behaviorism, sci-entific psychology had not only expunged all talk of “consciousness” but in-deed, had become a technocratic science of prediction and control.

Considering stereotypes of American pragmatism, its talk of the “cashvalue” of ideas, and Dewey’s association with Progressive political and socialtheory, it is easy to understand the taken-for-granted role of Dewey in the de-velopment of American psychology, especially when his remarkable 1896 es-say, “The Reflex Arc Concept in Psychology” is taken, remarkably, to be “oneof the most important arguments for the functional attitude toward the inter-actions (sic) between stimulus and response” (Hilgard 1987: 81), and evenmore remarkably, when we recall that in 1943, this paper “was chosen as oneof the most important articles ever published in Psychological Review”(Leahy, 1992: 282).

The confusions will take some unpacking. First, it is hardly clear what“functionalism” meant for those writers who got the label pinned on them.The distinction and labels evidently came from Titchener who got the ideafrom James (Boring, 1950: 542). Dewey’s orientation surely was “function-alist,” but that term carries a host of meanings. For Dewey (as for James),“functionalism” in psychology implied that there was an essential relation be-tween cognition and purpose. Indeed, this idea may be taken to be a definingidea of pragmatism. But more than this: For James and Dewey, it meant, aswith Wundt, that intentionality was the critical feature of minded behavior.Finally, functionalism entailed that mechanism in biology will not suffice.These come together, to be sure, in the remarkable essay on the reflex arc, tobe considered shortly.

But other senses were very much in the air. In addition to the mathemati-cal sense of “function,” Boring offers that Angell’s 1906 paper to the APA isuseful in this regard. Angell distinguished three conceptions of functionalpsychology. First, it may be regarded as “a psychology of mental operationsin contrast to the psychology of mental elements” (Boring 1950: 543, quotingAngell). Second, it may be thought of as the “‘psychology of the fundamen-tal utilities of consciousness,’ in which mind is ‘primarily engaged in medi-ating between the environment and the needs of the organism’” (543). Finally,“there is a broader view of functional psychology as psychophysics, that is tosay, the psychology of the total mind-body organism. Such a view leaves psy-chology room for the consideration of well-habituated acts, where conscious-ness has almost or entirely lapsed” (544). All three conceptions are open todiverging interpretations, from a mentalism to a dualism to a radical behav-

44 Chapter Two

Page 74: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

iorism. All, I think, would have made Dewey uncomfortable, even if the sec-ond, at least, gives attention both to the purposiveness of “mind” and of therelation of the “needs” of the organism in relation to its environment.9

Dewey’s 1896 Essay on the Reflex Arc

But the critical point is the failure to recognize the philosophical backgroundand fundamental point of Dewey’s famous essay on the reflex arc. It is trueand important that between perhaps 1891 and 1903, with the Studies in Log-ical Theory, Dewey had made a conversion to his distinctive version of natu-ralism. But it is equally true and important that this was as naturalism, whichcarried a huge Hegelian residue. Flower and Murphey say it well:

It is almost as if Dewey held off from naturalism until he should be able to in-tegrate with it those aspects of idealism which he regarded as philosophicallyimportant: the view of knowledge as organic and relational, the social characterof both self and knowledge, the unifying and purposive character of judgment.Dewey could not bring together those features with naturalism as long as thedominant model of the latter was atomistic . . . (1977: 820).10

This was precisely the burden of the reflex arc essay. Dewey acknowledgesthat “the idea of a reflex arc has upon the whole come nearer to meeting thedemand for a general working hypothesis than any other concept (EW, Vol. 5:96) and his essay is not intended “to make a plea for what it replaced.” Butthe new account, best intentions notwithstanding, suffered from all the fea-tures of the older account. “The dualism between sensation and idea is re-peated in the current dualism of peripheral and central structures and func-tions; the older dualism of body and soul finds a distinct echo in the currentdualism of stimulus and response” (96). Thus, “the sensory stimulus is onething, the central activity, standing for the idea, is another thing, and the mo-tor discharge, standing for the act proper, is a third.” But if so, it is impossi-ble to see how action can be thought-guided or how we can learn? Experienceshows not only that we do, but what is amiss: the reflex arc is not “a patch-work of disjointed parts, a mechanical conjunction of unallied processes.” Itwas not to be understood “mechanically” but “functionally.” Rather, it is “acomprehensive or organic unity” (97). Dewey could now reject mechanismand atomism from a fully naturalistic point of view.

Consider James’s familiar child-candle example. First, contrary to the pre-vailing view, it does not begin with a “stimulus.” The real beginning is an actof seeing. This act stimulates another act, the reaching, but both are bound to-gether, subordinate elements of a larger coordination so that the seeing controls the reaching and the reaching, in turn, stimulates and controls the

John Dewey and American Psychology 45

Page 75: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

seeing. It is now “seeing-for-reaching purposes.” At the next stage, there isanother sensorimotor coordination: Indeed, “only because the heat-pain qualeenters into the same circuit of experience with the optical-ocular and muscu-lar quales, does the child learn from the experience and get the ability to avoidthe experience in the future.” The act has become not merely seeing, but “see-ing-of-a-light-that-means-pain-when-contact-occurs” (98).

Dewey has given a description, a point of departure for a psychology. Buthe has not given us a psychology. While for Dewey, intentionality is a funda-mental feature of all learning, he surely has not given us an account whichprovides the mechanisms for this.11 Surely, it will be a very messy psychol-ogy including as it does a revisioning of the role and relationships of the all-critical elements. And it can be naturalistically implemented. But that is notthe point here. The point rather is that while subsequent S-R psychologyadopted the language of “function,” it failed utterly to take seriously Dewey’s1896 criticisms, and continued, happily, with the development of an atom-istic, mechanical and, ultimately, mindless psychology.

It is striking that Boring’s account of Dewey’s paper is both sympatheticand generally accurate, but that he fails to see the consequences of Dewey’scriticism for all later functional and behaviorist thinking. Instead, he remarks“Dewey was anticipating the position of Gestalt psychology” and “occupyinga position in the history of dynamic psychology (Boring, 1950: 554). Of per-tinence here is Boring’s view that American psychology protested Wundt,first with functionalism and then with behaviorism, while German psychol-ogy protested and got Gestalten. Boring concludes, “given everything else inAmerica as it was in 1920, the year of James’s death, you could not have hada protest against Wundt developing as Gestalt psychology—not there and notthen” (1950: 643f). Nor indeed on the present argument, does it seem thatDewey’s far more radical alternative protest could have been grasped andtaken seriously.

Others are even less clear as regards the radical character of Dewey’s po-sition. Hilgard in assessing the very much later loss of hegemony of S-R psy-chology, writes that “some dissatisfaction with the stimulus-response concepthad been expressed as early as Dewey’s (1986) criticism of the reflex arc con-cept and that Thurstone had reiterated some of Dewey’s points in 1923against behaviorism” (Hilgard, 1987: 224), but Hilgard does not reconsiderthe opinion expressed earlier that this paper “gave one of the important argu-ments for the functional attitude” (81). Given the flabbiness of the category“functionalist psychology,” one might be tempted to hold, with Leahy, thatthe ideas of reflex arc paper were “the commonplaces of functionalism.” Butthe temptation should be resisted, especially since, more important, Leahygoes on to conclude that “Dewey’s formulation was centrally important to

46 Chapter Two

Page 76: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

later American psychology” (1992: 282). His grounds for this claim are of in-terest. Dewey showed, he wrote, that psychology could dispense with an “aninaccessible ego” and instead “account for the control of perception and de-cision in terms of coordinated, ever-changing adaptive behaviors (282). Thereis a sense in which Dewey did show this, but unfortunately, nobody followedhis sort of solution. Indeed, in the effort to sustain a genuinely “scientific”psychology, anything that hinted that “mind” was doing the coordinatingwould be either eliminated, or as Boring had noted, made safe by appropriate“translation.”12

Scientism and Scientific Psychology

J. B. Watson’s “Psychology as the Behaviorist Views It” (1913) is generallytaken to be the founding document. Watson had little patience for philosophyand was a diligent experimenter. For him, the “new psychology” had pro-duced no systematic body of knowledge (Boring, 1950: 642) and introspec-tion was incapable of producing consensus on anything. It was “more inter-esting to study behavior for its own sake, describing it and noting itsfunctional use to the behaving organism” (1950: 641). One could observe“discriminatory behavior” in both animals and humans. All reference notmerely to “will” or “attention” could be dispensed with, but so too “sensa-tion” and “perception.” All could be restated in terms of “discriminatory re-sponse.” Imagery and feeling were another matter. Watson denied both.13

In the ensuing years, “stimulus” will be employed to suit nearly any pur-pose, from a physical input or physiological event to “a situation or an in-volved object with meaning encrusted on it” (Boring, 1929: 586). All of this,to be sure, was “scientific” defined exactly in positivist terms. By the 1930s,Percy Bridgman’s notion of an “operational definition” was absorbed by thereigning “logical positivism.” Thus, propelled and legitimated by the wed-ding of traditional Humean empiricism and the extensionalist logic of Prin-cipia Mathematica, logical positivism could vindicate the so-called Age ofTheory. With “intervening variables” and “hypothetical constructs, S-R psy-chology could even offer gestures in the way of central processes” (Koch,1964). Indeed, it was during the heyday of the Age of Theory that the Psy-chological Review applauded Dewey’s reflex-arc essay, exactly because itseemed at least to allow for both central processes, as per Hull-Spence andteleology, as per Tolman, of course, all properly “operationalized.”14 If itseems unlikely that the judges could have misunderstood Dewey’s paper, wecan note also that Watson fully admitted being perplexed by Dewey. Al-though he had been drawn to Chicago to study with Dewey, in his autobiog-raphy Watson wrote: “I never knew what he was talking about then, and

John Dewey and American Psychology 47

Page 77: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

unfortunately for me, I still don’t know (cited by Fancher, 1979: 312; Robin-son, 1981).

Philosophically grounded in nominalist logic and Humean empiricism,psychology has become “objective” and “instrumental” with a vengeance.And in that same 1913 programatic statement, having redefined “experimen-tal psychology” in behaviorist terms, Watson could also see the future ofAmerican psychology.

Those branches of psychology which have already partially withdrawn from theparent, experimental psychology, and which are consequently less dependentupon introspection are today in the most flourishing condition. Experimentalpedagogy, the psychology of drugs, the psychology of advertising, legal psy-chology, the psychology of tests, and psychopathology are all vigorous growths(Watson, 1963: 158).

It should be emphasized that it mattered less that introspection had been ex-punged, but that “atomism” had prevailed, and that the theoretical (sic) goalof scientific psychology had been transformed from an effort to understand“the mind in use” to “the prediction and control of behavior” (Samuelson,1979; Manicas, 1987; Danziger, 1990). We can see this most clearly in thecase of “experimental pedagogy” and “the psychology of tests.”

Although the tools were created in Europe, Boring is correct to argue, “thepsychology of tests is essentially American.” And, as Lewis Terman insisted,it “brought psychology down from the clouds and made it useful to men”(quoted by Samuelson, 1979: 106). Poor Dewey would be again credited, ifnot directly, at least by innuendo.15 In 1929, Boring was clear that it was thefunctional psychology of G. Stanley Hall and Cattell that prepared “the psy-chological soil” for tests and measurements, but as with educational psychol-ogy, it was James and Dewey who provided “its philosophical sanction” (Bor-ing, 1950: 570). Indeed, in the pages preceding this remark, Boring presumesto have shown “how this spirit of America was crystallized by Dewey andAngell at Chicago” (545–46)!!

Similar confusions attend Dewey’s work in educational psychology, espe-cially as that is often confounded with the work of his colleague at Columbia,E. L. Thorndyke. Unfortunately, space forbids an extended discussion ofDewey’s contributions—and the misunderstandings of them. Briefly, every-one recognizes that Dewey emphasized “intelligent problem solving, inwhich each child solves the problems that are confronted by selecting appro-priate materials and methods and by learning to adapt these materials andmethods to his or her needs” (Hilgard, 1987: 674). This emphasis derived di-rectly from insights already set out in his early psychological writings, and,

48 Chapter Two

Page 78: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

of course, they were developed extensively in the work in experimental logic.It is also recognized, remarkably perhaps, that his “intellectual heritage” ofDewey’s views on education was a critical appropriation from Europeanthinkers, including Rousseau, Pestalozi, Froebel, and Herbart, “particularlythe doctrine of interest, which he apparently permitted to cover also theHerbartian concept of apperception” (Hilgard, 1987: 674). Indeed, as Hilgardappreciates, although both preached “scientific method,” in marked contrastto Dewey, Thorndyke was conservative, aiming not at innovation in theschools, but at “quality control,” a consequence of the fact that “Thorndykewas first of all an experimenter and measurer who valued data above all else”(671). This was a result, of course, of his version of “functionalism,”—S-R“connectionism,” and made him an ally of the more general movement intesting. As Hilgard remarks: “Largely as a consequence of Thorndyke’s in-sistence on measurement of all aspects of education, supported by the promi-nence that intelligence and achievement testing had received just before andafter World War I, the decade of the 1920s was one in which educational psy-chology flourished” (Hilgard, 1987: 682). But to repeat: this was emphati-cally not an educational psychology that was in any sense Deweyan. To besure, adding to confusion, Dewey’s efforts at reform went in parallel to thisdevelopment—even if in fundamental ways, they were at odds.

The fundamental problem here, as with “functionalism,” is the flabbinessas regards “pragmatism,” especially its relation to “positivism.” For example,Boring is good on the early relation of psychology to Machist positivism, andproperly identifies the importance of 1930s positivism as regards behavior-ism, but he makes no mention, in either edition of his influential books ofDewey “instrumentalism.” Hilgard discusses positivism and pragmatism, butis not at all clear, offering both that “there are distinctions to be made amongvarious forms of positivism and pragmatism” (Hilgard, 1987: 777) and thatthe conception of “scientific method” of later behaviorism was “continuouswith the pragmatism of James and the instrumentalism of Dewey” (778).Leahy links James to Mach and thence to the logical positivists who influ-enced behaviorism” (Leahy, 1992: 148), but omits reference here to Dewey.Despite his welcome concern with the philosophy of science, Leahy pays noattention to Dewey’s theory of inquiry nor to his views of science. More gen-erally, for him, the dominant influence of pragmatism was its role in promot-ing a scientistic Zeitgeist for America: “American psychologists . . . offered ascience with pragmatic ‘cash value.’ Pragmatism demanded that ideas be-come true by making a difference in human conduct.” For him, this entailedthat psychology be applied: adjustment, testing and control (Leahy, 1992:342).

John Dewey and American Psychology 49

Page 79: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Dewey and Academic Psychology Part Ways

After the mid-1890s, Dewey wrote nearly nothing, which could be said tofind a place in the emerging discipline of American psychology.16 Several re-views give us some additional insight into why this was so. His 1898 reviewof Baldwin’s Social and Ethical Interpretations of Mental Development offersa critical distinction: between examining the individual from the standpointof psychical process and determining what of this is social, and examiningnot the process but the content of the individual’s experience to discover whatthis has in common with others (EW, 5: 385–86). For Dewey, the first belongsto psychology, the second to sociology. Baldwin confuses these question be-cause he falls into a trap: Both “the individual” and “the society” are taken asgiven. Accordingly, “when we want to know about the individual we are re-ferred to society; when we want to know about society we are referred to theindividual” (388).17

This theme is elaborated in the address, “The Need for Social Psychology”(MW, Vol. 10), given in celebration of the 25th anniversary of the APA in1916. Ludy Benjamin (1988: 419) well captures the characteristic misunder-standing of Dewey’s essay. He notes that the address, given shortly after Wat-son’s manifesto, “linked behaviorism with the development of social psy-chology in the service of social control.”18 This not only capitulates to myth,but utterly misses Dewey’s central point. Arguing that “anything which mayproperly be called mind or intelligence is not an original possession but is aconsequence of the manifestation of instincts under the conditions suppliedby associated life” (MW, 10: 59), Dewey endorses Tarde’s view that “all psy-chological phenomena can be divided into the physiological and the social,and that when we have relegated elementary sensation and appetite to the for-mer head, all that is left of our mental life is, our beliefs, ideas and desires,falls within the scope of social psychology” (MW, 10: 54).

Although both “the application of statistical methods” and “the behav-ioristic movement” were just getting started, Benjamin would seem to havebeen misled by Dewey’s optimistic belief that both would contribute towhat, he believed, was needed. Thus: “Social phenomena are of a kindwhich demand statistical mathematics” and the behavioristic movement“transfers attention from vague generalities regarding social consciousnessand social mind to the specific processes of interaction which take placeamong human beings” (Dewey, MW, 10: 57). Indeed, Dewey (remarkably!)foresees “a great reflex wave from social psychology back into general psy-chology” (58). “The net outcome of the newer type of psychologicalmethod” will then be “an unexpected confirmation of the insight of Tardethat what we call ‘mind’ means essentially the working of certain beliefs

50 Chapter Two

Page 80: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

and desires, and that these in the concrete—in the only sense in which mindmay be said to exist—are functions of associated behavior . . .” (59). Weneed to be clear that his suggestion that “from the point of view of the psy-chology of behavior all psychology is either biological or social psychol-ogy” (63), was both radical and unheard.

Dewey’s optimism regarding the future of psychology should have beentempered. Indeed, just three years earlier, in a paper read at a joint sessionof the American Philosophical and American Psychological Associationson “The Standpoint and Method of Psychology,” he expressed fears aboutthe direction of “the behavioristic movement.” It was quite a thing to throwout “consciousness” as private and open only to introspection. It was quiteanother thing to throw out “mind” in the sense just noted. “To conceive be-havior exclusively in terms of the changes ongoing on within an organismphysically separate in space from other organisms is to continue that con-ception of mind which Professor Perry has well termed, ‘subcutaneous’”(MW, 7: 54). His criticisms paralleled those made against S-R psychology:“In so far as behaviorists tend to ignore the social qualities of behavior,they are perpetuating exactly the tradition against which they are nomi-nally protesting” (54).

So far as I can tell, Dewey explicitly discussed ongoing work in psychol-ogy only twice more. In 1927, in his essay, “Body and Mind” (LW, Vol. 3),he argued that in consequence of neglecting the development and historicalcareer of an individual, “an account of the mechanism of a particular move-ment of behavior is converted into an account of behavior itself and of be-havior in its entirety” (LW, 3: 33). “The criticism may be broadened to takein the whole reduction of mental phenomena to the stimulus-response type asthat reduction obtains in current psychological theory, even among those whodo not call themselves behaviorists” (3: 33–34, my emphasis).

By this time it was clear to Dewey that there was very little about academicpsychology that he could endorse. Already by 1903, he saw that what hewanted to say did not need psychology—at least as it was then conceived, that“logic,” articulated within a thoroughgoing naturalism was the way to go. Ofcourse, “logic” for Dewey did not mean what it meant for most. It was, as Ishall insist, a strong form of an ecological psychology.

We need not review the development of Dewey’s views on logic, whichincludes not only the The Studies in Logical Theory (1903) and How We Think (1911), but Human Nature and Conduct (1922) and Experienceand Nature (1926), along with a host of pertinent essays. Instead, we can directly consider his 1938 Logic, the text, which is the culmination of thisdevelopment.19

John Dewey and American Psychology 51

Page 81: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


Principia Logic and Empiricist Epistemology

“Logic” for Dewey regarded the theory of inquiry and inquiry, for him, wasa problem-solving activity. In 1958, Allen Newell, Clifford Shaw, and Her-bert A. Simon published “The General Problem-Solver,” a computer “simu-lation” of human problem solving. This provoked an entirely new directionfor scientific psychology. Behaviorism in the form that Skinner had taken ithad utterly discounted “central processes.” Age of Theory S-R psychology,from Hull-Spence to Tolman to Estes, had made gestures in the direction ofthese, but with the new technologies, an entirely different approach was pos-sible: “an information-processing paradigm that has generality across artifi-cial and natural problems-solving systems” (Wagman, 1998: 11).

Remarkably, Dewey’s remarks in 1884 regarding his hopes for the “the newpsychology” expressed a prophetic caution: Would the new cognitive psy-chology make “mental life” “a theater for the exhibition of independent au-tonomous faculties, or a rendezvous in which isolated, atomic sensations andideas may gather, hold external converse and then forever part”? (EW, I: 56).Indeed, would it, like “the old psychology” hold to “a nominalist logic” andreinstitute” formal logic as a method and test” (I: 58). As with Hume, would itproceed on the basis of “purely logical models,” “abstract principles of differ-ence and identity . . . put in the guise of psychological expression”?

Amazing as it might seem, this is exactly what it did. And the reason forthis turns precisely on the question of whether we should or should not acceptthe still dominant empiricist epistemology of which the logic of Russell andWhitehead is an essential element. For Dewey “logic” is the theory of inquiry,naturalistic envisaged. It is sufficiently general “to explain the behaviors ofsimple biological systems but also those of, say, a human scientific commu-nity” (Burke, 1994: 23). It was intended to replace “epistemology” as that hadbeen conceived. Nor was there a “foundation” of knowledge problem.“Logic” or “inquiry into inquiry” was for him “autonomous” in that it was “acircular process” which did not need foundations, either in epistemology,metaphysics, or psychology. Indeed, it was the supposition that it did that hadforestalled an adequate understanding of knowing. In a wonderful under-statement, he notes that “a sound psychology” may be a great advantage andthat “unsound psychology has done great damage” (LW, 12: 29).

For the standard view, logic is basically a formal theory of linguistic syn-tax and insofar as “true belief” is a function of both the “knowledge base” andthe “inference mechanisms,” it plays an essential role in epistemology. Insum, it is the aim of traditional epistemology to establish the grounds for dis-

52 Chapter Two

Page 82: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

criminating beliefs as either true or false. Thus, the primary vehicle is the sen-tence. Second, the problem of the knowledge-base takes the form of identify-ing “the proposition or meaning of a sentence with the information con-veyed.” Quine notes “some epistemologists would catalog [the] alternativesby introspection of sense data. Others, more naturalistically inclined, wouldlook to neural stimulation.” Third, “if logic is to be centrally concerned withtracing truth conditions through the grammatical constructions, an artificialgrammar designed by logicians is bound to assign the truth functions a fun-damental place among its constructions. . . . The simple sentences are got bypredication, and all further sentences are generated from these by negation,conjunction, and existential quantification” (Quine, 1970: 35–36). That is,standard logic is extensionalist logic. A consequence of this is the incapacityto deal with intentionality or to provide a convincing analysis of causality andof lawfulness.20 Finally, empiricist epistemology is epistemologically indi-vidualist in holding that there are beliefs for which social causes are whollyirrelevant. (See Chapter 6.)

I specifically select Quine here to represent current empiricist epistemol-ogy since the point to be made holds whether one is positivist, logical empiricist (neo-positivist) or post-positivist, a la Quine, whether or not, thatis, one holds to a firm analytic/synthetic distinction, to a nonholist verifica-tion theory of meaning or indeed, to one of the more recent varieties of “reliabilism,” “internalism,” or “externalism.”21 It can hardly be doubtedthat the blossoming of AI was profoundly constrained by assumptions takenuncritically from standard empiricist epistemology. Here is an early formu-lation:

The human brain is an information-processing system whose memories hold in-terrelated symbol structures and whose sensory and motor connections receivedencoded symbols from the outside via sensory organs and send encoded sym-bols to motor organs. It accomplishes its thinking by copying and reorganizingsymbols in memory, receiving and outputting symbols, and comparing symbolstructures for identify and difference.22

And this depends upon logic. As Zenon Pylshyn argues, ‘there is good reasonwhy computers can be described as processing knowledge.” This good rea-son, which owes, he says, to Hilbert, Gödel, Russell and Whitehead, and Tur-ing and Church, was this:

Reasoning about meaningful things—about things in the world or in the imagination—could be carried out by a process that itself knew nothing of theworld or of meanings, did not know what ‘thoughts’ were about . . .

The idea that logical inference can be carried out by a process of examiningmeaningless symbols leads directly to the foundational assumption of cognitive

John Dewey and American Psychology 53

Page 83: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

science, which is that thought is species of computing. . . . The bridge from for-mal symbol manipulation to computing was completed in 1936 by the mathe-matician Kurt Gödel who showed that anything that could be described in termsof manipulations of symbols could be carried out by a simple machine (latercalled a Turing machine), which became the defining property of reasoning andlater of intelligent action (Lepore and Pylshyn, 1999: 6–7).

For AI theory, “the intelligent organism is a sentential automaton, whose be-havior is the outcome of a sequence of mental states (beliefs that p, desire thatp, etc.) and the processing will be described in terms of the semantic and syn-tactic relations among the content-specifying sentences” (Churchland, 1980:188; Wagman, 1998: 25). Accordingly, AI inquiry can proceed, not only in-dependently of organism/environment relations, but independently of neuro-physiology as well.

This paradigm has not, to be sure, proceeded without criticism.23 An im-portant challenge comes from the so-called connectionist or neural networkparadigm. This approach assumes that sensory, motor and, ultimately, cogni-tive processes are explained in terms of inhibitory or excitatory connectionsbetween “nodes” which differ in strength. “The language of connectionism isdifferential equations rather than mathematical logic” (Wagman: 26). A brainmodel of the mind replaces the computer model of mind. Cognition is viewedas “the emergence of global states in a network of simpler components. In-stead of symbols, meaning resides in these emergent global states. Instead ofprocessing ‘information’ provided to mind by senses, minds create informa-tion for their own uses” (Varella, Thompson, and Rosch, 1991).24

An Ecological Psychology?

It is clear that while “the symbolic” approach is wholly inconsistent with aDeweyan approach, the connectionist approach is not inconsistent with suchan approach. I will round out this discussion with some brief remarks on this.First, and critically, in connectionist theory,

there will be nothing that corresponds to the classical symbolic data-structures.Instead, context-sensitive shifting coalitions of units will correspond to singleclassical representations. . . . Since there are thus no neat analogues to the clas-sical symbolic structures, the system cannot (not even tacitly) embody knowl-edge of transition rules defined over these very structures (Clark, 1990: 297).

Put bluntly, the constraints on an intelligent problem-solving device are “nolabels in the world, no external semantics, and no internal, unexplained ho-munculus in the loop to provide meaning” (Franklin, 1995: 301, explicating

54 Chapter Two

Page 84: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

the work of Edelman, 1992). These constraints were all fully articulated anddefended by Dewey.

Second, as Dennett has argued, there is nothing in connectionist theory, norin Dewey, which forbids that human symbolic capacities are a “recent addi-tion,” an evolutionary “enhancement” of mammalian cognitive architecture”(Dennett, 1991: 28). Finally, there is the question of whether attention to neu-ral networks will be sufficient to glean an understanding of problem-solvingintelligence.

The early chapters of Dewey’s Logic set out the naturalistic basis for“logic” in his sense, beginning with the obvious fact that when people in-quire, they “employ their eyes and ears, their hands and their brains” and that“these organs, sensory, motor or central are biological” (LW, 12: 30). Hugechunks of this read as if they had come directly from his reflex arc essay. Forexample, in decided contrast to the then current S-R theory, he wrote:

When the stimulus is recognized to be the tension in the total organic activity(ultimately reducible to that between contact activities and those occasionedthrough distance receptors), it is seen that the stimulus in its relationship to spe-cial activities persists throughout the entire pursuit, although it changes its ac-tual content at each stage of the chase. As the animal runs, specific sensory ex-citations . . . alter every change of position . . . (LW, 12: 37).

That is, there is no way to disconnect seeing and acting, nor to disconnectthese from the situation which is changing as the consequences of acting.Other passages sound like arguments in Human Nature and Conduct.

Habits are the basis of organic learning. According to the theory of independentsuccessive units of excitation-reaction, [both then current S-R and later learningtheory] habit formation can mean only the increasing fixation of certain ways ofbehavior through repetition . . .

Developmental behavior shows, on the other hand, that in the higher orga-nisms excitations are diffusely linked with reactions that the sequel is affectedby the state of the organism in relation to environment. In habit and learning thelinkage is tightened up not by sheer repetition but by the institution of effectiveintegration of organic-environing energies—the consumatory close of activitiesof exploration and search (38).


Even the neuro-muscular structures of individuals are modified through the in-fluence of the cultural environment upon the activities performed . . .

This modification of organic behavior in and by the cultural environment ac-counts for, or rather is, the transformation of properties with which the present

John Dewey and American Psychology 55

Page 85: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

discussion is concerned. . . . Any theory that rests upon a naturalistic postulatemust face the extraordinary differences that mark off the activities and achieve-ments of human beings from those of other biological forms (49).

As Dewey rightly emphasized, while lower organisms are proficient problem-solvers, there was no problem acknowledging that much human problemsolving does require “symbols” (and thus, linguistic capacities). But as he in-sisted, it was an “intellectualist” fallacy to impose this on all problem-solvingactivity. Moreover, as part of his powerful account of language, Dewey (andMead) insisted that the plateau of coordinated animal behavior is not irrele-vant to communication at the linguistic level even if it cannot be reducible toit. (See Chapter 4.)

As these quoted texts imply, Dewey’s approach is not “psychologistic” ifthat means (as it usually does) that “mental life” is “a theater for the exhibi-tion of independent autonomous faculties, or a rendezvous in which isolated,atomic sensations and ideas may gather, hold external converse and the for-ever part” (EW, I: 56). It is thus, if anything, an “ecological psychology.” AsBurke rightly notes, Dewey’s views compare to J. J. Gibson’s theory of per-ception, which stands in marked contrast to standard “psychologistic” theo-ries. “Ecological psychologists and Dewey share the view that perception isnot mediated by internal representational processes, which is not to hold thatit is not mediated by something. Perception is mediated, rather, by establishedattunements to lawlike relations among ways of acting in the world, that is byhabits . . . ” (Burke, 1994: 93); and more generally: “Perception and cogni-tion in general do not happen somewhere up in the head, but rather they in-volve an interactive information-processing mesh that cuts across a simplis-tic organism/environment distinction” (95). As cognitive scientist StanFranklin notes (following Skarda and Freeman, 1987; Edelman, 1992; andVarela et al., 1996), information is created not from sensory input but fromstructural coupling, the dynamical relations between “subsumption architec-ture,” accomplished by “competences,” “a series of incremental layers, eachlayer connecting perception to ‘action’” (Franklin, 1995).25

As the foregoing suggests, not all-current connectionist inquiry wouldseem able to deal with the problem initially posed by Dewey’s reflex-arch es-say. As Hanson notes: “Connectionist models have been fundamentally aboutsystem-level brain accounts.” And, indeed,

Without appreciating that commitment, it is hard to understand how a simpli-fied neuron model and synaptic connectivity could be informative for actualbrain function. . . . It is a common experience in the neurosciences to discovercells that behave in some orderly way without at the same time understandingwhat their larger purpose might be in terms of system-level function, which in

56 Chapter Two

Page 86: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

turn requires a deep understanding of the way cells interact and what emergentproperties might arise when millions of cells that code for spatial, temporal orstructural properties of the world begin to compute something (Hanson, 1999:425).

Hanson is surely correct that this problem does not go away by measuringmore cells, measuring them more precisely or measuring the molecular prop-erties of cells. But he is overly optimistic in supposing that the problem is suf-ficiently addressed by a systems-level theory “that takes into account somesimplified assumptions about systems of cells and simplified self-organizingprinciples such as learning.” That is, if Dewey is correct, we need to be talk-ing about an intact organism with a brain, including, ultimately, an organismwith a mind in Dewey’s sense, acting in an environment.26 This cannot besimulated by a computer. But it might be simulated by a robot, which could“explore, work in, and communicate results of its ongoing activities in distantplanetary environments” (Burke, 1994: 262). Burke offers that Dewey’s log-ical theory cannot tell us how to build such a machine, “but it does offer anumber of design principles which would, in his view, have to be treated asfundamental, not as goals to be achieved later, once other preliminaries aretaken care of” (263). Indeed, “such an approach to AI and robotics is actuallynot so foreign to work lately found in the cognitive science literature.”27 Fi-nally, one might want to claim that

Whether or not Dewey’s theory of knowledge is acceptable in every detail, histype of theory, namely, a naturalistic operation-based theory geared to explain-ing problem solving in concrete contexts, is the only one which holds any prom-ise for handling issues in the cognitive sciences which hinge on our knowingwhat knowledge is (265).

This essay argued that there are two ironies as regards Dewey’s relation to thediscipline of psychology. The first regards the belief that his pragmatism in-fluenced the development of psychology in America. The second irony thenis this: Having abandoned even the more refined forms of behaviorism, thecutting edge of current work in psychology is so-called cognitive psychology.But, remarkably, not only does Dewey’s Logic, misunderstood when it is notignored, give us prophetic insights into the most fruitful of these approaches,an ecologically oriented, biologically grounded cognitive science, but showsus decisively why “symbolic” AI models must fail. One wonders whetheDewey will be ill served once again? More generally, Dewey’s type of theorywould seem to be the only one that holds any promise for understanding theremarkable capacities of sentiment beings—including homo sapiens.

John Dewey and American Psychology 57

Page 87: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


1. That Wundt ultimately divorced psychology from physiology is now generallyagreed. See my account, A History of the Philosophy of the Social Sciences (Oxford:Basil Blackwell, 1987), 182f. which follows S. Diamond, “Wundt before Leipzig,” inR. E. Reiber (ed.), Wilhelm Wundt and the Making of a Scientific Psychology (NewYork: Plenum, 1980). In the last edition of the Gründzüge, he wrote:

Of the two tasks that are . . . implied by the name of physiological psychology—onemethodological, relating to the use of experiment, the other amplificatory, relating to thecorporeal basis of mental life—it is the former that is more essential to psychology itself,while the latter has value chiefly with respect to the philosophic question about the unityof life processes (quoted by Diamond, 1980: 169).

2. Flower and Murphey (1977) see rightly that his essay, “The New Psychology,”“reads like a preliminary comment for Experience and Nature” and that “armed withpost game wisdom,” all the “idealist” works have the promise of “the thoroughly nat-uralistic direction of his pragmatism” (820).

3. Hilgard refers to a Berlin newspaper account of the 1896 Third InternationalCongress of Psychology in Munich which described Wundt as “the psychologicalPope of the Old World” and James as “the psychological Pope of the New World”(Hilgard, 1987: 37). He offers a useful comparison of the two and rightly concludesalso “neither the psychology of Wundt nor the psychology of James persisted inAmerica in anything like their original forms (1987: 65).

4. Owen J. Flanagan, Jr. (1984) gives a sympathetic reading of James’s Principlesas “the first formulation of the naturalistic position in the philosophy of mind”(23–24).

5. Spencer’s “Transfigured Realism” (developed in his Principles of Psychology,first edition, 1855, and many thereafter) also had no influence in the development ofa scientific psychology for reasons similar to those that explain the rejection of Wundtand James. Briefly, Spencer distinguished sharply between physiology, an “objectivescience” and psychology whose data were “subjective.” But contrary to the mental-ism of associationist (and Wundtian psychology), and to materialism, inquiry couldnot be restricted to the laws of successive states of the mind. One needed also to knowhow these were connected with changes in the central nervous system (“inner rela-tions”) and then to the external environment (“outer relations”). He called this“aestho-physiology.” Similar moves were made by Gustav Fechner’s Psychophysics(1860). For James’s decisive criticism of these, see Chapter 1, note 15, and my “Mod-est Realism, Experience and Evolution,” in Roy Bhaskar (ed.), Harré and His Critics(Oxford: Basil Blackwell, 1990), 23–40.

6. Letter to Henry Holt, his publisher, May 1990, quoted from Ralph Barton Perry,The Thought and Character of William James, Two Vols. (Boston: Little, Brown,1935). Hilgard usefully discusses Gordon Allport’s 1943 article, “The ProductiveParadoxes of William James.” He sees six: the relation of mind and body, positivismand phenomenology, the self, freedom and determinism, association and individual-

58 Chapter Two

Page 88: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ity. James, suggested Allport, “did not make them productive by synthesizing his con-tradictory views, at least not the Principles, but left them as unsolved paradoxes”(Hilgard, 1987: 59). Chapter 1, argued that “radical empiricism” was, ultimately,James’s preferred solution, a solution similar to the one offered by Mach. But this “so-lution” similarly had no effect on the development of American psychology.

7. The history of the use of “variable” in psychological discourse was criticalhere. See Kurt Danziger (1997: 163–79). Causal relations (and hence lawfulness)could now be construed extensionally, as “relations of variables,” or “functions” inthe mathematical sense.

8. There is now available a three volume collection, The Chicago School of Func-tionalism, edited by John R. Shook (Bristol: Thoemes Press, 2001).

9. Dewey must take some responsibility. In his 1922/1925 “Development ofAmerican Pragmatism” (LW, Vol. 2), he did associate pragmatism with behavior-ism, but as the context makes clear, it was the idea that the brain was “an organ forthe co-ordination of sense stimuli” which led him to make the association.“Functionalism/mechanism” are still unsettled issues. See the essays by Stephen JayGould, Christopher Bourse, and William Wimsatt in Eliott Sober (ed.), Conceptual Is-sues in Evolutionary Biology (Cambridge: MIT Press, 1984).

10. On their view, the “drift” he reported in his autobiographical essay was “aneminently reasonable one in terms of the very ‘new question’ of the relation of psy-chological methods to philosophy” (Flower and Murphey: 819). In addition to thepsychological articles, including several not discussed here, e.g., “The Theory ofEmotion” (1884, 1885), “The Psychology of Infant Language” (1894) and “Interpre-tation of the Savage Mind” (1902), Flower and Murphey trace the “drift” also in therevised Study of Ethics: A Syllabus (1894), propelled by James’s Principles. The re-vision, remarked Dewey, was “in no sense a second edition. . . . On the contrary, [thenew studies] undertake a thorough examination of the process of active experience,and a derivation from this analysis of the chief ethical types and crisis—task , so faras I know, not previously attempted” (EW, 4: 221).

11. Mead was fully in agreement with the central issue. Similarly, in arguingagainst both Wundt and Watson, his problem was precisely to explain “mind” andmeaning in terms consistent with Darwin. See Chapter 4.

12. That is, if “behavior” is not merely “movement” then it involves intentional de-scriptions: reference to the object, goal or meaning that it has for the agent (Taylor,1964; Margolis, 1985). As we now appreciate, efforts to eliminate intentionality (asin Watson) or to “operationalize” it (as in Tolman), not only failed, but must fail.

Hilgard argues “there is a family resemblance between Dewey’s position and Skin-ner’s operant behavior, in which responses are coordinatd with the stimuli to whichthey lead” (Hilgard, 1966: 298 note). But, of course, Skinner’s “reinforcers” are vac-uous. As Taylor writes: “But although the property of having been reinforced is cer-tainly a property of the object itself separate from other properties of size, shape, etc.,it cannot as a stimulus property be separated. For in order for it to hold as a stimulusproperty the animal has to recognize the object as that object which was in fact rein-forced, one has to know some other description true of it beside simply that of ‘hav-ing been reinforced’” (Taylor, 1964: 133).

John Dewey and American Psychology 59

Page 89: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

13. It is true that many were not attracted to the more extreme physicalism of Wat-son. They could be “experimentalists” who, employing the standard S-R jargon coulddeal with “behavior” ambiguously understood. (See Kurt Danziger, 1997, Chapter 9).

14. See Estes et al., 1954 and for a critique, S. Koch, 1964, “Psychology andEmerging Conceptions of Knowledge as Unity.”

15. To my knowledge, Dewey never offered his views on the psychology of men-tal tests. There is an 1889 review of “Natural Inheritance, Galton’s Statistical Meth-ods,” in which Dewey both encourages the use of the new techniques and notes itslimitations.

16. One might want to include Dewey’s Human Nature and Conduct (1922),which attracted some attention from mainstream psychologists. But the combinationof Allport’s individualist social psychology, laboratory experimentation, for example,as in Thurston (1928) and Freud was sufficient to marginalize Dewey’s book amongpsychologists.

17. This, it may be noticed, is Dewey’s version of the rejection of what is nowtermed, “agent/structure dualism.” (See Chapter 4.)

18. See also Leahy’s comments on this paper. “[Dewey] offered his pragmaticistconception of mind as a social creation as the proper foundation for an experimentalpsychology. Since, on Dewey’s view mind was created by society, it could be delib-erately molded by society, and psychology, the science of the mind, could take as itsgoal, the scientific management of society” (1992: 342).

19. The fundamental work has been done, by Tom Burke, in his groundbreaking,Dewey’s New Logic (Chicago: University of Chicago Press, 1994).

20. See my essay “W. V. Quine” (2004), and M.T. Turvey, R. E. Shaw, E. S. Reed,and W. M. Mace, “Ecological Laws of Perceiving and Acting: In Reply to Fodor andPylyshyn, Cognition Vol. 9 (1981): 237–304.

21. See Susan Haack (1995) for the most thorough assessment of variants of “an-alytic epistemology.”

22. Morton Wagman, Cognitive Science and the Mind-Body Problem (Westport,Conn.: Praeger, 1998): 11, quoting J. A. Anderson, The Architecture of Cognition,1983. In an earlier pertinent context, Dewey had wisely remarked, “Those who areconcerned with ‘symbolic logic’ do not always recognize the need for giving an ac-count of the reference and function of symbols. . . . Any theory of logic has to takesome stand on the question of whether symbols are ready-made clothing for mean-ings that subsist independently, or whether they are necessary conditions for the ex-istence of meanings . . .” (LW, 1938: 27). The former view neatly characterizes“symbolic AI theory,” the latter idea is developed both in Dewey’s Experience andNature and in Mead’s Mind, Self and Society. (See Chapter 4.)

23. This is hardly the place to review this. For an excellent review, see Franklin,1995. In addition to the various writings of Churchland, 1980, see Flanagan, 1984 andSearle, 1992.

24. This approach has the much longer lineage going back to late nineteenth cen-tury neurophysiological speculations underpinning “associationist” psychology.Hebb’s account (1949) is “purely associationist and, in the form of simple vector dotpotentials, takes on the presumption of neuronal activations and synaptic potentials”

60 Chapter Two

Page 90: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

(Stephen Jose Hanson, “Connectionist Neuroscience: Representational and LearningIssues in Neuroscience,” in Lepore and Pylyshyn (eds.), What Is Cognitive Science?:404). McCulloch and Pitts (1943) offered a model of the brain as a specialized com-puting device and Rosenblatt developed a perception-learning machine, the Percep-tron, 1962.

AI inquiry (in contrast to “natural intelligence”) in the 1970s and 1980s over-whelmed these beginnings which more lately have been rediscovered and have issuedin a variety of new efforts, including so-called hybrid symbolic, connectionist mod-els.

25. See Rodney A. Brooks, Flesh and Machines (New York: Vintage, 2003). Seealso his website: Brooks’s work is discussed in auseful essay by Robin Marantz Henig, “The Real Transformers,” New York TimesMagazine,7 July 2007.

26. For an account of the limits of connectionism see Hans Radder, The World Observed/The World Conceived (Pittsburgh: University of Pittsburgh Press, 2006.

27. See the review by Judith Effken and Robert E. Shaw, “Ecological Perspectiveson the New Artificial Intelligence,” Ecological Psychology, Vol. 4 (1992): 247–70.See also references to Rodney A. Brooks and the review in Wagman (1998: 83–95).

John Dewey and American Psychology 61

Page 91: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 92: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


It has not been an easy matter to judge John Dewey’s relation to the social sci-ences in America. Most writers have held that his influence was significant.Some of these think that this influence was a good one; others are critical,since for them it contributed to what is seen to be a technocratic version ofsocial science.1 It is easy to infer what seems to have propelled this view:Pragmatism was an important and culturally influential philosophical move-ment in the United States Dewey was at Michigan and then at Chicago (withG. H. Mead) at what was the crucial period in the genesis of the social sci-ences in America; Dewey was distinctly interested in promoting a view whichincorporated science and the scientific frame of mind; hence, Dewey’s prag-matism must have left its mark on American social science. Those who findthat this “influence” was salutary also believe, I think, that on the whole aca-demic social science provides us with much needed knowledge.2

Although the premises are all true, the argument doesn’t work. It doesn’tmainly because the key ideas are mostly either mushy or ideological (orboth?). To see this, one must be clear not only about Dewey’s version of prag-matism, still very much contested but also about the character of social sci-ence in America. This last requires some concrete history governed by aphilosophically sophisticated understanding of science and its possible goals.For me, what may be termed “mainstream” social science is generally a dis-aster for substantially the reasons pointed to by Dewey’s erstwhile colleague,Thorsten Veblen. Veblen insisted that social science had as its task, “inquiryinto the nature and causes, the working and the outcome, of [the] institutionalapparatus.”3 Such inquiry need bear “no colour of iconoclasm,” since even ifit did not, its outcome “will disturb the habitual convictions and preconcep-tions upon which they rest.” Instead, “usages and conventions that have by

Chapter Three

John Dewey and American Social Science

Page 93: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

habit become embedded in the received scheme of use and wont, and so havebeen found to be good and right” are given scientific legitimation. The resultis “a ‘science’ of complaisant interpretations, apologies, and projected reme-dies” (1957: 136). Veblen’s objection, like mine and Dewey’s, is not that so-cial scientists were reformers but rather that they were not good scientists.

As regards Dewey, we are just now, I think, beginning to get clearer abouthis instrumentalism, despite the ill-conceived effort to appropriate it for post-modernist purposes, and even if the two most recent full length accounts, oneby an historian, the other by a philosopher, say precious little of any use abouthow it bears on Dewey’s conception of science, including social science.4 Inthis essay, accordingly, I want to develop Dewey’s scattered views on socialscience, both as he came to understand what they had become, and what theymight be. Much of what he did say gave ample room for both misunder-standing and misappropriation.5 Still, there remains in Dewey’s philosophysome untapped resources for reconstituting social science.

The Origins of Social Science in America

It is of considerable importance to notice that the modern disciplines of the so-cial sciences are an American invention that European universities had nothinglike what we now take for granted as social science. Indeed, some of the disci-plines were not part of European higher education until after World War IIwhen, as with so much else, Americanization became the order of the day.6

America provided the nearly perfect conditions for the modern idea of thesocial sciences.7 There was, first of all, “the social problem” produced duringthe Gilded Age by rapid industrialization, urbanization, and massive immi-gration. Second, America had a “weak state” in the sense that it lacked bothsignificant state bureaucracies and a strong central government. This pro-moted responses from “civil society,” but especially from the private collegesand universities. Third, lacking a feudal past, America was “bourgeois” fromits beginnings: As Bledsoe put it, “Americans lacked tradition as a source ofauthority, but they did not lack ‘science.’” Before Johns Hopkins became auniversity in 1876, there were no universities in America—the educationalupshot of the absence of a feudal past. Educational enterpreneurs could con-vince the John D. Rockefellers, Carnegies, and Mellons that science was justwhat was needed and that it could be produced with good effect in the newinstitutions. Finally, as science had itself been industrialized, a group of Eu-ropean philosopher/physicists had articulated a thoroughly positivist under-standing of the successful sciences, from the practically irrelevant idea of“science” as theoria, to a practically relevant productive and predictive in-

64 Chapter Three

Page 94: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

strument whose ultimate vindication was its capacity to generate technologies“for the relief of man’s estate.”

Indeed, for the men (sic) in “the institutions of higher learning,” the prob-lem was not class war, but ignorance. Social problems surely were no lesssubject to “scientific” solutions than other problems. Moreover, since forthem there was nothing fundamentally wrong with America’s basic institu-tions, these problems could be dealt with as technical questions in a piece-meal, ameliorative fashion. But if social scientists were to be professionalwith legitimate claims to authority and autonomy, they must mark out theirscientific territories, clear away all that was nonscientific, and establish theirown system of credentialing. What this meant was clear enough: It meant es-tablishing distinct disciplines exactly in the terms which they believed anytrue science must be constituted. The outcome, settled between the wars, wasthe disciplines of the social sciences, as we know them today. This was, then,the context in which Dewey reflected. Where, we may ask, did he fit in?


Dewey said very little about the social sciences and although one findsthroughout the corpus references to science, one finds too little in the way ofa systematic account of science. Most of the terms descriptive of science andin general use were—and are—vague and uncritically employed: for exam-ple, cause, law, theory, explanation, and experimental method. Dewey, likemost writers today could take these terms for granted even if, as I would in-sist, one can get contradictory conceptions of science from different analysesof them. This unclarity should not surprise us. What we now think of as animportant subdiscipline of philosophy, philosophy of science, emerged onlyin the 1950s and it was only in the 1970s that there has been a genuine com-petitor to the positivist interpretation of science.

Logic: The Theory of Inquiry, published in 1938, is surely the main excep-tion to the overall absence of texts on Dewey’s theory of science. What isthere is very important, but there are many important questions which Deweydid not address and, typically, he does not make much effort to place his ef-fort in the context of other writers on science, Vienna positivism, for exam-ple. When Logic was published, as Ralph Sleeper has argued, it was both ig-nored and misunderstood, so thoroughgoing were entrenched assumptionsabout logic and science. Moreover, by this time, systematic misunderstandingof Dewey was also well entrenched.8 Accordingly, it was not then and is notnow a genuine competitor for the received view.

John Dewey and American Social Science 65

Page 95: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

At bottom is an ill-developed conception of science, which is distinctlyDeweyan. On the positivist, technocratic conception, the aim of science isprediction and control. To achieve this one needs only confirmed regularities,laws or lawlike statements. On this view, since they can be no part of science,ends are given or assumed and the only question is means. Since it is notwithin the realm of social science to decide on goals, the social scientist (quasocial scientist) is “neutral” regarding who his work serves.

For Dewey, none of the foregoing was true, but for reasons noted, is not aneasy matter to get a firm grip of Dewey’s alternative view. There is a sense inwhich it is utterly unique, the consequence of his radical position in episte-mology. His “instrumentalism” involves a rejection of the “epistemologicalproblem,” and thus of the fact/value dichotomy. It offered a unique concep-tion of “control” and a confusing conception of the character of theory and ofthe goals of science. I try to keep my account focused, as much as possible,on what Dewey had to say about social science.


Modern philosophy, responding to the new science, has been haunted by “theepistemological problem,” the problem of justifying true belief. As argued inChapters 1 and 2, Dewey rejected the assumptions, which generated the prob-lem. Failure to see this has misled many otherwise astute commentators.Dorothy Ross, for example, singles out Dewey’s (1897) lecture, “The Signif-icance of the Problem of Knowledge” as a critical intervention on the side ofthe technocrats. But this is far from being the case: Its thrust is against tradi-tional foundationist epistemology: rationalist, sensationalist, and Kantian.Dewey writes:

Knowledge can define the percept and elaborate the concept, but their union canbe found only in action. The experimental method of modern science, its erec-tion into the ultimate mode of verification, is simply this fact obtaining recog-nition (EW, Vol. 5: 21).

Contrary to the epistemologists, there is no problem of knowledge in gen-eral: philosophy is “not an original fountainhead of truth.” And this meansthat for answers to questions about how knowledge is possible we need to look to psychology and social ethics—“including in the latter term all the related concrete social sciences, so far as they may give guidance toconduct” (22). Dewey’s project was to naturalize epistemology and moraltheory.9

66 Chapter Three

Page 96: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Psychology is naught but the account of the way in which conscious life is . . .progressively maintained and reorganized. Psychology is the attempt to state indetail the machinery of the individual considered as the instrument and organthrough which social action operates (23).


The sociologist, like the psychologist, often presents himself as a camp followerof genuine science and philosophy, picking up scraps here and there and piecingthem together in somewhat aimless fashion. . . . But social ethics represents theattempt to translate philosophy from a general and therefore abstract methodinto a working and specific method; it is the change from inquiring into the na-ture of value in general to an inquiry of the particular values which ought to berealized in the life of everyone, and of the conditions which shall render possi-ble this realization (23).

This is the stunning research program for social science, stunningly ignored.We need to be clear about this. Dewey believed, rightly, that a human sciencescould help us to understand ourselves: how we think and inquire and why,when thinking and inquiry is successful, it is successful. They would give usinsight into what were our genuine interests and purposes and their relations,and most obviously, they would give us an understanding of the obstacles inpresent arrangements, which keep us from realizing our genuine interests andpurposes. The human sciences would be emancipating in exactly the sensethat they would clear away misconceptions about ourselves and our arrange-ments and empower us to reconstruct the social world more in accordancewith our wants and aims.

Central to this project was the rejection of the bifurcation of fact and value,a further consequence of the mistaken assumptions that had generated “theepistemological problem.” In his Logic, Dewey argued that “most current so-cial inquiry” was marked by “the separation of theory and practice” (LW, 12:487). It is sound principle, Dewey says, that one should avoid making socialjudgments “on the ground of moral preconceptions, conceptions of what isright and wrong, vicious and virtuous” (489). But this is mistakenly convertedto the principle that one should make no evaluations about ends. These are,accordingly, precluded from inquiry. But “only recognition in both theory andpractice that ends to be attained (ends-in-view) are of the nature of hypothe-ses and that hypotheses have to be formed and tested in strict correlativitywith existential conditions as means, can alter current habits of dealing withsocial issues” (491).

If one wants a ready current example, consider poverty. What indeed, arethe possible ends-in-view of current policy and what, accordingly, are the

John Dewey and American Social Science 67

Page 97: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

existential conditions that are demanded for their satisfaction? It is easy todeal here with high abstractions, “getting people off welfare,” “getting peo-ple to work,” “ensuring that people can acquire skills and knowledge whichwill make them employable,” and to leave up in the air, unexamined, the req-uisite “existential conditions.” Although I cannot prove this here, I would in-sist that the modern social sciences must take large measure of responsibilityfor the shallowness of the usual understanding of problems like poverty andcrime.10

Moreover, it is easy to assume that “the problems which exist are alreadydefinite in their main features,” and if so, then inquiry could be aimed at find-ing the best methods of solution. The result is that “methods for resolvingproblematic situations are proposed without any clear conception of the ma-terial in which projects are to be applied and to take effect,” with often aworsening of the situation which generated the inquiry (LW, 12: 487). Theanalogy between current modes of inquiry in social science and prescientificmedicine was apt. As Dewey noted elsewhere, such practice was a combina-tion of empiricism and quackery: Without analysis, symptoms were re-sponded to in terms of handed down remedies. Of course, these sometimesworked. But as regards medicine at least, “it is now recognized that choice ofremedial measures looking to restoration of health is haphazard until the con-ditions which constitute the trouble or disease have been determined as com-pletely and accurately as possible” (488).

The poverty example again illustrates this: It is held that people are notworking and that present arrangements make them welfare-dependent. Thesolution is obvious: eliminate welfare. But it does not take much to see thatthe conditions which constitute the trouble begin with the absence of jobswhich would pay enough to take a family out of poverty and that one wouldneed here to be clear about a host of other attending steps and conditions tomake this possible.

The self-imposed constraints of “allegedly scientific social inquiry” alsoexplain the positivist penchant for “fact-gathering.” Dewey had attacked thisidea in his 1931 essay, “Social Science and Social Control.” Dewey offered“the existing limitations of ‘social science’ (Dewey’s quotation marks) aredue mainly to unreasoning devotion to physical science as a model, and to amisconception of physical science at that” (LW, 6: 64). In Logic, Dewey heldthat methods adopted “in the professed name of social science” are merely theform of genuine science since they fail “to observe the logical conditionswhich in physical science give the techniques of observing and measuringtheir standing and force” (LW, 12: 492). There are many places where Deweyassessed current social science as deficient. Moreover, it is surprising that theforegoing explanation of the deficiency is overlooked by Ross and other writ-

68 Chapter Three

Page 98: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ers who accuse Dewey of contributing to “scientism.” In this essay (as in theLogic), Dewey held: “. . . [T]he facts of social ‘fact-finding’ remain a mis-cellaneous pile of meaningless items.” “Since their connections with humanwants and their effect on human values are neglected, there is nothing whichbinds them together into an intelligible whole” (LW, 6: 65).

Dewey was surely aware that his colleagues, among them Merriam atChicago and Ogburn at Columbia had by then established “fact-gathering” as thegoal of social science.11 This was, of course, a main target of Robert Lynd’sKnowledge for What? (1939), a book that was both very Deweyan and verymuch out of the mainstream. Indeed, in a related section of Logic, Dewey de-veloped an argument that C. W. Mills will pick up in his 1959 SociologicalImagination. Dewey saw two one-sided distortions. The “positivist” school (histerm) singlemindedly directs itself as “fact-finding”—what Mills had called “ab-stracted empiricism.” But the opposing tendency “places its entire emphasis onconceptions” (LW, 12: 497)—what Mills called “Grand Theory.” “Facts are sub-sumed directly under ‘principles,’ the latter being regarded as fixed norms thatdecide the legitimacy or illegitimacy of existing phenomena and that prescribethe end toward which endeavor should be directed” (497).

There is another issue, part of his more general instrumentalist theory of in-quiry that needs to be introduced if we are to have any hopes of graspingDewey’s thoughts on science.


Dewey’s commitments to scientific method, his persistent attacks on inquirydetached from human concerns and his extensive use of technologicalmetaphors have caused enormous confusion, almost certainly because asDewey himself saw, modern science had not been the salvific force that it wasonce hoped to be. (See Chapter 1.)

Surely the most far-reaching attempt to illuminate Dewey’s philosophy interms of “technology” is Larry Hickman’s John Dewey’s Pragmatic Technol-ogy (1990). It may be that Hickman goes too far in asserting that “late in hislife, technology became a synonym for the very method of inquiry” (1991: 1);but Hickman wisely glosses Dewey’s “instrumentalism” by arguing that“Dewey goes beyond theory and beyond praxis to production: his concern iswith the making and testing of new entities including extra-organic tools aswell as goals and ideals” (15). “Science” in this sense is a more refined anddeveloped form of all inquiry.

Thus, in Logic, Dewey insists, “there is no sharp dividing line between common sense and science.” “Gradually and by processes that are more or less

John Dewey and American Social Science 69

Page 99: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

tortuous and originally unplanned, definite technical processes and instrumen-talities [were] formed and transmitted.” It was just these, which allowed for“control.” “Control”—as Hickman says, a synonym for knowledge—does notrefer to the subordination or domination of something. Rather, as Deweymakes clear enough, “control” refers to our capacity to apply intelligence suc-cessfully: to produce, adapt, adjust, accommodate, achieve, institute, identify,order, discriminate, and to “resolve” problems in many other sorts of ways.“Control” has been achieved when the problem, which generated inquiry, hasbeen resolved.

It is in this sense, also, that “practical” must be understood. These “techni-cal processes and instrumentalities” then become “the background of materi-als and operations which we term science” (LW, 12: 77). And, indeed,

Genuine scientific knowledge revived when inquiry adopted as part of its ownprocedure and for its own purpose the previously disregarded instrumentalitiesand procedures of productive workers. This adoption is the radical characteris-tic of the experimental method of science (LW, 12: 99, 388–89).12

But this does entail a collapse of science into technology in the sense that allinquiry has some immediate practical aim and surely not in the sense that wecan and should seek to dominate nature. All knowing is technological in thesense that if the problematic situation is to be brought under “control,” lan-guage, mathematics and/or artifacts of various kinds are required. Indeed,more generally, this is consequence of Dewey’s attack on the “spectator the-ory of knowledge.” But the difference between science and common sense isexactly that while commonsense inquiry “occurs for the sake of settlement ofsome issue of use and enjoyment,” scientific inquiry occurs “for its own sake”(LW, 12: 66–67).

Dewey’s position here is almost always overlooked. Dewey did not rejectthe (Greek) idea that inquiry could be aimed solely at understanding. He re-jected the bifurcation of theory and practice, the idea that one could under-stand anything without “tools” and without “experimental operations, involv-ing definite techniques” (LW, 12: 151, 420, 455). Of course, it would be hardto deny that understanding may well promote the development of technolo-gies—a key feature of late nineteenth century industrialized science. Thisleaves open the question of whether this was, as Dewey would sometimes atleast seem to suggest, the ultimate justification of science.13

I want to say more about “experimental operations,” but we need here tonotice that the continuity between science and common sense creates a veryspecial burden for social science. Cultural conditions impact all inquiry—acritical point for a sociology of science, but because “the physical” is “rela-tively independent of social issues,” “the influence of cultural conditions” is

70 Chapter Three

Page 100: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“indirect.” For example, “it is not possible . . . to separate nineteenth centurydevotion to exclusively mechanical conceptions from the needs of industry ofthat period.” In social science, by contrast, “prejudices of race, nationality,class and sect play such an important role that their influence is seen by anyobserver of the field” (LW, 12: 482). It is, however, more than annoying tonotice that Dewey did not, as far as I can tell, say much about how such prej-udices were influencing the social sciences.


Critical to any understanding of science is the conception of law and causal-ity. We can here briefly summarize the relevant conclusions of the previouschapters of this volume. First, Dewey rejected the most characteristic, evendefining features of empiricist philosophy of science: that “scientific laws areformulations of uniform and unconditional sequences of events,” and thatcausality must be defined in terms of such sequences (LW, 12: 437). Of all thedoctrines, which currently inform mainstream social science, these are surelythe most pernicious. Once accepted, we are committed to an event ontologyand a regularity determinist view of the universe: Whenever this, then that. Itis then also easy to assume a covering law model of explanation, and thus tohold also that prediction and explanation are symmetrical. One final conse-quence is the inability to conceptualize agency: the fact that persons makethings happen. But as Dewey rightly sees, “there are no such things as uni-form sequences of events” (LW, 12: 445).

Second, he argued, “atoms and molecules show a selective bias in their in-differences, affinities and repulsions to other events” (LW, 1: 162). These “se-lective biases,” he says, define their “essence,” a term Dewey used withoutprejudicing his fully processual view of the universe. But since on a realistview, the “things” of the universe are always related to other “things,” out-comes are never guaranteed. Thus, “iron as such exhibits characteristics ofbias or selective reactions,” but “iron as a genuine constituent of an organizedbody acts so as to tend to maintain the type of activity of the organism towhich it belongs” (195). In a living organism, it functions not to produce iron-oxide—as it would in a hinge—but to contribute to metabolism.

Moreover such moves are quite consistent with his idea that commonsenseinquiry is continuous with advanced science. Dewey gives some examples:“A good rain will cause the seeds that have been planted to grow.” The ex-pectations are “explained” by the unscientific person by attributing a powerto rain. The empiricist disallows this, but content with an effort to establishthe validity of the expectations, he does not seek to understand the “power.”

John Dewey and American Social Science 71

Page 101: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Dewey sees, rightly, that “from the standpoint of scientific inquiry, these ex-pectations are but material of problems” (LW, 12: 446). He may, however,miss the main point. That is, if the scientific problem is to try to provide amore refined regularity or to fill in ever-larger numbers of “variables,” thenhe has succumbed to the regularity determinist conception.14 On realistgrounds, the scientific problem is not, as positivists would have it, to close thesystem in order to make better predictions. Rather the scientific problem is toidentify what it is about the nature of water and of seeds such that a good rainwill (ceteris paribus) cause the seeds to grow. One needs a theory about per-tinent causal mechanisms, not a better analysis of the “variables.”

In Quest for Certainty, he argued against empiricist ontology, both of thenaive realist sort characteristic of Greek science and of modern sensationalistversions. “The experimental method,” he writes, “substitutes data for objects”(LW, 4: 79). “By data is signified subject-matter for further interpretation;something to be thought about. . . . Hot and cold, wet and dry, light and heavy,instead of being self-evident matters with which to explain phenomenon,were things to be investigated; they were ‘effects,’ not causal principles” (LW,4: 80). Hot, for example, is surely an effect of what is a most complicated“causal nexus,” a nexus that includes not only the properties of bodies, but organisms.

Nevertheless, Dewey’s view needs to be distinguished from a scientific real-ism, which holds that “things” have “causal powers.” For Dewey, causality is alogical category, not an ontological one. For Dewey, the empiricist rightly ruledout “occult” qualities, but then offered “a hybrid notion which took from com-monsense the idea of succession and from the science the idea of invariability ofconjunction.” But “the contents which are invariably related in a law are notevents, and . . . their relation is not one of sequence” (LW, 12: 446). As a rejec-tion of regularity determinism, this seems right. And while I do not think thatDewey’s positive account of causality is satisfactory, his rejection of regularitydeterminism was all that is needed to distinguish his views from the prevailingpositivisms in social science. This is generally missed.

For example, Ross (1991: 253) holds that Dewey’s “Psychology and SocialPractice” is another place where he endorses technocracy. Dewey argues thatthe teacher has a psychological theory, like it or not. “Teachers tell you that achild is careless or inattentive in the same final way that they would tell youthat a piece of paper is white.” But, insists Dewey,

it is only through some recognition of attention as a mechanism, some aware-ness of the interplay of sensations, images and motor impulses which constituteit as an objective fact that the teacher can deal effectively with attention as afunction (139).

72 Chapter Three

Page 102: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Dewey’s point is exactly that unless teachers have an understanding of thestudent as a psychosocial being, all their efforts are bound to be misdirected,ineffective, even destructive. It is only by understanding the psychological“mechanisms” of attention, memory, cognition, and judgment and the social“mechanisms” implicated in all experience and behavior that the teacher cancultivate the powers of the student.15 This is for Dewey a research program tobe satisfied. We are, he says, discussing the question of the role of psycho-logical science in education only because “we have as yet made so little head-way” (144).

Dewey’s use of the term “mechanisms” here is notable and suggests howfar he is from a regular determinist view. This is made even clearer in a 1918essay entitled “A New Social Science,” one of the very few places whereDewey explicitly discusses social science. (The Logic is the other notableplace.) Dewey argues against the idea, inherited from Comte and Spencer—and still current—that “the existing social order is the product of natural lawswhich are expounded in a rational, scientific way” (MW, 11: 89). Dewey in-sists that World War I should finally have exposed this idea as myth: “. . . Thewar has revealed that our existing social situation is in effect the result of aconvergence of a large number of independently generated historic accidents”(90). Indeed,

Any science which pretends to be more than a description of the particularforces which are at work and a descriptive tracing of the particular consequenceswhich they produce, which pretends to discover basic principles to which socialthings conform, and inherent laws which “explain” them is, I repeat, sheermythology (90).

Dewey acknowledged radical contingency in the universe, a universe thatwas both “precarious and stable.” There were uniformities—a consequence of“selective biases” and there were plenty of surprises, a consequence of theopen systematic character of the world. But such a metaphysic calls for a his-torical and concrete social science. The “description of particular forces” atwork are the analogue of the “selective biases” discoverable by physical sci-ence. The “particular consequences” which they produce are not guaranteedin advance because the relations of such “mechanisms” are complex and his-torically contingent. There are no “general laws” under which we can sub-sume and thereby explain wars, revolutions or, for that matter, hurricanes orthe genesis of a species (Manicas, 2006, Chapter 5).

Dewey concludes this brief but rich essay by remarking, “there is . . . animmense amount of empirical subjectmatter contained within the confines ofexisting social sciences. The only trouble is that it has been ‘framed up’ and

John Dewey and American Social Science 73

Page 103: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

betrayed by its mythical and apologetic setting” (MW, 11: 91). He does not,unfortunately, elaborate on this very pregnant idea.


Dewey’s views on experimentation certainly did not help clarify his position.I noted that, for Dewey, one could not understand anything without “experi-mental operations, involving technique.” There is, of course, a paradigm,characteristic of the laboratory, but how far can this be extended? A long wayit seems.16 Thus, he insists “there is no ground whatever upon which a logi-cal line can be drawn between the operations and techniques of experimenta-tion in the natural sciences and the same operations and techniques employedfor distinctly practical purposes” (LW, 12: 434). But what counts here as the“same operations and techniques”? This text continues with what may be hismost general definition:

Experimentation is a form of doing and making. Application of conceptions andhypotheses to existential matters through the medium of doing and making is anintrinsic feature of scientific method (ibid).17

As before, if this is a consequence of his general criticism of the spectator the-ory of knowledge, there is no problem. On the other hand, Dewey did not, Ithink, have a clear understanding of the laboratory experiment as it is actu-ally practiced in the successful sciences and this allowed him to give the ideaa very extended sense. Only sometimes does he suggest that the main use ofan experiment is to test a well-articulated theory. On this realist view, theidea, roughly, is to deduce what the theory entails and then to establish ex-perimental closure to see if what was “predicted” by theory under closuredoes, in fact, obtain.18

But if we think of an experiment in this sense, as a situation in which a the-ory of a “mechanism” is to be tested, then, as is very plain, this is never pos-sible in social science—putting it at a distinct disadvantage. This is not, how-ever, what Dewey seems to have in mind when he speaks of experimenting insocial science. In the Logic, he remarks, “every measure of policy put into op-eration is, logically, and should be, actually, of the nature of an experiment”(LW, 12: 502; see also LW, 12: 486). Insofar as we should make the effort tosee as clearly as we can what consequences obtained after a policy was in-troduced, there is good sense to this. We know, for example, that people didn’tstop drinking alcoholic beverages when prohibition was enforced. But this isa test of a policy not of a theory of social behavior, exactly because, as Deweyclearly recognized, there are always a host of connected and interacting

74 Chapter Three

Page 104: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

processes involved which conjointly produced the actual outcomes. In thiscase, as we now know, “demand” for alcohol was satisfied by illegal produc-ers and distributors so that, if anything, the policy served to create crimi-nals—including law enforcement officers—and to deprive the society of anyeffective control of the production and distribution of alcohol.

In his 1931 “Social Science and Social Control,” alluded to earlier, Deweydid indeed sound technicist. He there offered that “The Five-Year Plan ofRussia . . . whether noble or the reverse, has many of the traits of a social ex-periment, for it is an attempt to obtain certain specified social results by theuse of specified definite measures, exercised under conditions of consider-able, if not complete, control” (LW, 6: 65). This is, in my mind, so much non-sense: Despite totalitarian methods of “control,” the outcomes were, as theymust be, conjoint products of a myriad of interacting activities of whichsome, at least, were directly contradictory to the intentions of the planners.Here “experiment,” and “control” get Dewey into unnecessary difficulty.

The example raises, as well, the question of the relation of democracy tosocial scientific knowledge. For the technocrats, one “controls” the condi-tions and gets “predictable results,” and because “experts” have knowledgethat the “masses” lack, democracy must give way.


It is easy enough to establish that World War I had a tremendous impact onDewey and that one of the consequences was his readiness to believe that thewar had brought forward “the more conscious and extensive use of sciencefor communal purposes.” (See Chapter 7.) It had “made it customary to uti-lize collective knowledge and skill of scientific experts of all kinds, organiz-ing them for community ends.” The warfare state, remarkably, had laid thefoundations for the Nationalist Liberalism, which became the political agendaof Dewey’s associates at the The New Republic. But when Walter Lippman,already persuaded of a technocratic version of social control, published hisPhantom Public in 1925, Dewey finally came to grips with the problem ofscientific knowledge and democracy.

In The Public and Its Problems (1927), Dewey agreed that there were ahost of “technical” questions which could be answered by “experts”: “sanita-tion, public health, healthful and adequate housing, transportation, planningof cities, regulation and distribution of immigrants, selection and manage-ment of personnel, right methods of instruction and preparation of competentteachers, scientific adjustment of taxation, efficient management of funds and so on” (LW, 2: 313). But the idea that such knowledge was sufficient was

John Dewey and American Social Science 75

Page 105: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

profoundly in error. Those who hold to such views “ignore forces which haveto be composed and resolved before technical and specialized action cancome into play” (LW, 2: 313). The problem is deep: “It is in the first instancethe search for conditions under which the Great Society may become theGreat Community” (327). The public is lost, eclipsed, inchoate, bewildered,caught in a drift, which it cannot grasp and therefore cannot overcome. Lipp-mann (and later C. W. Mills) was not wrong in diagnosing that the Americanpublic was a mass, but he was wrong in thinking that social scientists shouldnow rule. Dewey was clear that such “experts” lacked the knowledge that wasneeded. Indeed, “the prime condition of a democratically organized public isa kind of knowledge and insight which does not exist” (339). Citizens neededto understand what was happening and why. Some technical knowledge wasneeded, to be sure, but in the absence of a widely shared understanding of the“forces” at work, no democratic public could emerge.

Dewey is clearly correct in this analysis, but he is not as radical as he mightbe in assigning the causes of this. I put aside here the problems of distribut-ing “the kind of knowledge which does not exist,” for example, problems ofthe corporate control of mass communication, and concentrate here on therole of the social sciences themselves. In particular, while he acknowledgesthe limits of the special sciences in generating such knowledge, he does notseem to see that they contribute mightily to the mystification of what needsto be known. Instead of illuminating and emancipating, too much contempo-rary social science obscures and misleads.

Dewey gets his hands on some of the reasons for this. He notes that the“backwardness of social knowledge is marked in its division into independentand insulated branches of learning” (171).19 But this is more than a “mark” ofits “backwardness”: It guarantees backwardness. It is not merely, as he says,that there is lacking “continuous cross-fertilization,” but that fragmentationprevents us from grasping causes and connections. Thus we are told thatpoverty is a “psychological” or “cultural” problem: People lose initiative,lack ambition, look for the easy way. The sociologist assumes that this is“fact” (it is not!) and then tries to explain it. We are told the cause is “thebreakdown of the family” or “welfare dependency.” Moreover, our social sci-entist can, without risk, ignore an economic or historical analysis. She can,for example, altogether ignore the lack of decent-paying jobs and the reasonsfor this. The reasons can be left to the economists who tell us that markets areself-correcting and that, accordingly, a political analysis which calls for stateintervention is self-defeating.

Dewey notes also that specialized knowledge aims to be “abstract” whichpractically means that “it is not conceived in terms of its bearing on humanlife” (171). Plainly, the commitment to value-neutrality requires this. The up-

76 Chapter Three

Page 106: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

shot, of course, is not value-neutrality, but as Veblen insisted, scientific legit-imation of “usages and conventions that have by habit become embedded inthe received scheme of use and wont, and so have been found to be good andright.” Social science happily conspires in persuading us that the poor haveonly themselves to blame.

He argues forcefully that what counts as “news” in our daily papers is ren-dered completely unintelligible in terms of its connections but fails to arguethat this tendency is reinforced by “fact-gathering” social science. He is cor-rect that “a genuine social science would manifest its reality in the daily press,while learned books and articles supply and polish tools of inquiry” (347), butof course, it is precisely because “we” are not journalists but “social scien-tists” that we write jargonized “learned” books and articles. As Lynd said, weare either “scholars” or “technicians”—working for whoever will pay the bill.

Finally, for all of Dewey’s interest in education, he makes no mention ofthe disastrous consequences of current patterns of education in the social sci-ences. Instead of cultivating what Mills called “the sociological imagination,”we offer student’s textbooks, which guarantee disciplinary fragmentation,empty abstractions and uncritical thought. Instead of seeking causes and in-sisting on making connections, we require “disciplinary” integrity. Instead ofraising questions about “habits embedded in the received scheme of things,”we seek “relations of variables.”

Dewey was surely on the right track when, as early as his essay on Renan(see Chapter 1), he offered some reasons for these patterns of ideology anddisinformation. He then wrote that we do not yet appreciate “the dead weightof class interest which resists all attempts of science to take practical formand become a “social motor.” I conclude by saying that we still do not—itselfa function of the failure of the present practices of the social sciences.


1. For evidence that Dewey is often charged with holding to a technocratic versionof psychology and the social sciences, see Chapters 1 and 2. As regards Dewey andsocial science more generally, see, most recently, Dorothy Ross, The Origins of Amer-ican Social Science (Cambridge: Cambridge University Press, 1991). For Ross,Dewey’s position was “scientistic.” See the note that follows.

2. Ross does not celebrate academic social science in America. She argues that itis “scientistic.” “[Scientism] was the result of a long-standing commitment perenni-ally deferred, an effort to make good on the positivist claim that only natural scienceprovided certain knowledge and conferred the power of prediction and control. Withscience now defined by its method, scientism demanded that the requirements of nat-ural scientific method dominate the practice of social science” (Ross, 1991: 390).

John Dewey and American Social Science 77

Page 107: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

While I very much agree with Ross as regards much academic social science, herexplanation is very different than the one I offered in my History and Philosophy ofthe Social Science (Oxford: Basil Blackwell, 1987). She seems wrongly to assumethat the positivists provide a generally correct understanding of natural science. Forher, scientism arises only when the social sciences are constituted in positivist terms.Ross seems to favor interpretative models “available in history and cultural anthro-pology” or “the generalizing and interpretative model offered by Max Weber” (Ross,1991: 473). I argued in my 1987 book that there was a third, realist alternative. It al-lows us to incorporate the historical and hermeneutic and also to give social sciencean emancipatory role. See also my A Realist Philosophy of Social Science (2006). Isuggest in what follows that Dewey seems to have stumbled toward a fourth alterna-tive, neither positivist, realist nor “interpretative.”

3. Thorsten Veblen, The Higher Learning in America: A Memorandum on the Con-duct of Universities by Businessmen (New York: Sagamore, 1957: 132). This waswritten before World War I and published in 1918. It remains a wonderful account.Indeed, things have changed little. Dewey and Veblen agreed on much but it is hardto discern how the influences ran. Both had struggled with the views of Peirce.

4. See Robert B. Westbrook, John Dewey and American Democracy (Ithaca, N.Y.:Cornell University Press, 1991) and Alan Ryan, John Dewey and the High Tide ofAmerican Liberalism (New York: Norton, 1995). Westbrook and Ryan are helpful asregards Dewey’s views on democracy and politics. It is a bit surprising that Ryan paysso little attention to issues in the theory of science, given his interests. Ryan findsDewey’s Logic: The Theory of Inquiry “somewhat baffing” (1995: 309). This is a bitremarkable (though not unusual) since this is the place where Dewey makes his mostfundamental assault on the conventional wisdom.

5. I think also that his influence was minimal even though many parties, often con-flicting, were fond of drawing on him to suit their purposes. Early Chicago school soci-ology has some Deweyan marks, but it too moved toward positivism. See Lester R.Kurtz, Evaluating Chicago Sociology (Chicago: University of Chicago Press, 1984). G. H. Mead, of course, was the direct influence on symbolic interactionism and Mead’srelations to Dewey remains unclear. Symbolic interactionism, in any case, was always aminor competitor to mainstream social science. Perhaps C. W. Mills, for all his unhappi-ness with Dewey—for many of the right reasons—is the most Deweyan social scientist.

6. The most comprehensive comparative overview of the emergence of modern so-cial science is P. Wagner, B. Wittrock, and R. Whiteley (eds.), Discourses on Society:The Shaping of the Social Science Disciplines, Sociology of Sciences Yearbook, 1991(Dordrecht: Kluwer, 1991).

7. See my account, A History and Philosophy of the Social Sciences, Chapter 11,and Ross, Origins, Chapter 3.

8. See Ralph W. Sleeper, The Necessity of Pragmatism (New Haven, Conn.: YaleUniversity Press, 1986), especially Chapter 6, and Thomas Burke, Dewey’s NewLogic (Chicago: University of Chicago Press, 1994).

9. For a very useful account of Dewey’s theory of inquiry as epistemology see, H. S. Thayer, “Objects of Knowledge,” in John J. Stuhr (ed.), Philosophy and the Re-construction of Culture (Albany: State University of New Press). (See Chapter 6.)

78 Chapter Three

Page 108: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

10. See Stephen Steinberg, Turning Back (Boston: Beacon Press, 1995) and Her-bert Gans, The War against the Poor (New York: Basic Books, 1995) for review andcriticism of the contribution of recent mainstream social science to this impoverish-ment. More recently, see Alice O’Connor, Poverty Knowledge: Social Science, SocialPolicy, and the Poor in Twentieth Century History (Princeton, N.J.: Princeton Univer-sity Press, 2001) and my brief essay, “The Sociology of Poverty,” in Encyclopedia ofPoverty (Beverly Hills and London: Sage, 2006).

11. In 1929, Herbert Hoover assembled a distinguished group of social scientists“to examine the feasibility of a national survey of social trends . . . to undertake theresearches and to make . . . a complete impartial examination of the facts.” This wasfunded by the Rockefeller foundation with support from the SSRC, one of Merriam’sinspirations. Four years of work by hundreds of social scientists filled 1,600 pages ofquantitative research. The document called “the Ogburn Report” after its director,William F. Ogburn of Columbia, may be taken to signal the full maturation of American-style social science.

12. It is not altogether clear what Dewey has in mind here. He may assume that thescientific revolution of the seventeenth century owed to incorporation of techniquesderived from the workshops of craftmen. But it has been argued by Koyre, Butter-field, and Kuhn that as regards the development of the classical sciences: physics andastronomy, the scientific revolution was not a consequence of new experimental tech-niques, but of “new ways of looking at old phenonena.” Baconian experiments did,later, give rise to a large number of newer scientific fields which had their roots in thecrafts and in alchemy. See Kuhn (1977).

13. Having abandoned the quest for foundations, Dewey had to justify science asa mode of fixing belief “pragmatically.” But this need not give technology a privi-leged position. For discussion, see Chapter 6.

14. He writes “scientific inquiry proceeds by introducing qualifications. Theamount of arsenic has to be specified. . . . The conditions of the system into which itis taken have to be determined. . . . The presence or absence of ‘counteracting con-ditions’ has to be taken into account . . .” (LW, Vol. 12: 446–47).

15. Dewey very early on insisted that all behavior and experience was social, buthe did not say much about what this meant or entailed. See Chapter 4. For some al-ternative conceptions, see John Greenwood (ed.), The Mark of the Social (Rowmanand Littlefield, 1997).

16. In some places, he suggests that a mind experiment may be quite sufficient; inothers he seems at least to deny this. For example:

Experimental operations change existing conditions. Reasoning as such, can providemeans for effecting the change of conditions but by itself cannot effect it. Only executionof existential operations directed by an idea in which ratiocination terminates can bringabout the re-ordering of environing conditions required to produce a settled and unifiedsituation (LW, 12: 121).

17. See also LW, 12: 458. As a consequence of his rejection of “the spectator the-ory of knowledge,” it is clear that all inquiry requires experimentation for Dewey. He

John Dewey and American Social Science 79

Page 109: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

says also that there are at least three needs satisfied by experimentation: the institu-tion of data, the elimination of material that is irrelevant to the problem at hand, andthe generation of new existential materials.

He generally seems to have in mind something much like the “methods” that Millhad provided in his Logic (LW, 12: 190). But sometimes, experiment seems to be ex-ploratory in its aim (LW, 12: 317), close indeed to the Baconian idea of “twisting thetail of the lion.”

18. For a systematic discussion of “experiment” see Roy Bhaskar, A Realist The-ory of Science, 2nd Edition (Atlantic Highlands, N.J.: Humanities Press, 1978).

19. This feature of Dewey’s concern with the existing “disciplines” of social sci-ence is rarely acknowledged. See also Logic (LW, 12: 501–2).

80 Chapter Three

Page 110: Dewey Rescuing Dewey Essays in Pragmatic Naturalism



My title intends to associate my effort with Dewey’s Experience and Nature,better titled, he later thought, Culture and Nature. My main interest is to re-consider naturalism in the light of recent debate in the philosophy of the so-cial sciences. Since at least Dilthey, the question of a human science had beenvery much contested. Could one hold, for example, that as there were physi-cal laws, there was social laws? Could one argue that explanation proceededin terms of these, just as, presumably, it does in the physical sciences? Couldone hold that, even ideally speaking, the terms of the social sciences could be“reduced” to terms in the “physical language”?

Indeed, antinaturalism could be defined as the view that epistemologicaland ontological differences in the domains of nature and culture demand awholly different methodology. Beginning with the Kantian cleavage betweenan empirical (phenomenal) realm subject to knowable law and an intelligiblerealm where agents are free, late nineteenth-century thought dichotomizedeklaren (causal explanation) and verstehen (interpretative understanding), thenomothetic and the idiographic, the domain of nature and the domain of his-tory. For antinaturalists, then, even if the methods of the natural sciences areapt for the investigation of nature, by virtue of the meaningful, linguistic, orconceptual character of the human sciences, the methods of the human sci-ences need to be toto coelo different. They require a hermeneutic, phenome-nological approach.

Chapter Four

Culture and Nature

Page 111: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


This bifurcation was the operative idea behind Maurice Natanson’s 1963much used anthology, The Philosophy of the Social Sciences. In his forward,Natanson wrote:

Two distinctly opposed philosophical attitudes are taken as polar positions un-derlying the social sciences: let us, for want of satisfying alternatives, call them“objective” and “subjective” Weltanschauungen (Natanson, 1963: viii).

As the book unfolds, sociologist and naturalist, George Lundberg is putagainst Simmel’s neo-Kantianism, essays by Ernest Nagel and C. G. Hempelon concept and theory formation in the social sciences are paired with one onthe same topic by Alfred Schütz, an essay by A. J. Ayer is juxtaposed with oneby Merleau-Ponty. But perhaps most striking was the exchange generated byThelma Lavine’s incisive essay, “Note to Naturalists on the Human Spirit.”

Lavine sharply criticized any naturalism that was “content to be defined bya principle of continuity of analysis conceived of in terms of experiment andempirical verifiablity. . . .” This amounted, she insisted, “to forfeiting its sta-tus as a positive, i.e., constructive philosophy” (Natanson, 1963: 252). Natu-ralists not only exaggerated experimentalism but they confused the method ofnaturalism with methods stipulated by naturalism for inquiry into all types ofsubject matter. Finally, Lavine charged that naturalists had “thus far been ableto satisfy its new-found concern with the human spirit by recommending themethod of experimentation to the social sciences.” By default, they had failedto show that naturalisms were not, finally, materialisms. On her view, theseweaknesses could be overcome, but only if naturalists developed a naturalis-tically reconstructed method of verstehen (254, 258).

Lavine’s essay brought sharp rebuttals from both Nagel and Natanson.Nagel found that “the ‘difficulties’ she claims to find in current naturalism areonly doubtfully genuine; and the specific recommendation . . . of question-able worth” (262). Natanson much approved of Lavine’s criticism but foundthat her recommendation was, finally, incoherent: “To reinvoke naturalisticcriteria as correctives for a reconstructed naturalistic method is to take a stepforward and follow with a step back” (282). For Natanson, since verstehenwas “foundational,” the “way out” was “the transcension of naturalism in fa-vor of a phenomenological standpoint” (283). Indeed, after saying that W. I.Thomas, Cooley, and Mead were “all representatives of the phenomenologi-cal standpoint,” Natanson offered that this “transcension” could be achievedby adopting the phenomenological stance of Edmund Husserl.

In what follows, I argue that both Nagel and Natanson were wrong and thatLavine was correct. But to do this requires rejection of mainstream, empiri-

82 Chapter Four

Page 112: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

cist, neopositivist philosophy of science, including especially its characteris-tic philosophy of language and the (still dominating idea) that explanationproceeds by subsumption under law. Instead, I draw on Dewey and recentwork in the realist philosophy of science.1 In turn, I offer that in terms of thisview of science, a human science may be secured with a robust naturalism ofthe sort defended by John Dewey and George Herbert Mead.2 On the presentview, antinaturalisms thrive because, beginning with the debates of the lastdecades of the nineteenth century, both sides of the argument have shared inassumptions about both nature and culture and about what natural science is.They still do.


Nature exists independently of human activity. Society (and, not trivially,knowledge of nature) does not.3 On the present view, society is best con-strued as a relatively enduring ensemble of social relations, relatively en-during because social relations are incarnate in the activities of persons(Manicas, 2006; Giddens, 1984). There would be no society without hu-man activity. There would be no human activity without “culture,” broadly,everything which has meaning, including then, language and “the totalrange of material objects that are regularly used by people in mediatingboth their social and their environmental actions” (Byers, 1991: 3). Al-though the activity-dependent character of society has implications for in-quiry in the social sciences, this fact does not, emphatically, call for anti-naturalism.

The philosophical basis for such a naturalistic (yet, nonreductive) view ofsociety is hinted at by Marx (especially in The German Ideology), and devel-oped by G. H. Mead and Dewey. Alternative—and on the present view, badlymistaken “naturalisms”—are offered in the nineteenth-century positivist for-mulations of Spencer, Haeckel, and Engels and in more recent “eliminativematerialism.”4

For Dewey and Mead, life and mind are emergent evolutionary products;but as Tiles has argued, it is critical to see that most theories under this ban-ner amount “to little more than dualism back from the laundry” (Tiles, 1988:49). Characteristically, it is acknowledged that life and mind evolved, butthen argued that mind is consciousness and that its contents are “ideas” (or in-tentions), qualities directly known only to “subjects.” Within, then, this Carte-sian framework, meaning and communication require either reductioniststrategies, for example, verificationism, behaviorism, or they remain miracu-lous, at the very least wanting of some nonnaturalistic solution.

Culture and Nature 83

Page 113: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

A fully naturalistic posture will not merely allow mind (culture, meaningand society) an evolution from the sentient, but will reject what Dewey called“intellectualism,” “the two-fold error of operating with an incomplete (ab-stracted) picture of what is to have [an] experience of seeing and recognizing,and on the other hand, [imposing] on all experience, specifically sentient ex-perience, a structure which is present only in sophisticated (cognitive) devel-opments of sentience” (Tiles, 1988: 55).

One cannot, I think, underestimate the hold of “intellectualism” on philoso-phers, psychologists, and social scientists that seek an understanding of hu-mans and society. Semantic theories of language, current “cognitive science,”talk about “the cultural system” as in Parsonian-influence theory or more rad-ically, in the cultural work of Geertz or his opposite, Lèvi-Strauss, is anti-naturalist, even platonist in this way. Similarly, the Parsonian conception ofthe affective as providing only motivating, noncognitive role in action, recentrational choice theory and the idea that all knowledge is discursive knowl-edge each thrive on “intellectualism.”

For Dewey, three general “plateaus” are easily—and empirically—discerned, “each of which incorporates the function and relations of those be-low it, and is such that it cannot be understood in isolation from the level (orlevels) below it . . . ” (Tiles, 1988: 56). The first plateau is inanimate nature.In strongly realist terms, Dewey writes “atoms and molecules show a selec-tive bias in their indifferences, affinities and repulsions, when exposed toother events.” 5 Following James in his Principles, Dewey argues that “thingsare defined by means of symbols that convey only their consequences withrespect to one another. ‘Water’ in ordinary experience designates an essenceof something which has familiar bearings and uses in human life. . . . But H2Ogets away from these connnections, and embodies in its essence only instru-mental efficacy in respect to things independent of human affairs” (E&N:160). In the foregoing terms, H2O represents a theoretical entity, real but not“empirical.” For realist theory of science, “things,” both the things of ordi-nary experience and the highly abstracted theoretical things of advanced sci-ence, are metaphysical “compounds.” Ordinary table salt is a compound ofdifferent kinds of molecules even while it is mostly NaCl. NaCl, of course, isa theoretical entity, an item of the current ontology of science. Realist philos-ophy of science is strongly naturalistic in holding that “nature” exists inde-pendently of mind, even if the nature of nature is a scientific problem, to besettled by inquiry.6

In contrast to empiricisms, “laws” are not statements about empirical reg-ularities but assertions about the dispositional powers of things—their “se-lective bias’s”—and these are understood as “natural necessities,” in Dewey’sterms, “essences.” Theories are conceived as representations of enduring

84 Chapter Four

Page 114: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

structures or mechanisms. On this view, theories have essential nonsententialdimensions, what Harré termed “imagined paramorphs,” models of causalmechanisms at work in the world.7

The second general plateau, life, is distinguished by “the way physico-chemical energies are interconnected and operate. Animate bodies seek “tomaintain a temporal pattern of activity . . . to utilize conserved consequencesof past activities so as to adapt subsequent changes to the needs of the inte-gral system in which they belong.” Thus, “iron as such exhibits characteris-tics of bias or selective reactions” but “iron as a genuine constituent of an or-ganized body acts so as to tend to maintain the type of activity of theorganism to which it belongs” (E&N: 192). In an organism, it functions notto become iron-oxide (as it would in a hinge), but to contribute to metabo-lism. As Dewey sees, how some element of a concrete composite behaves de-pends upon its (theorized) dispositional properties, on how in the integratedsystem, it is related to other “things,” and on how the composite is related to“things” external to it. It is because iron—Fe—is what it is that it has prop-erties, which enable it to function differently in different relations. Comparehere hinges in New Mexico and Honolulu.

We experience patterns not invariances. Patterns are the result of relativelystable configurations of causal mechanisms. Salt (usually) dissolves in water;for human percipients, gold is—almost always appears—yellow; and to shiftthe example to the domain of society, there is a strong positive correlation be-tween poverty and one-on-one crime. Indeed, in terms of Dewey’s most ba-sic metaphysical category, there is both precariousness and stability because“the world” is not, as empiricists have it, a determined concatenation of con-tingent events but a contingent concatenation of ensembles of complexly re-lated natural necessities, a world of genuine change and novelty. The impli-cations of this for a proper understanding of science are, withoutexaggeration, simply enormous.


But if vitalism in biology is no longer persuasive, mind remains a problem,not only in the persistent mind/body dualism (and epistemological individu-alism) of most general psychology, but also in the social sciences, in what is,effectively, a radical bifurcation of nature and culture. As Dewey says, “uponthe whole, professed transcendentalists have been more aware than have pro-fessed empiricists of the fact that language makes the difference betweenbrute and man.” “The trouble is,” he continued, “that they have lacked a nat-uralistic conception of its origin and status” (E&N: 140). For Dewey (and

Culture and Nature 85

Page 115: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Mead), society, meaning and mind are tightly linked and a genetic account isindispensable if we are not to fall into “the philosophic fallacy,” the conver-sion of “eventual functions into antecedent existence.” “The fallacy convertsconsequences of interactions of events into causes of the occurrence of theseconsequences” (200). As Dewey recognized, this fallacy is widespread, butnowhere more vivid than in accounts of mind and of society. The former areepistemologically individualist in positing as given, an available language,beliefs expressed in this language and rationality independently of the socialrelations, which generate these. Accounts of society are methodologically in-dividualist in believing that social relations are not presupposed in action.Dewey’s move, to shift the problem of mind to the problem of language,sounds remarkably au courant. But his naturalistic account of its origin andstatus has yet to be taken seriously. We can usefully supplement Dewey’s ac-count with G. H. Mead’s.

Creatures that lack language nevertheless “gesture.” Thus the perceptionby a dog that another “is ready to attack becomes a stimulus . . . to change hisposition or his own attitude. He has no sooner done this than the change of at-titude . . . causes the first dog to change his attitude.” “We have here,” Meadnotes, “a conversation of gestures” (Mead, 1967: 43). To be sure, it would bean error to say these acts have meaning for the animals. Dewey and Mead in-sist that meanings do not come into being without language” and these crea-tures lack language (Dewey, E&N: 212), On the other hand, as Tiles writes,“animals which do not already respond to each other’s behavior cannot re-spond to each other’s intentions to produce modifications in their behavior”(Tiles, 1988: 89). The plateau of coordinated animal response is not irrelevantto communication at the linguistic plateau even if it is not reducible to it. Con-sider, then, a linguistically apt creature.8

Gestures can become “significant symbols.” That is, vocal gestures canarouse in an individual making them the response that they explicitly arouse,or are supposed to arouse, in other individuals (Mead, 1934: 71). They cancome to “stand for” a particular act or response. Significant symbols aremeanings. Mead wrote that the difference between a gesture and a significantsymbol is that “the individual is conscious of the meaning of his own ges-ture.” Indeed, Mead often refers to intentional acts, which “entail an elaboratemental process.” David Rubenstein (1977: 212) calls this an inconsistency inMead and says that it was the reintroduction of the “psychical entities” hetried to eliminate which invited interpretation of him as a phenomenologist.But this (not uncharacteristic) reading is a huge error, in Dewey’s terms, astraightforward product of “the philosophic fallacy.” Neither Dewey norMead deny that persons are conscious or that they have intentions. Rather, itis their claim that meaning cannot be explained in terms of intentions (or in-

86 Chapter Four

Page 116: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

tentional objects, psychic or otherwise). Thus, if someone is to be taken as,e.g., making a request, as Tiles writes:

he has to be taken to have responded to the object not as a stimulus but from thestandpoint of the [other]. And what establishes the possibility of thus adoptingthe standpoint of the other is the recognition of the regularity of the relationshipholding between gesture and completed act (1988: 93).

These perceived “regularities” are the foundation of socially constructed lin-guistic universals. In the absence of this plateau, meaning cannot be made in-telligible. Thus, semantic theories that try to define meaning in terms of truthconditions without acknowledging that linguistic acts (or their vehicles) pre-suppose social activity fail to explain how a linguistic vehicle could getmeaning. They must, finally, either beg the question or postulate meanings.

It is also to deny nominalism, the dominating posture of Hume-inspiredempiricist philosophies of language. As consequences of social interactions,which depend upon regularities that can become habitualized and standard-ized, neither meaning nor essence is “adventitious and arbitrary.” Yet, as im-portant, linguistic universals are not platonic entities or formulae, which pre-scribe their application. By explaining meaning, Dewey can also account forphilosophy’s enduring fallacies regarding it: Thus, “meanings that were dis-covered to be indispensable to communication were treated as final and ulti-mate in nature itself. Essences were hypostatized into original and constitu-tive forms of all existence”—the philosophic fallacy at work (E&N: 155).There is no objection to talk about either meanings or essences—as long asone fully appreciates them to be nothing more than relatively enduring socialproducts.

On the other hand, exactly because meanings (and essences) are groundedin regularities of interaction and are the product of these, they are both ob-jective and remain revisable. Thus, meaning is not “a peculiar kind of thing,”a Platonic Idea, a “subsisting concept” or “‘logical’ in a style which separateslogic from nature.” For Dewey, “meanings are rules for using and interpret-ing things; interpretation being always an imputation of potentiality for someconsequence.” As before, gestures depend upon expected outcomes that pre-suppose the regularities of past experience. Accordingly, use is constrained,neither “adventitous nor arbitrary,” but because in acting, agents decide, useis revisable. In noting “the scope and limits of application are ascertained ex-perimentally [practically] in the process of application” (E&N: 156). Deweyanticipates what has come to be called a “finitist” conception of rules, the ideathat since there is no universal or “natural” scale for weighing similarityagainst difference, the application of rules (including meaning-rules) are

Culture and Nature 87

Page 117: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

contingent judgments by actors using materials at hand.9 As I argue subse-quently, the implications of seeing that meanings are both “objective” and re-produced and transformed by practice are critical for social science.


The failure to see that meaning is to be found in transactions is propelled bythe failure to see that there is a radical difference between “individuals withminds” and “individual minds,” the characteristic posture of epistemologicalindividualism. To avoid a solipsism of the present moment, epistemologicalindividualists need to hold that the experienced world is shared by individu-als; but it is not a social product, this world needs to be the world of the naiverealist where things are, pretty much, as they appear. Thus, in the mind inde-pendent world, there are red apples, even if we must learn to call them “redapples.” If, however, we take modern physical science and the evidence ofhistory and anthropology seriously, we need to acknowledge that the capac-ity to identify the most mundane things of experience requires, in addition toour evolved natural capacities, a massive system of meaning which has beenhistorically, regionally, and locally bequeathed. Indeed, the failure to ac-knowledge this would seem to be consequence of both “intellectualism” andthe conversion of eventual functions into antecedent causes. As Dewey insists,

The whole history of science, art and morals proves that the mind that appearsin individuals is not as such individual mind. The former is in itself a system ofbelief, recognitions, and ignorances, of acceptances and rejections, of expectan-cies and appraisals of meanings, which have been instituted under the influenceof custom and tradition (E&N: 180).

A more esoteric example may make the point clearer and show also its rele-vance to the present essay. According to Bulmer, the terminal taxa of the Karamcorrespond very well with some 70 percent of the cases with species identifiedby a scientific zoologist.10 The cassowary is an instance of noncorrespondence.Karam have the taxon “yakt” for birds and bats, but the cassowary in not placedin this taxon. Instead, it appears in a special taxon, “kobity,” making it a non-bird/non-bat. For Bulmer, this is an error explained by Karam willingness to al-low “culture” to supercede “objective biological facts.” But on the present view,what counts as an objective scientific fact depends upon practices which maydiffer from culture to culture. If we think that Karam taxonomy is wrong, it isbecause we have reason to think that the practice of science generates tax-onomies, which better serve our purposes.

88 Chapter Four

Page 118: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

It should be clear enough that the text just quoted from Dewey does not be-tray an Hegelian, objective idealist prejudice. On such views, mind is severedfrom natural existence and individuals merely “participate” in mind. Here Irefer you to the customary bifurcation of cultural system and social system,to the standard notion of socialization wherein selves are empty vessels intowhich meaning is poured, and to the normative determinism so characteristicof mainstream sociology. The foregoing account of meanings gives us the resources to hold that the idea of “individuals with minds” admits of a fullynaturalist interpretation. The difference between it and an idealist alternative,unfortunately, is easily missed.

The idea of social super-mind is a philosophical nightmare because itprecludes agency. It makes agents merely “bearers” of cultural systems that,in the last analysis, determine action. By contrast, for Mead and Dewey, be-cause meanings are “modes of natural interaction,” culture is the continu-ous evolving product of recognitions and ignorances, acceptances and re-jections, and expectancies and appraisals that are themselves the mediumand product of conscious activity. Mind is social, not in the sense that cultural meaning is intersubjective—between subjectivities—but in thesense that meanings are public, in the world, and not (only or merely) in ourconsciousness.11


Philosophical positions have never been irrelevant to the practice of socialscience—most often, I am afraid, for ill. Attacks on empiricist philosophy ofscience, beginning in the 1950s joined with phenomenological criticisms ofempiricist social science. Both were unsettling to the mainstream, dominated,then—and still, if less so, by the “objectivist” Parsonian synthesis. But thesecriticisms did not encourage a rethinking of naturalism in nonempiricistterms. Rather, they encouraged antinaturalism. I must be brief.

The work of Schütz, no doubt, was critical. On the one hand, his work con-tained many valuable insights, for example: that sociological constructs areconstructs of social constructions, that the “stock of knowledge” is held in“typfied form” and dispersed, and that commonsense knowledge is “a patch-work” in which “clear and distinct experiences are intermingled with vagueconjectures; suppositions and prejudices cross well-proven evidences; mo-tives, means and ends, as well as causes and effects, are strung together with-out clear understanding of their real connections.”12 This contributed heavilyto undermining the Parsonian theory of action, including, critically, the stillstandard theory of rationality.13

Culture and Nature 89

Page 119: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Similarly, Schütz’s idea, derived and extended from Husserl, of a “life-world” which is the taken-for-granted beginning for inquiry (the epoché of“the natural attitude”), was a strong solvent for the naive realism of main-stream social science.14 On the other hand, as Farber insisted, the life-worldfell “like manna from heaven, as an unexpected answer to the prayer of per-sons seeking an alternative to the worldview of a scientific philosophy, but forwhom the existentialist bill of linguistic fare [was] inpalatable” (Farber, 1967:122).15 Even a “mundane phenomenology” could not (as Natanson agreed)consistently be grounded in naturalism. The trouble was not that Schütz re-mained committed to Husserl’s transcendental project, but that he retained, asGiddens noted, “the unbilical tie to the subjectivity of the ego.” For Schützthe social world is “strictly speaking, my world” (Giddens, 1976: 31). Schützdid acknowledge that to study the social world, it was necessary to “abandonthe strictly phenomenological method.” But while he was comfortable to as-sume the existence of the social world, not only did intersubjectivity remain,philosophically, a problem, but the social world seemed, at least, to be noth-ing more than a construction of consciousness. Put in other terms, it was dif-ficult to see how to incorporate either the “natural” context or the relativelyenduring consequences of action into the account. Moreover, interpretativesocial science was restricted to describing and clarifying “what is thoughtabout the social world by those living it.” This thoroughly descriptivist,ethnographic orientation, even in the hands of sensitive inquirers, lost eventhe hint of causal explanation and in consequence, any capacity for critique.Finally, while Schütz often said that it was the aim of sociology to obtain or-ganized knowledge about “the world of cultural objects and social institu-tions”—leaving unclear what exactly this meant, the discovery of “in orderto” motives became central task for sociological explanation. That is, phe-nomenology encouraged a highly psychologized notion of social science.16

Garfinkel, a student of Parsons, seems to have begun with Schütz, and beforehe was finished, offered a radical and powerful alternative to the Parsonian “ac-tion frame of reference.” This included a strong emphasis on agency, and re-jection of “motive analysis” in favor of inquiry into “situated actions.” This wasprofoundly propelled by his generalized use of “indexicality” (indirectly owedto Peirce) and by finitism (derived from Wittgenstein).

Garfinkel, however, was but ambivalently naturalistic. On some readings,e.g., his principle of “ethnomethodological indifference” was not merely arecommendation to bracket temporarily aspects of the empirical world, butwas converted into an ontological commitment wherein, as Giddens wrote,“social phenomena ‘exist’ only in so far as lay actors classify or identify themas ‘existing’” (Giddens, 1976: 42)—an intellectualist dip into voluntarismand idealism. Moreover, like Schütz, Garfinkel was preoccuppied with the

90 Chapter Four

Page 120: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

conditions of action, ignoring almost utterly the consequences of action, in-tended and unintended. Thus, neither could he sustain an adequate notion ofsocial structure. While he acknowledged that actors had resources which werethe medium of their actions, his actors became so thoroughly disconnectedfrom their bodies and the larger preexisting contexts in which they acted thatthese resources reduced to abstractly detached meaning-rules. In effect, hisintellectualism led him to ignore the fact that his agents were fleshly, inter-ested actors contexted in a geographical environment and embedded in so-cially sustained, but not always transparent social relations of power. Worse,betrayed by epistemologically generated worries, he often suggested thatthese meaning-rules were not “about” anything. As in current antirealisms,participants could cooperatively reconstitute them at will.

By the 1970s, phenomenologically inspired social science remained on themargin, but the unsatisfactory character of both mainstream and Marxist ap-proaches was more than noticed. The response was “cultural studies,” in-spired by the structuralism of Lèvi-Strauss and Saussure and then, in responseto this, both poststructuralism and the hermeneutical approach of CliffordGeertz, and, finally, by Marxists writers responding to Althusserian struc-turalism.17 Powerfully encouraged by the epistemological criticisms of em-piricism, including here Kuhn’s ambivalent Structure of Scientific Revolution,the work of Schütz, Garfinkel, and Erving Goffman, the hermeneutics of Ri-coeur and Gadamer, and the more radical poststructuralism of Foucault andDerrida, cultural studies betrayed a decided shift toward idealism. As Alexan-der rightly pointed out,

Insofar as [sociological theory] seeks a purely hermeneutic analysis—not onlyis there always cultural reference for every action but . . . there is only a culturalreference. Every change in action, every source of stability, everything thatworks for the good, everything that works for the bad—all must be explained interms of the search for meaning itself. Every culturalist theory is . . . a form ofsociological idealism (Alexander, 1987: 311–12).

Nor did Marxists escape the drift toward idealism—especially those taken bystrong readings of Gramsci (a student of Croce) and by postmodernist episte-mology.18 The response, in my view, is neither a return to materialism, nor tosome pseudosolution that demands that we think “dialectically” about cultureand “material life.”19 On the present view no such dialectic is possible be-cause divorced of culture, material life is utterly empty. Here we are betrayedby systematic ambiguity as regards the very idea of “culture.” At one time,culture was used inclusively to refer to forms of life, to ensembles of mean-ingful patterns of activity that included work, play, marriage, and worship.

Culture and Nature 91

Page 121: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

More recent cultural studies, however, have conceived culture far more nar-rowly, in terms of “mentalitiés,” “values,” “symbolic codes,” “signs,” “texts,”and “discourses” that are effectively, if not explicitly autonomous.20 We needto return to the older idea; but we need to do this, as already suggested, witha strong agent centered naturalistic conception of mind and society. That is,instead of supposing that meanings have independent existences, we need tosee them in contexts of action.


More particularly, we move in the right direction by putting the insights ofSchütz and Garfinkel onto the naturalistic footing provided by Dewey andMead. As Rubenstein rightly said, it was a major motive of phenomeno-logical and verstehen approaches to social science to describe action interms of mental components in order to combat the naturalistic inclinationto treat action and social phenomena in the same way one treats the mean-ingless properties and events of nature. On the other hand, it was a majormotive of empiricists to argue, “reliable knowledge cannot be establishedabout what is essentially private to the actor” (Rubenstein, 1977: 232). Butif the foregoing is sound, there is no reason to be suspicious of an approachthat insists that the category of meaning is indispensable to the under-standing of human behavior, and for the same reason, there is reason to besuspicious of those philosophies of language which inform empiricist phi-losophy of science. As work by Kuhn, Polanyi, and more recent sociologyof scientific knowledge make clear enough, the meaning of scientific termsdepends not on “operational definitions” and other semantic devices, but intaking account of science as a social activity in the sense of Dewey andMead.

First, with Schütz, we can endorse verstehen and the idea that sociologicalterms are constructions of what are already social constructions, what Gid-dens has called “the double hermeneutic.” Social science (in contrast to thephysical sciences) is involved in theorizing and communicating about an al-ready meaningful social world. But because verstehen is not some form ofempathetic understanding—indeed, is a presupposition of any human activ-ity, including, then the practices of natural science, social science requires nospecial observational methods.

Second, pace Garfinkel, instead of a one-sided emphasis on action asmeaning, we can shift to action as praxis, as Giddens writes: “the involve-ment of actors with the practical realization of interests, including the mate-rial transformation of nature through human activity” (Giddens, 1976: 53). So

92 Chapter Four

Page 122: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

construed, “culture” is not bifurcated from “material activity,” but is under-stood as inseparable, substantively and analytically, from it.21

Third, in consequence, “all social research has a necessarily cultural,ethnographic or ‘anthropological’ aspect to it” (Giddens, 1984: 1984). Inother terms, qualitative research is an indispensable component of social sci-ence. This will be largely descriptive, even though, inevitably, it will be the-oretically informed, and thus, not only is literary style not irrelevant to the ac-curacy and communicability of such descriptions, but social scientists mustdraw on the same sources of “mutual knowledge” drawn on by novelists andjournalists.22

But, fourth, because meaning is not “in the head,” and “experience” is notreducible to conscious contents, we need to distinguish practical knowledgefrom discursive knowledge. That is, while the present view centers agencyand, unlike most mainstream and Marxist views, acknowledges that actorshave complex skills and knowledge which they employ in acting and inter-acting, if we are to avoid the intellectualist fallacy, much of this knowledge isnot discursively available. As Bhaskar says, it is “tacit and implicit, sponta-neous and not reflective, a matter of know-how rather than know-that”(Bhaskar, 1986: 163). Accordingly, even if what is discursively available (ormade available) is true, acquiring knowledge of the beliefs of actors will notbe sufficient to establish an understanding of their social world. That is, evena good ethnography must go well beyond “what people think about theirworld.”

Of course, since social activity cannot be described at all unless the in-quirer knows what actors know, accounts from actors are necessary and playa critical role in enabling us to assess accounts offered inquirers. But indeed,there is good reason to hold that beliefs discursively available are not alwaystrue! While practical activity is skillful and intended, it does not require truebelief as regards the conditions of action. In part this is because action alwayshas an unintended consequence, viz., the reproduction and transformation ofthe very conditions of action. We do make history but, as Marx insisted, notwith a plan and not with materials of our own choosing. Unacknowledgedconditions, unknown and unintended consequences, self-deception and otherobstacles limit our ability to cognize fully and accurately the social worldwhich our own actions sustain. Were it otherwise, there would hardly be apoint to human science.

This means, fifth, not only that social science can enlarge the understand-ing of members, but by so doing, it can have a critical and emancipatory di-mension. That is, because the domain of the social sciences comprises socialobjects, e.g., institutions, social practices, and social relations which are theproduct of social activity, and because this domain includes beliefs about

Culture and Nature 93

Page 123: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

these activities, when these beliefs can be shown to be false, distorted or oth-erwise inadequate, agents have grounds to change these social forms. Con-sider the belief that males are superior. If this belief is constitutive of the re-lations which define the nuclear family, then if (as women increasinglyappreciate!) this belief is false, people have good reasons for altering these relations—as indeed, they have been doing! That is, eklaren, the effort to explain how these forms have come to be and why people have the beliefsthey have is an essential part of the task of social science.

Sixth, such explanation is not via subsumption under law. On the present, re-alist view, success in the theoretical sciences depends upon the capacity to ab-stract a strata of the world and to identify, theoretically, causal mechanismswithin that strata. Such theory gives us an understanding. We gain, thus, an un-derstanding of tensile strength by appeal to physical theory. But because the the-oretical powers of “things” are never operating in a closed system—other causesare always operating on them—there are patterns, but no invariant empirical reg-ularities. Everything that happens is caused, but it is complexly caused, a func-tion of the causal powers of the things of the world and their continuously chang-ing relations and configurations. For example, one explains the collapse of alanding gear by appeal to the tensile strength of the materials and a host of otherpertinent causes, including, perhaps, the historical effects of the maintenanceschedule and the decision to make an emergency landing on a field not hereto-fore used by such aircraft. Because causal conjunctures are contingent, we are of-ten in a position to explain something that happened when we could not have pre-dicted it. Stellar mechanics is the worst possible paradigm for a science exactlybecause, as regards the pertinent “variables,” the solar system is relatively closed.

Explanation in social science has the same form, involving, on the onehand, the effort to identify the social mechanisms or structured processes be-ing sustained by the activities of agents, and on the other, the effort to grasp,concretely, the capacities which they have and the constraints to which theyare subject, what they know and understand, and, finally, the uses to whichthey put their capacities and knowledge. Because all these are historicallyvariable, social science, in contrast to the most successful of the physical sci-ences, is inevitably concrete and historical—and for the same reasons, itcould never be finished (Manicas, 2006).

For example, one begins to understand a capitalist society by identifyingthe “logic” of capital. That is, given the (very different) resources made avail-able by capitalist social relations, as a consequence of their actions, intendedand unintended, actors will promote a tendency toward overaccumulation. Tobe sure, because between Japan and the United States, or between the UnitedStates in 1929 and the United States today, there will be immense differencesin the concrete forms of these relations and their relation to other structured

94 Chapter Four

Page 124: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

practices, there will be differences in the capacities, constraints, and forms ofknowledgeability between actors in these different times and places.

On the other hand, the tendency to overaccumulation will surely figure inany account of the Great Depression, even if, as noted, any plausible accountwill need to integrate a host of other processes, contingent events, and deci-sions by persons, acting and interacting, as always, as cultural beings with be-liefs and a range of meaningful material objects.

I conclude with what for me is the most important idea: The foregoing im-plies that the antinaturalistic Kantian bifurcation of “freedom” and “deter-minism” needs to be thoroughly rejected. The problem of human freedom,naturalistically understood, is the problem of possessing the capacity to act inrealizing one’s genuine interests; and this involves understanding the sourcesof constraint and limitation, and then transforming these to “needed, wantedand empowering sources of determination.” Indeed, it is to generate a socialscience that would have satisfied Dewey.


1. To be clear, the realism assumed here is not of the variety associated with HilaryPutnam or Richard Boyd. It is, roughly, the “policy realism” of Rom Harré, an explicitlypragmatic version of realism, still too little appreciated in the USA. See Rom Harré, ThePrinciples of Scientific Thinking (Chicago: University of Chicago Press, 1970); RomHarré and Edward Madden, Causal Powers (Oxford: Basil Blackwell, 1975) and mostrecently, Harré, Varieties of Realism (Oxford: Basil Blackwell, 1987). See also RoyBhaskar, A Realist Theory of Science, 2nd Edition (Brighton: Harvester, 1979) and ThePossibility of Naturalism (Brighton: Harvester, 1979), my A History and Philosophy ofthe Social Sciences (Oxford and New York: Basil Blackwell, 1987) and A Realist Phi-losophy of Social Science (Cambridge: Cambridge University Press, 2006).

2. I say “of the sort” here since my interest is not primarily to represent eitherDewey or Mead. I use them to develop a position that is both naturalistic and realis-tic. Although influenced by R.W. Sleeper’s defense of Dewey as a “transactional re-alist,” I am not clear whether Dewey would be entirely happy with the sort of realismdefended here. See R. W. Sleeper, The Necessity of Pragmatism (New Haven, Conn.:Yale University Press, 1986).

3. It is thus that naturalisms need to be committed to strong versions of the sociol-ogy of knowledge. See P. T. Manicas and Alan Rosenberg, “Naturalism, Epistemo-logical Individualism and ‘The Strong Programme’ in the Sociology of Knowledge,”Journal for the Theory of Social Behavior, Vol. 15 (March 1985). See Chapter 6.)

4. Thus Rorty’s physicalist idea that

every speech, thought, theory, poem, composition, and philosophy will turn out to be com-pletely predictable in purely naturalistic terms. Some atoms-and-the-void account of

Culture and Nature 95

Page 125: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

micro-processes within individual human beings will permit the prediction of every soundor inscription which will be uttered (Richard Rorty, Philosophy and the Mirror of Nature([Princeton, N.J.: Princeton University Press, 1979]: 387).

5. John Dewey, Experience and Nature, The Later Works, 1925–1953 (Carbon-dale: Southern Illinois Press), hereafter E&N.

6. We need to distinguish philosophical ontology, what is presumed by inquiry,from scientific ontology, the result of (ongoing) inquiry. See Roy Bhaskar, A RealistTheory of Science.

7. See Harré, Principles of Scientific Thinking (1970) and Varieties of Realism,(1986). On the present view, it is an error of considerable importance to think of the-ories as “interpreted” deductive systems.

8. See Derek Bickerton, Language and Species (Chicago: University of ChicagoPress, 1990) for a superb evolutionary account of the genesis of linguistic capacity.

9. The idea is also critical to Wittgenstein in the Philosophical Investigations, al-though it may well be that Wittgenstein’s version is not entirely free of nominalist fan-tasy. As we note below, finitism is a central part of Garfinkel’s theory of action. Mycriticism of his use of the idea is that he ignores the powerful constraints of enduringsocial relations, well recognized by Dewey and Mead.

10. I take this example from Barry Barnes, “On the Conventional Character ofKnowledge and Cognition,” Philosophy of the Social Sciences Vol. 11 (1981): 303–33.See R. Bulmer, “Why Is the Cassowary Not a Bird?” Man, Vol. 21 (1967): 4–25.

11. On the other hand, to say that mind is social is not deny the individuality of in-dividuals with minds. “Personality, selfhood, subjectivity are eventual functions thatemerge with complexly organized interactions, organic and social” (E&N: 71). AsDewey wrote:

Mind denotes the whole system of meanings as they are embodied in the workings of or-ganic life; consciousness in a being with language evolves denotes awareness or percep-tion of meanings. . . . The greater part of mind is only implicit in any conscious act or state;the field of mind—of operative meanings—is enormously wider than that of conscious-ness. . . . Mind is, so to speak, structural, substantial, a constant background and fore-ground; perceptive consciousness is process, a series of heres and now (247).

12. See John Heritage, “Ethnomethodology,” in A. Giddens and J. H. Turner (eds.)Social Theory Today (Cambridge: Polity Press, 1987): 230, quoting Schutz.

13. Roughly, actors are rational in the sense that they distinguish means and ends,articulate consequences of alternative courses of action and assess their costs and ben-efits. As rational, they optimize benefits and minimize costs. For some discussion, seeJohn Heritage, 1987.

14. More recently, of course, “deconstruction” has been an even more powerfulsolvent.

15. It is some interest to note that Farber and Schütz were very close colleaguesduring the early years of Philosophy and Phenomenological Research. Schütz’s biog-rapher writes: “. . . it was the close connection and collaboration with Farber, morethan anything else, that was responsible for the early realization of Schütz’s intention

96 Chapter Four

Page 126: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

to establish contacts with American philosophers and find an opportunity to addressAmerican philosophical audiences” (Helmut R. Wagner, Alfred Schütz: An Intellec-tual Biography, [Chicago: University of Chicago Press, 1986]). One can give at leastSchütz’s later writings a naturalistic reading.

Peter Hare reports that the archives of PPR show that Farber struggled hard to keepthe pages of the journal open.

16. Some avowed followers of Weber also often express what is at least a tensionhere, between a psychologistic explanation of some act and a sociological explanationof acts of that sort. Thus, one may need an “in order to” motive to explain why someparticular person commits a crime, but understanding what structures criminal be-havior will require more and other than this. See my A Realist Philosophy of SocialScience (2006).

17. For a useful critical overviews, see Anthony Giddens, “Structuralism, Post-structuralism and the Production of Culture,” in A. Giddens and J.H. Turner (eds.),Social Theory Today (Cambridge: Polity Press, 1987). For Geertz, see The Interpre-tation of Culture (New York: Basic Books, 1973) and criticism by Jeffrey Alexander,Twenty Lectures, Lectures 16 and 17. As regards Marxism, see Perry Anderson, Ar-guments within English Marxism (London: Verso, 1980); Stuart Hall, “Cultural Stud-ies: Two Paradigms,” in T. Bennett et al. (eds.) Culture, Ideology and Social Process(Berkshire, UK: Open University Press, 1981). See also my “The Rise and Fall of Sci-entism,” in William Outhwaite and Stephen P. Turner (eds), Handbook of Social Sci-ence Methodology (Thousand Oaks, Calif.: Sage, 2007). This essay includes discus-sion of other critical figures including Bourdieu, Giddens, and Foucault.

18. On Gramsci, see Paul Piccone, Italian Marxism (Berkeley: University of Cal-ifornia Press, 1983). A useful compendium is Cary Nelson and Lawrence Grossberg(eds.), Marxism and the Interpretation of Culture (Urbana and Chicago: University ofIllinois Press, 1988). As the editors note, “as little as twenty years ago, it would havebeen impossible to imagine such a project and such a volume” (2). See especially theessays by Gayatri Chakrovorty Spivak, Chantal Mouffe, and Ernesto Laclau.

19. The most powerful nonreductionist Marxisms come from Raymond Williamsand E. P. Thompson. On the present view, Williams’s Marxism is far and away to bepreferred. Hall (“Cultural Studies”) quotes Thompson that “the dialectical intercoursebetween social being and social consciousness—or between ‘culture’ and ‘not culture’—is at the heart of any comprehension of the historical process within the Marxist tra-dition. . . . The tradition inherits a dialectic that is right but the particular mechanicalmetaphor through which it is expressed is wrong.” But it is hard to see how the “di-alect” is right in the absence of clarity about what could be the “right” metaphor? On the other hand, Thompson is quite correct to bring together “the two elements—consciousness and conditions—around the concept of experience.”

For Williams, there is no interesting sense of “dialectic.” Indeed, it nowhere ap-pears in his important book, Marxism and Literature (1977). Instead, Williams insiststhat talk about “base/superstructure,” “economy,” “culture,” and then, problems of “determination” and “mediation” are predicated on reifying abstractions. This is,he insists, particularly ironic since Marx’s central emphasis was on a conception of productive activity in which “labour and language, as practices, can be seen as

Culture and Nature 97

Page 127: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

evolutionary and historically constitutive” (33). As Hall says, disapprovingly,“Williams so totally absorbs ‘definitions of experience’ into our ‘ways of living,’ andboth into an indissoluble real material practice-in-general, as to obviate any distinc-tion between ‘culture’ and ‘not-culture’” (26). This is, of course, very reminiscent ofDewey.

20. As Roy D’Andrade has remarked: “When I was a graduate student, one imag-ined people in a culture; ten years later culture was all in their heads (Andrade, “AColloquy of Cultural Theorists,” in Richard A. Schweder and Robert A. LeVine (eds.),Culture Theory, [Cambridge: Cambridge University Press, 1987]: 7).

For a representative sample of work that puts culture “all in their heads,” see K. C.Alexander and S. Seidman (eds.), Culture and Society (Cambridge: Cambridge Uni-versity Press, 1990). The conjunctive “and” in the title betrays the problem of muchrecent work.

21. As Paul Willis has insisted, “there is no question . . . of counterposing the ‘cul-tural’ with the ‘productive’ or the ‘real,’ as if the former had no actual constitutive rolein the basic social relations which govern the form of . . . society” (in Alexander andSeidman (eds.), Culture and Society: 84). Thus, the class relations of British prole-tariat toward the end of the century and of jute workers in British India in this centurywere fundamentally different. See Dipesh Chahkrabarty, Rethinking Working ClassHistory: Bengal, 1890–1940 (London: Verso, 1991). In a nutshell, the jute workers ofBengal were wageworkers but were culturally not proletarian in Marx’s sense.

22. Indeed, it is fair to say that as regards communicating an understanding of cultural milieu, novelists and journalists often do a better job than do academic socialscientists!

98 Chapter Four

Page 128: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Part Two


Page 129: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 130: Dewey Rescuing Dewey Essays in Pragmatic Naturalism



In order, eventually, to make some claims about realism and naturalism, I be-gin with a view of philosophy and with a very brief sketch of “the philosophyof the recent past.” For me, philosophical labels, like the term “philosophy”itself, do not discriminate natural kinds. Accordingly, with Jonathan Ree, Ithink of philosophy as part of literature and general history (Ree, 1988).1 Andwith Rorty, I think that we need to appeal to “contingent arrangements” to ex-plain both what counts as philosophy and its “problems” and to understandthe meaning of philosophical terms of art in use at any given time and place(Rorty, 1984). Even given some historical continuity in the use of some terms,this implies that what counts a philosophical problem in one period needs notbe one in some other. The present period owes directly to “contingentarrangements” in the late nineteenth century—as I shall argue, and it is im-portant to acknowledge this if we are to understand philosophy’s current sit-uation. Rorty comments: “We need to realize that the questions which ‘thecontingent arrangements’ of the present time lead us to regard as the questionsare questions ‘which may be better than whose which our ancestors asked,’but need not be the same.” Of course, they may be better, but they may alsobe worse.

I would not go so far as to say that there are no perennial philosophicalproblems, but they are ethical and political. Dewey had something like this inmind, I would guess, in urging philosophers to forego struggle with “prob-lems of philosophers” for struggle with the problems of humans.

Many who are not recognized by anybody’s canon as “philosophers”have spoken, sometimes wisely, to my two perennial problems. But those

Chapter Five

Naturalism and Subjectivism:Philosophy for the Future?

Page 131: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

traditionally deemed philosophers have tended to provide a ground for theirviews on questions in ethics and politics, generally metaphysics and some-times an epistemology. Two further observations: First, recent philosophyhas tended to be antimetaphysical, and much more interested in epistemol-ogy than in ethics or politics. In the limiting case characteristic of much“analytic” philosophy, ethical and political claims are deemed, on episte-mological grounds, “noncognitive.” I am especially interested in denyingthis, but here I only assume that the arguments of “naturalists” like Dewey,Abraham Edel, and Marvin Farber can be sustained. Second, by virtue ofthis, it is not implausible to follow Farber and assert that the only seriousalternatives for philosophy are “naturalism”—or alternatively, “material-ism,” and “subjectivism”—in his special senses.2

Farber used materialism and naturalism interchangeably, “having the ad-vantage of flexibility” and acknowledged, of course, that there are alternativeversions of both. He included as “subjectivisms,” all forms of idealism, in-cluding absolute idealism and phenomenology, and various types of existen-tial philosophy. But what is most striking about his dichotomy is the absenceof reference to positivism. We need some history.


Mandelbaum argued that during the nineteenth century, there were “only twomainstreams of philosophical thought, each of which possessed a relativelyhigh degree of continuity, and each of which tended to deal with similar prob-lems, although from opposed points of view” (Mandelbaum, 1971: 5). Theproblems regarded “knowledge” but especially the nature and role of “sci-ence.” The two positions were “metaphysical idealism and positivism.” Hisdefinition of idealism is useful: “metaphysical idealism holds that within natural human experience one can find the clue to an understanding of the ul-timate nature of reality, and this clue is revealed through those traits whichdistinguish man as a spiritual being” (1971: 6). Epistemology, a new philo-sophical discipline that derived from Kant, was the critical feature of theforms of argument of idealism but as Mandelbaum argues, the movement was“part of a more general rebellion against the conceptions of man and naturewhich characterized the Enlightenment” (1971: 7). As in Kant, idealism wasmotivated by distress that God, freedom, and immortality were being under-mined by the new “science.” Neither the forms of argument nor “the generalrebellion” have gone away, even while, especially as regards the “rebellion,”there are now at least two important anti-Enlightenment postures, premoderntheisms which, for the most part, are not defended by academic philosophers,

102 Chapter Five

Page 132: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

and ethical nihilisms, aided and abetted both by positivist epistemology and subjectivist “postmodernisms,” very much academic positions. But weneed to fill in positivism, and to get back to Farber, to define materialism andnaturalism.


The nineteenth-century battleground was over science and what it had to say about man and nature. Positivisms and materialisms both stood on “science”—critically, a very much-contested concept until very late in thecentury. And both positivisms and materialisms were opposed to traditionaltheologies. But it would be wrong to suppose that nineteenth-century posi-tivisms were all materialisms.3 Engels and then Lenin, who had an antiposi-tivist conception of science, was closer to the truth in asserting that positivistswere each covert idealists.

Farber’s dichotomy, of course, follows Engels who had argued that thereare but “two great camps” in philosophy: idealists and materialists. “Thosewho asserted the primacy of spirit to nature and, therefore, in the last instance,assumed world creation in some form . . . comprised the camp of idealism.The others, who regarded nature as primary, belong to the various schools ofmaterialism” (Engels, 1935: 32). For Engels, “the great basic question of allphilosophy, especially of modern philosophy, is that concerning the relationof thinking to being.” Engels’s dichotomy was Lenin’s point of departureagainst the Machists, and the entire cast of “empiricist” philosophers of science of the period. In his infrequently read Materialism and Empirio-criticism (1908), Lenin defended Engels’s materialism against those “boldwarriors, who proudly allude to the ‘modern theory of knowledge,’ ‘recentphilosophy’ (or ‘recent positivism’), the ‘philosophy of the natural sciences’or even more boldly, ‘the philosophy of natural science of the twentieth cen-tury’” (Lenin, 1970: 7).4 Lenin was quite correct to identify positivist theoryof science as hegemonic, to link it with “modern theory of knowledge,” andto see that it rejected the realism of materialisms.

Mandelbaum well summarized the distinct feature of positive philosophy.

Since positivism confines all human knowledge to what has been experiencedor can be experienced, it claims that a science which has freed itself from meta-physical preconceptions will restrict itself to discovering reliable correlationswithin experience. . . . According to this view, a scientific explanation does notinvolve appeal to any immanent forces nor to any transcendent entities; to ex-plain a phenomena is to be able to subsume it under one or more laws of whichit is an instance” (1971: 11.)

Naturalism and Subjectivism: Philosophy for the Future? 103

Page 133: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Positivists are manifestly empiricists, and it is easy enough to see how, asMandelbaum put it, the positivist interpretation of science “even came to beabsorbed into the idealist tradition” (11). As he notes in another place, com-mon to all forms of idealism and phenomenalism is “subjectivism,” under-stood as the idea that all we can know are the contents of consciousness (whatRorty later referred to as the “veil of ideas”). If, then, reference to “transcen-dent reality” is to be rejected and all knowledge is restricted to what is “in ex-perience,” then what is to be gained by holding that “the objects” of experi-ence exist independently of it? As Farber well said: “the methodologicalrestriction of the objects of reality to a relationship with an experiencing sub-ject—the subject-object limitation—serves as a wheelhorse for idealistic ar-guments at critical points” (Farber, 1984: 130).

It is very critical to see also that causality figures hugely in the positivistvision. What is rejected as causation, of course, is any sort of “metaphysical”notion of causes as productive powers. Instead, we have a Humean concep-tion of causality as an empirically available constant conjunction. Hence evenin the “transfigured realism” of Spencer, reality was “unknowable” exactlybecause, on his empiricist premises, there was no way to get from “objectivereality” to “experience.” As Mandelbaum summarizes:

Science would be transcended and metaphysics would set in if one tried to formany conception of how motions in the nerves “produce” sensations, or how com-plex associations of ideas can lead to those efferent nerve-impulses which even-tuate in action. To attempt to go beyond the verifiable correlations between theseutterly different types of concept would be to introduce notions, which it is notin any way possible to verify within experience (1971: 304).5

Accordingly, despite large differences between them, British empiricists,for example, J. S. Mill, and Spencer, Kirchoff, Mach, Avenarius, Duhem,and Poincaré, and, of course, the “logical positivists,” “logical empiricists,”“neopositivists,” and “antirealists” of today are all “positivist” (in Comte’ssense). Indeed, since their conception of science is perfectly comfortablewith the extensionalist logic of Whitehead and Russell, it came to definelogical positivism; and despite fatal challenges, remains the unquestionedassumption of both most current discussions in epistemology and the phi-losophy of science. “Naturalists” take their stand with science, but the crit-ical point then is precisely how science is to be understood. To anticipate,the naturalism (and realism), which I will defend, follows Rom Harré andEd Madden’s groundbreaking, Causal Powers and Harré’s revolutionary as-sault on “deductivism” in the philosophy of science.6 But before pursuingthis idea, we need to comment on the third nineteenth-century contender,materialism.

104 Chapter Five

Page 134: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


Although materialism was a widely held view in the eighteenth-century,Mandelbaum notes that despite some confusion on this issue, there werevery few materialists in the nineteenth-century, and most of them were Ger-man. The obvious materialism, and from the point of view of later globalphilosophy, is, of course, Marxism. We can usefully begin with Mandel-baum’s definition:

Taken in its broadest sense, materialism is only committed to holding that thenature of that which is self-existent is material in character, there being no enti-ties which exists independently of matter. Thus, in this sense, we would class asa materialist anyone who accepts all of the following propositions: that there isan independently existing world; that human beings, like all other objects, arematerial entities; that the human mind does not exist as an entity distinct fromthe human body; and that there is no God (nor any other non-human being)whose mode of existence is not that of material entities (1971: 22).

This is plainly Farber’s sense and explains why naturalism and materialismare usefully interchangeble. Thus, “reductionist” forms of materialism, as inErnest Haeckel, or Moleschott and Büchner, are easily distinguished from En-gels’s “dialectical” reading of science—and even more important from thepresent point of view from the nonreductionist naturalisms of Marx andDewey.

As important, the definition leaves open questions about the nature or char-acter of the material world, including whether “matter” is a “substance” (asper Descartes) or an underlying “substratum” (as per Locke). As is wellknown, these positions are subject to Berkeleyan criticisms. But the nature orcharacter of the “material world” may be understood as an entirely scientificquestion. If so, a “realist” and anti-positivist theory of science will be re-quired. Materialisms, of course, are realisms in the first important sense that,in contrast to “idealisms” which make “reality” mind-dependent, for the ma-terialist, the world exists independently of minds, God’s included.


As the twentieth century began, positivism had won the battle over the char-acter of the physical sciences.7 But the existential status of “the externalworld” had not been resolved. It informs, of course, the problem articulationof Moore and Russell, James and Dewey, and all the American realists (intheir several varieties), with variant forms of positivism confounding matters

Naturalism and Subjectivism: Philosophy for the Future? 105

Page 135: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

further. I shall not attempt a review of the often-puerile debates that charac-terize this battle and why it has been so difficult to be clear about the perti-nent issues.

But I would insist that this debate is a philosopher’s problem in the sensethat first, “the things that we see, hear, and touch—and to a significant degreealso what we taste or smell—appear as independent of our seeing, hearing, ortouching them” (Mandelbaum, 1964: 222).8 Only a philosopher could raisequestions about this. Second, we cannot doubt that we can learn from experi-ences in this world, even if philosophy and science remain puzzled as to justhow this occurs. Thus, in no human community, however different, have per-sons failed to make judgments about the resources and dangers of this world.Papaya was identified as nourishing before there was any understanding ofmetabolism, and cold was avoided before we understood the mechanisms.But even more than this, only a philosopher could doubt that the modern nat-ural sciences produce genuine knowledge of this world. I take it as funda-mental not merely that nature exists independently of at least human experi-ence, but that it is structured in some way independent of human inquiry andthat we can have some knowledge of it.9

But this does not mean that debate between idealists and their opponentsleft no residual problems. I find two, the one that motivates Farber and theother, the paradigmatic philosopher’s problem (or nest of philosophers’sproblems). I begin with the latter.


Epistemological problems, either in traditional “foundationist” or in more re-cent nonfoundationalist, “analytic” variations, are philosopher’s problems.The “discipline” is of recent vintage, achieving self-consciousness only afterKant, indeed, as part and parcel of the emergence of metaphysical idealism.Since eighteenth-century thinkers did not distinguish science and philosophy,in the Enlightenment vision of man and nature, metaphysics and physics werenot distinguishable. Thus Newton, Boyle, and Locke could assume that sci-entific inference could offer evidence that there were nonperceivable, inde-pendently existing objects, which could be known.10 And since they workedprior to the development of empiricist criticisms of causality, they could alsoseek causal explanations of experience. Evidently, these could not be under-stood as “correlations” of directly observed sequences.

After Kant, claims about knowing would preempt claims about being(what Roy Bhaskar called “the epistemic fallacy”) leaving room only for an

106 Chapter Five

Page 136: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

idealistic metaphysics or a positivism, including forms of Kantianism, phe-nomenalism, naive realism, or doctrines of “pure experience.” And, as part ofthis, despite struggles by Helmholtz, Spencer, and James (in his Principles),“science” could offer nothing of interest about knowing. The question, “Howis our knowledge possible?” was, thereafter, nothing like the question: “Howare telephones possible.” As a philosophical problem, it could be answered ineither one of two ways: Either by taking a “transcendental turn” (phenome-nology) or in terms of evidential relations between basic and nonbasic propo-sitions (Rorty, 1979).11


There have been some modern philosophers who tried to avoid this regressivepursuit. Indeed, one can argue that this is what is most distinctive aboutAmerican pragmatism, from Peirce to Dewey.12 But none, including Dewey,entirely escaped the Kantian epistemological problematic. This explains theodd character of Peirce’s philosophy, James’s shift to “radical empiricism,”the frustrating debates that Dewey had with the American realists of varioussorts, his frustration with them, on grounds that he was not offering but an-other “epistemology,” and the failure to see also that he was not offering justanother “scientism.” (See Chapter 1.)

Dewey was surely correct to reject “the spectator theory of knowledge”and to deny the idea that truth was to be determined by its relation to the in-dependently existing reality—the assumption of at least some “realisms,” andhe was correct in his effort to displace epistemology for “inquiry into in-quiry,” comprehended as a practical, social activity which made science con-tinuous with common sense. But his “naturalism” was burdened by his com-mitment to “experience.” The problem was not, however, his defense of naiverealism or even his “postulate or criterion of immediate empiricism” (prop-erly understood). Rather the problem was his unwillingness to accept a strongversion of scientific realism, necessary if I am correct, to carry out the pro-gram of his groundbreaking and little understood Logic: The Theory of In-quiry. And this amounts to saying that he could find no grounds on which toassert that the “thing-in-itself” was knowable and causally pertinent.

Experience had been corrupted by the tradition, which gave us epistemol-ogy. As John Shook points out, by the time he was ninety-one years old,Dewey saw this. As a good empiricist, he had intended “to liberate philoso-phy from desiccated abstractions” (a task also set by Marx). But experience“had become effectively identified with experiencing in the sense of the psy-chological, and the psychological had become established as that which is

Naturalism and Subjectivism: Philosophy for the Future? 107

Page 137: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

intrinsically psychical, mental, private.” Accordingly, his insistence that “‘ex-perience’ also designates that what is experienced was a mere ideologicalthundering in the Index for it ignored the ironical twist which made this useof ‘experience’ strange and incomprehensible” (LW, 1: 362). If indeedDewey’s instrumentalism was an epistemology, then this move was “strangeand incomprehensible” exactly because it denied the starting point of “theepistemological problem.” Either Dewey, like Moore in his famous refutationof idealism, missed the point or he was a covert idealist, perhaps a Hegelianof some sort.13

There is nothing fatal about the “postulate of immediate empiricism,” that“things—anything, everything, in the ordinary or non-technical use of theterm ‘thing’—are what they are experienced as” (MW, 3: 159). The postulatenot only allowed for, but required that we recognize that the experiences ofindividual’s may well differ, so “if it is a horse which is to be described, orthe equus which is to be defined, then must the horse-trader, or the jockey, or. . . the paleontologist tell us what the horse is which is experienced.” Theseaccounts may differ, but none is privileged as “real” against others, which aredeemed “phenomenal.” For Dewey, it is clear that each account is from somepoint of view and that the conditions necessary for understanding the differ-ences as well as the agreements can be provided. This plainly will be a prob-lem for psychology and the sociology of knowledge—an inquiry demandedby Dewey’s theory of knowledge and welcomed by me. But, presumably,there is something independent of each of these experiences which is causallypertinent to the having of them—and, if so, why not independent of any-body’s? And if not, why was this not an idealism? Indeed, on Dewey’s ownnaturalistic premises, cats, bats, and beetles each have “worlds” which are en-abled and constrained by their particular sensory (and “mental”) capacities.14


Tom Burke is quite correct, I think, to argue that the naturalism of Dewey’sLogic joins with the ecological psychology of J. J. Gibson (Burke, 1994:83–96). Burke summarizes:

In contrast with a classical empiricist view of perception (involving so-called,sense data, sense impression, stimulations or nerve endings, irritations of bodysurfaces, and so forth), ecological psychology emphasizes a different array oftheoretical concepts; one being the concept of “invariants” and another the con-cept of affordances . . .

Ecological psychology treats the perceiving agent as a dynamic organism/environment system, continually engaged in various sorts of actions designed

108 Chapter Five

Page 138: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

for exploring the world and utilizing its resources. Controlled sampling of theworld gives evidence of possible uses of things (and of ways to orchestrate sub-sequent actions) by virtue of the agent’s being attuned to lawlike relations whichinvolve stable associations of different sorts of possible experiences (84).

Now, my point is just this: the idea of “invariants”—lawlike relations, and theconcept of “affordances”—possibilities as determined by invariants require arealist theory of causal powers.

Affordances are dispositional properties of things, which refer to a thing’spowers construed as per Harré and Madden (1974) and Bunge (1970). It is toassert a categorical referring to the nature (structure) of the thing and to “ten-dencies” true of the thing by virtue of its nature. Dispositions manifest them-selves (minimally) in pairs: salt dissolves in water, clay is molded with thehands. Affordances are dispositions in an organism-populated world. As Tur-vey et al. write: “Possibilities for action, or more precisely, things with pos-sibilities for action, are among the kinds of things that populate an animal’sniche and are, contrary to received wisdom, things to be heard, or smelt, etc.”This is most easily seen with an example:

Sharks electrically detect things to eat and things that impede locomotion. . . .An edible thing such as flatfish differs in ionic composition from the surround-ing water, producing a bioelectric field partially modulated in the rhythm of theliving thing’s respiratory movements. A flatfish that has buried in the sand willbe detectable by a shark swimming just above it. Reproducing the bioelectricfield of the flatfish artificially, by passing a current between two electrodesburied in the sand, invites the same behavior. The shark digs tenaciously at thesource of the field departing from the site when the act fails to reveal an ediblething. . . . Now there is no intelligible sense in which it can be claimed that thesource ought to have appeared edible if the shark’s perception of affordanceswere direct. In the niche of the shark, “edible thing” and “electric field of, say,type F” are nomically related. To predicate of the shark (a) “detects electricalfield of type F” and (b) “takes to be an edible thing” is not to refer to two dif-ferent states of affairs, one (viz. (b)) that is reached from the other (viz. (a)) byan inference. Rather, it is to make reference to a single state of affairs of theshark-niche system. The linking of (a) and (b) is not something that goes on inthe “mind” of the shark, as the Establishment would have it. The linking of (a)and (b) is in the physics of an ecological world . . . (Turvey et al., 1981: 276–77).

Dewey would, I think, strongly agree that ecological psychology picks up onthemes that he articulated, especially, in the Logic. And it is perfectly clearthat for Dewey, even perception is profoundly affected by the fact that hu-mans are social beings, a fact which raises immense problems for empiricalpsychology (Manicas and Secord, 1984). But we need to ask: How did

Naturalism and Subjectivism: Philosophy for the Future? 109

Page 139: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Dewey stand on what is now termed “scientific realism?” Unfortunately, aswith earlier “realism” debates, it is not perfectly clear what this implies. Alarge part of the problem, moreover, turns on whether the claims mean to pro-vide an account of the actual practices of the physical sciences, especiallyphysics, chemistry, biochemistry, and whether if they do, the accounts areconstrained by traditional epistemological assumptions, for example, as inQuine, whose understanding of “empirical” and of “logic” (as extensionaland providing the canonical form of scientific sentences) gives his under-standing of “scientific realism” a most distinctive “empiricist” caste (Mani-cas, 2004). So as to be as clear as possible on my position, let me merely as-sert one of Margolis’s conclusions (as I understand them).

Margolis has argued convincingly, I think, that a strong form of scientificrealism need not be either “foundationist” or “cognitivist” as he explicatesthese. To do this, one needs to assert “ontic externalism, the view that ‘theworld consists of some fixed totality of mind-independent objects,” that “thequestion the way the world is” makes sense “relative to one conceptual themeor another,” and finally, that “objectivity in the cognitive sense is only ‘ob-jectivity for us’” (Margolis, 1985: 285). Dewey would, I think, agree with allthree, even while taking what amounts to an antirealist position regarding un-observables and even if he denies that causality is an ontological category.

We can notice, first, that Dewey’s prose leaves us with some questions onthe pertinent issues. This results, in part at least, from his willingness to in-corporate into his own highly idiosyncratic theory of science, elements fromcompeting historical traditions, but especially “empiricism” and “rational-ism.” Thus, it is clear enough that he was not a Humean (although Mill is usually his target), that he joined “nominalism” and “realism,” and that hesupposed that one could settle most of the questions about inquiry, and, ac-cordingly, about science, by paying close attention to the function of proposi-tions in use in science.

He clearly rejected the “regularity determinist” ontology of events so char-acteristic of Hume and positivism. For him, “there are no such things as uni-form sequences of events” (LW, Vol. 12: 445) and hence “scientific laws” couldnot be “formulations of uniform and unconditional sequences of events” (LW,12: 437. This would seem to encourage the view that, for him, science assumedan ontology of “things.” Similarly, in Experience and Nature, he held that“atoms and molecules show a selective bias in their indifferences, affinities andrepulsions . . . to other events” (LW, 1: 162). “Selective biases” are surely “ten-dencies” in the sense of Harré and Madden, and “atoms,” if not molecules, arenot “observable”—at least as ordinarily understood.

On the other hand, he denies explicitly that causality is an ontological cat-egory; for him, it is a “logical category” (in his special sense) and “the term

110 Chapter Five

Page 140: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

‘causal laws’ is, . . . in spite of its general use, a figure of speech,” “a case ofmetonymy” (LW, 12: 440). Indeed, he gave an account of what he took to bethe “confusion” regarding causality. There is, upon reflection, “a qualitativegap” between “gross qualitative objects (which are the objects of direct per-ception),” for example, the lighted match and the burning of the paper.“Forces” were introduced to get over this difficulty. Thus, “the match wassupposed to have a certain calorific power” (LW, 12: 445). But “the timecame when it was seen that forces by definition are such as to be incapable ofscientific observation. They were then ruled out of science. . . . ”15 “Thenthere grew up the hybrid notion which took from common sense the idea ofsuccession and from science the idea of invariability of conjunction” (LW, 12:445). If Dewey is not a Humean, neither, it seems, would he accept the ideathat “things” have causal powers.

But if so, his alternative is anything but clear. It turns, I think, on his criti-cal distinction between “generic propositions” and “universal propositions.”Generic propositions, for example, “sugar is sweet,” “iron rusts,” are “exis-tential” and (as with singular propositions, for example, “this is sweet”)“predicates represent potentialities which will be actualized when certain fur-ther operations are performed . . .” (LW, 12: 251). “Universal propositions”for example, “if a particle at rest is acted upon by a single moving particle,then . . .” (LW, 12: 254) and, ambiguously, sentences of the form, “All A isB” (rendered as in modern logic as conditionals) lack existential import. Theyare “valid, if valid at all,” because they express “a necessary relation of ab-stract characters” (LW, 12: 255).

Ernest Nagel was correct, it seems, in saying that the “function” ofgeneric propositions “is to organize perceptual materials . . .” (LW, 12: xvi).They are, accordingly, the heart of our commonsense understanding of na-ture. The formulation just quoted suggests a reading of them as dispositions,nonrealistically analyzed: If X is tasted, then if X is sugar, X will tastesweet. But “will be actualized” (even ceteris paribus) suggests also thatthere is some sort of necessity attached to them. If so, this is an odd mix. Itis easy to see how one could have “natural necessity” if generic proposi-tions are analyzed realistically.16

“Universal propositions,” by contrast, formulate “necessary relations” be-tween “abstract characters” and “their function in inquiry is to propose possi-ble operations which, if carried out, might solve the problem under inquiry”(LW, 12: xvi). This, of course, grapples with the medieval problem of real-ism/nominalism. Nagel quite understandably is puzzled by the putative “ne-cessity” in such “laws.” Such necessity surely is not a priori for Dewey, evenwhile he terms the relation “logical” and “definitional,” nor does it seem to square with standard “logicist” efforts (unsuccessful!) to discriminate

Naturalism and Subjectivism: Philosophy for the Future? 111

Page 141: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

between “accidental generalizations” and “laws.” But neither does it representwhat Harré and Madden termed “natural necessity,” for this is ontological.

Dewey seems to think that the pertinent issues are resolved once we acceptthat “conceptual subject-matter is [to be] interpreted solely and wholly on theground of the function it performs on the conduct of inquiry” (LW, 12: 462).We can, he says, then reject as spurious an exclusive dichotomy between“conceptions [as] mere devices of practical convenience,” or as “descriptiveof something actually existing in the material dealt with” (462). The formeris an “instrumentalist” reading; the latter, realist. And, of course, dependingon what they are “devices” for, they may be both “descriptive” and “of prac-tical convenience,” perhaps useful also as guides to inquiry. But are theseconceptions “descriptions”? And if so, what are they descriptions of?

Dewey sees rightly that the notion of “abstraction” is part of the problem.As he sees it, if conceptions are descriptions, “abstract characters” are “ab-stracted” from “existents” in the sense of “selective discrimination.” But, heinsists, this is quite impossible as regards an abstract character as a “scientificconception.” He gives an example: “smoothness, as an instance of a scientificconception, is not capable of observation and hence of selective discrimina-tion” (462). Hence, as scientific conceptions, such abstract characters are notdescriptions.

But the scientific realist, not bound by positivist predilections, will agreethat while “abstraction” is part of the problem, we should not be looking atabstract characters at all, but at models of “things” as abstractions from thereal, concrete. We experience water as fluid and clear and capable of manysorts of transactions. H2O, an abstraction, identifies the model for a moleculeof water, and molecular chemistry develops the theory, which explains thesepowers. The model is not a fiction, but an abstracted real structure. Experi-enced water is H2O but it is not only H2O. The water of immediate experiencedoes what it does by virtue of being H2O. Hence, ceteris paribus, because itis NaCl, ordinary (experienced) table salt must dissolve in the water in myboiling pot.17

Perhaps Dewey’s account can be rescued, and perhaps it is sound as itstands. John Shook18 seems to bite the bullet. He has argued that while Deweyallowed, “the sciences should be permitted to postulate unexperienceable,transcendent entities that permit scientific explanation of experiencedevents,” he also “refuses to take a realistic stand towards such ‘objects,’ whileQuine [e.g.,] encourages realism here.” But if as he says, “scientific theoriesare used to guide inferences toward predictions,” and universal propositions“function in science regardless of whether their terms actually refer to any-thing at all,” then as Mach, Poincaré, and Duhem each insisted, why cling tothe idea that science seeks to explain? It is the core of realist theories of sci-

112 Chapter Five

Page 142: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ence that science explains and could not unless we accept that inquiry givesus knowledge of the causal powers of the things, which exist independentlyof us.


I noted that there were two residual problems of the idealism/realism debate.The second regards not philosopher’s problems, but problems of how weshould live. I want to support Dewey’s theory of inquiry as a naturalism be-cause, as Farber insisted, there are only two ways to address these problems.One is either a naturalist who holds that naturalist inquiry can answer thesequestions or one is antinaturalist and denies this. Today, antinaturalism hastwo main forms: the appeals to authority of traditional theology, and the sub-jectivisms of positivism and postmodernist theory.

Our daily papers are filled with examples of the first.19 But a central issueis identified by Hare and Madden in their little book, Evil and the Concept ofGod (1968). They argue that, however understood, evils should be eliminatedas far as humanly possible; but if indeed, they are not remediable, and if,worse, they serve some theological values not obvious to us, then why makethe effort? Or as Parry writes in his short rejoinder to Father Clark’s defenseof theism:

There is no need to blame Jupiter for the lightening, nor a jealous god for natu-ral death. Violent homicide is indeed blameworthy, especially wholesale slaugh-ter. Though “the system” is undoubtedly faulty, yet it operates only through in-dividuals, who must be held morally responsible. The rulers of the world, on myview, must be held primarily responsible for such horrors as burning civilians bygas chambers, atom bombs, and napalm; and all of us are jointly responsible tothe extent that we support our rulers (Parry, 1968).

That this needs saying is, itself, shameful.Positivisms accept science, but on its understanding of knowledge, science

becomes irrelevant to questions of morals and politics. So, for example, theeminent Harvard zoologist, Stephen Jay Gould, argues that “religion” and“science” are complementary: “Science tries to document the factual charac-ter of the natural world. . . . Religion, on the other hand, operates in theequally important, but utterly different (my emphasis) realm of human pur-poses, meanings and values. . . .”20

Postmodernisms deny nature and hold that “science”—generally misunderstood—is but one among many “discourses,” including, then, the“discourses” of multiplied “communities,” “faith,” “ethnic,” and otherwise.

Naturalism and Subjectivism: Philosophy for the Future? 113

Page 143: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Like the “New Age” quest for a new “inwardness” with its “metaphysicaldissolvent,” “Transcendental Individualism,”21 the postmodernist obliter-ates “objectivity” and licenses equally whatever beliefs are shared by theseself-defined communities, however belief gets fixed. Moreover, positivismand postmodernism are consistent with and propelled by “capitalist am-biance,” “flexible accumulation,” and consumerism.22 Marvin Farber had itright:

The philosophical Pandora’s box [of subjectivism] is one more fairy tale. . . . Itis, however, a fairy tale with sociohistorical linkage and consequences, for it isan ingenious philosophy of renunciation that leaves the status quo unexaminedand unchallenged and that may even be accommodated to reactionary ideas(1984: 130).

Dewey is pertinent here. But bringing me full circle back to Marvin Farber,the naturalism of Marx is even more pertinent:

The great thing in Hegel’s Phenomenology . . . is simply that Hegel grasps the self-creation of man as a process, objectification as loss of the object, as alienation andtranscendence of this alienation; that he thus grasps the nature of work, and com-prehends objective man, authentic because actual, as the result of his own work.The actual, active relation of man to himself as a species-being or the confirmationof his species being as an actual, that is, human, being is only possible so far as hereally brings forth all his species-powers—which in turn is only possible throughthe collective effort of mankind, only as the result of history. . . .

We see here how a consistent naturalism or humanism is distinguished fromboth idealism and materialism, as well and at the same time the unifying truthof both. We also see that only naturalism is able to comprehend the act of worldhistory . . . (Marx, in Easton and Guddat 1967: 321–25).

Evidently, although I would need at least another paper to elaborate thesemost pregnant insights and to demonstrate their connection to alienation, theproblem of democracy and the analysis of capitalist society, nothing that I amlikely to say would add much to what is, by now, a rich and still relevant lit-erature. (See Chapter 10.)


1. In his review of the multivolume Dictionary of Eighteenth Century BritishPhilosophers, James Harris remarks: “just as it is usually hard to distinguish “philos-ophy” from science in the eighteenth century, then so also it is difficult to hold apartscience and theology for long. That is why, if the character and significance of their

114 Chapter Five

Page 144: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

work is to be properly understood, men such as Hume and Reid have to be surroundedby so many relatively obscure figures from disciplines which today have little or noth-ing to do with philosophy. For the truth is that there are no ‘purely philosophical ques-tions in eighteenth century Britain’” (Times Literary Supplement, May 5, 2000). Seebelow as regards Locke, Boyle, and Newton.

2. Especially, Marvin Farber, Naturalism and Subjectivism (Springfield, Ill.:Charles C. Thomas, 1959); The Search for an Alternative (Philadelphia: University ofPennsylvania Press, 1984). The extraordinary volume, Philosophy for the Future (NewYork: Macmillan Co., 1949) edited by Roy Wood Sellars, V. J. McGill, and Farber, wason the reading list of Farber’s marvelous lecture course, “The Philosophy of the Re-cent Past.” A defense of materialism, the essays, many authored by the distinguishedlist who contributed regularly to Science and Society, are remarkably pertinent.

3. Mandelbaum notes that the confusion persists despite explicit disavowals on thepart of Comte, Spencer, Bernard, Huxley, and Mach—and the positivists of the veryrecent past.

4. Farber and I would agree with Lenin’s attack on the Machists, but it seems clearthat neither Engels nor Lenin provided a plausible answer to the question of the rela-tion of thought to being: the “reflection theory” surely will not do. Nor, tragically inmy view, did either Engels or Lenin provide a convincing alternative philosophy ofscience. I have discussed this, along with Engels’s relation to competing materialisms,in my “Engels’s Philosophy of Science,” in Terrell Carver and Manfred Steger (eds.),Engels after Marx (College Station: Pennsylvania State University Press, 2000).

For a beautifully wrought criticism of logical positivism—from a Marxist perspec-tive, see V. J. McGill, “An Evaluation of Logical Positivism,” in Volume 1, Number1 of Science and Society: A Marxian Quarterly (Fall 1936). Parry and Albert Blum-berg are thanked by McGill who was, of course, a close associate of Farber’s. See alsoLewis Feuer’s excellent account, “The Development of Logical Empiricism,” Scienceand Society, Vol. 5, No. 3 (Summer 1941).

5. See my “Modest Realism, Experience and Evolution,” in Roy Bhaskar (ed.),Harré and His Critics (Oxford: Basil Blackwell, 1990), 23–40, and the discussion ofJames, Chapter 1, who was caught in the same dilemma. Mandelbaum notes, cor-rectly, that having established that science had demonstrated that what we directly ex-perience never gives us the characteristics of what exists independently of us, bothSpencer and Helmholtz “reversed themselves and spoke as if it were a defect inknowledge that we do not directly experience the world as it exists independently ofus” (1971: 362). The solution, available to both, was to admit that “transdiction, or in-ference to what is in principle not experienceable is scientifically justified. McGillgave a very similar argument, briefly that one cannot argue coherently from the causalargument that “sensations cannot be regarded as copies or direct representations of . . . the material object” (which McGill holds to be true) to either “agnosticism” or“phenomenalism” (1936: 51).

6. See Causal Powers (Oxford: Blackwell, 1974); Principles of Scientific Thinking(Chicago: University of Chicago Press, 1970). One should cite here also, MarioBunge’s infrequently noticed, Causality and Modern Science, First Edition, 1959(New York: Dover, 1979) and Michael Scriven’s essays in Minnesota Studies in the

Naturalism and Subjectivism: Philosophy for the Future? 115

Page 145: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Philosophy of Science. Parry is acknowledged by Harré and Madden. As noted, BillParry’s critique of extensionalist difficulties with “entails” and the contrary-to-factconditional was a lasting influence on me, but I do not remember whether he raisedthis with particular reference to causality. See also Roderick Chisholm, “The Contrary-to-Fact Conditional,” (1946), reprinted in Manicas (ed.), Logic as Philoso-phy (New York: D. Van Nostrand, 1971).

7. The story of the human or social sciences is different and more complicated.See my A History and Philosophy of the Social Sciences (Oxford: Basil Blackwell, 19 87).

8. Mandelbaum notes that this, probably, catches the element of truth in Moore’sfamous refutation. “In direct experience we are all realists and cannot avoid being so.”He insists, rightly, that this is only the beginning of an argument, for me, a philoso-pher’s argument. Moreover, as part of this, it is not true that “everything we experi-ence exists precisely as we experience it” (1964: 2).

9. As Farber many times insisted, “the philosophical problem of existence . . .arises when a method is adopted that does not proceed from the basic fact of experi-ence,” Phenomenology and Existence: Toward a Philosophy within Nature (NewYork: Harper Torchbooks, 1967): 70. It is a “methogenic problem.” Indeed, “the factof nondependent existence is basic of philosophical thought. Not to recognize thatfact is to incur the error of illicit ignorance . . .” (72).

See also the work of another student of Farber’s, Wilfred Sellars. But we should notgo as far as Sellar’s “machorealism” (Roy Bhaskar’s term) and argue that if “the sci-entific image” is true, then “the manifest image” is false. See Science, Perception andReality (London: Routledge and Kegan Paul, 1963): 96. Farber might argue that Sel-lars’s startling conclusion is also a “methogenic” result.

10. It is a serious error to read Boyle and Newton as “positivists.” Their “corpus-cularism” depended on their perfect comfort with “transdictive” inferences: inferenceto what lies beyond the scope of all possible experience (Mandelbaum, 1964: Chap-ter 2).

11. See also his useful footnote on the historiography of philosophy (1979: 132).12. As argued in Chapter 1, Peirce recast the epistemological problem by reject-

ing the transcendental move but by accepting the Kantian “insulation” against skep-ticism. See also, Murray G. Murphey, The Development of Peirce’s Philosophy (Cam-bridge, Mass.: Harvard University Press, 1961) and the essay by my former associateat Buffalo, R.G. Meyers, “Peirce on Cartesian Doubt,” Transactions of the Charles S.Peirce Society, Vol. 3 (1967).

13. See Kenneth R. Westphal’s excellent Hegel’s Epistemological Realism (Dor-drecht: Kluwer Academic Publishers, 1989).

14. As Burke notes, we must distinguish “operational perspectivity from subjec-tivity.” The former is impossible to avoid; the latter in a Deweyan frame is not thestarting point, but needs to be explained. See Tom Burke, Dewey’s New Logic: A Re-ply to Russell (Chicago: University of Chicago Press, 1994).

15. Perhaps it is unnecessary to note here that Dewey endorses this “ruling out,”and that it was precisely this move which defines positivism and which burdenedSpencer, Helmholtz, and James.

116 Chapter Five

Page 146: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

16. A power ascription can be analyzed as: “X has the power to A” means X cando A, in the appropriate conditions, in virtue of its intrinsic nature (Harré and Mad-den, 1974: 86). Empirical investigation is needed to fill in the italicized clause. Thiswill require theory and, as well, construction of a model, perhaps detailing the micro-structure of the “thing.” See below. In contrast to nonrealist ascription, “things” havepowers even if never exercised—as was held by Peirce. See also Everett J. Nelson’spowerful “The Category of Substance,” in Sellars, McGill, and Farber, Philosophy forthe Future.

17. We need theory to fill in the CP clause, and we experiment to test the model.See Roy Bhaskar, A Realist Theory of Science (Sussex: Harvester Press, 1978). If wedrop the CP clause, this becomes a “tendency.” On models, see Harré, Principles ofScientific Thinking, esp. Chapter 2.

Derek Sayer has offered a reconstruction of Marx’s theory of science along theselines. Thus, Marx criticizes Ricardo and others as engaging in “violent abstraction.”He summarizes:

It conveys the idea of precipitate abstraction from manifest phenomena to theiralleged essences, without the mechanisms by means of which the latter cause theformer to assume the forms they do being adequately specified; or, to use differ-ent terminology, an idea of immediate identification of phenomena as supposedinstantiations of general laws, when in fact these laws operate only in mediatefashion through a series of intervening links which the analysis ought to specify.

“True abstract thinking” . . . entails elaborating the mechanisms linking lawsand phenomena in such a way that their apparent divergence is consistently ex-plained (Sayer, 1979: 121–22).

See also the several essays in Craig Dilworth (ed.), Idealization IV: Intelligibility inScience (Amsterdam: Rodopi, 1992), including my essay, “Intelligibility and Ideal-ization: Marx and Weber” and references therein.

18. In addition to his book, see the extended discussion in his unpublished paper,“Dewey and Quine on What There Is.” The following quotations are from this man-uscript, hopefully permitted by Shook.

19. Writing in the New York Times (June 19, 2000), the president of the SouthernBaptist Theological Seminary held that arguments over creation, women’s roles, ho-mosexuality, abortion, etc., are, for his 16 million parishioners, “settled by the wordof God.”

20. Quoted by Jerry A. Coyne, “Is NOMA a No Man’s Land?” Times Literary Sup-plement (June 9, 2000). Gould seems not have noticed that the idea that religion andscience “complement” one another is both factually false and founders on the as-sumption that facts and values can be bifurcated. One may hope that the surveys areflawed, but Coyne notes that “nearly 50 percent of Americans believe that humanswere directly created by God within the past 100,000 years, and 40 percent think thatcreationism should replace [not just be taught!] evolution in the biology classroom.”The New York Times helps this along when it publishes an essay by Richard Rothstein(June 7, 2000), which argues that “facts are only what we observe.” Evolution is not

Naturalism and Subjectivism: Philosophy for the Future? 117

Page 147: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

a fact: “There could be other theories.” Perhaps Rothstein took a course in philoso-phy at one of our more distinguished institutions?

21. The term is Irving Kristol’s. See his excellent “Faith a la Carte.” Times Liter-ary Supplement (May 26, 2000).

22. The best treatment is David Harvey’s, The Condition of Postmodernity (Ox-ford: Basil Blackwell, 1987).

118 Chapter Five

Page 148: Dewey Rescuing Dewey Essays in Pragmatic Naturalism



It is surely plausible to think of the histories of humankind as a series of dis-continuous, and sometimes continuous, intersecting movements marked byaccidents, some benign, some fortuitous, and some disastrous. In this regard,if nothing else, history is radically contingent—even if looking back, we canoften provide altogether satisfactory explanations of what happened andwhy?1 One such legacy is the intertwined conceptual and institutional lega-cies of science and academic philosophy.

Yet not all is well as regards these. Dewey could lament that we had failed toreplace old habits of thought with more scientific ways. This was one aspect ofthe reconstruction in philosophy for which he called. We wonder, not unrea-sonably, whether the authority of science was but Western provincialism, the ra-tionale for the imperialist obliteration of non-Western cultures. On the otherside, while Dewey was aware that science had been misappropriated and mis-applied, he remained optimistic that this could be changed, that democraticprocesses could be brought to bear on expert claims to authority. This, too, wasan aspect of his call for reconstruction. Yet, as above, Dewey’s hopes strike usas naive. Wholly disjoined from experts, we stand in terror of their so carefullyconsidered decisions. What, heaven help us, will be the unintended conse-quences of genetic engineering or the disaster of Iraq? If the nineteenth centuryhad Frankenstein, we have energy-obsessed Dr. Strangeloves.

This chapter pursues the idea of reconstructing philosophy; thus, if very in-directly, of reconstructing culture. With Dewey as both guide and foil, the fo-cus is on the implications of the current debate over the effort to naturalizeepistemology, that is, to study knowledge scientifically. Dewey was surely

Chapter Six

Naturalizing Epistemology:Reconstructing Philosophy

Page 149: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

correct that we need an alternative to dogmatism and to skepticism; but asperhaps Dewey did not clearly see, we cannot take science for granted. A sec-ond goal of this chapter, then, is to raise some questions both about currentscientific practices and our understanding of these.


We can begin with Barry Stroud’s critique of Quine’s influential essay, “Epis-temology Naturalized.”2 Quine argues that naturalized epistemology is “theempirical study of a species of primates, or, in the particular case, of an indi-vidual human subject in interaction with his environment” (in Kornblith,1985: 77). Thus:

This human subject is accorded a certain experimentally controlled imput—certain patterns of irradiation in certain frequencies, for instance—and in the full-ness of time the subject delivers as output a description of the three-dimensionalexternal world and its history? (Kornblith, 1985: 77).3

The story continues: we observe the subject as she interacts with her envi-ronment. Given then that we know her environment and have an adequatepsychology, we then explain her “output,” seeing that, in the fullness of time,what she says is true.

But, of course, the situation just described is not the situation of our natu-ralizing epistemologist, since, of course, like our knower, she is (as every-one!) utterly denied that independent, theory-neutral access to the worldwhich could be the only basis for determining whether inputs from it ever re-sult in outputs which are true. Stroud concludes that Quine simply fails to ad-dress what he takes to be the traditional question of epistemology: How dowe know that the external world is what anybody says it is?

This problem is not the problem of whether there is an external world or,for that matter, whether it has some structure. As Peirce and Dewey insisted,we can call into question any particular version of how the world is, butCartesian skepticism cannot be reasonably motivated.4 But as the foregoingseems at least to show, one can assume an external and structured world, themethod of science, and still ask if what is presumed to be known is known.

Dewey might add there that this “problem” presumes an absolute concep-tion of reality.5 On this (commonsensical) view, reality means reality as it isindependently of you and me, independently of what it is known as. My skep-tic demands that we show that knowledge of this reality is possible. If theonly knowledge we can have is from some viewpoint, how can we knowwhether it—our or some other—is valid? That is, even given an absolute con-

120 Chapter Six

Page 150: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ception of reality, it would seem that we are forced to accept a relative con-ception of knowledge. It may be, of course, that we can justify some view-point. Perhaps (as I argue below) there are modes of fixing belief, whichshould be preferred. Such a relationalism and fallibilism could then be con-trasted to relativism: understood as the thesis that no viewpoint, no mode offixing belief, can claim privilege over any other.

It may be doubted that we need an absolute conception of reality. But wesurely do—if we want to avoid relativism and to anchor our fallibilism.6 Evenif we can privilege some mode of fixing belief, we will need to aspire to theideal of grasping the world as it is, independently of what you and I might believe. The problem of historical knowledge is, perhaps, the clearest case.We must acknowledge that we shall never have more than a fragment of thepossible evidence and that alternative histories are always possible. Butsurely what transpired transpired independently of these. It is just this, whichgrounds the limits of all perspectives and thus our fallibilism.

More, we need an absolute conception of reality if criticism and persuasionis not to collapse into sophistry, to be merely a struggle to win opinion. Asrhetoricians know, truth-talk plays a vital role in argument, persuasion, andcriticism. Indeed, if we could dispense with the conception of an absoluteconception of reality, truth-talk might be dispensable, replaced by pragmatic“works/does not work” or “predicts/does not predict.” Language (and theo-ries about the world) are surely (our) tools for coping with our world; but forsocial animals, they could not serve if they did not have a rhetorical function.7

There is another form of answer to Stroud. We don’t need independent ac-cess to the world if we can assume there is a necessary connection betweensome method, say, the method of science and truth. Thus, with persistent ap-plication of our method, our (un-Peircean) individual, given a (Peircean) full-ness of time, will arrive at truth. But even this act of faith does not help us.Since we will always lack independent access, we can never know whetherwe have arrived!

Here we can pause to consider, even if too briefly, an argument put forwardby Michael Friedman in his essay, “Truth and Confirmation” (Kornblith,1985). He points out (rightly) that there is no necessary connection betweenconfirmation and truth and that what traditional philosophy of science has tooffer on the relation cannot be sustained.8 In particular, he argues that if sci-entific method (or any other) is to show that it achieves (or even approxi-mates) truth, the Tarskian theory of truth (shared by all traditional candidates)must be supplemented to include a causal theory of reference. Since ourmethods cannot guarantee success, “we have to know facts about the actualworld if we are to know which method is best; and we have to know factsabout the actual world to know even that any given method has any chance at

Naturalizing Epistemology: Reconstructing Philosophy 121

Page 151: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

all of leading to truth!” (Kornblith, 1985: 155–56). To do this in a nonvi-ciously circular way “we need general laws connecting physics and psychol-ogy [sic] with the theory of truth; and it is precisely this kind of generalitythat a theory of reference tries to provide” (161).9

There are two fairly obvious objections. First, even given these laws, it isby no means clear what scientific method is and thus what and how it is to betested (of which more below). But, second, as Friedman says, we lack utterlysuch general laws. Worse, I believe that there is little reason to suppose thatsuch are possible. To anticipate, Friedman’s program seems, at least, to fol-low Quine’s in being committed to an epistemological individualism.10


There are, I believe, about nine (or forty?) ways into this. One way is to ob-serve that experience requires concepts and that most of our concepts—atleast—are social products. Thus, I cannot “see” a tree unless I have the con-cept “tree” and this is surely learned in a social process.11

The epistemological problem that is at issue here was surely propelled bymodern science, but contrary to what Quine (and Dewey) imply, acknowl-edging this is no advantage for the naturalizing epistemologist. It is, indeed,because of modern science that, as William James rightly saw (versusSpencer), the epistemological problem is so intractable. Naive realism could(and does!) sustain epistemological individualism: if you don’t believe thereare red apples in the world, then just look and see! But if you take modernphysical science and Dewey’s views of experience seriously (as I think wemust), then it is a rather gigantic system of belief, recognition, and the rest,which has been instituted under the influence of custom and tradition. Giventhis, how can we be so confident that our beliefs correspond to a world thatexists independent of either you or me?

This does not mean that I am deluded in saying “there is a red apple” whenI see a red apple. That is not the issue. Plainly, science did not undermine ourordinary ways of thinking and speaking. When a G. E. Moore says, “I knowthat there are red apples,” and a neuroscientist says “The experience of redapples is the product of physical and biochemical transactions between some-thing and us” and the skeptic says, “Nobody could know that there are red ap-ples,” the same words are being used differently (Stroud, 1985: 76). WhatMoore and the rest of us say, even if true, is not decisive as regards either theepistemological or the scientific investigation of knowledge.

We began with Quine’s psychological program; this is the appropriatepoint to refer to recent sociology of knowledge, to the so-called strong pro-

122 Chapter Six

Page 152: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

gram (Manicas and Rosenberg, 1985; 1988). Its key insight is what Barry Barnes calls “the naturalistic equivalence of the knowledge of differentcultures.”

Naturalism . . . implies the most intensely serious concern with what is real. . . . Everything of naturalistic significance would indicate that there is indeedone world, one reality, “out there”; the source of our perceptions if not their to-tal determinant [that is, though not their total determinant], the cause of our per-ceptions being fulfilled or disappointed, of our endeavors succeeding or beingfrustrated. But this reality should not be identified with any linguistic account ofit, or needless to say, with any way of perceiving it, or pictorial representationof it. Reality is the source of our primitive causes, which, having been presup-posed by our perceptual apparatus, produces changes in our knowledge and theverbal representations of it which we possess. All cultures relate symmetricallyto this reality. Men [sic] in all cultures are capable of making reasonable re-sponses to the causal inputs they receive from reality—that is, are capable oflearning. That the structure of our verbal knowledge does not thereby necessar-ily converge upon a single form, isomorphous with what is real, should not sur-prise us. Why should we ever expect this to be a property of our linguistic andcognitive capacities? (Barnes, 1977: 25).

Because, like Quine, Barnes and Bloor take science seriously, they believe (a)there is an independently existing world, but (b) they also believe with Deweythat human cognition is always socially mediated. (See Chapter 4.)

The idea that knowledge of different cultures is naturalistically equivalentis both a premise and a conclusion of strong program science.12 Strong pro-grammers are interested in understanding belief, and therefore, for scientificpurposes, beliefs which we think of as rational—including accordingly, thosewhich are fixed scientifically—must be treated as on the same footing as allothers.13 The belief “I see a Panda now” involves language. Hence social con-siderations are relevant. Just because our only access to a world is causal, andepistemological individualism cannot be sustained, a naturalistic epistemol-ogy interested in explaining knowledge must appeal to social facts. Not onlydo these enter into concept formation (enormously complicating empiricalpsychology), but we need to acknowledge that the problem of reproducing thecognitive order could not possibly be explained without a sociological under-standing of the relevant social mechanisms, for example, how belief is au-thorized and stabilized.

On the other hand, because our best science implies that all we can have isa representation of reality, and because there is no way to measure any repre-sentation against reality-in-itself, we cannot escape a relationalism. But itdoes not follow from this that all truth claims are equally good. At this point,we can turn to Dewey.

Naturalizing Epistemology: Reconstructing Philosophy 123

Page 153: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


It is not perfectly clear what Dewey would say to a Karam defender of hissystem of classification. Who here has the truth? He might say that thequestion is badly posed. He might say that since “true” presupposes an ab-solutist conception that is neither necessary nor possible, both are right.Although the world is structured, it does not allow us to discriminate be-tween contrary taxonomies. It does constrain these: A culture could not, forexample, treat what we have identified as poisonous as foods, for their bi-ology will not allow them to survive. But there is nonetheless plenty ofroom for alternative and contrary schemes depending on a host of alterna-tive contingent factors regarding beliefs about the gods, the good, etc. Onthis view of the matter, there are too many truths and it is idle to supposethat any can be privileged.

There is an attractive aspect to this move. Given that peoples have differ-ent interests and different ideas about the gods and the good, we need to ac-knowledge that they may very well be able to justify their beliefs about theway the world is—however strange these may seem to us. There is, unfortu-nately, an unattractive aspect to this posture. Not only does it disavow an at-tempt to give any special credence to the claims of science (perhaps not sucha bad thing?), but it disavows any effort to provide guidance about how weought to go about finding out what to believe, including here, lest we forget,beliefs about what is good and right. I think that we can do better. So, too, didDewey.


Dewey wrote that “the methods of knowing practiced in daily life and scienceare excluded from consideration in the philosophical theory of knowing”(MW, Vol. 10: 37). Presumably, “the actual process of knowing;” involves“operations of controlled observation, inference, reasoning, and testing.”While this seems true enough, it does not help us in the present instance.Surely, Karam do all these things even as they are arriving at a different tax-onomy than ours. Are the Karam going about the business of inquirywrongly? Perhaps what is needed is a more systematic attempt at discrimi-nating the special features of successful knowing. Dewey set this as the goalof inquiry into inquiry. Thus, he writes:

The position here taken holds that since every special case of knowledge is con-stituted as the outcome of some special inquiry, the conception of knowledge as

124 Chapter Six

Page 154: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

such can only be a generalization of the properties discovered to belong to con-clusions which are outcomes of inquiry. Knowledge, as an abstract term, is aname for the product of competent inquiries (LW, 12: 16).

That is, by examining the outcomes of inquiries, we can discover why theseare knowledge and, conjunctively, by examining inquiries we can discoverwhat makes for competence. As Dewey insisted: “Through examination ofthe relations which exist between means (methods) employed and conclu-sions attained as their consequence, reasons are discovered why some meth-ods succeed and other methods fail” (LW, 12: 17).

It is important to see what this program is and is not. Not only was Deweynot altogether clear regarding what is involved, but more troublesome, theprogram does not neatly join with work being carried on today in philosophy,psychology or sociology as they are generally practiced in today’s academicdisciplines. Perhaps Dewey was off the mark, or perhaps the fault is with thedisciplinary division of labor.


It may be best to proceed indirectly and to begin by noting that Dewey’s pro-gram is not akin to the psychologically oriented programs of Quine or, for ex-ample, William Lycan (1988). Quine has not said very much about the sort ofpsychology he assumes will explain how we know, but we may guess that itis some sophisticated version of behaviorism. By contrast, Lycan is very clearin his commitments to a Fodor/Dennett-inspired homuncularism, a currentlyfashionable version of cognitive psychology.14

But in either case, Quine, Lycan, and the psychologies they presume areepistemologically individualistic and Dewey’s psychology was not. More-over, these writers and the psychologies they want to include are committedto a logical theory which Dewey found to be misdirected.

Throughout his career, from his brilliant essay on the reflex arc, throughthe 1903 studies in experimental logic, to How We Think (1910), to his ill-understood Logic, Dewey developed a naturalistic theory of inquiry that to-tally went against the dominating and now taken-for-granted Frege-Russellconception of logic.15 In this view, logical relations hold between abstractpredicates and inference (deductive and inductive) depends on there beingsome sort of objective relation between propositions.16 Because this as-sumption was a feature of what Dewey called “intellectualism”; he lookedat the matter entirely differently. As Thomas Burke says, Dewey’s concep-tion of inquiry “has to be understood not so much as cognitive problem

Naturalizing Epistemology: Reconstructing Philosophy 125

Page 155: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

solving but more generally in terms of an adaptive stabilization propensityof organism/environment relations.”17 This basic naturalistic starting pointled Dewey to totally refashion “inference,” “propositional content,”“kinds,” and other critical terms in standard logical theory. Thus, as I un-derstand Dewey, “inference” is fundamentally a way of handling informa-tion, which does not require human language. The logician’s concept of in-ference is not, of course, to be abandoned; it is rather to be seen as a highlyuseful abstraction, regimented for particular purposes.18

This is hardly the place to develop the radical implications of Dewey’s re-vision. Continuing along lines just suggested, one example will have to suf-fice. Enormous effort in psychology has been directed at solving Meno’s par-adox: “If inputs require concepts to be meaningful, then concepts mustprecede ‘inputs’ as in nativism; but if concepts (to be at all useful in the realworld) require ‘input’ for their content, then ‘inputs’ must precede conceptsas in empiricism (either of the ontogenetic or phylogenetic variety).”19 Forexample, behaviorist-learning theory needs to assume that the organism hasmade the relevant abstraction if it is to be reinforced. But this assumes exactlywhat needs to be explained. On the other hand, recent cognitive psychology,by conceiving of mind as an information processing system, assumes that in-formation comes sententially prepackaged, ready-made for use by the lin-guistically apt learner.

If I am correct we can now identify three obstacles to an adequate under-standing of knowing: a pervasive intellectualism, an epistemological individ-ualism, and third, a pervasive assumption, shared by the main contendingviews, that, as Kelly and Kreuger put it “the only relations between contentsof cognitive states which makes a process involving those states a cognitiveprocess are the sorts of logical functions used in classical experiments”(Kellyand Kreuger, 1984: 64). But, of course, on the standard view of logic, logicalfunctions can hold only between abstract predicates. Meno’s dilemma is theninescapeable! Indeed, until psychology breaks from those philosophical dog-mas that have formed it, we shall not have an adequate psychology of learn-ing, and we shall not naturalistically understand knowing. Dewey’s path, if Iam right, was the right one.

But even if we accept completely Dewey’s picture of inquiry (including ahost of details yet to be filled in), this would not, of itself, respond to the prob-lem of judging between the belief systems of the Karam and the Western zo-ologist. Presumably, Dewey’s account applies to both. If only the zoologistgets it right, then something more is being assumed, likely that somethingcalled the method of science privileges the findings of the zoologist. But wehave yet to see the argument for this.

126 Chapter Six

Page 156: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


There is an entirely different program that is rightly construed as inquiry intoinquiry. Dewey offered (as we noted) that “through examination of the rela-tions which exist between means (methods) employed and conclusions at-tained as their consequence, reasons are discovered why some methods suc-ceed and other methods fail” (LW, 12: 17). This might be understood asmeaning that the task is not only to frame theory which describes and ex-plains the general feature of inquiry, but consistent with this, to consider em-pirical science empirically. It might then be possible to generate warrantedmethodological rules, thus satisfying the demand that we be able to judge be-tween contrary beliefs and belief systems. This version of a naturalistic pro-gram has had some recent advocates, among them, most outstandingly,Nicholas Rescher and Larry Laudan. There are, I think, three main features,which distinguish this approach.

First, the Deweyan inspiration is in the effort to avoid a vicious circularityin which one either justifies outcomes by assuming that means are competent,or warrants the means by assuming the truth of the outcomes. If as Rescherputs it, “justification is here an essentially two-way process—its results legit-imate the method as proper and appropriate, and the method justifies its re-sults as ‘correct’” then one needs either to break the circle or to show that itis nonvicious (Rescher, 1975: 27).

Briefly, Rescher argues that “any experiential justification of a truth crite-rion must pull itself up by its own bootstraps—it needs factual inputs, but yetfactual inputs cannot at this stage already qualify as truths.” To meet thisneed, accordingly, Rescher appeals to “truth candidates”; “data which are nomore truths than candidate-presidents are presidents . . .” (28).20 Rescher thenenvisages “a feed-back loop” in which “the reasonableness of the over-allprocess . . . rests not only on the (external) element of success inherent in thefactor of pragmatic efficiency, but also on the (internal) factor of intrinsic co-herence and the mutual support of self-substantiation that the various stagesof the whole are able to lend to one another” (36).

Laudan offers what he calls a “reticulated model of scientific rationality”which explicitly introduces values:

The reticulational approach shows that we can use our knowledge of the availablemethods of inquiry as a tool for assessing the viability of proposed cognitiveclaims. . . . Equally, the reticulated picture insists that our judgments about whichtheories are sound can be played off against our explicit axiologies in order to re-veal tensions between our implicit and explicit value structures (Laudan, 1984: 62).

Naturalizing Epistemology: Reconstructing Philosophy 127

Page 157: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

For Laudan, fully in the spirit of Dewey, “axiology, methodology, and factualclaims are inevitably intertwined in relations of mutual dependency” (63).

Second, both writers (Rescher explicitly, Laudan implicitly) reject a thesis(or propositional) pragmatism in which the problem is to vindicate particulartruth claims. Instead, they opt for a methodological pragmatism in which theproblem is to justify methods. Thus, “pragmatic considerations are neverbrought to bear on theses directly. The relationship becomes indirect and me-diated; a specific knowledge claim is supported by reference to a method,which in turn is supported on pragmatist lines” (Rescher, 1975: 73). Beliefsarrived at with warranted methods may very well be false. The aim, however,is to find methods which are reliable in the sense that they answer to humanpurposes, critically assessed.

Thesis pragmatism is highly vulnerable to a wholly idiosyncratic mode de-termining what counts as warranted assertibility. This is, of course, a long-standing objection to pragmatism. Methodological pragmatism offers hopesince beliefs are warranted only insofar as they are the outcome of an explicitmethod that has been warranted independently of this or that particular belief.As Rescher points out:

Considerations of the suitability and effectiveness of methods introduce an inher-ently rational orientation, which serves to assure the logical properties. Moreover,methods are intrinsically public, interpersonal, and communal. A method is not asuccessful method unless its employment is generally effective—otherwise we aretalking about a knack or skill rather than a method. A skill can only be shown, itcannot be explained. . . . This line of thought indicates the fundamentally social di-mension of methods. . . . They can be examined and evaluated in abstracto, with-out any dependency on particular practitioners (Rescher, 1975; 73).

These are certainly desirable features of this program, even if as I shall sug-gest, instead of methods, we are better advised to try to warrant practices. Butwe should first notice that if the program carries, we will have escaped sub-jectivism, we will have warranted our method and thus beliefs which are de-termined by means of these methods, but we will not have secured truth. But,of course, Dewey’s shift to warranted assertibility was a rejection of thesearch for truth (understood, as always, in the absolutist sense).21


It will be important to notice the bearing of the foregoing on the question ofmoral relativism. In my view, this is surely the most important of the trou-

128 Chapter Six

Page 158: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

blesome questions raised by relativism. Plainly, I cannot pursue this here. Yet,it seems to me that it is just here that the foregoing is most helpful. The rea-son is clear enough. The skeptical objection forecloses the possibility of se-curing a perspectively neutral truth about a world, which exists independentlyof you and me. But since on naturalistic grounds, our ideas about what is goodand right are our ideas, the skeptical objection has no force. All that we needas regards questions of the good and the right is warranted assertibility. More-over, in this context, the public, interpersonal, and communal aspects are fun-damental. While the effort to secure ever-inclusive representations of the ex-ternal world cannot secure truth about it, the effort to secure ever-inclusivegoods is exactly what is called for as regards moral matters.


There is a third aspect to methodological pragmatism. It presumes that an em-pirical study of science will yield clarity about aims and methods, and thatthere is a way to reflexively test methods against aims, once identified. As faras I know, Laudan (and his associates), have been in the forefront of actuallyengaging in such research.22 But I think that on this count, there are some verydifficult problems.

First, there is the abstraction science. It is easy to suppose that althoughthere are manifest differences in the sciences, the term, “science” is mean-ingful because the sciences share in goals, for example, prediction andcontrol, and/or because there is something called “scientific method,”again, usually defined in terms of a series of abstractions about the forma-tion, deductive elaboration, and testing of hypotheses. Dewey was, I be-lieve, utterly uncritical in this.23 Laudan acknowledges that goals do differand that, pertinent to this and to subject matter, methods (not merely tech-niques) may vary. Still it would seem that an adequate empirical picturewould show some fundamental differences, not only in the sites and goalsof the practices of the sciences, but in their methods and standards aswell.24

Consider first the idea of the goals of these practices. Even a cursory ex-amination would show, I believe, that there are at least four fundamentallydifferent goals currently operative in the sciences.

1. Description: for example, ethnographic work in anthropology; quantita-tive research in economics or demography; much geography; and taxo-nomic work in botany and zoology (motivated, I believe, by very differentgoals than the prescientific taxonomy of the Karam!).

Naturalizing Epistemology: Reconstructing Philosophy 129

Page 159: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

2. Prediction and control: behavioral social science, most psychology, mete-orology, the engineering sciences, and applied sciences.

3. Understanding: basic science, including work in space/time theory, quan-tum mechanics, evolutionary theory, some (but surely the smallest part) oftheoretical work in psychology and the social sciences.

4. The explanation of concrete events: history and the historical sciences (in-cluding here, some social science, some geology, some evolutionary biol-ogy, some psychology).

It has been easy to collapse these cognitive goals. For example, by means ofthe idea that the discovery of laws is the goal of science, it has been easy tobelieve that explanation, understanding, and prediction are of a piece. Butwhile the point cannot be pursued here, it is easy to show that these are con-ceptually, and in practice, distinct aims (Manicas, 2006). Empirical examina-tion of practice would show, I believe, that those practices which aim at pre-diction and control (implicitly or explicitly) offer nothing in the way ofunderstanding or in the explanation of concrete events. For example, behav-iorist psychology gives no understanding of learning; and it cannot explainthe most elementary concrete act, for example, my response, “fantastic,” toseeing Guernica for the first time. But—and this is not to be minimized—be-haviorist psychology has been an effective tool for manipulation and control.

But the point of this sketch is not to settle issues, but to raise questions for theempirical program that I have called “Inquiry into Inquiry: II” Speaking nowwithin its frame of reference, if the goals are different, then we can expect themethods to be different. For any particular goal, there still will be methods thatare most effective and suitable. And we can still endorse the basic Deweyan ef-fort to self-referentially bootstrap. But not all these goals will be pertinent to theproblem with which we began. The skeptical objection is plainly irrelevant if the aim is prediction and control. Moreover, it is very easy to justify science asthe preferred mode if prediction and control is the goal. Indeed, this is a majormotive for continuously attractive instrumentalist theories of science. If, how-ever, one is interested in understanding or in explaining what happens, then in-quiry into the practices of sciences with those aims will be pertinent. Laudan andhis colleagues look at the “basic” natural sciences (despite their antirealism). Ifthey had looked at most mainstream psychology, indeed at most mainstream so-cial science, things would look very different.

But let us assume that the intention is to get an “empirically well-groundedpicture” of those practices that in fact (and not merely in intention) aim at giv-ing an account of how the world is. Of course, it will not do, as Laudan hashimself so strenuously insisted, to accept those descriptions of science thatare written with manifest assumptions imported from Carnap or Popper, the

130 Chapter Six

Page 160: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

two dominating traditions of twentieth-century philosophy of science. Thesestudies cannot count as tests because they are self-authenticating. Nor can we,uncritically, accept what practitioners say are their methods (or for that mat-ter, their standards and goals). As Einstein remarked, “If you want to find outanything from the theoretical physicists about the methods they use, I adviseyou to stick closely to one principle: Don’t listen to their words, fix your at-tention on their deeds.”25 We have the best chance of getting some under-standing about what they do if we can study activities directly. For standardhistoriographical reasons, things get much more difficult when we considerpast practices. Indeed, in the face of these problems, there may be a tempta-tion to assume methods and then to assume that they determine outcomes. Wethen enter history less problematically, with an eye merely to these.

Laudan’s program risks this. Thus, he seems (at least) to begin with hy-potheses regarding methodological rules that derive from philosophies of sci-ence. The idea then is to enter into history and seek either confirmation or dis-confirmation of these. But there are now two additional problems. First, if asLaudan asserts, science changes, it is not clear how much generalizing will bepossible26 and thus, to what degree we can regard conclusions drawn fromsuch inquiry as tests. For example, consider but the institutional differencesbetween Newton’s scientific research and the big science of today. Assuming(what is likely contrary to fact) that the goals are comparable, can we be confident that abstracted methodological rules effective then would now beeffective?

Second, as recent studies surely show, agents making decisions in scienceare complexly affected. Not only are they capable of self-delusion (likeeveryone else), but rules, even if they are crisply formulated and form a con-sistent set, need to be applied concretely. This is hardly to say that methodsare irrelevant. Rather it is to assent to Kuhn’s view, rejected by Laudan, thatmethodological criteria rarely if ever determine choices between rival theo-ries. As Kuhn (and strong program writers) have insisted, this is not to denyrationality; it is to affirm that rationality is both changing (as Laudan admits)and concrete, exactly in a more Deweyan sense that we cannot explainchoices by subsuming them under rules.

One thrust of my argument has been against philosophers (and those influ-enced by these) who, despite the best intentions, have been unable to freethemselves from the shackles of traditional epistemology. Another thrust hasbeen to sympathize warmly with pragmatic approaches, but to suggest thatamong the most outstanding of these, there are serious problems to be faced.Before concluding, I summarize:

First, if we are to understand knowing naturalistically, we need to rethink,in Deweyan terms, the psychology and logic of knowing. This will require, if

Naturalizing Epistemology: Reconstructing Philosophy 131

Page 161: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

I am correct, some important changes in the mainstream practices of empiri-cal psychology. (See Chapter 2)

Second, since knowledge is inescapeably a social product, we need to wel-come the efforts of recent sociology of scientific knowledge. Our “natural-ism” cannot be “half-hearted.” (See Chapter 4.)

Third, having achieved a better understanding of the production ofknowledge, if we are to seek warrant for beliefs (or to prescribe norms forbelief), we will need to embrace a Deweyan approach to the fact/value di-chotomy. We need to acknowledge, straight out, as Sleeper puts it, thatsince “all judgment is practical . . . there is no gulf between intellectual andpractical judgment.”

Fourth and finally, instead of trying to warrant methods, we will be betteradvised to try to warrant practices. I conclude with a sketch of what I meanby this.


Practices are, roughly, ways of doing. Practices include the beliefs of practi-tioners, the tools they use, their explicit goals, and much else besides. Prac-tices are institutionalized (structured) activities, activities that presupposehabits in Dewey’s sense, dispositions that carry the legacy of training and cus-tom. The shift from methods to practices has consequences:

First, we will not be stymied if, as seems to be the case, much of what isknown by practitioners is not formulable in terms of rules, but is tacit, craft-like, and learned at the side of experienced mentors. One learns how to usethe tools, not merely the instrumentation, but the special languages, for ex-ample, the mathematics, and the standards for employing them. One learnswhat counts, what are the pitfalls, what are the ongoing standards of ade-quacy. Indeed, understanding these is precisely what would count as under-standing a practice.27

Second, the shift to practice allows us to acknowledge that structured ac-tivities have unintended consequences. This includes not merely the uses towhich basic work can be put, but the potential that outcomes may be surpris-ing and hence not subject to control, and that intentions may be frustrated andtransformed. Third, as part of the picture, we can include the real possibilitythat actors engaged in a practice can have false consciousness: they may havebeliefs which are essential to the practice in the sense that if they have believed otherwise, they would not do what they do, but these beliefs mightbe false in the sense that actors may not fully understand just what they aredoing. For example, they may believe that they are Popperians or instrumen-

132 Chapter Six

Page 162: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

talists when in fact they are not; they may believe that they are explainingwhen they are merely controlling; they may believe that their research is un-contaminated by interests foreign to their aims, etc.

A fourth advantage of this shift is that it allows us to incorporate as criticalthe fact that scientific practices are enmeshed in, effecting and effected by, ahost of other practices: economic, educational, and political. Thus, the polit-ical economy of “big science” is critically relevant to understanding how itsproblems get defined and how it approaches and resolves them.28 The issue isnot merely that the goals and methods of the practice of big science are notautonomous, but that nonscientific factors are playing critical roles in consti-tuting these practices.

Finally, we can be sensitive to the fact that scientific practices are very dif-ferently constituted, not merely between and among disciplines, but acrosstime. Given this, a global defense of science may not be possible. On theother hand, we do not need a global defense. We need only to learn by inquirywhat it is that makes a practice warrantable.


With these considerations in mind, there is at least one recent work to whichwe can point. Rom Harré is a trained physicist/philosopher who takes fully toheart the idea that (a) we had best look at scientific practices; (b) that a de-fense of practice in theoretical/experimental physics is not, tout court, a de-fense of science or of scientific method; and (c) that such a defense must rec-ognize the skeptical challenge with which we began. Plainly, this is not theplace to detail Harré’s important work, but I believe that (with some minoremendations), it is entirely congenial to the views of this chapter.

Begin with (c). Harré rejects what he calls “truth realism;” roughly that abelief is true if and only if it corresponds to reality. Sensitive to argumentsfrom Hume to Laudan, he defends “referential realism;” roughly, the idea that“some of the substantive terms in a discourse denote or purport to denote be-ings of various metaphysical categories such as substance, quality, or relation,that exist independently of that discourse (Harré, 1986: 67). In terms of ourearlier discussion, not only is there an external world, but given what weknow, the most plausible causal theory of perception is Gibsonian. That is,“while one must concede that there could not be psychological laws whichexplain how someone came to see a pencil, it does not follow that there couldnot be psychological laws which explain how someone came to see long, thinthings, causal sequences, and other generic perceptibles” (154). On this view,

Naturalizing Epistemology: Reconstructing Philosophy 133

Page 163: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

these are natural “affordances,” possibilities of action ecologically offered tonaturally evolved species and found in the exploration of the ambient array.But plainly, this story will not suffice to explain our relevant taxonomiessince, as above, Karam see (mediately) kobity and we see birds (or if we arebirdwatchers, we see cassowary).

We need to make room for the social component of knowledge. Harré ex-ploits Dretske’s explication of “seeing that. . . .” Thus,

1. S sees b.2. The conditions under which S sees b are such that it would not look the

way it does look, say L, unless it were P.3. S, believing 2, takes b to be P.

Here, b is a Gibsonian invariant; condition 2 introduces S’s corpus of priorbelief. That “b is P” is knowledge, but, plainly, it is relative to the corpus ofbeliefs held by S, and there is no way to find some original, terminal, or foun-dational belief! We have found a toehold on the world, but we have not se-cured an absolutist conception of knowledge. Nor have we secured science.

We can imagine a discourse, Harré calls it Realm I discourse, which madereference only to the states and relations of beings known in actual experience(the heaven of empirical realisms!). Could such a discourse sustain a science?No doubt, human communities have put considerable attention on classifyingbeings in Realm I discourse, but as is now sufficiently clear, the boundaries“which serve to maintain discrete groupings in any human classificatory prac-tice cannot be justified without reference to unobservable properties andstructures of the beings in question” (179).29

Harré’s problem is now clear. Can he justify theoretical/ experimentalphysics as a preferred mode of fixing belief about the external world?Grasping fully the idea that “the science we consume, so to speak, is thefinal product of the complex interplay of social forces and cognitive andmaterial practices” (and not the product of a “logic engine”), he argues thatone must acknowledge that scientific communities control their products“by the informal yet rigorous maintenance of a moral order” (12). Indeed,on Harré’s view, a great deal of the best work by philosophers of scienceis most usefully understood as sketching an ideal moral order, not an ideal(or still less, real) epistemic order. On this reading, the (epistemic realist)manifesto, “Scientific statements should be taken as true or false by virtueof the way the world is” as a moral principle becomes: “As scientists, thatis, members of a certain community, we should apportion our willingnessor reluctance to accept a claim as worthy . . . only to the extent that we sin-cerely believe that it somehow reflects the way the world is.” Similarly, the

134 Chapter Six

Page 164: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

idea that we should seek falsification cannot be sustained as an epistemicprinciple. But it has manifest merits as encapsulating moral injunctions:for example, “However much personal investment one has in a theory, oneshould not ignore contrary evidence” (90).

For Harré, then, “facts” are socially constructed, but not only are they not constructed from whole cloth (the burden of referential realism), but to the extent that the ideal moral order is functioning, then the results are to bepreferred—on strict pragmatic grounds.


Harré appeals to empirical studies of scientific practices to identify what, per-tinent to the problem of knowledge, is distinctive of these and, then, why, ide-ally speaking, they should be preferred. Speaking as a naturalizing episte-mologist, this is all that we can demand. He does not, to be sure, say verymuch about the social conditions that would seem to be requisite to sustain-ing the ideal moral order. In general terms, these are the ideas that we famil-iarly associate with Peirce and Dewey, critically, the ideas of publicity and ac-cess.30 But it is also clear that under conditions of industrialized science, it isjust these conditions that are now under threat.

“Shoddy science” becomes possible when published papers are not be-ing read and thus not subjected to critical scrutiny.31 But since they easilybecome part of the construction of facts, how can we know what to trust?“Entrepreneurial science” allows contractors to establish huge mission-oriented, capital-intensive enterprises in which researchers lose all inde-pendence and everyone else is denied access. Since these products are notassessed by consensus, why should they be trusted? “Runaway technol-ogy” can produce “reckless science.” Here ready access to millions of dol-lars aimed at some specific technical power, for example, the manipulationof genetic materials, can produce shoddy science now accompanied by therisk of catastrophic consequences. Finally, there is “dirty science” in whichopportunities to fund research projects aimed at realizing understandingare converted into technologies for state purposes of destruction, or con-trol, or manipulation.

We thus come full circle. We are stuck with our history. With the inventionof modern philosophy as a discipline pretending not only autonomy but aprivileged role in the intellectual division of labor, philosophers unwittinglyconspired in mystifying a world in which science has played a profoundly im-portant role. Seventy years after Dewey’s call for reconstruction, the need is,if anything, even more urgent.

Naturalizing Epistemology: Reconstructing Philosophy 135

Page 165: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


1. I have made two efforts at this, in A History and Philosophy of the Social Sci-ences (New York and Oxford: Basil Blackwell, 1987) and War and Democracy (NewYork and Oxford: Basil Blackwell, 1989).

2. Quine’s 1969 essay and Stroud’s 1981 “The Significance of Naturalized Episte-mology” are reprinted in Hilary Kornblith’s influential anthology, Naturalizing Epis-temology (Cambridge: MIT Press, 1985), cited in what follows.

3. Quine’s picture, presumably, is that, in Carnapian fashion, our knower can con-tinually revise C-functions or in Popperian fashion, she can continue indefinitely toconceive hypothesis, which she tries to falsify. Note also the epistemological individ-ualism. See below. These two programs in philosophy of science have been the mostinfluential epistemologies in our century, but, as Laudan observes in a wonderful under-statement, they “have run into technical difficulties which seem beyond theirresources to surmount” (“Progress or Rationality: The Prospects for Normative Nat-uralism,” American Philosophical Quarterly, Vol. 14, No. 1 [January 1987]: 19).

4. See John Dewey, “The Existence of the World as Logical Problem” (MiddleWorks, 8: 94–95). Of course, there are other forms of skepticism. See, for example,Thomas Nagel, The View from Nowhere (New York: Oxford University Press, 1986):71. J. E. Tiles, in Dewey (London: Routledge and Kegan Paul, 1988), quotes Russell’scomplaint that “Professor Dewey ignores all fundamental skepticism. To those whoare troubled by the question: ‘Is knowledge possible at all?’ he has nothing to say”(14). Tiles retorts that this is not fair: “what Dewey had to say was that the questionlacked foundation” (14). But “the question” is ambiguous. Whether there is an exter-nal world that is at least partly structured cannot be motivated. But whether knowl-edge is possible, given our history, does have a point. See below.

5. See Tiles, Dewey: 70–76, 116–23, 127–29. My account departs, however, fromTiles (and from Dewey?). Holding to an absolute conception of reality does not com-mit one to an absolute conception of knowledge. It is not my contention that we coulddescribe the world “from no point of view.” Knowing is necessarily a relation be-tween a situated knower and “the world.” But unless being depends upon knowing,this does not make whatever is at the object end either featureless or unknowable. Onthe other hand, Dewey was correct to insist that objects of knowledge (the characterof things as known) were produced by inquiry. But because they are not producedfrom whole cloth (either by individuals or groups!), the skeptical problem arises.

6. Fallibilism, according to Nagel in The View from Nowhere, holds that our beliefs“go beyond their grounds in ways that make it impossible to defend them againstdoubt” (68). Nagel here is defining skepticism, not fallibilism! It is hard to say howmuch disagreement in epistemology turns on different usages.

Peirce’s limit conception of truth provides an anchor, but at a cost. See below.Dewey seemed at least to subscribe to Peirce’s conception. See John Dewey, Logic:The Theory of Inquiry (LW, 2: 345).

7. See David Bloor, Knowledge and Social Imagery (London: Routledge andKegan Paul, 1974): 32–33; Barry Barnes, Interests and the Growth of Knowledge(London: Routledge and Kegan Paul, 1977). In his Dewey, Tiles holds that Dewey’s

136 Chapter Six

Page 166: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

fallibilism was secured by the idea that inquiries comprise a continuum and suggeststhat Dewey was correct to de-emphasize Peirce’s limit theory (Tiles, 1988: 106–8). Idoubt this. Where there is no doubt, there is no inquiry. But the limit notion of truthdoes, at least, give the dissenter a rhetorical tack that is otherwise lacking.

8. Friedman offers what seem to me to be fatal objections to positivist views andto those views, like Peirce’s, which seek to ensure a connection between confirmationand truth by giving a special meaning to truth. This would include Popper and at leastsome contemporary versions of instrumentalism—perhaps Larry Laudan. As regardsthe theory of reference, see my sketch of Harré’s approach.

9. Notice that sociology is omitted. Presumably, what it has to offer is irrelevantto epistemology?

10. Quine waffles on just what he is claiming. Susan Haack holds that Quine is“ambivalent” between a reformist “Modest Naturalism” in which epistemology is anintegral part of empirical belief and a revolutionary “Scientistic Naturalism” “ac-cording to which epistemology is be conducted wholly within the natural sciences.”See her “The Two Faces of Quine’s Naturalism,” ms., n.d. On his more notorious am-biguities regarding the “validation” of claims to knowledge, see Ken Geme, “Episte-mological vs. Causal Explanation in Quine, or Quine: Sic et Non,” ms., n.d.

11. Another way into epistemological individualism is to observe (versus Quine)there is no way (as far as we can know) to go from “molecules upon our sensory sur-faces” to the rat perception of, e.g., an edible object, to the (linguistically modeled)belief that there are red apples in the world. We return to this.

12. There are important differences between those doing sociology of science asregards questions in philosophy, between (say) Barnes, Harry Collins, Steve Woolgar,and Latour. Confusion over the claims of Barnes and Bloor is now joined by confu-sion over these differences. See also the excellent more recent collection of essays ed-ited by Andrew Pickering, Science as Practice and Culture (Chicago: University ofChicago Press, 1992).

13. Thus drawing the rage of philosophers. It is presumably one thing to explain irra-tionally fixed belief by appeal to sociological facts; it is quite another thing to supposethat rational belief needs these. Presumably, one must contrast my belief that some fig-ure presently in my vision is the Virgin Mary with my belief that some figure presentlyin my vision is a panda. Anthony Flew, now speaking for countless epistemological in-dividualists, thinks that the former belief admits of a sociological explanation, but that ifit is being argued (and it is!) that “intrusive, non-social, physiological, and biologicalfacts” are not sufficient to explain this latter belief, then the view is “manifestly prepos-terous and in its implications, catastrophically obscurantist” (Anthony Flew, “A StrongProgramme for the Sociology of Belief,” Inquiry, Vol. 25 [1982]: 366–67).

14. According to Lycan, “occurent beliefs are sentencelike representations storedand played back in our brains” (1988: 6). A belief, then, “is epistemically justified ifand only if it is rated highly overall by the set of all-purpose, topic neutral canons oftheory-preference that would have been selected by Mother Nature for creatures ofour general sort . . .” (160).

15. See Ralph W. Sleeper’s important The Necessity of Pragmatism (New Haven,Conn.: Yale University Press, 1986); and Thomas Burke, “Dewey on Defeasibility,” in

Naturalizing Epistemology: Reconstructing Philosophy 137

Page 167: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Situation Theory and Its Applications, eds. R. Cooper, K. Mukai, and J. Perry, (Stanford,Calif.: CSLI Publications, 1990): 233–68 and Chapter 2.

16. Confirmation theory is the skeleton in the closet of empiricist epistemologiesof this century. For some of the key papers, see P. T. Manicas, ed., Logic as Philoso-phy (New York: D. Van Nostrand, 1972).

17. As Burke writes:

The basic scenario is that a given organism/environmental system is constantly perform-ing certain operations as a matter of course employing sensory mechanisms, scanning,varying, probing, and otherwise moving about and altering things. Inquiry is initiated bysome unsettling perturbation. . . . None of this needs be “deliberate:’ Dewey’s picture ofinquiry is supposed to describe general architectural and dynamic features of virtually anyconstituent subsystem of living animals, characterizing the simplest cellular life-functionsas well as the most complex motor activities” (1990: 236).

Classical epistemology is intellectualist in that it miscontrues “experience” andthen conflates “having of an experience” with knowledge. Experience is “an affair ofthe intercourse of a living being with its physical and social environment”; it is notprimarily “psychical;’ nor “a knowledge affair”; and it is “pregnant with connexions”and “full of inference” (MW, 10: 6).

18. Compare Barnes and Bloor, “Rationalism and the Sociology of Knowledge,”in Rationality and Relativism, eds. Hollis and Lukes (Cambridge: MIT Press, 1982):44. For a provocative treatment in the context of recent philosophy of mathematics,see Mary Tiles, Mathematics and the Image of Reason (Cambridge: Cambridge Uni-versity Press, 1991).

19. M. T. Turvey, R. E. Shaw, E. S. Reed, and W. M. Mace, “Ecological Laws of Per-ceiving and Acting: In Reply to Fodor and Pylyshyn” Cognition, Vol. 9 (1981): 285. Seealso W. B. Weimer, “The Psychology of Inference and Expectations: Some PreliminaryRemarks,” in Minnesota Studies in the Philosophy of Science, eds. G. R. Maxwell and A.R. Anderson, Vol. 6 (Minneapolis: University of Minnesota Press, 1975).

20. See Guy Axtell, “Logicism, Pragmatism and MetaScience,” in Philosophy ofScience, (1990).

21. Of course, the pragmatist, rejecting the problem of knowledge überhaupt andalive to differences in aims, cognitive and otherwise, is open to the possibility that dif-ferent communities with different aims, cognitive and otherwise, might well be justi-fied in their beliefs about the world. Thus, it is not clear that Karam methods, perhapsinformed by and tested against goals that, for example, emphasize harmony with thenatural world, are not justified.

22. Expanding on work in his Progress and Its Problems, Laudan has provided a“test” of “realist axiology and methodology” in his Science and Values, Chapter 5. Heconstrues realism as a truth-realism and then argues that a great deal of what physi-cists have believed to be true has been given up. Accordingly, realist methodologicaladvice cannot be historically vindicated. However, as Harré says, this conclusion isvulnerable to a “modest” objection: “While physicists perhaps have not been able tokeep their stock of deep fundamental theories unscathed by later developments, therehas been a continual refinement and growing repertoire of very plausible items of in-

138 Chapter Six

Page 168: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

formation about many kinds of being whose existence can no longer be seriouslycalled into doubt” (Varieties of Realism, Oxford and New York: Basil Blackwell,1987: 41). Indeed, once realism is construed as by Harré, not only can Laudan’s bul-let be dodged but an excellent case for what Harré calls policy realism can be made:“If a substantive term seems to denote a being of certain natural kind (and some spe-cial conditions are satisfied by the theory in which that term functions) it is worth set-ting up a search for that being” (59). That is, by including in their working vocabu-lary a robust referring expression, there are “features of theories which historicalexperience shows are good bets for having anticipated experience . . .” (60).

For other suggestions for rules worth testing, see Rachel Laudan, Larry Laudan,and Arthur Donovan, Scrutinizing Science (Holland: Kluwer Academic Publishers,1988).

23. Dewey inconsistently offered an instrumentalist theory of science in whichprediction and control were emphasized as the goals of the sciences. (See Chapter 1).In his antirealism as regards theoretical terms, he shares much with my former col-league, Larry Laudan. (See Chapter 5.)

24. Think of astrophysics at Princeton’s Advanced Institute and at Rome ARDC,research in solid state physics at Stony Brook or at Roswell, N.M., DNA/RNA re-search at Cold Spring Harbor and at Texas Medical Center; biochemists working atMax Factor, or on bonding metals ions to antibodies at Scripps Clinic, or neurotrans-mitters at the University of Hawai’i; economists at the Bureau of Labor, the Ameri-can Enterprise Institute, or Cambridge University, England; psychologists at MerrillLynch, in the social welfare services of the City of New York, at the New School forSocial Research, at MIT; unfunded anthropologists in Thailand and anthropologistsworking for AID in Thailand. One could easily go on.

25. Quoted by G. Holton, “Mach, Einstein, and the Search for Reality,” in ErnstMach, Physicist and Philosopher, eds. R. S. Cohen and R. J. Seeger (Dordrecht: Rei-del, 1970). The text is from Einstein’s 1933 Herbert Spencer Lecture.

26. Feyerabend surely goes too far here. Harré points out that Feyerabend aims hisguns at “the logicisms of the alleged inductive method and the fallibilism of Popper,”but this target is too restricted. More importantly as regards the present context, “theremay be more than one but not indefinitely many contexts of enquiry, in each of whichdifferent methodological and metaphysical principles, each cluster of which could betaken as defining a scientific inquiry, could be rationally defended” (Varieties of Re-alism: 24–25). This would, I think, still undermine Laudan’s program.

27. Compare, of course: Michael Polanyi, Personal Knowledge (London: Rout-ledge and Kegan Paul, 1958); Jerome Ravetz, Scientific Knowledge and Its SocialProblems (New York: Oxford University Press, 1971).

28. Merely by way of illustration, what are we to make of the fact that sixty bil-lion dollars was spent in the 1990s on a half-dozen projects—a space station, a hu-man genome project, a supercollider. And what are we to make of the criticism that“big science has gone berserk,” that “good minds and a lot of money are going intoareas that are not relevant to American competitiveness, American technologicalhealth, or even the balanced development of American science” (New York Times,Sunday, May 27, 1990: 1).

Naturalizing Epistemology: Reconstructing Philosophy 139

Page 169: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

29. They will, for example, require dispositional attributions. The clearest exam-ples are in everyday discriminations, for example, salt and sugar. But consider alsoclassifications of modern zoology, sustained (or not) by appeal to beliefs derivingfrom neo-Darwinian theory (just as the Karam classifications are sustained [or not] byappeal to beliefs which run past Realm I discourse).

30. See Dewey’s remarkable The Public and Its Problems (LW, 2) and Chapter 3.Dewey pointed out that the conditions for assessing claims were, in general, beingeroded. Thirty years later, C. W. Mills picked up this theme: As “experts” constraineddialogue, “publics” were being converted to “masses.”

31. The term “shoddy science,” the analysis, and the other categories that follow areowed to Jerome Ravetz, Scientific Knowledge and Its Social Problems, Chapter 10.

140 Chapter Six

Page 170: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Part Three


Page 171: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 172: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


John Dewey believed that Americans “were not an over-agile peoplemorally” (Middle Works,Vol. X: 261).1 Like many, then and since, he wasstruck by the stunning turn in their attitudes toward the Great War in Europe.

Our remoteness from the immediate scene of international hatreds, the bad after-taste of the Spanish American war, the contentment generated by success-ful industrialization, the general humanitarianism of which political progres-sivism was as much a symptom as social settlements, the gradual substitution ofcalculating rationalism for the older romantic patriotism, all these and more hadcreated an American sense that war was “the supreme stupidity” (MW, X: 260).

When the Great War came, some managed the shift in attitude easily. It was,in fact, depressing that so many “who when war was actually declared merelyclumsily rolled their conscience out from under the imperative of ‘Thou shaltnot kill’ till it settled under the imperative of ‘Obey thy law,’”—and this de-spite the fact that “they still saw the situation exactly as they had before”(MW, X: 263). For others, seeing the situation exactly as they had before, “thepacific moral impulse” remained steady. Now a troublesome minority, Deweyworried that they were being treated badly. Not yet clear that this was but atip of the iceberg, he wrote that they deserve “something better that accusa-tions, varying from pro-Germanism and the crime of Socialism to traitorousdisloyalty” (MW, X: 61).

For others, a “moral wrench” had been necessary, a “moral adjustmentwhich if not involving a tragedy of the inner life has been effected only withsome awkward trampling of what has been cherished as the finer flowers ofthe soul.” Indeed, Dewey could “hardly believe [that] the turnover could havebeen accomplished under a leadership less skillful than that of President

Chapter Seven

American Democracy: A New Spirit in the World

Page 173: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Wilson, so far as he succeeded in creating the belief that just because themoral impulse retained its full validity Germany must be defeated in order tofind full fruition”—the war to end all war.

Dewey’s view that it took the “skillful leadership” of Wilson must not betaken lightly. It may be that few others could have so successfully fostered the belief among Americans that by entering the war they could fulfill aunique historical task. Moreover, in these passages, Dewey was not suggest-ing that Wilson had created a false belief. On the contrary, right along Deweyhad been insisting that “this is not merely a war of armies, this is a war of peo-ples.” Accordingly, “there is no aspect of our lives to which this war does notcome home or which it does not touch.” In his judgment, “we ought not to beneutral when the war comes home in one form or another and to talk of be-ing neutral is to talk foolishness” (MW, X: 158). “There is,” he insisted, “sucha thing”“ as interests being vitally affected without a vital interest being af-fected” (258).

Unlike most Americans, Dewey had convinced himself early on that theUnited States had to be in the war. He was also confident that the Allieswould win the war. But he was not thoroughly convinced that the aims ofAmerican entry would be realized. The United States could not enter the war“with full heart and soul though we join with unreserved energy.” “Not un-til the almost impossible happens,” he continued, “not until the Allies arefighting on our terms for our democracy and civilization, will that happen”(259; my emphasis).

That Dewey should himself have believed, with Wilson, that this war wasa great opportunity to further “our democracy and civilization” is stunning,given what we now know. The texts quoted above are from Dewey’s “In Timeof National Hesitation,” written with relief that “at last we were in it our-selves.” Dewey concluded this remarkable text by offering that “the war hasshown that we are no longer a colony of any European nation nor of them allcollectively.” “We are,” he continued, “a new body and a new spirit in theworld” (259). This was surely true. But as he later came to see, he had mis-read that “new spirit.” Dewey did not then know that “our democracy” and“our civilization” were not what he had thought them to be. He did not thenappreciate that this Great War would prove that the chauvinists, pacifists, in-ternationalists, and cynics were correct. The “new body” would be a globallypowerful America, which would occupy its “rightful” place in what Deweywas to call “the war system.”

In this chapter, the focus is on the role and thought of two of America’smost important political philosophers and analysts, John Dewey and WalterLippmann. The idea is to use them to recover a critical moment and argumentin American history over democracy in the epoch of modern war. The perti-

144 Chapter Seven

Page 174: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

nence of this moment to the present cannot, perhaps, be overstated. Deweyand Lippmann played important roles as regards American entry into the warand, in different ways, as regards the terms of the peace. Although little hasbeen made of this, for Dewey the war was a transforming experience. Itforced him to articulate a philosophy of democracy, which was profoundlyradical. Lippmann was not transformed; but the war forced him to becomeclear on some critical themes, which he had previously left unclear. In theprocess, the two of them came to share a diagnosis of the problems of Amer-ican society, even if their prescriptions were leagues apart. Their positions onscience, government, mass society, democracy, and war say much aboutAmerica, before and after the Great War.


The first Republic, of course, had been written by Plato. Reflecting on theproblem of war in his world, Plato had envisioned a radically reformed polis,capable of dealing with both war and, of even greater importance, “civilstrife” (stasis). The second republic was that of Cicero, who, while approvingof empire, nonetheless wanted Rome to return to its more virtuous republicandays. The “New Republic” referred to in this part of the book, however, isneither Athens nor Rome, but, with appropriate equivocation, the magazinewhich was created in 1914 by Herbert Croly and funded by Willard Straightand his wife, Dorothy. Straight was a Morgan banker and an “arch-exponentof American imperialism”; Dorothy Straight was a Whitney with the benefitsof ample Standard Oil royalties. The young Walter Lippmann—he was butfour years out of Harvard College, the first president and founder of its So-cialist Club—was a key member of the magazine’s original staff. The mucholder John Dewey, born in 1859, was an enthusiastic contributor. There werea host of other notables who were close to the new magazine, including as aneditor, Walter Weyl, author of The New Democracy (1912) and Felix Frank-furter, who was then teaching at Harvard Law School. Among the first con-tributors were Van Wyck Brooks, the youthful author of America’s Coming ofAge (1915). Indeed, the list of early contributors reads like a who’s who ofEnglish-speaking intellectuals. More interesting, perhaps, Theodore Roo-sevelt looked on the magazine—with the encouragement of its editors—“ashis own personal stepping-stone back to the White House.”2

On the face of it, this array of personalities seems like a disparate group;but they shared in thinking that the United States was to be a new republic.Croly had been a Harvard philosophy student and had become instantly fa-mous with his The Promise of American Life (1909), the perfect title for a new

American Democracy: A New Spirit in the World 145

Page 175: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

vision of American liberalism, a “nationalist liberalism” which would give “ademocratic meaning and purpose to the Hamiltonian tradition and method”(Forcey, 1961: 129). Croly had savaged the ]effersonians as individualists,isolationists, and defenders of a bankrupt laissez-faire economic philosophy.He had argued that Big Business was here to stay, but that with Big Govern-ment and Big Labor, there were totally new possibilities. He had insisted thatthis would call for strong leadership—the answer to the “devil-take-the-hind-most” individualism of the Jeffersonians (38). As Forcey notes, although“clouded by a certain horror of the former Rough Rider’s lusty militancy,”Croly had a “deep and abiding admiration of Theodore Roosevelt” (40). Butthen there are always trade-offs in politics.

Lippmann’s first book was his 1913 A Preface to Politics. It was an icono-clastic book, influenced by Croly and by his old teachers at Harvard, WilliamJames and especially Graham Wallas, the famous British Fabian and politicaltheorist. But the then fashionable Freud, along with Nietzsche and Sorel, wereeven more in evidence. In the background, usually unnoticed, is WoodrowWilson’s new theory of democracy.

Lippmann distinguished between “routineers” and “inventors” and arguedthat government, dominated by “routineers,” had failed. “The trusts had ap-peared, labor was restless, vice seemed to be corrupting the vitality of the na-tion. Statesmen had to do something” (Lippmann, 1913: 35). Their trainingwas legal and therefore “utterly inadequate.” But it was all they had. As “rou-tineers,” they panicked and “reverted to ancient superstition. They forbadethe existence of evil by law.” Lippmann insisted that what was needed was anentirely different approach. It was necessary to put this restless, untamed en-ergy to work. The impulses were “like dynamite, capable of all sorts of uses.”“Instead of tabooing our impulses, we must redirect them” (49–50). Accord-ingly, the United States needed a “real government that has power and servesa want, and not a frame imposed upon men from on top” (45).

But the United States was no Greek polis:

Plato and Aristotle thought in terms of ten thousand homogeneous villagers; wehave to think in terms of a hundred million people of all races and all traditions,crossbred and inbred, subject to climates they have never lived in before,plumped down on a continent in the midst of a strange civilization. . . . Nor canwe keep the problems within our borders. Whether we wish it or not we are in-volved in the world’s problems, and all the winds of heaven blow through ourland (105).

In the face of this, “improvements in knowledge seem meager indeed.” Whatis demanded is a different conception of government and different kind ofstatesmen.

146 Chapter Seven

Page 176: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Princeton political scientist Woodrow Wilson had been concerned since the1890s with “leaderless govemment.”3 He was correct in arguing that theAmerican Founding Fathers, “with genuine scientific enthusiasm,” had fol-lowed Montesquieu in yearning for “equipoise and balance in a machinelikesystem” (Constitutional Government: 56). But on his view, if what had beenwrought by them had been good enough for their sons, it was not goodenough for the sons of their sons or, presumably, for their daughters either.Sharply critical of a disjointed, hence incapacitated government, Wilson re-sponded to the same currents that had moved Max Weber; but he seemed toendorse what Weber had feared. Modern societies were mass societies, com-plex and inchoate. Wilson offered that “policy—where there is no absoluteand arbitrary ruler to do the choosing for the whole people—means massedopinion, and the forming of massed opinion is the whole art and mastery ofpolitics” (“Leaderless Government”: 339).

This was truly a remarkable idea, pregnant with implications that Lipp-mann was shortly to pursue—with a vengeance. Wilson argued that since thepresident was the only official with a national mandate, he had this specialrole. The air of German philosophy still present in America surely influencedhis next move. In a text that Hegel could have endorsed, he argued that lead-ership is “interpretation:” “The nation lay as it were unconscious of its unityand purpose, and [the leader] called it to full consciousness. It could neveragain be anything less than what he had said it was. It is at such moments andin the mouths of such interpreters that nations spring from age to age in theirdevelopment” (Constitutional Government: 21).4

In A Preface to Politics, Lippmann fully shared in rejecting “the machineconception of government” (Lippmann, 1913: 13). He insisted that “the ob-ject of democracy is not to imitate the rhythm of stars but to harness politicalpower to the nation’s need” (21). “Our choice,” he maintained, “lies betweena blind push and a deliberate leadership, between thwarting movements untilthey master us, and domesticating them until they are answered” (286). Butif Wilson had grasped what was needed, Lippmann was unsure about whetherWilson could fill the bill. “Woodrow Wilson has a talent which is [WilliamJennings] Bryan’s chief defect—the scientific habit of holding facts in solu-tion” (102–3). On the other hand, “Wilson understands easily, but he does notincarnate: he has never been part of the protest he speaks. You think of himas a good counselor, as an excellent presiding officer.” “Roosevelt hasseemed to me the most effective, the most nearly complete. . . . He is a fore-taste of a more advanced statesmanship” (103). Indeed, “Roosevelt in histerm did much to center government truly. For a time natural leadership andnominal position coincided, and the administration became in a measure areal sovereignty” (23).

American Democracy: A New Spirit in the World 147

Page 177: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

A Preface to Politics is surely criticism of “tradition”—the “record and machine-like imitation of the habits; that our ancestors created”—and of thetraditional view of politics in a republic. Lippmann criticized the “mysticaldemocrats” who believe that “an election expresses the will of the people, andthat that will is wise” (115). Like Wilson, he was struggling to rearticulate democracy for a complex mass society. But he stepped back from the Nietzschean implications of the new language of masses and urged instead a“break-up of herd politics.” What was needed, on his view, was a more robustpluralism. Accordingly, he saw the reformation of party politics somewhatdifferently from either Wilson or Max Weber. He condemned “the rigidity ofthe two-party system.” For him, it “ignores issues without settling them, dullsand wastes the energies of active groups, and chokes off the protests whichshould find a civilized expression in public life” (262). And he appealed tojust those mechanisms, which Weber had rejected, saying that “the initiativeand referendum will help” (263).

Like Wilson, Lippmann wanted leadership and saw the leader as an “inter-preter;” although for him the relation of leader to mass was more dialecticalthan it had been for Wilson. “Social movements” had tendencies and energy,but they needed an “inventor” if they were to be “imbued with life” (Lipp-mann, 1913: 63). At the same time, “to govern a democracy you have to ed-ucate it: . . . contact with great masses of men reciprocates by educating theleader.” He was optimistic, indeed enthusiastic. “In a rough way and withmany exceptions, democracy compels law to approximate human need”(116). Given all that he himself said, it is not clear what could possibly be themechanism for this.

Lippmann’s second book, Drift and Mastery, published in 1914, representsa decided shift. It carries forward some of the earlier themes, but in manyways it is a more democratic book; and, with its enormous emphasis on theapplication of science to politics, it is much closer to the vision which we nowassociate with Dewey. Indeed, one is tempted to say that if James had everwritten a political book, it would have been Drift and Mastery! Both the titleand the main argument are Jamesian: “A nation of uncritical drifters canchange only the form of tyranny, for like the Christian’s sword, democracy isa weapon in the hands of those who have the courage and skill to wield it; inall others it is a piece of junk” (Lippmann, 1961: 16). The book “begins withthe obvious drift of our time and gropes for the conditions of mastery” (19).

What is this “obvious drift”? “We have lost authority. We are ‘emanci-pated’ from an ordered world” (111).

We are all of us immigrants in the industrial world, and we have no authority tolean upon. We are an uprooted people, newly arrived, and nouveau riche. As a

148 Chapter Seven

Page 178: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

nation we have all the vulgarity that goes with that, all the scattering of soul. Themodern man is not yet settled in this world. It is strange to him, terrifying, al-luring, and incomprehensibly big. . . . We make love to ragtime and die for it.We are blown hither and thither like litter before the wind. Our days are lumpsof undigested experience (118).

Worth emphasizing is the fact that for Lippmann, America is “the modernworld,” the land of immigrants and the land where “all of us are immigrantsspiritually.” Surely industrialization and urbanization are part of this; but ofthemselves, they do not make a people “modern.” Americans, we are askedto believe, are modern because they are the first people to acknowledge thefailure of all absolutes. The theme runs throughout the book: “Our ancestorsthought they knew their way from birth through all eternity: we are puzzledabout the day after tomorrow.” “The guardianship of the master and the com-fort of the priest” have evaporated. “The iconoclasts didn’t free us. Theythrew us into the water, and now we have to swim” (112).

But this remarkable diagnosis, along with its “existential” tone, is Jame-sian, in that it allows for the most characteristic of Jamesian themes: the pur-posive effort to shape the environment, to make relations, to create and re-create an unfinished world. “Mastery,” an ill-chosen word, is possible; but we must be clear about what it means:

When we cultivate reflection by watching ourselves and the world outside us,the thing we call science begins. We draw the hidden into the light of con-sciousness, record it, compare phases of it, note its history, experiment, reflecton error, and we find that our conscious life is no longer a trivial irridescence,but a progressively powerful way of domesticating the brute.

This is what mastery means: the substitution of conscious intention for un-conscious striving. . . . You cannot throw yourself blindly against unknown factsand trust to luck that the result will be satisfactory (148).

This is a distinctly Jamesian view in which there is nothing “inhuman aboutthe scientific attitude” (158). By now the idea has all but been lost, a victimof the distance of esoteric language, unintelligble to all but specialists, the im-age of anonymous men in white coats “experimenting,” the modem magic oftechnology and the Bomb. For James, as for Lippmann, there was nothingabout science properly understood which “need make it inevitably hostile tothe variety of life” (161), nothing putting the scientific attitude at odds withimpulse, intuition, imagination, creativity, or indeed, religious belief:

There have been hasty people who announced boldly that any interest in the im-morality of the soul was “unscientific.” William James, in fact, was accused oftreason because he listened to mystics and indulged in physical research. Wasn’t

American Democracy: A New Spirit in the World 149

Page 179: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

he opening gates to superstition and obscurantism? It was an ignorant attack.For the attitude of William James toward “ghosts” was the very opposite ofblind belief. He listened to evidence. No apostle of authority can find the leastcomfort in that (161).

This was no scientism or positivism. Science was controlled inquiry; but, aswith Dewey, it was “creative intelligence,” purposive, and in the service ofhuman impulses. Neither was it a technologism.5

Accordingly, it did not call for a technocracy. “The method of a self-governing people is to meet every issue with an affirmative proposal whichdraws its strength from some latent promise” (174). For Lippmann, “mastery,whether we like it or not, is an immense collaboration, in which all the prom-ises of today will have their vote” (75). Indeed, “there is nothing accidental . . . in the fact that democracy in politics is the twin-brother of scientific think-ing. They had to come together. As absolutism falls, science arises. It is self-government. . . . The scientific spirit is the discipline of democracy, the es-cape from drift, the outlook of the free man” (151). To be sure, Lippmann isusefully unclear here as to exactly how the leadership which was so impor-tant in Preface functions in this scheme of things; and as before, he is unclearabout the mechanisms which might join the scientific spirit with the machin-ery of a democracy. Moreover, as he became clear, after the war, he alsochanged his mind.

This vagueness, as well as certain other strands in the book, make it easy tosee a continuity which has thrown off more than one commentator. It also makesit easy to see why the book could receive adulation from almost all sides. Lipp-mann had not given up on Croly’s notion that organized labor could be a “coun-tervailing power,” to adopt Galbraith’s term. To this he added the idea that con-sumers would be a power to be reckoned with; but, more than that, inanticipation of Adolf Berle, he argued that “the real news about business . . . isthat it is being administered by men who are not profiteers” (42). The “estab-lished” magazines and newspapers, ready to accept Lippmann’s deflation of so-cialism and his celebration of America’s uniqueness could easily be enthusiastic.

So too could the women of America. In a chapter devoted exclusively tothe topic of “the Women’s Movement,” Lippmann saw that there had to beconfusion and conflict within the movement, precisely because “every step inthe woman’s movement is creative” (123). Randolph Bourne, who wasshortly to lay blame for America’s entry into the war on “the war intellectu-als,” said that he would have given his soul to have written Drift and Mastery.And with much less good reason, even the revolutionary magazine, TheMasses, concerned perhaps with the good relations between Reed and Lipp-mann, received it with warmth.6

150 Chapter Seven

Page 180: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Drift and Mastery was a “pragmatist’s” political book and important; but itwas anything but Lippmann’s last word on the topic under discussion. Weneed first to see what, during this same period, the active, influential prag-matist, John Dewey, was up to. For some time prior to the publication of Driftand Mastery, Dewey had been insisting on the application of “experimentalmethods” to social change, arguing that in crucial ways the problem of a bet-ter society was a problem of knowledge. Yet the most striking thing aboutDewey’s thought up to World War I was the absence of a political philosophy.Until that time, as an intellectual and a theorist, Dewey was a psychologist, amoral philosopher, an educationist, a defender of his new “instrumentalist”version of pragmatism, a “philosopher” engaged in philosophical problemswhich, if they would touch “the problems of men,” had not yet issued a clearvision of the good society.7

One here suspects a kind of characteristic American innocence. Dewey,reared in the town-meeting atmosphere of Vermont, had no reason to doubtthat if there were problems in America, “the American way,” erected on solidfoundations, was essentially sound. Of course, he had always been more thana theorist. He had always been involved practically, in Chicago with the workof Jane Addams and the experimental lab school at the University of Chicago,and in New York, especially with the schools.

The Great War prompted Dewey’s first systematic political work, GermanPhilosophy and Politics (1915). But if the book was motivated by the war, itwas guided by Dewey’s conviction that there is a “mutual relationship of phi-losophy and practical social affairs” (Dewey, 1915: 13). As the title suggests,the book is an effort to grasp the German politics of “World Policy” by astudy of German philosophy, from the esoteric inquiries of Kant’s Critique ofPure Reason to the philosophy of history, the state, and of war in the philos-ophy of Hegel.8 Both German philosophy and politics come off badly. LikeMarx, Dewey had been nurtured on Hegel, but had long since broken withthat tradition. So far as politics is concerned, German Philosophy and Politicsis Dewey’s German Ideology.

At the root of German politics, Dewey finds the “two worlds” of Kantianphilosophy and its subsequent “correction” by Hegel. Thus:

The division established between the outer realm, in which of course acts fall,and the inner realm of consciousness explains what is otherwise so paradox-ical to a foreigner in German writings: The constant assertion that Germanybrought to the world the conscious recognition of freedom coupled with theassertion of the relative incompetency of the German folk en masse for po-litical self-direction. To one saturated with the English tradition which iden-tifies freedom with power to act upon one’s ideas, to make one’s purposes

American Democracy: A New Spirit in the World 151

Page 181: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

effective in regulation of public affairs, the combination seems self-contradictory. To the German it is natural. (34)

General Friedrich von Bernhardi’s frank and immodest Germany and theNext War of 1912 is extensively quoted. Drawing on the Reformation andKant, Bernhardi had concluded: “To no nation except the German has it beengiven to enjoy in its inner self ‘that which is given to mankind as a whole.’. . . It is this quality which especially fits us for leadership in the intellectual

domain and imposes upon us the obligation to maintain that position” (quotedby Dewey, 1915: 35, emphasis in the original). This was no metaphor, ofcourse. In this book, Bernhardi had called for “the elimination of France” (dieAusschaltung Frankreichs), the foundation of a Central European federationunder German control, and the acquisition of new colonies.

In a masterful understatement, Dewey comments: “Outside of Germany,cavalry generals who employ philosophy to bring home practical lessons are,I think, rare. Outside of Germany, it would be hard to find an audience wherean appeal for military preparedness would be reinforced by allusions to theCritique of Pure Reason” (35). Dewey does not stop at bashing German phi-losophy, however. He draws more general conclusions. There is a real differ-ence between “a theory which is pinned down to belief in an Absolute beyondhistory and behind experience, and one that is frankly experimental. For anyphilosophy that is not consistently experimental will always traffic in ab-solutes no matter in how disguised a form. In German political philosophy,the traffic is without mask” (89). America, unsurprisingly, is said to be ex-perimental: “America is too new to afford a foundation for an a priori philos-ophy. . . . For our history is too obviously future” (129). On the other hand,“our country is too big and too unformed . . . to enable us to trust to an em-pirical philosophy of muddling along. . . . We must have system, constructivemethod. . . . I cannot help but think that the present European situation forceshome the need for constructive planning” (129–30).9

Indeed, there is a pressing need “to clarify and guide our future endeavor”;but to do this, we need to “articulate and consolidate the ideas to which oursocial practice commits us” (130). Current American social practices weresound. They needed to be discerned so as to provide leverage on the future.He allows himself one “illustration:” “The present situation presents the spec-tacle of the breakdown of the whole philosophy of Nationalism, political,racial and cultural” (130). The philosophy of “isolated national sovereignty”will no longer suffice. But just for that reason, neither will those remedieswhich were then in the air.” Arbitration treaties, international judicial coun-cils, schemes of international disarmament, peace funds and peace move-ments are all well in their way. But the situation calls for more radical think-

152 Chapter Seven

Page 182: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ing” (130–31). There is an unacknowledged “depth and width of human in-tercourse,” and this needs to be applied “without and within our national life.”As to the remedies just mentioned, “an international judicial tribunal willbreak in the end upon the principle of national sovereignty” (131).

For Dewey, political, racial, and cultural nationalism will take on an in-creasing prominence and urgency during the next three years. We find thetheme in Democracy and Education, published in 1916. Because it is rich inthe philosophy and practice of education, readers usually fail to notice the un-derlying tension created by Dewey’s insight into the problem of nationalistand statist politics. He asks, “Is it possible for an educational system to beconducted by a national state and yet the full social ends of the educativeprocess not be restricted, constrained, and corrupted?” (Dewey, 1966: 97).The problem is plain enough. As Plato knew, no polity can escape from thedemands of creating citizens, and no educational system can escape from thefact that to be a citizen is to value and honor the distinct features of the polity.In Gernan Philosophy and Politics, Dewey had both applauded and con-demned the German system of education: “Germany is the modern statewhich provides the greatest facilities for general ideas to take effect throughsocial inculcation. Its system of education is adapted to that end” (14–15).Surely, American schools had to make Americans. But what was an Ameri-can? The solution presented itself. If we are talking about education “in andfor a democratic society,” then it is possible for an educational system to beconducted by a national state without a corruption or restriction of “the fullsocial ends of the educative process.” While the “if” was a big one, Deweydid not yet have any doubts that education in America was in and for a dem-ocratic society, and that in this regard America was leading the way.

Democracy and Education introduced a critical Deweyan distinction, thatbetween democracy as “a mode of associated living” and democracy as “aform of government.” He had little to say about the latter except to notice thatthe two ideas went hand in hand. Education in a political democracy has asits aim “sustaining and extending” democracy as a mode of associated living,of “conjoint communicated experience.” As a way of life, democracy was“the extension in space of the number of individuals who participate in an in-terest so that each has to refer his own action to that of others, and to considerthe action of others to give point and direction to his own” (87).

There was an unnoticed difficulty. Given the increasing “depth and widthof human intercourse,” it would seem that democracy as a way of livingwould present an increasingly difficult—perhaps even intractable—problem.This surely was the conclusion Lippmann had already come to. It would, in-deed, be the basis of Lippmann’s incisive analysis of the postwar Americanpolity. But in 1916, at least, Dewey seems not to have been in the least

American Democracy: A New Spirit in the World 153

Page 183: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

disturbed. He noted that there was definitely a “widening of the area of sharedconcerns.” He concluded, optimistically, that these would break down “thebarriers of class, race and national territory.” Moreover, such widening wasnot “the product of deliberation and conscious effort.” On the contrary, it wasthe result of “the development of modes of manufacture and commerce,travel, migration and intercommunication which flowed from the commandof science over natural energy” (87). Like Lippmann, he saw that “all thewinds of heaven blow through our land”; or, to adopt Graham Wallas’s influ-ential term, that there was now a “Great Society” which was international andinterdependent. For Dewey, as for Marx and Engels in the Communist Mani-festo, if for different reasons, the implications of this were entirely hopeful.

We can, no doubt, instantly agree with Dewey that these forces were giv-ing rise to problems whose solutions were increasingly intractable to local in-strumentalities; but it was far from clear that these forces were in fact gener-ating the appropriate instrumentalities for their own solution. As he came toappreciate, the problem of democracy was to provide ways whereby prob-lems could be recognized as shared concerns and to provide the means andthe instruments for dealing with them as shared. In 1916 he seems still to havebeen a victim of an element of that same German philosophy which had vic-timized Marx. Just as Marx had supposed that capitalist modes of productionwould destroy national boundaries, make for international proletarian soli-darity, and politicize workers, so Dewey seems to have thought that “the ma-chine age,” once hooked to genuine American experimentalism, would pro-pel democracy as a way of life.10

The United States, forced to invent, had invented well. The idea that theOld World was just that—old—had always been a feature of Americanthought, to be sure. But the new psychology, new experimental philosophy,new nationalism, new democracy, new freedom, and new internationalism of-fered a new promise. When America entered the war, Wilson and the “war in-tellectuals” were ready to commit Americans to a fight for “our democracyand our civilization.”11


The editors of The New Republic were democratic nationalists, not demo-cratic socialists. But, as Forcey comments, “to be a nationalist amidst the car-nage that followed Serajevo [sic], . . . was no longer so easy as it had been inthe innocent days that gave birth to the new liberalism” (Forcey, 1961: 221).This was especially difficult, since it was hard for anyone to imagine whyAmerica should have entered the war.

154 Chapter Seven

Page 184: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

We cannot here review the difficulties on this score faced by the New Re-public nationalists, except to notice that, to begin with, they probably did bet-ter than most of the violently partisan press (Forcey, 1961: 231). Theyridiculed the pacifists, to be sure, and on this they got ample assistance fromDewey. In the January 1916 New Republic, Dewey offered the classic prag-matist response to pacifism: “Until pacifism puts its faith in constructive, in-ventive intelligence instead of its appeal to emotions and in exhortation, thedisparate unorganized forces of the world will continue to develop outbreaksof violence” (MW, X: 214). In any case—and more fundamentally—the is-sue of whether force was justified depended on the consequences. If war can-not be shown “to be the most economical method of securing the resultswhich are desirable with a minimum of undesirable results, it marks wasteand loss” (MW: 214–15). But plainly, it might be so shown.12

The editors also resisted Roosevelt’s shrill calls for “preparedness,” asking,reasonably, “Preparedness for What?” In this they also got considerable sup-port from Dewey, who wrote a series of essays against the idea that compul-sory military service would overcome the admitted “defects” in our educa-tional system. Indeed,

when Mr. Roosevelt writes with as much vehemence about national aid to vo-cational education, national aid to wipe out illiteracy, and national aid forevening and continuing schools for our immigrants, as he now writes in behalfof military service, I for one shall take him more seriously as an authority on theeducational advantages of setting-up exercises, firing guns and living in camp(MW: 186).

But most striking, perhaps, are the Orwellian terms which the men of the NewRepublic helped to create and promote as they moved closer and closer tomilitancy. They called for “differential neutrality.” Their “constructive radi-calism” became “constructive patriotism.” With the sinking of the Lusitania,they called for a “new kind of war,” “armed neutrality,” forgetting that theyhad only recently been telling Americans who sailed on British ships that theydid so “at their own risk.”13 It was now perfectly justifiable for the UnitedStates to use defensive convoys, confiscate German assets, and intern its ship-ping. Wilson’s actions showed that he appreciated that the option was not todo nothing. Similarly, the New Republic men had been ecstatic when Wilsoncalled for “Peace without Victory,” believing, with good cause, that Wilsonhad got the idea from them (Forcey, 1961: 365–68). The only question waswhether Wilson really understood “Aggressive Pacifism.” It turned out thathe did.

Historians remain in disagreement over the explanation of America’s entryinto the Great War: whether, as American schoolbooks have it, the United

American Democracy: A New Spirit in the World 155

Page 185: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

States (by which they must(?) mean the President and the Congress) was forced into war,14 whether there were “deep causes” having to do with securing imperial interests, or whether, as Randolph Bourne had sug-gested, it was the result of failure of some distinct aspects of the AmericanWeltanschauung.

However, there are some facts upon which everyone can agree. For one, asMorison writes, “No citizen of a neutral state lost his life as a result of theBritish blockage, and all neutral cargoes seized were paid at war prices.” Onthe other hand, “U-boat warfare took a toll of some 200 [actually 118] Amer-ican lives on the high seas while America was still neutral” (Morison, 1965:851). And it is surely the case that the infamous Zimmerman telegram15 wascritical as regards Wilson’s final decision—whether as the last straw or theperfect excuse. Still, it is hardly self-evident that the German proposal of aGerman-Mexican alliance was good reason to send troops 3,000 miles acrossthe Atlantic to France.

Nor could there be any doubt that some Wall Street interests were servedby war, and that many of their spokesmen, such as Teddy Roosevelt andWillard Straight, wanted war. Still, unless we assume that Wilson was a dupefor these interests, more needs to be said.16

But however this may be, although it is often supposed that in the 1916campaign Wilson was “the peace candidate,” this is very far from being thetruth. The slogan “He kept us out of war” was part of the campaign, to besure; but not only was no promise ever made that Wilson would continue tokeep the United States out of the war, but the importance of the slogan hasbeen magnified by Republicans who, retrospectively, like to believe that theywere right all along and that Wilson had played a game of duplicity (Paxson,1966, I: 347). After all, nearly everyone was for peace, motherhood, and ap-ple pie. Even those who, like the editors of the Chicago Tribune, were vigor-ous in their support of Teddy Roosevelt were saying that “preparedness” wasthe only hope for peace.


We need here to look, if only briefly, at American domestic politics. By thetime of the 1912 presidential campaign, the Republican Party, whose right torule had scarcely been questioned since the Civil War, had collapsed intoopen schism. The result had been the election of Wilson as a minority presi-dent over William Howard Taft and Teddy Roosevelt’s splinter ProgressiveParty candidacy. Wilson’s presidency had not satisfied the Progressives, and

156 Chapter Seven

Page 186: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

the war had not made things any easier for Wilson. For one thing, ethnic pol-itics had taken on entirely new dimensions.

The census of 1910 showed some 91,972,266 Americans of whom9,827,763 were Afro-Americans. Politicians could safely ignore them. Ofthe remaining 82,144,503, 32,413,723 were foreign-born or had a foreign-born parent, and 10,984,614 had either been born in Germany or Austro-Hungary or had a parent who had—that is, around one in eight. But therewere also 4,505,360 who were Irish-born or of Irish parentage. Displeasedwith Wilson’s policy of “differential neutrality,” they joined a chorus of anti-British sentiment, which grew louder after the unsuccessful Irish Rebellionat Eastertime of 1916. Wilson had never been happy with the “hyphenatedmovement,” and he probably did as much as anyone to give currency to theterm. In his widely quoted May 1914 remarks at the unveiling of a statue inmemory of a revolutionary hero, he said impatiently: “Some Americans needhyphens in their names because only part of them has come over. But whenthe whole man has come over, heart and thought and all, the hyphen dropsof its own weight out of his name” (Paxson, 1966, I: 205). In his “Leader-less Government” address to the Virginia Bar Association, an address which,with appropriate changes, he evidently repeated many times, he was dis-tressingly frank: “We have the immemorial practice of the English race it-self, to which we belong. Nowhere else has the pure strain of the nationwhich planted the colonies and made the independent government underwhich we live been kept so without taint or mixture as it has been in Virginia,and hitherto in all the South” (Public Papers: 337). Of course, Americans ofless “pure strain” could find little to attract them in Roosevelt either, an atti-tude then shared by the big-shots in the Republican Party who were notabout to forgive Roosevelt’s recent, disastrous bolt of the party. Nonetheless, Roosevelt’s hopes were decidedly boosted by the vocal—if minority—“preparedness” sentiment.

When the Republicans met for their convention, there were those whocounted on a stampede for Roosevelt. This group almost certainly wanted warand knew that Roosevelt would not disappoint them. When the movement toRoosevelt did not materialize, efforts to seal the schism led, on the fourth day,to the nomination of Charles Evans Hughes, who at least seemed electable.Roosevelt’s well-publicized dislike of him, along with his ancestry, made hima plausible friend of the Germans. Moreover, since he had said nothing to in-dicate that he leaned toward the cause of the Allies, the Irish might go alongas well. So might Catholics and others who were angry with Wilson’s con-fused interventions in the Mexican civil war and pacifists and militarists un-happy with “armed neutrality.” More than that, he might appeal to the Pro-gressives, the “hyphens,” and the nonhyphens. After all, Hughes had been

American Democracy: A New Spirit in the World 157

Page 187: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“one of the craftsmen of the movement” (Paxson, 1966, I: 342), Dewey, writingin the New Republic, seems to have got it right: “Were one to judge from thestyle of campaign undertaken by the Republican managers, one would concludethat there are no issues in the present campaign—unless the business of oustingMr. Wilson from the Presidency be called an issue” (MW, X: 252). As Deweysaw, Hughes’s “undiluted Americanism” was but “the mask” for a “contradic-tory medley.”17 Nevertheless, this perfectly clear perception of the nature ofAmerican party politics seems not to have disturbed Dewey. Or at least, it didnot disturb him enough to deter him from saying, “I find myself, along withmany others who have not been especially enthusiastic in the past about Mr. Wil-son, warming up to him more and more every day” (X: 253).

Worth noting, the Socialists did not even have a convention. For them amail primary was sufficient. Senator Bob La Follette optimistically prophe-sied, in April 1916, “the day is coming when the people, who always pay thefull price, are going to have the final say over their own destinies. . . . Theywho do the fighting will do their own deciding” (Quoted in Paxson, I: 274).Wilson’s campaign was a combination of exploiting “undiluted American-ism” and a belated, energetic shift to progressivism. Jeremy A. O’Leary’s“American Truth Society” had been saying, reasonably, that Wilson’s neu-trality was fraudulent. Knowing that the group had been cultivated by Countvon Bernstorff, the German ambassador, Wilson’s response to O’Leary wasbrief: “I would feel deeply mortified to have you or anybody like you vote forme. Since you have access to many disloyal Americans and I have not, I willask you to convey this message to them” (Quoted in Paxson, I: 350). But thatwas still not the end of the matter. The Democratic National Committee dis-covered that Hughes “had been in conference” with O’Leary’s group. Withgood effect on the campaign, they accused Hughes of being in secret alliancewith disloyal “hyphens.” On the other side, Wilson nominated the notoriouslypro-labor Louis Brandeis to the Supreme Court; and in the few closing weeksof the congressional session, a raft of Progressive measures, stemming frompresidential initiative, got passed: the Federal Farm Loan Act, the creation ofthe United States Tariff Commission, an important new labor bill restrictingchild labor, and the Adamson Act, a novel piece of legislation which had beenthe favorite of railroad labor. Still, Wilson won only barely. In the electoralcollege, the victor needed 266 out of 531 votes. With California still out,Hughes had 254 and Wilson 264. The thirteen California electoral votes madethe difference. Paxson comments: “Indeed, with a well-placed smile Hughesmight have won the needed four votes in a thousand from the opposition,nearly half of them Republican at heart, but they had been snubbed” (Paxson,I: 363). The fact that Governor Johnson of California had combined the non-Democratic vote to win in that state makes this a convincing argument.18

158 Chapter Seven

Page 188: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Nevertheless, it had been clear enough to Lippmann that Wilson would bea war president. Steel reports a summer 1916 interview granted to him byWilson, then strenuously cultivating Progressive support. After almost twohours on domestic questions and “benevolent neutrality,” the discussionturned to war. Steel writes:

Wilson knew what Lippmann wanted to hear. Neutrality, “benevolent” or other-wise, Wilson said, was becoming more difficult. “Let me show you what Imean,” he added, and dramatically pulled out a cable from the embassy in Berlinpredicting that the Germans would resume unrestricted submarine warfare afterthe American elections in November. “It’s a terrible thing to carry around withme.” The implication was clear. When the Germans sank the Sussex five monthsearlier, Wilson had said that he would break relations if they resumed unre-stricted submarine warfare. Now he had to either back down or go to war.

Lippmann hurried back to New York to meet Croly and Straight. “Now we’ll ‘have to face it,’” Lippmann told them. “What we’re electing is a warPresident,—not for the man who kept us out of war. And we’ve got to make upour minds whether we want to go through the war with Hughes or with Wilson(Steel, 1980: 106–7).

We shall never know, of course, whether Hughes, too, would have been“forced” into the war—any more than we can be sure about Wilson’s true mo-tivations and beliefs either prior to the election or up to his call for war. His-torically, these are of little consequence. Far more important is the fact thatup to the day on which Wilson delivered his stirring war message to Con-gress, 2 April 1917, few Americans could have found any reason to enter thewar in Europe. Wilson himself offered but one reason, and that one reason,ironically, had been a gift of the Russian Provisional Government just onemonth before: “We desire,” he declared, “no conquest, no dominion.” “Weare but one of the champions of the rights of mankind. . . . America is privi-leged to spend her blood and her might for the principles that gave her birth.”“The world must be made safe for democracy.”

As important, surely, is the fact that he decided that America should go towar, and that he could have decided otherwise. But this is not a statement thatWilson had “free will.” Rather, unlike the similar judgment by Thucydides inreference to the Spartans, it is to say that he could have decided otherwisewithout in any obvious way compromising the “interests” of the UnitedStates. Of course, this claim may be contested. One might argue, for exam-ple, that a victorious Germany would have been a threat to the United Statesand to its interests. It is hard to know what sense to make of this sort of de-fense, especially in light of the fact that Germany was defeated and Hitler hadeven greater aspirations than the Kaiser!

American Democracy: A New Spirit in the World 159

Page 189: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

The German High Command had blundered disastrously in believing thatAmerica was not disposed to enter the war and that therefore it could renew thesubmarine campaign with impunity.19 This gave Wilson all the justification—or excuse?—he needed. To be sure, America could have adopted a consistentlyneutral stance, could have abided by the principles of a neutral country underinternational law, marked her ships appropriately, and engaged only in shippingthat was pacific. Wilson knew that he was within his authority in protectingAmerican rights at sea, even if that meant abandoning a consistently neutralposture and, thus, encouraging war. But he preferred “not to act upon generalimplication” (quoted in Paxson, I: 399). Nonetheless, his request for an imme-diate bill arming merchant ships was blocked in committee by La Follette. OnWilson’s view, “a little group of willful men representing no opinion but theirown, have rendered the great Government of the United States helpless andcontemptible” (quoted in Paxson, I: 401). Indeed, this same “little group ofwillful men” had been doggedly demanding a national referendum on war be-fore any further step toward war be taken. We can only guess what its outcomemight have been. But there is no doubt that the Congress supported Wilson: thedeclaration passed 82 to 6 in the Senate and 373 to 50 in the House. With lessgood reason than the German Social Democrats, these men were not going tobe parties to a posture of cowardice which would make the United States look“helpless and contemptible.”

Finally, and not to be overlooked, it is by no means clear that they did theright thing in endorsing Wilson’s war policy. They could not know, of course,the consequences of the American entry; nor can we make any sort of sensi-ble judgment about the consequences of that war had they acted otherwise.But surely it was as clear then as now, that the proffered reason for war was,at best, highly dubious.


Dewey was instantly distressed by official and unofficial responses to the ex-igencies of war. In December 1917 he addressed a group at DeWitt ClintonHigh School in New York City. Three teachers had been charged with disloy-alty and suspended. Dewey saw that it was no coincidence that the three hadbeen active in promoting the new Teachers’ Union. There had been no trial,no opportunity to present evidence or weigh testimony. It was, said Dewey,an inquisition. He offered that he was pro-Ally; but to be so, it was not nec-essary “to be in favor of establishing Prussianism in New York City” (MW,X: 159). Repression justified by war was not a novelty in the liberal democ-

160 Chapter Seven

Page 190: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

racies and had not been so in America. What was new was a new capacity touse repression, now coupled with a fantastic new ability to obliterate the dis-tinction between “information” and “propaganda.” It is not clear whether atthis time Croly, Lippmann, Dewey, or indeed anyone, appreciated this or itsimplications or how far it would go. But as the repression increased, so didthe protest that issued from these men, especially from Dewey. The Espi-onage Acts (1917) and the Sedition Act (1918), aided by Justice Holmes’s fa-mous articulation of a “clear and present danger,” provided the “legal”ground for slapping a ten-year jail sentence on Eugene Debbs for an anti-warspeech and for the mass indictment of the leaders of the IWW and othersequally protected by “due process.” The editors of the Masses went on trial;an issue of Oswald Garrison’s Nation was suppressed; the Jewish Daily For-ward was threatened; and Simon Patten and Charles Beard joined the list ofacademics who lost their posts. The atmosphere of suspicion resulted in thewithdrawal of the subsidy that had kept Bourne’s The Seven Arts in print.Even the New Republic came under government surveillance until GeorgCreel, the chief of President Wilson’s new propaganda bureau, the Commit-tee of Public Information, pointed out to federal agents that the magazine wasa supporter of the Administration!20

In order to satisfy the need for “an authoritative agency” for the dis-semination of “facts about the war,” the Committee of Public Information(CPI) had been created seven days after the declaration of war. It had nocongressional authority and was primarily supported out of the President’sfund. Its efforts were monumental. Not only did Creel’s office “release thenews,” which meant that it had control over what Americans learned aboutthe war; but, as Creel said, there was “no medium of appeal that we did notemploy”: films, posters, cartoons, prepared speeches, and widely distrib-uted pamphlets. A sample of their titles gives the flavor: How the WarCame [sic] to America, The War Message, [Wilson’s speech before Con-gress] and the Facts [sic] Behind It, Why Working Men [sic] Support theWar, and the Official Bulletin, a novel experiment in “government journal-ism.” In his summary Paxson concludes that “the Wilson doctrine was thedoctrine of his CPI. It was elaborated in the war of pamphlets and was ex-plained out of the history of the United States and of the world. It was rationalized as a reasonable outgrowth of United States experience. Itwas grounded in the ideas implicit in the phrase, ‘a world safe for democ-racy.’” (Paxson, II: 48). Plenty of people saw the films, read these tracts,and passed on what they had read. Still more were influenced by the CPIeffort at disseminating “information.” The pieces of printed matter andpresentations number in the millions.

American Democracy: A New Spirit in the World 161

Page 191: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Moreover, with the Government providing the example, private organiza-tions, especially hate mills like the National Security League, could go towork. Seeing traitors everywhere, their strident, irresponsible publications at-tacked “liberals” as traitors, especially those with Teutonic names, at the sametime giving the CPI a benign, centrist, responsible look.

The February Revolution in Russia had made Wilson’s war for democracyplausible, but the Bolshevik Revolution now gave the defenders of reactionwonderful new fuel. Because it was quite impossible to get any clear pictureof what was happening in Russia; because, since at least the Haymarketbombing, American WASPs had associated socialism with “foreigners,” es-pecially Slavs and Jews; because, in turn, these were “terrorists”; because theGerman High Command and German “Social Democrats” had chimed in withthe threat of “Bolshevism,” it was easy for Americans to believe whateverthey were told about the Russian Revolution. Lippmann, who surely knewthis, could observe that “the people are shivering in their boots over Bolshe-vism, and they are more afraid of Lenin than they ever were of the Kaiser.”He concluded by noticing what may be a characteristic trait of American cul-ture in this century: “We seem to be the most frightened lot of victors that theworld ever saw” (quoted by Steel: 156). Nor have those “truths” about Rus-sian history been erased.

The Red scare had by then begun. The Palmer raids and the mass deporta-tions of “dangerous” Americans—without trial—made previous efforts at re-pression seem sweet. Lippmann had confessed to Colonel Rouse that he had“no doctrinaire belief in free speech,” but that he could not be sanguine overthe hysteria. Dewey, who had been on platforms with many of the “radicals,”including the deported anarchists Emma Goldman, and Alexander Berkman,responded, in 1920, along with Roger Baldwin, Norman Thomas, andClarence Darrow, by forming the American Civil Liberties Union (ACLU).Dewey had insisted that Goldman’s “reputation as a dangerous woman wasbuilt up by a conjunction of yellow-journalism and ill-advised police raids.”The “trial” of Sacco and Vanzetti, in 1921, propelled Dewey to the convictionthat the ACLU was anything but sufficient, a discovery that would motivatehis thought from then on.

But there were other unintended consequences of the war. One was the re-alization of at least the main features of the New Republic image of a nation-alist liberalism. Perhaps Dewey was in the best position to see this clearly. Ina remarkable essay entitled “What Are We Fighting For?” of June 1918, hespelled it out. The war had brought forward “the more conscious and exten-sive use of science for communal purposes.” It had “made it customary to uti-lize collective knowledge and skill of scientific experts of all kinds, organiz-ing them for community ends.” This was, he concluded prophetically, “the

162 Chapter Seven

Page 192: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

one phase of Prussianism . . . which is likely permanently to remain” (XI:98–99). The warfare state had laid the foundations for the new liberalism!Still, “Prussianism” was hardly a democratic image.

Further consequences of the war were “the formation of large politicalgroupings” (MW, XI: 99) and “domestic integration within each unit” (XI:101). But this too seemed to propel democracy. “Production for profit” hadbeen “subordinated to production for use.” On Dewey’s view, “the war has. . . afforded an immense object lesson as to the absence of democracy in mostimportant phases of our national life, while it has also brought into existencearrangements for facilitating democratic integrated control” (XI: 102). It didnot matter what you called this, “state socialism,” “state capitalism,” “social-ization,” or something else. The fact of “deeper import” was “the creation ofinstrumentalities for enforcing the public interest in all the agencies of mo-dem production and exchange” (MW: 102). This was the key. At this time,Dewey surely believed that the instrumentalities of representative govern-ment were being extended and that they could do the job. But he was incred-ibly vague on the possible dangers.

He did not deny that the “absorption of the means of production and dis-tribution by government” and “the replacement of the present corporate em-ploying and directive forces by a bureaucracy of officials” led to centralizedgovernment. Moreover, “so far as the consequences of war assume this form,it supplies another illustration of the main thesis of Herbert Spencer that acentralized government has been built up by war necessities and that such astate is necessarily militaristic” (XI: 104). Dewey did not even comment onthe idea that “such a state is necessarily militaristic.” He seemed satisfied topoint out merely that

in Great Britain and this country, and apparently to a considerable degree evenin centralized Germany, the measures taken for enforcing the subordination ofprivate activity to public need and service has been successful only because theyhave enlisted the voluntary cooperation of associations which have been formedon a non-political, non-governmental basis (XI: 104).

The workplace too was being “democratized”: “The wage-earner is more likelyto be interested in using his newly discovered power to increase his own shareof control in an industry than he is in transferring that control over to govern-ment officials” (MW: 105). Still, these words, published just as the workers’and soldiers’ councils in Russia had begun to solidify and just before a work-ers’ revolution had come to Germany seemed hardly true of America.

Indeed, Randolph Bourne’s appraisal of the situation was very much closerto the truth. In a series of articles published between June and October 1917,

American Democracy: A New Spirit in the World 163

Page 193: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Bourne had offered an extraordinary criticism of “the war intellectuals,” andespecially of Dewey’s New Republic articles.21 Bourne had noted that for“war intellectuals,” “democracy remains an unanalysed term, useful as a callto battle, not as an intellectual tool, turning up fresh sod for the changing fu-ture” (Bourne, 1919: 123). He asked, rhetorically, “Is it the political democ-racy of a plutocratic America that we are fighting for, or is it the social de-mocracy of the new Russia? Which do our rulers really fear more, the menaceof Imperial Germany, or the liberating influence of a socialist Russia?”(123–24).

“To those of us who have taken Dewey’s philosophy almost as our Amer-ican religion,” he noted that “it never occurred that values could be subordi-nated to technique.” He agreed that “the young men in Belgium, the officers’training corps, the young men being sucked into the councils of Washingtonand into war organization everywhere have among them a definite element,upon whom Dewey, as veteran philosopher, might well bestow a papal bless-ing” (128). Liberal and enlightened, they had “absorbed the secret of scien-tific method as applied to political organization.” “Creative intelligence” wasindeed “lined up in the service of war technique.”

“We were instrumentalities,” he admitted; “but we had our private utopiasso clearly before our minds that the means fell always into its place as con-tributory. You must have your vision and you must have your technique. Thepractical effect of Dewey’s philosophy has evidently been to develop thesense of the latter at the expense of the former” (130–31).

Bourne was a pragmatist. He began his essay “Twilight of Idols” by evok-ing James and concluded by again evoking him:

A more skeptical, malicious, desperate, ironical mood may actually be the signof more stirring life fermenting in America today. It may be a sign of hope. Thatthirst for more of the intellectual “war and laughter” that we find Nietzsche callus to may bring us satisfactions that optimism-haunted philosophies could neverbring. Malcontentedness may be the beginning of promise. That is why I evokedthe spirit of WIlliam James, with its gay passion for ideas, and its freedom ofspeculation, when I felt the slightly pedestrian gait into which the war hadbrought pragmatism. It is the creative desire more than the creative intelligencethat we shall need if we are ever to fly (138–39).

Bourne’s remarks are, perhaps, a confession of his rude awakening, not somuch to the traps and ambiguities of instrumentalism, but to the nature ofthose “nebulous ideals” which so many had presumed to be instantiated inAmerican democracy. It would take Dewey a bit longer before he would getclear on the critical issues. But, contrary to many of his later critics andepigones, get clear he eventually did.

164 Chapter Seven

Page 194: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


As the war labored to an end, Lippmann and then Dewey had the chance tobe engaged in projects, which put their pragmatic philosophies to work. Lipp-mann’s project was an enormous success; Dewey’s was not.

Colonel House, Wilson’s powerful man for all seasons, initiated Lipp-mann’s project in September 1917. So secret that it had no name, the groupwhich formed its directorate decided on “the Inquiry.”22 The name, a mem-ber noted, would be “blind to the general public, but would serve to identifyit among the initiated”(Steel, 1980: 128). The directorate included SidneyMeses, House’s brother-in-law and a philosopher who was then president ofCity College, David Hunter Miller, a law partner of House’s son-in-law, Gor-don Auchincloss, Columbia historian James T. Shotwell, and geographer Isa-iah Bowman. Eventually it came to number some 150 academic experts, in-cluding Samuel Eliot Morison; but Lippmann was its general secretary and,as it turned out, its motivating spirit. Evidently, Dewey was tempted byLippmann to head a Moscow branch, but in the end decided against the plan.It is interesting to speculate on how that experience might have affected hispolitical philosophy.

The mandate of the inquiry was broad. It was to consist “not only of a studyof the facts but of quiet negotiation, especially among the neutrals, so thatAmerica could enter the peace conference as the leader of the great coalitionof forces” (129). It was just at this time that the Bolsheviks published the se-cret treaties. Because Wilson feared, rightly, that these would adversely affectAmerican public opinion—how would Wilson maintain the fiction that thewar was not “an unholy alliance of bribes and rewards”?—he tried, but failed,to prevent their publication in America. In consequence, there was an urgencyin House’s early December invitation to Lippmann to come to his home. Wil-son had to disconnect himself from the manifest imperialism of his Alliedpartners and to set out a peace of his own. The terms had to “purge andpacify” the Allied cause and, at the same time, be so tempting to the Germanpeople that they would reject their own leadership.

This rather incredible mandate was brilliantly managed by Lippmann. OnDecember 22, 1917, Lippmann presented House with a document entitled“The War Aims and Peace Terms It Suggests.” The president got it on Christ-mas Day. On January 2, Lippmann responded to requests for clarificationwith a revised memorandum. Wilson accepted the recommendations, addingsix points of his own, and on January 8 he assembled Congress to offer themhis historic Fourteen Points. Lippmann was rightly exultant.

Dewey’s inquiry, by contrast, bore absolutely no fruit. It had been initiatedin the summer of 1918 by Albert Barnes, a self-made millionaire who had

American Democracy: A New Spirit in the World 165

Page 195: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

been a student in Dewey’s social philosophy course at Columbia. Barnes hadasked Dewey “whether he would like to come down to Philadelphia and at-tempt to put his social theory into practice,” to address a large group of Pol-ish immigrants on questions of national identity and democratic pluralism. Itturned out that Philadelphia was just the starting place for “the-ranging in-quiry with direct pertinence to the terms of the peace.”

Dewey and his graduate students quickly discovered that Poles in America—like other groups, presumably—were caught in a set of intractable dilemmas. Byvirtue of their understandable affection for their own language and traditions,they quickly became isolated, reinforcing their otherness versus the mainstream.This prevented them, as the Handlins remark, “from getting their fair share of[America’s] rewards.”23 At the same time, it made them vulnerable to exploita-tion and manipulation. As Dewey said, “We discovered much fear and intimida-tion in a certain part of the Polish population, much manipulation and exploita-tion in another part, together with much criticism of leaders who they werenominally following with much enthusiasm” (MW, XI: 260–61). To get at thiscomplex web of fear, manipulation and exploitation, and contradictory responsesto leadership, however, Dewey saw that it would be necessary to extend thestudy “to European and international relations.”

The problem was focused by a forthcoming convention in Detroit, whichaimed to unite the Poles in America behind the faction of IgnacePaderewski, the famous pianist and prospective first president of the newPolish Republic. The United States had already set in motion its plans tomake Paderewski’s group, exiled in Paris, “the official representatives ofthe Polish people.” But Dewey quickly came to believe that this factionwas not terribly interested in democracy, that Paderewski represented atiny minority in his homeland, and that the leaders of the KON—theacronym (from the Polish) of the Congress (or Committee) of National De-fense which opposed Paderewski—had a far broader democratic base. ThePoles in America were being manipulated with the full, if unintended, co-operation of American media and officialdom.

As to the European aspect, Dewey saw that the struggle went way back intohistory. It was between a party “whose chief policies were monarchical, re-actionary and clerical and a party which was radical, often revolutionary andsocialistic, anticlerical and republican” (XI: 262). This, of course, was thecharacteristic form of struggle going on in all those “nations” which, throughno fault of their populations, had achieved neither modern, industrial civi-lization nor, in consequence, republican institutions.

Neither side was “especially favorable to the cause of the Jews but therecord of the Conservative Party is much the more aggressively anti-Semitic.” Similarly, “both parties share the tendency among all Poles to

166 Chapter Seven

Page 196: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

exaggerate territorial claims based upon events of past history, some of themas old as the twelfth century.” Still, Dewey continued, “the conservative partyis the more imperialist and extreme, being ‘Pan-Polish.’ Since the Russianrevolution particulary, the radical party has moderated its claims” (XI: 262).

As to the American aspect, there existed an alliance between Polish clergy,“opposed to and admittedly afraid of Americanization,” and the conservativeEuropean faction. The radical group, on the other hand, had suffered contin-uous accusations of pro-Germanism, anarchism, Bolshevism, and antidemoc-racy, all this despite the fact that one of their leaders, General Pilsudski, wasthen in a German prison, and that “the adherents of K.O.N. in this countryhave been officially expelled from the socialist party” (XI: 293)! Indeed,based upon his personal knowledge of the leadership, Dewey asserted that hecould not “speak too strongly of the malicious campaign of insinuation, mis-representation and personal attack carried on against the leaders of theK.O.N.” (XI: 294).

The problem, or better, the set of problems, which Dewey had diagnosed,were anything but unique in domestic and international American politics.They have reoccurred steadily, from World War I to the present. (Compare,more recently, immigrant Salvadoreans, Vietnamese, and Iraqis.) The officialresponse to Dewey’s efforts is perhaps typical. Dewey tried desperately to gethis detailed seventy-five-page report into the hands of pertinent officials, andeventually to House himself. Yet, although he finally managed this, he did nothave Lippmann’s success. This is hardly surprising, for Dewey said nothingthat House wanted to hear. Indeed, House was later to write: “He[Paderewski] came as the spokesman of an ancient people whose wrongs andsorrows had stirred the sympathies of the entire world. This artist, patriot, andstatesman awakened the Congress to do justice to his native land, and soughtto help make a great dream come true” (quoted in XI: 406).

The “great dream come true” is summarized by the conservative historianPaul Johnson: “Of the beneficiaries of Versailles, Poland [the Paderewski fac-tion?] was the greediest and the most bellicose, emerging in 1921, after threeyears of fighting, twice as big as had been expected at the Peace Conference.”The Polish government had, of course, exploited Western fears of Bolshevismand interests in a cordon sanitaire around Russia. But now, with the largestminorities problem in Europe, “with a third of her population treated as vir-tual aliens,” it would not be long before “she maintained an enormous policeforce, plus a numerous but ill-equipped standing army to defend her vast fron-tiers” (Johnson, 1985: 39).

There is no doubt that Dewey’s inquiry was, for him, a profoundly educa-tive experience, perhaps, indeed, the decisive turning point as regards hishopes for a more democratic world. Lippmann’s direct experience in Europe

American Democracy: A New Spirit in the World 167

Page 197: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

as a propagandist for the Military Intelligence Bureau had already given himall the education he needed when, in Public Opinion, he returned to the topicof democracy. But before turning to this Lippmann-Dewey argument, weshould look at another Lippmann-Dewey argument, this one regarding theLeague of Nations and the campaign to outlaw war.


As Lippmann quickly saw, the Versailles Peace was a sham, whatever hemight at one time have hoped. With Allied troops on their way to intervene inRussia, the Fourteen Points was a vehicle for constructing a cordon sanitairearound Bolshevik Russia. “We’ve got no business taking part in unauthorizedcivil war in Russia. We’ve got no business either in law or morals or human-ity trying to starve European Russia in the interests of Kolchak, Denikin andthe White Finns” (quoted in Steel, 1980: 164).

Central Europe was “balkanized”; but millions of people, including Germans,were forcibly put under alien rule. The reparations imposed on Germany werecontrary to anything Lippmann had expected, and the League of Nations, al-though it was not a defense pact, incredibly excluded an unarmed Germany. “Forthe life of me I can’t see peace in this document,” he wrote (quoted in Steel,1980: 158). Lippmann got it exactly right: “Unless the bridges to moderate rad-icalism are maintained, anarchy will follow.” It was not just that American pol-icy was illegal and immoral; it was also counterproductive. Now Lippmannfound himself saying things that no one wanted to hear.

But surely he had to share the blame. Like Dewey, he had contributed toWilson’s ideological war politics, now so successful that those who were stillable to distinguish between reality and ideology were suspect—if not accusedof downright disloyalty. Nor has it become easier since then to distinguish re-ality from ideology. After a hostile Senate committee caught the administra-tion in a host of highly dubious claims and outright lies, the Treaty, with itsprovision for the League, might still have passed had Wilson, its creator, beenwilling to compromise. In any case, it is clear that Europe got a League whichit did not want, and that Wilson, who seemed to believe throughout that theLeague was as he said it was, was thoroughly discredited.

In Dewey’s first comments on the League, he was enthusiastic about theidea of “permanent international government whose powers shall be evenmore executive and international than judicial” (MW, XI: 138). In another es-say, of November 1918, he had argued that the League was not, as he under-stood it, merely to “enforce peace.” This betrayed the same logic of the “oldmilitary-political system.” What distinguishes Wilson from the “other states-

168 Chapter Seven

Page 198: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

men of the epoch,” he said, was “his prompt recognition that, given the con-ditions of modern life, no adequate defense and protection of the interests ofpeace can be found except in a policy based upon positive cooperation for in-terests which are so universal as to be mutual” (XI: 128). This was not merelya shared interest in peace, of course. Rather, it involved interests which “grewout of common everyday necessities, and which operated to meet the com-monplace needs of everyday life with respect to food, labor, securing raw ma-terials . . . and so on.” He continued, “An organization which grew out ofwants and met them would, once formed, become so indispensable thatspeedily no one could imagine: the world getting on without it” (XI: 129).The idea had been important to Dewey since at least his German Philosophyand Politics.

Dewey was surely correct about the limits to the idea that peace could beenforced. But, as with his remarks regarding shared concerns in Democracyand Education, because global interdependence had accelerated and mecha-nisms were needed to respond to the new problems generated by inter-dependence, were we to suppose that the League proposed by Wilson wouldbe such an instrument? Might we not suppose that “sovereignty” would standequally in the way of what Dewey had hoped for?

In another essay of the same month, he linked the League to “the NewDiplomacy” and argued that the question was whether the end of the war willreverse the “relative eclipse” of democracy, whether “the efforts of a nationthat entered the war to make the world safe for democracy will effect a trans-formation of sentimental valuations.” In particular, the question was whetherwe continue with “an unconscious adoption of the older morale of honor anddefense of status,” or whether the democratic movement has “the intellectualcourage to assert the moral meaning of industry, exchange and reciprocal ser-vice” (MW, XI: 132).

For sometime, Dewey was relatively silent in judging which of the tworoads had been taken. But by March 1920, he had decided. “There is no usein blinking the non-democratic foreign policy of the democratic nations ofFrance and Great Britain.” Did Dewey forget to include the United Stateshere, or was this simply taken granted? He continued, “The Versailles Con-ference was not an untoward exceptional incident. It was a revelation ofstanding realities” (MW, XII: 5). “Diplomacy is still the home of the exclu-siveness, the privacy, the unchecked love of power and prestige, and one maysay the stupidity, characteristic of every oligarchy. Democracy has nottouched it” (XII: 7). The distinction between peoples and governments wasimportant, but by 1923, he could write that “the League of Nations is not aLeague of Nations but of governments, and of the governments whose poli-cies played a part in bringing on the war and that have no wish to change their

American Democracy: A New Spirit in the World 169

Page 199: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

policies” (MW, XV: 79–80). Like Lippmann, he seems not to have remem-bered that he was once an important soldier in the army of writers whosewords had served to obscure these important facts.

His whole posture had indeed changed. One shift regarded the little understood—and now forgotten—campaign to outlaw war.24 The idea de-rived from Salmon O. Levinson, and by 1923, in addition to Dewey, it hadpicked up the support of Senator William Borah, Republican leader of the as-sault on the League and sponsor of Senate Resolution 411 (introduced Feb-ruary 14, 1923). Borah’s Resolution had three parts: a universal treaty mak-ing war “a public crime,” the creation of a “code of international law ofpeace,” and the creation of a “judicial substitute for war . . . in the form of aninternational court” (XV: xvi). At first blush, the idea seems incredibly naiveand utopian. To most of its critics, it also seemed impossible that its support-ers could at the same time be such adamant opponents of the League and theHague World Court. Yet the idea was neither naive nor utopian; nor were itssupporters inconsistent. Indeed, the main idea is startlingly reasonable.

“There is no such thing as an illegal war” except the kind of war that appears to most persons the most justifiable from the moral standpoint—internal wars of liberation” (XV: 62).25 In denying the sovereignty of an im-perial power or the authority of a regime, a group or a people must makethemselves “outlaws.” But for a sovereign state in its relation to other states,“war is the most authorized method of settling disputes between nationswhich are intense,” the “ultima ratio of states” (54).

Yet we insist that individuals in conflict face some sort of mechanism fornonviolent resolution, that they engage in some kind of negotiation adjudica-tion. The point has nothing to do with the justice or injustice of a particularwar. The point is that by not making war illegal, we utterly abandon the ideaof nonviolent resolution of conflict. Nor is Dewey saying that a law makingwar illegal, signed by all, will end war. Even given the heavy sanctions avail-able to lawmakers within states, crime does not cease. Nor is Dewey sayingthat the international mechanisms to be created should include coercive sanc-tions. “The measure is logical—not merely formally logical but substantiallylogical in its adherence to the idea that war is a crime” (XV: 94). The use ofpolice power against an individual is not at all like its use against a nation.“The latter is war, no matter what name you give it. . . . You cannot coerce anentire nation save by war. To outlaw war and in the same measure to providefor war is to guarantee the perpetuation of the war system” (XV: 94). This, ofcourse, is one of the implications of Dewey’s old objection to the League, onewhich also applies to the court at The Hague. Both lack coercive sanctions.Yet, incredibly, if one thinks about it, “they operate under an international lawwhich sanctions recourse to war” (XV: 96).

170 Chapter Seven

Page 200: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Nor is this a form of pacifism. It is not claimed that force is never to beused or that nations should disarm. On the contrary, they may well have tofight a war. It is to say, by contrast, that unless and until war is outlawed andthere are alternative mechanisms for settling disputes, we will not have takenone single stride toward lasting peace. Indeed, all “steps” taken within “thewar system” are useless; for “it is not a step we need, it is a right-about-face”(XV: 98).

The choice, then, is between “political methods based upon a system whichlegalizes war, and political methods which have as their basic principle that waris a crime, so that when diplomacy and conferences cannot reach agreement thedispute shall be submitted to a court” (XV: 119–20). But isn’t this naive?

Consider the possibilities. Suppose that we choose the second alternativeand outlaw war and create an international tribunal. Then there are three al-ternatives: first, issues are settled by the open inquiry of the court; second,one party (or both) refuses to assent to the court’s decision; or third, one party(or both) refuses to even submit the dispute to the court. In the latter case,then, assuming that there are mechanisms for publicity, it should be clear tothe people of the world, including the people of the nations involved, that theparty has no case, that the regime’s rationale for war cannot stand up to thepublic scrutiny. The refusal to submit to publicity indicts them. In the secondcase, the world, including the people of the recalcitrant regime, can judgewho, if anybody, has right on their side. Dewey argues that the proposition tooutlaw war has never been put to the people of the world. If they do not wantwar, they will respond. Similarly, if such a mechanism existed, the peoplecould decide whether they wanted the particular war, which they were beingasked to fight.

At the risk of pedantry, I have tried to make the foregoing absolutely clear.But Walter Lippmann, surely one of the most perceptive men around, neverseemed to quite grasp what Dewey and the others had in mind. This is strik-ing. Dewey’s position is open to criticism, of course; but Lippmann’s criti-cisms are utterly off the mark. He argued, for example, that Dewey’s proposalwas a plan to “enforce peace” (XV: 405), that it committed people “to a codeso radical that it destroys the patriotic code which they are accustomed to as-sociate with their security and their national destiny,” that nations couldnever. agree on the code to which they would be bound (XV: 409–10), thatany test would require an abrogation of sovereignty (XV: 411), that the ad-vocates “propose to continue to legalize all kinds of wars” (XV: 412), that theidea calls for the elimination of diplomacy and other voluntary mechanismsof adjudication (XV: 414), and more.

Dewey fielded these objections in two essays in response to Lippmann’spolemic, and in each instance it was easy to show that Lippmann had been

American Democracy: A New Spirit in the World 171

Page 201: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

mistaken, that he had distorted the text or missed the point.26 The main ob-jection to Dewey’s plan, of course, was that no government, not even in thedemocracies, was willing to submit its foreign policy aims to anybody’sscrutiny, still less the scrutiny of its own people. It seems fair to say that Lipp-mann simply took this for granted.

There was nothing wrong with Dewey’s logic, even if his position madehim a holdout for the “new diplomacy.” Perhaps he did not see that in thisworld, vested interests were, if anything, more powerful than they ever hadbeen; that, if anything, people would have less say with regard to war thanever in the past. By 1923 there was urgency in his posture. In his 1927 re-sponse to Lippmann’s mature views on democracy, he converted urgency intoradical analysis.


It is no exaggeration to say that Lippmann’s Public Opinion is one of the mostimportant books in modern democratic theory. Published in 1922, it is a mas-terful account of the epistemology, conditions, and mechanisms of mass-opinion formation in a modern mass society. It also includes a brilliant chap-ter that annihilates the individualist’s image of democracy, of “theself-centered man” and the “self-contained community.” As Dewey saw, theonly disappointing aspect of the book was Lippmann’s constructive sugges-tions. In Public Opinion Lippmann did not draw the deep implications of hisanalysis for democracy, although they were clear enough. He did this in hisPhantom Public of 1925. And this is the book that prompted Dewey to hisfull-dress response, in The Public and Its Problems of 1927.

Part I of Public Opinion sets the parameters: “The World Outside and Pic-tures in Our Heads.” Lippmann does not doubt that there is “a World outside,”and that in some sense it is knowable. People in modern mass societies have“direct acquaintance” with their milieus; but even the latter involve “the se-lection, the rearrangement, the tracing of patterns upon, the stylizing of, whatWilliam James called ‘the random irradiations and resettlements of ourideas’” (Lippmann, 1954: 16). Knowing is through “the medium of fictions.”But fictions are not lies. A fiction is “a representation of the environmentwhich is in lesser or greater degree made by man himself.” It may have “al-most any degree of fidelity,” depending on its construction. The “persistentdifficulty,” he concludes, “is to secure maps on which their own need, orsomeone else’s need, has not sketched in the coast of Bohemia” (16). Theproblem is that most people believe that they have a good map without hav-ing any way to know this.27 The materials of “public opinion” are the “pic-

172 Chapter Seven

Page 202: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

tures inside the heads . . . of themselves, of others, of their needs, purposes,and relations.” “Those pictures which are acted on by groups of people, or byan individual acting in the name of groups, are Public Opinion with capitalletters” (19).

In successive chapters, Lippmann develops the mechanisms for the forma-tion of public opinion. Part II, “Approaches to the World Outside,” has chap-ters on “censorship and privacy,” “contact and opportunity,” “time and atten-tion,” and “speed, words and clearness.” Members of modern mass societiesare not polis-dwellers directly engaged in a world where the causes and con-sequences of acting can be used to check one’s maps. Nor have they the timeor the opportunity to range across the spaces of indirect involvement. Alwayssubject to mediation by others, from the childhood books put into their handsto the representations of the Official Bulletin, they have no way to discrimi-nate among the representations set before them or to judge whether some-one’s need has not “sketched in the coast of Bohemia” on their map. More-over, it has now become possible, when necessary, to create something “thatmight almost be called one public opinion all over America” (47).

Lippmann gives a devastating account of the CPI and its unwitting con-spirators in manipulation. He argues that the Committee drew in a host ofwilling helpers, from the Boy Scouts who delivered the president’s anno-tated addresses to doorsteps to the 600,000 teachers who received the fort-nightly periodicals and passed on the “information” contained therein totheir pupils to “Mr. Hoover’s far reaching propaganda about food” to theRed Cross, the YMCA, the Salvation Army, and other groups who carriedout the campaigns. Largely voluntary, the Committee’s effort was insidi-ous, an achievement which far outran the hopes of the small group who satat its center.

In Part III, Lippmann illuminates the overwhelming role of stereotypes informing thought. “We see a bad man. We see a dewy mom, a blushingmaiden, a sainted priest, a humorless Englishman, a dangerous Red, a care-free bohemian, a lazy Hindu, a wily Oriental, a dreaming Slav, a volatileIrishman, a greedy Jew, a 100% American” (119–20). Taken as an ordered en-semble, they provide “a picture of a possible world to which we are adapted.”“No wonder,” Lippmann concludes, that “any disturbance of the stereotypesseems like an attack upon the foundation of the universe” (95). Indeed, whenthese are reproduced ad nauseam, in authoritative histories, magazines, sto-ries, cartoons, movies, radio shows, television productions and more,—theirgrip is irresistible.

Part IV is an insider’s account of the role and limits of newspapers. Theydeal with news, not truth. “The news does not tell you how the seed is ger-minating in the ground, but it may tell you when the first sprout breaks

American Democracy: A New Spirit in the World 173

Page 203: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

through the surface” (341). Moreover, where there is a good machinery ofpublic record, statistics on crime, stock prices, election returns, and the like,the modem news service is excellent. But where information is “spasmodi-cally recorded,” unclear, explanatory, contestable, or “hidden because of cen-sorship or a tradition of privacy,” the service fails.

Worse, “news and truth are not the same.” “The function of news is to sig-nalize an event, the function of truth is to light the hidden facts, to set themin relation with each other, and make a picture of reality on which men canact” (358). The Press is “like the beam of a searchlight that moves restlesslyabout. . . . Men cannot do the work of the world by this light alone.” And, crit-ically, the press cannot do otherwise. The newspaper is neither a church nora school. It is a business. Indeed, “the citizen will pay for his telephone, hisrailroad rides, his motor car, his entertainment. But he does not pay openlyfor his news” (322). Unreported scandals by dry-goods merchants who ad-vertise are not the problem. “The real problem is that the readers of a news-paper, unaccustomed to paying the costs of newsgathering, can be capitalizedonly by turning them into circulation that can be sold to manufacturers andmerchants” (324) Lippmann neatly summarized the point. He wrote: “To getadvertisers [a paper] must get readers. To get readers it must defer to theirown experiences and prejudices as setting the standard; it must adapt itself tosell newspapers” (341). Dewey thought that Lippmann had given up tooquickly. But the problem is not the immorality of editors or publishers. It isstructural, a self-reproducing, closed, causal loop. Indeed, it may be a loop,which is well nigh impossible to break!

Lippmann illustrates the mechanisms of the “making of a common will”with a case study of the building of the Wilsonian picture of the Great War.It joins all the previous themes and deserves a full airing here. We must settle, however, for only its flavor. Well before the CPI geared up and wellbefore Wilson’s dramatic congressional speech, the Republican candidateHughes unwittingly contributed. At the critical moment of the campaign of1916 he did what was expected of him. He played politics. His first speechset the tone. Lippmann summarizes it thus: “On the non-contentious record,the detail is overwhelming; on the issue everything is cloudy” (210). “Whatcannot be compromised must be obliterated, when there is a question onwhich we cannot all hope to get together, let us pretend that it does not ex-ist” (201). With regard to Wilson, the “experiment” of the Fourteen Points,“addressed to all the governments, allied, enemy, neutral, and to all the peo-ples,” would have been impossible “without cable, radio, telegraph anddaily press” (207). And there was the opportunity. By the end of 1917, “theearlier symbols of the war had become hackneyed, and had lost their powerto unify. Beneath the surface a wide schism was opening in each Allied

174 Chapter Seven

Page 204: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

country” (209). Moreover, “the whole Allied cause had been put on the de-fensive by the refusal to participate in Brest-Litovsk” (210). Wilson filledthe gap. But, of course, the Fourteen Points served precisely because “noone risked a discussion.” Indeed, on pain of exposing their roles, they couldnot. “The phrases, so pregnant with the underlying conflicts of the civilizedworld, were accepted. They stood for opposing ideas, but they evoked acommon emotion” (215).

Lippmann then turns to democracy. The never true, fanciful, democraticimage of “the self-centered individual” autonomously and directly con-fronting the world as it makes no sense. Nor does the idea that the commu-nity is “self-contained,” and that, accordingly, there is no “unseen environ-ment” which escapes everyone’s “direct and certain knowledge.” We are notpolis-dwellers. We are a mass.

Yet Lippmann does not for a second pretend that the Founding Fatherswere democrats, mystical or otherwise. On the contrary, “when they went toPhiladelphia in May 1787, ostensibly to revise the Articles of Confederation,they were really in full reaction against the fundamental premise of Eigh-teenth Century democracy” (277). They were “determined to offset as far asthey could the ideal of self-governing communities in self-contained envi-ronments. The problem, as they saw it, was to restore government as againstdemocracy” (278). To be sure, “the American people came to believe thattheir Constitution was a democratic instrument, and treated it as such.” More-over, “they owe that fiction to the victory of Thomas Jefferson, and a greatconservative fiction it has been. . . . It is a fair guess that if everyone had al-ways regarded the Constitution as did the authors of it, the Constitution wouldhave been violently overthrown, because loyalty to the Constitution and loy-alty to democracy would have seemed incompatible” (284). (See Manicas,1989, Chapters 6, 7, and 8.)

What then is the upshot? What is the solution? One might guess here thatLippmann believes that all is well, on the grounds that, mythology notwith-standing, the people do not rule anyway. Representatives rule, and surely theyhave good maps. But Lippmann thinks otherwise. Everywhere in the world,he says, representative bodies are discredited. And there is a good reason: “Acongress of representatives is essentially a group of blind men in a vast un-known world” (288). Indeed, for Lippmann, one of the preconditions of a“strong parliament” can never be satisfied: “There is no systematic, adequate,and authorized way for Congress to know what is going on in the world.” Thepresident “tells Congress what he chooses to tell it” (289).

Recurring now to his earlier views, he concludes that this is why the pres-tige of presidents has grown in modern democracies. He seems to throw abouquet to the Congress, but it ends up being more like a crumb. He writes,

American Democracy: A New Spirit in the World 175

Page 205: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

“There is no need to question the value of expressing local opinions and ex-changing them.” Accordingly, “Congress has a great value as the market-place of a continental nation” (288). But since the president, “presiding overa vast collection of bureaus and their agents, which report as well as act,”frames and directs policy, the congressional “marketplace” is effectively thecongressional talking shop!

Although it would appear that Lippmann did not know what Max Weberhad recently said on the subject, he has further limited the capacities of a rep-resentative body and, without any apparent fear of the consequences, has cel-ebrated the singular importance of leadership in the modern mass state. As anAmerican, he could still have special faith in experts. Predictably, this was the“entering wedge” which allowed him to join “knowledge” and power.

As Dewey and the pragmatists had been saying all along, the problem ofthe modern state was the problem of “organized intelligence.” Lippmann,having forgotten his William James, now gives an unabashed elitist, techno-cratic version of this. “Gradually . . . the more enlightened directing mindshave called in experts who were trained, or had trained themselves, to makeparts of this Great Society intelligible to those who manage it” (370). Thoughthese “enlightened directing minds” knew that they needed help, they were“slow to call in the social scientist” (371). Lippmann hopes that the lesson hasbeen learned. What is needed is presidential leadership responsive to the bestof “social scientific knowledge”!28

Lippmann ended Public Opinion by referring to Plato’s parable of the shipat sea. “In the first great encounter between reason and politics, the strategyof reason was to retire in anger,” “leaving the world to Machiavelli” (1954:412). Whenever one makes an appeal to reason in politics, the parable recurs.But Lippmann’s answer is not the one just given. His answer is to combinePlato and Machiavelli: “Even if you assume with Plato that the true pilotknows what is best for the ship, you have to recall that he is not so easy torecognize, and that uncertainty leaves a large part of the crew unconvinced,”(413). Worse, during a crisis at sea there is no time “to make each sailor anexpert judge of experts.”

It would be altogether academic, then, to tell the pilot that the true remedy is,for example, an education that will endow sailors with a better sense of evi-dence. . . . In the crisis, the only advice is to use a gun, or make a speech, uttera stirring slogan, offer a compromise, employ any quick means available toquell the mutiny, the sense of evidence being what it is.29

Indeed. By the time of the Phantom Public, Lippmann had groped his way toa clearly articulated, novel conception of democracy. He could now insist thatthe democratic ideal is a false ideal because it is unattainable, “bad only in the

176 Chapter Seven

Page 206: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

sense that it is bad for a fat man to try to be a ballet dancer” (39). Things aretoo complicated, too changing, too obscure, and too difficult; and ordinarypeople simply have no time to get the information they need in order to makeintelligent judgments. Even if each person is equipped for 1925, this “will notequip him to master American problems ten years later.” “That is why theusual appeal to education as the remedy for the incompetence of democracyis so barren” (26).

The individual man does not have opinions on all public affairs. He does notknow how to direct public affairs. He does not know what is happening, why itis happening, what ought to happen. I cannot imagine how he could know, andthere is not the least reason for thinking, as mystical democrats have thought,that the compounding of individual ignorances in masses of people can producea continuing directing voice in public affairs (39).

What, then, is democracy?

To support the Ins when things are going well; to support the Outs when theyseem to be going badly, this, in spite of all that has been said about tweedledumand tweedledee, is the essence of popular government. Even the most intelligentlarge public of which we have any experience must determine finally who shallwield the organized power of the state, its army and its police, by a choice be-tween Ins and Outs (126).30

But is there, then, any important difference between democracy and dictator-ship? “A community where there is no choice [between Ins and Outs] doesnot have popular government. It is subject to some form of dictatorship or itis ruled by the intrigues of the politicians in the lobbies” (126).


Tweedle-dum and Tweedle-dee did not satisfy Dewey. Eleven incredibleyears had separated Democracy and Education from The Public and Its Prob-lems, Dewey’s direct response to Lippmann and almost certainly the besttwentieth- century defense of the idea of democracy. The distinction betweendemocracy as a way of living and democracy as a form of government re-mained. However, not only had Dewey developed a critique of democracy asa form of government, but, aided and abetted by Lippmann, he had come tosee that democracy as a way of life was not being fostered by the new inter-dependencies and the new capacities of technological society. On the con-trary, democracy as a mode of associated living was being profoundly under-mined by these forces. Since the problem was deep, Dewey was driven to a

American Democracy: A New Spirit in the World 177

Page 207: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

radical solution. Indeed, as Chapter 9, argues, Dewey’s version of democracyis so strong that it bears little comparison to the very weak forms of what isnow called democracy. Dewey did not give up the simple and fundamentalidea that democracy requires that interdependent individuals must actuallyparticipate in decisions that affect them all. Indeed, here comparison is bestmade to some versions of anarchism (Chapter 8) and to the version of social-ism sketched in the writings of Marx (Chapter 9). In sum, Dewey saw that hehad erred in supposing that the institutions created in the American Foundingwere adequate to the new property relations, the new forms of commerce andindustry. These had indeed brought about the forms of democratic govern-ment—general suffrage and executives and legislators chosen by majorityvote—but these same forces have thrown huge barriers in the way of the re-alization of democratic publics. Woodrow Wilson’s “‘new age of human re-lationships’ has no political agencies worthy of it” (Dewey, 1954: 109).

Lippmann’s analysis of the mechanisms of the formation of public opinionwas surely not wrong; but in describing the public as “a phantom,” he drewthe wrong conclusions. As Dewey had it, “The democratic public is inchoateand unorganized”; it is “lost,” “eclipsed,” “confused,” and “bewildered.”There is a Great Society, but organized into a war system of states, individu-als, who are impersonally dependent, commodified, alienated, and disem-powered, are prevented from identifying themselves as members of publics:“Where extensive, enduring, intricate and serious indirect consequences ofthe conjoint activity of comparatively few persons traverse the globe,” the ab-sence of publics is a catastrophe. Surely the Great War is “a convincing re-minder of the meaning of the Great Society” (128).

The problem admitted no easy solutions; for surely it did not involve per-fecting the institutions of political democracy. “The old saying that the curefor the ills of democracy is more democracy is not apt if it means that the evilsmay be remedied by introducing more machinery of the same kind as thatwhich already exists, or by refining or perfecting that machinery” (1954:144). The problem was much deeper and concerned the disintegration of theconditions for democracy as a way of life: the incapacity of interdependentpeople even to perceive the consequences of “combined action,” still less toperceive shared goods and to act on them.

Dewey is at pains to emphasize the role of knowledge and participation inthe constitution of democratic community. Although as individuals we are in-terdependent, there is at present no way for the countless “I’s” to become“we.” Moreover, as Rousseau and then Marx had discerned, “interdepen-dence provides just the situation which makes it possible and worthwhile forthe stronger and abler to exploit others for their own ends, to keep others in astate of subjection where they can be utilized as animated tools” (115).

178 Chapter Seven

Page 208: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Community requires both communication and knowledge; but Lippmannand the technocrats failed to realize that the kind of knowledge which is “theprime condition of a democratically organized public is a kind of knowledgeand insight which does not yet exist” (166: my emphasis). For Dewey, suchknowledge is knowledge of the causes and consequences of activity; but it isknowledge which funds experience by transforming needs and wants intomutually understood ends, knowledge which can be used in the conscious di-rection of conjoint activity.

War is surely the clearest case. It takes the combined energies of many, and the sufferings of many, many more; but unless there is someone—thephilosopher-king—who knows what all the rest cannot know, they have arightful claim to the requisite information and to being parties to the decision.It is one thing to argue that the people lack the knowledge they need to makea decision, quite another to argue that they cannot have it. In rejecting thedemocratic ideal as a false ideal, Lippmann, like so many before and afterhim, takes everything as it is and offers us, dangerously and naively, an un-accountable technocracy. By contrast, Dewey refuses things as they are. Heis a democrat, and his vision is immense. Assuming that

The Great Society is to become a Great Community; a society in which the ever-expanding and intricately ramifying consequences of associated activities shallbe known in the full sense of the word. . . . The highest and most difficult kindof inquiry and subtle, delicate, vivid, and responsive art of communication musttake possession of the physical machinery of transmission and circulation andbreathe life into it. When the machine age has thus perfected its machinery itwill be a means of life and not its despotic master. Democracy will come into itsown, for democracy is the name for a life of free and enriching communion. Ithad its seer in Walt Whitman. It will have its consummation when free social in-quiry is dissolubly wedded to the art of full and moving communication (184).

This vision is Jeffersonian and cosmopolitan.If today it seems utopian in a vicious sense, it is worth remembering that

not long ago, it seemed not only possible, but imminent! On the other hand,Dewey remained hopeful that people could find one another and act for them-selves. The Soviet Union provided a fatal test.


The year after the publication of The Public and Its Problems, Dewey visitedthe Soviet Union. In a series of six articles published in the New Republic, heoffered his impressions (LW, Vol. 3). In the first, “Leningrad Gives the Clue,”

American Democracy: A New Spirit in the World 179

Page 209: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

in terms reminiscent of Tocqueville, he wrote that he was “inclined to think thatnot only the present state of Communism (that of nonexistence in any literalsense), but even its future is of less account than is the fact of this achieved rev-olution of heart and mind, this liberation of a people to consciousness of them-selves as a determining power in the shaping of their ultimate fate” (204). Thespirit of democracy remained irresistible. Here, as throughout these essays,Dewey notes right away that, given conventional beliefs about Bolshevism andBolshevik Russia, what he says may “seem absurd.”

In the second essay, “A Country in a State of Flux,” he insists that anythingsaid about Russia must be dated, since Russia was, again, rapidly undergoingchange. “From the World War, the blockade and the civil war,” the govern-ment did “practically take over the management of co-operatives,” evenwhile, as he remarks parenthetically, the legal forms of the cooperatives were“jealousy guarded.” But, he reports, “this state of affairs no longer exists: onthe contrary, the free and democratically conducted cooperative movementhas assumed a new vitality—subject, of course, to control of prices by theState” (209–10).

In the third, “A New World in the Making,” he writes of “the sense of en-ergy and vigor released by the Revolution . . . a sense of the planned con-structive endeavor which the new regime is giving this liberated energy”; andhe says, “I certainly was not prepared for what I saw; it came as a shock”(217). And in his concluding essay, “The Great Experiment and the Future,”he sees an “experiment” with two purposes:

The first and more immediate aim is to see whether human beings can have suchguarantees of security against want, illness, old age, and for health, recreation,reasonable degree of material ease and comfort that they will not have to strug-gle for purely personal acquisition and accumulation, without, in short, beingforced to undergo the strain of competitive struggle for personal profit. In its ul-terior reaches, it is an experiment to discover whether the familiar democraticideals—familiar in words, at least—of liberty, equality and brotherhood will notbe most completely realized in a social regime based on voluntary cooperation,on conjoint workers’ control and management of industry, with an accompany-ing abolition of private property as a fixed institution—a somewhat differentmatter, of course, than the abolition of private possessions as such (244).

Dewey must have known that the Western democracies had played nosmall role at a critical moment in 1918 in derailing the “experiment.” He hadgone to Russia almost indifferent to this. Now in language as strong as he canfind, he expressed his altered perception: “I came away with the feeling thatthe maintenance of barriers that prevent intercourse, knowledge and under-standing is close to a crime against humanity” (249).

180 Chapter Seven

Page 210: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

In 1933, Lippmann also commented on the future of the Soviet Union. IfDewey’s hopes for democracy were crushed by the experience of the SovietUnion, Lippmann’s understanding of this failure has perhaps not been sur-passed. For Lippmann, while the Bolsheviks made errors, the existence ofpowerful enemies was the decisive fact:

This is, I believe, a crucial point in any and every effort to understand the in-wardness of the communist regime. The circumstance which compelledLenin to depart from the Marxian idea of controlling the economy organizedby capitalists, and to adopt the idea of organizing a new economy, was thecivil and international war which broke out in July 1918 and lasted until No-vember 1920.

The proof is to be found in the fact that the two Five-Year Plans have had astheir primary objective the creation of heavy industries in the strategically in-vulnerable part of Russia, and that to finance this industrial development theRussian people have been subjected to years of forced privation. . . .31

Of course, people are “more afraid of Lenin than they ever were of theKaiser” (quoted by Steel, 1980: 156). And, of course, as he had also so pow-erfully argued, it had become easy to provide people maps, which satisfiedsomebody else’s need.


1. Unless otherwise indicated, citations from Dewey are from John Dewey: TheCollected Works, Ann Boydston (ed.) (Carbondale and Edwardsville: Southern Illi-nois University Press, 1976–1983), cited by volume and page numbers in parenthesesin the text.

2. The foregoing draws on Ronald Steel, Walter Lippmann and the American Cen-tury, Chapters 6 and 7, and Charles Forcey, The Crossroads of Liberalism: Cro/y,Weyl, Lippmann and the Progressive Era, 1900–1925 (New York: Oxford UniversityPress, 1961), Chapter 5. The New Republic was to be “radical without being socialis-tic.” How radical is arguable, of course. The magazine ran a deficit, except toward theend of the war, when it was selling more than 40,000 copies. For four years, theStraight subsidy ran to $100,000 per year.

3. This is the title of an oft-given speech and essay. A version may be found in RayStannard Baker and William E. Dodd (eds.), The Public Papers of Woodrow Wilson(New York: Harper and Row, 1925). See also Wilson’s very influential CongressionalGovernment: A Study in American Politics (Boston: Houghton Mifflin, 1885), which,by 1900, was in its fifteenth edition; and his later Constitutional Government in theUnited States (New York: Columbia University Press, 1908). These are cited in thetext respectively as “Leaderless Government,” Congressional Government, and Con-stitutional Government.

American Democracy: A New Spirit in the World 181

Page 211: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

According to Wilson, “we have in this country . . . no real leadership; because noman is allowed to direct the course of Congress, and there is no way of governing thecountry save through Congress which is supreme” (ibid. 205).

Jeffrey K. Tulis has given an excellent account of Wilson’s transformation of thepresidency. See his The Rhetorical Presidency (Princeton, N.J.: Princeton UniversityPress, 1987). The Great War gave Wilson the chance to bring to realization his ideathat the systemic difficulties of “mechanical government” could be overcome by apresident who had the capacity to form mass opinion. See below.

4. The idea of the leader as “interpreter” has a distinguished German history. Itruns from Wilhelm von Humboldt to Hegel, Ranke, Droysen, Treitschke, Dilthey, andMeinecke. The fundamental premise is well expressed by Ranke: “No state ever ex-isted without a spiritual basis and spiritual content. In power itself a spiritual essencemanifests itself. An original genius, which has a life of its own, fulfills conditionsmore or less peculiar to itself.” By means of “interpretation,” the historian discernsthis “genius” and thereby makes history intelligible; and the leader (Der Führer) who“expresses” it becomes, as for Hegel, a “World-Historical Individual.” I have dis-cussed these remarkable notions in my A History and Philosophy of the Social Sci-ence: 86–96, 117–24.

On General von Bernhardi’s use of these ideas, see below. We do not need to be re-minded here that these ideas also had a remarkable future in fascist and Nazi ideol-ogy. See, e.g., Benito Mussolini, Fascism: Doctrine and Institutions (Rome: Ardita,1935).

5. David Hollinger properly insists that James had an importantly distinct notionof science and its relation to culture, a notion thoroughly grasped by Lippmann. SeeHollinger’s “Science and Anarchy: Walter Lippmann’s Drift and Mastery,” AmericanQuarterly (1977), and idem “William James and the Culture of Inquiry,” MichiganQuarterly Review (1981), both repr. in Hollinger, In the American Province (Bloom-ington: Indiana University Press, 1985).

6. Reed and Lippmann, friends since Harvard, had been part of a group which had re-cently put on a Madison Square pageant dramatizing the situation of striking IWW silk-workers in Paterson, New Jersey. But in Drift and Mastery, Lippmann had applauded“conservative unions” and condemned the IWW as preferring “revolt to solidarity” and,in practice, being “ready to destroy a union for the sake of militancy” (p. 62).

7. The 1908 Ethics, a collaboration with James H. Tufts, contains much socialphilosophy and sensible social philosophy at that. Still, these parts were Tufts’s con-tribution. See my discussion, Chapter 10.

8. Dewey seems to have liked Marx’s little joke which he cites in a note attachedto Hegel’s famous reference to “the bird of Minerva which takes its flight only at theclose of day”: “Marx said of the historic schools of politics, law and economics thatto them, as Jehovah to Moses at Mt. Sinai, the divine showed but its posterior side”(ibid.: 110)!

9. The text is worth calling to the attention of critics of Dewey who say thatDewey’s experimentalism kept him from insisting on the need for an inclusive plan.The idea stayed with him throughout. But, like his associates at the New Republic, he

182 Chapter Seven

Page 212: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

was no socialist—at least until later. Accordingly, he waffled regarding the key fea-tures of a “constructive plan.”

10. Dewey was reluctant to refer to the capitalist epoch by its name, preferring in-stead “machine age,” “industrial order,” “new forms of commerce and industry,” andso forth. This had some severe consequences, especially after his “radical turn.” SeeChapter 9.

11. For the early period in American culture, see Sacvan Berkovitch, The Ameri-can Jeremiad (Madison, Wis.: University of Wisconsin Press, 1978).

12. See also Dewey’s April 1916 essay for The International Journal of Ethics,“Force and Coercion” (Middle Works, X: 244–51). In terms of Dewey’s altogethersensible moral posture, the pacifist case against World War I, of course, was muchstronger than it was against World War II.

13. In June 1915, the issue had brought about the resignation from Wilson’s Cab-inet of Secretary of State William Jennings Bryan. Bryan had wanted the UnitedStates to bring all disputes between it and the thirty countries with which it hadtreaties before an international commission. He wondered, moreover: “Why shouldan American citizen be permitted to involve the country in war by travelling upon abelligerent ship when he knows that the ship will pass through a danger zone?” Fi-nally, he could not understand how American passenger ships were permitted to carrycargoes of ammunition, a policy that plainly and provocatively threatened the shipsand encouraged war. Wilson would not submit the issue to impartial inquiry; norwould he disclaim responsibility for the precipitous actions of private citizens. See“Bryan’s Letter of Resignation,” in S. Cohen (ed.), Reform, War and Reaction:1912–1932 (Columbia: University of South Carolina Press, 1972): 58.

14. Forcey, Crossroads: 265–68. The conventional opinion regarding the Ameri-can entry is ably represented by Samuel Eliot Morison, The Oxford History of theAmerican People: 851–56. See also Frederic L. Paxson, American Democracy and theWorld War, Three Volumes (repr. New York: Cooper Square, 1966).

15. In January 1917 Arthur Zimmermann, the German Foreign Minister, had senta telegram to Mexican president Carranza which had been intercepted by British in-telligence. It read (in part):

We intend to begin unrestricted submarine warfare on the first of February. We shall en-deavor in spite of this to keep the United States neutral. In the event of this not succeed-ing, we make Mexico a proposal of an alliance on the following basis: make war together,make peace together, generous financial support, and an understanding on our part thatMexico is to reconquer the lost territory of Texas, New Mexico and Arizona.

Given the circumstances, the idea was hardly shocking. Not only was the UnitedStates arming ships that carried munitions to Britain; but also if the United States wereto be in a full-scale war with Germany, then Germany would obviously hope for allthe help she could get. For anyone wanting war, of course, the note was a gift fromheaven. Wilson released it on 28 February and in April received a vote in favor of war.See below.

American Democracy: A New Spirit in the World 183

Page 213: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

16. It is certainly true that American foreign policy was imperialist. It had been sofrom the beginning. But its imperialist designs had been confined to the WesternHemisphere and, more recently, to the Pacific.

17. Ronald Steel quotes a “spleen-filled passage” from Lippmann’s report of theRepublican convention. A piece of this gives the flavor:

I think that there were fifteen nominations plus the secondary orations. It was a nightmare,a witches’ dance of idiocy and adult hypocrisy. . . . The incredible sordidness of the con-vention passes all description. It was a gathering of insanitary callous men who blas-phemed patriotism, made a mockery of Republican government and filled the air with sod-den and scheming stupidity (Steele: 103).

Lippmann gives a brilliant analysis of Hughes’s speech in his Public Opinion. See below.

18. The foregoing is influenced by Walter Karp, even though it departs from his ac-count. See Walter Karp, The Politics of War: The Story of Two Wars Which Altered For-ever the Political Life of the American Republic (1890–1920) (New York: HarperColophon, 1980). Karp argues: “If the interests of the country or even the desire to winthe elections had shaped the policy of the Republican Party leaders, Wilson’s diplomacywould have provided a political target impossible to miss” (216). No doubt it is true thatfor the Republicans, “straightforward warmongering was out of the question. It wouldhave brought not war, but political disaster to the agitators” (220). “Wilson’s diplomacy,”he writes, had “opened up the prospect for war, and war was what the Republican oli-garchy wanted and needed” (216–17). But they wanted war “to undo the deep damageof the preceding ten years.” As Bourne saw, “they wanted war . . . because they saw inwar the opportunity to become the great captains of an industrial war machine and part-ners, once again, in the governance of the country” (219).

19. Commentators agree that the German High Command were completely confi-dent that the U.S. government would not confound its plans, and that even if it choseto do so, America would be unable to raise, train, and send much of a force. See Pax-son, American Democracy, I: 394, and Fritz Fischer, Germany’s War Aims in the FirstWorld War: 307. These views were very much the product of German thinking aboutthe war-making capacity of democracies!

20. Since “exclusion from the mails was near-equivalent for silencing,” the Espi-onage and Trading-with-the-Enemy Acts permitted Postmaster-General Burleson torepress by administrative fiat (Paxson, American Democracy, II: 286). Not only didsecond-class mail come under his autonomous purview, but he had the power to ex-amine private correspondence as well. Even before the war had begun, the AttorneyGeneral had developed a vast network of agents, law-enforcement officers, and vol-untary coadjutors, who persistently prosecuted complaints, sometimes malicious,sometimes hysterical, against accused “saboteurs,” and “traitors.” The Sedition Actprescribed language that was “disloyal, profane, scurrilous, or abusive.” It did nottake many successful prosecutions before Americans got the idea.

It is still very much worth reading George Creel’s enthusiastic How We AdvertisedAmerica (New York: Harper and Row, 1920).

184 Chapter Seven

Page 214: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

21. Randolph Bourne, Untimely Papers (New York: Huebsch, 1919). These beau-tiful essays, so Jamesian—and Sartrean!—in style and thrust, have been too soon for-gotten. Page references are given in the text in parentheses.

22. See Lawrence E. Gelfand, The Inquiry: American Preparations for Peace,1917–1919 (New Haven, Conn.: Yale University Press, 1963).

23. Dewey, Middle Works, vol. XI, contains the full materials on Dewey’s study ofPolish conditions, including a valuable note, presumably by Lillian and Oscar Han-dlin, who introduce the volume.

24. Middle Works, vol. XV, contains all Dewey’s essays on the campaign to outlawwar, along with Walter Lippmann’s polemical rebuttal to Levinson and Dewey. CarlCohen’s introduction to the volume is also most useful.

25. After World War II, the United Nations made both war and imperialism illegal.In terms of the UN code, wars of national liberation are the only legal wars. The up-shot, not foreseen by Dewey and Levinson, was that thereafter there would be no de-clared wars, only “police actions” sent to suppress national liberation movements!

26. See Lippmann’s “The Outlawry of War” (XV: 404–17) and Dewey’s rejoin-ders: “What Outlawry of War Is Not” (XV: 115–21) and “War and a Code of Law”(XV: 122–27). Sovereignty is not denied, since the state will decide whether to sub-mit its claims and whether to abide by the judgment of the court. No third party ex-ists to enforce decisions. No complicated code is required beyond the ordinary, vagueconventions governing international law. The plan hardly denies that diplomacy isnecessary for maintaining peace.

27. Dewey was later to make the point vividly:

Schooling in literacy is no substitute for the dispositions which were formerly provided bydirect experience of an educative quality. The void created by lack of relevant personal ex-periences combines with the confusion produced by impact of multitudes of unrelated in-cidents to create attitudes which are responsive to organized propaganda, hammering inday after day, the same few and relatively simple beliefs asseverated to be the “truths” es-sential to national welfare. (Freedom and Culture [New York: Capricorn, 1963], p. 46).

28. It is not irrelevant here that the social sciences as we now know them had justthen completed their institutionalization in the universities of America. See my A His-tory and Philosophy of the Social Sciences, Part II. Perhaps thinking here of Veblen’scritique of the social sciences, Lippmann writes that if so much social science is“apologetic rather than constructive, the explanation lies in the opportunities of socialscience, not in ‘capitalism’” (Public Opinion: 373). Lippmann, who usually sees thepertinence of the fact that practices are influenced by conditions external to them,clearly forgets this here.

29. Lippmann’s unabashed Machiavellianism is also clear in the Phantom Public(New York: Harcourt, Brace and World, 1925). There he writes:

We do know, as a matter of experience, that all the cards are not laid face up upon thetable. For however deep the personal prejudice of the statesmen in favor of truth as amethod, he is most certainly forced to treat truth as an element of policy. Insofar as he has

American Democracy: A New Spirit in the World 185

Page 215: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

power to control the publication of truth, he manipulates it to what he considers the ne-cessities of action, of bargaining, morale and prestige (158).

30. Compare Robert A. Dahl’s important Preface to Democratic Theory (Chicago:University of Chicago Press, 1956). Dahl agreed with Lippmann that there is no wayfor populations in democratic mass states to influence policy; yet he supposed that hehad shown that “elections are a crucial device for controlling leaders,” and thus that“the distinction between democracy and dictatorship still makes sense” (131–32). Butwhat, apart from what Lippmann says, can “control leaders” mean? It is true, I be-lieve, that two-party representative systems are important in preserving hard-woncivil liberties, and that these are fundamental and not to be scorned. Nevertheless, weshould not confuse civil freedom with democracy. Compare Dewey, below.

31. Walter Lippmann, The Good Society (Boston: Little, Brown, 1936: ix). Seealso Manicas, War and Democracy, Chapter 11. Indeed, in his 1936 book, Lippmannprovided convincing grounds that it was already “evident” that “the world was mov-ing toward a gigantic war.”

186 Chapter Seven

Page 216: Dewey Rescuing Dewey Essays in Pragmatic Naturalism



John Dewey’s social and political philosophy has been as much interpreted asit has been praised and condemned. Thus, to merely illustrate the spectrum ofopinion, his philosophy of democracy has been called “a Jeffersonian provin-cialism,” nostalgic and irrelevant, and a pluralist federalism fully pertinent tothe prevailing American political order. His theory of inquiry is construed asessential to his social philosophy, as independent of it, and as inconsistentwith it. Finally, his basic philosophy is understood as an independent elabo-ration of “the best elements in Marx’s thought” and, remarkably, as “the phi-losophy of American imperialism.”1

Dewey is not altogether blameless. There are many strands in his thought,sometimes conflicting strands. He is sometimes unclear, sometimes justwhere one wants a clear statement most of all. But these are not the mainproblems in coming to grips with Dewey as a social philosopher. Rather, mostof the difficulty derives from the style and range of his thought. This makesit difficult, if not impossible, to associate him in a clear way with any of the“isms” by which we tend to identify a political and social theory. And, in turn,this makes him fair game for ideological purposes. Political writing, after all,is itself a political act.

Dewey was, of course, both a liberal and a democrat, and he was not aMarxist. Yet, as I shall argue, his liberalism and democratic philosophy weredecidedly radical, more socialist than libertarian, indeed, more anarchist thancommunist or liberal. But let me not be misunderstood. Dewey was no anar-chist (however amusing it is that Sidney Hook should have written that, inlooks at least, Dewey resembled “a cross between a philosophical anarchist

Chapter Eight

John Dewey: Anarchism and the Political State

Page 217: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

and Robert Louis Stevenson”). Dewey was not, for example, exactly clear onthe future role of the state and he seems to have had little taste for anarchist“direct action”—however peaceful. I do not argue, accordingly, that he wasan anarchist without knowing it or that anarchism is itself so vague that histhought, as he worked it out, is easily subsumed. Nevertheless, we can ad-vance our understanding of Dewey and of the problems of political philosophy, if we take a fresh look at his writings from the vantage point ofanarchism.

This will be the main aim of this paper. Still, there is a historical argumentto be made, an argument that seems to me to be very important, but which canbe but hinted at here. That is, not only is it not farfetched to juxtapose Deweyand anarchism, but perhaps more fundamentally, on the present view, the FirstWorld War and the period immediately following were crucial years for rad-ical political thought. Dewey was caught up in this, and as Chapter 7 has ar-gued, it left decided marks on his thought.

We need, perhaps, to be reminded that Dewey was twenty-seven when thebomb was thrown at Haymarket Square (1886), already fifty-eight when theGreat War was coming to an end and the Bolshevik revolution erupted. Ac-tive in public issues at least from his Chicago days (from 1894–1904), Deweynonetheless wrote no political philosophy until perhaps 1908, the severalchapters that he wrote as part of Ethics, with James H. Tufts. World War Iseems to have been critical for him as for many others.

In German Philosophy and Politics (1915), his first systematic politicalwork, Dewey traced the philosophic basis of patriotic statism in Germany andconcluded that “the present situation presents the spectacle of the breakdownof the whole philosophy of Nationalism, political, racial and cultural” (GPP:130).2 Opening a theme to which he returned repeatedly, he attacked the ideaof “national sovereignty” and argued that “the situation calls for a more rad-ical thinking,” more radical than “arbitration, treaties, international judicialcouncils, schemes of international disarmament, peace funds and peacemovements” (130). Dewey was correct in calling for “more radical thinking”of the problems, but unfortunately, he was entirely wrong if he hoped that sta-tism was dead. As is well known, of course, Dewey supported the allied wareffort, as did Kropotkin and a host of other “internationalists” in the radicalparties of Europe. Indeed, patriotic statism and imperialist war had hardly runits course.

In 1919, a host of radicals, including the anarchists Emma Goldman andAlexander Berkman, became the victims of a virulent Americanism and weredeported. With Dewey, Norman Thomas, Clarence Darrow, Roger N. Bald-win, and others as founders, the American Union Against Militarism became,in 1920, the American Civil Liberties Union (ACLU). The Red Scare was by

188 Chapter Eight

Page 218: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

now in full flower. Dewey had known Goldman and Berkman, having sharedplatforms with them on several occasions. Baldwin has reported that “she hada very high regard for [Dewey’s] ideas,” a view which, as I shall argue, ishardly surprising given the coincidence of so many of their ideas. For his part,Dewey found that Emma Goldman’s “reputation as a dangerous woman wasbuilt up by a conjunction of yellow-journalism and ill-advised police raids”—a fact which Dewey would find increasingly ominous as the decade pro-ceded.3 It was during this time, as Hook has noted, that Dewey “made thegreat turn” and came to believe in the essential correctness of socialist diag-noses of America’s ills.4

The ACLU was hardly sufficient. The 1921 trial of the acknowledged an-archists, Sacco and Vanzetti, and their subsequent execution in 1927—fortheir beliefs and not for the unproved charges against them—had a pro-found effect on Dewey. As he saw it, the events had put America on trial.5

Almost certainly, they forced him to considerably temper whatever opti-mism he might have had regarding the use of intelligence in conditionswhere emotions are so easily mobilized and manipulated in the service ofreactionary politics. As I shall develop, these considerations became an in-creasingly important part of Dewey’s incisive analysis of the failures ofpresent arrangements.

In 1928, Dewey visited the Soviet Union and reported that his “own an-tecedent notions—or, if you will, prejudices, underwent their most completereversal” (C&E, I: 425). Assessing the revolution as “an experiment to dis-cover whether the familiar democratic ideals—familiar in words, at least—will not be most completely realized in a social regime based on voluntary co-operation, on conjoint workers’ control and management of industry. . . .”Dewey concluded enthusiastically that “its future is of less account than is thefact of this achieved revolution of heart and mind, this liberation of a peopleto consciousness of themselves as a determining power in the shaping of theirultimate fate” (380, 424).

Two key books were written during this period, The Public and Its Prob-lems (1927) and Individualism Old and New (1929). But with the Great De-pression and the failure of the existing major political parties to respond to thechallenge, Dewey—as a radical—faced a dilemma.

The socialist and anarchist radical traditions, reflective of their nineteenth-century European roots, had always been revolutionary in the sense that rad-ical social change was seen to involve a mass insurrection against the pre-vailing order of things. But in contrast to socialists, communists, anarchistsand most American radicals of the 1920s and 1930s, Dewey saw no evidencefor this view. As he put it, “I do not . . . hear the noises of an angry proletariat”(ION: 78). But—and this must be understood—it was not because they were

John Dewey: Anarchism and the Political State 189

Page 219: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

drowned out by “shouts of eagerness £or adventurous opportunity”—howevermuch this was the official gospel of America. Rather, for Dewey, “the murmursof discontent are drowned” by “the murmurs of lost opportunities, along with thedin of machinery, motor cars and speakeasies” (78).

These are powerful metaphors and suggest an analysis, to be developed be-low, which is far subtler than the ones offered by the mechanical formulas ofthe period’s revolutionaries. But if so, then Dewey’s question at this time was,how to be radical and still be relevant? This question has always been espe-cially difficult for Americans and it was so for Dewey. His own quandaries onthis score go some way, indeed, toward explaining his efforts—ultimately un-successful—to generate (beginning in 1931) a genuine radical, mass-based,third-party alternative. Although it may be now forgotten, Dewey was argu-ing, by the 1929 writing of Individualism Old and New, that “our presidentialelections are upon the whole determined by fear” and that neither of the ma-jor parties could be vehicles for radical change—even if this change was tobe incrementally won. On the other hand, Dewey could not align himself un-ambiguously with the Socialist Party either. Like the anarchists, the socialiststoo were isolated.6 Kropotkin’s sadly prophetic letter to Lenin, written in1920, identified a significant reason, a reason that Dewey fully appreciated.Kropotkin wrote: “If the present situation continues, the very word ‘social-ism’ will turn into a curse . . .” (Kropotkin, 1970: 337). Indeed, in one sense,the problem of the present essay, and, I believe, still a problem of our time, isto recover an idea: Dewey referred to it as “the idea of democracy,” but oth-ers have called it “socialism,” and still others have called it “anarchism.” Inthis regard, Dewey’s idea of democracy is neither a nostalgic ]effersonianismnor a liberal pluralism. I shall argue that it is anarchist insofar as it contains:

(1) a view of an ideal, noncoercive, nonauthoritarian society;(2) a criticism of existing society and its institutions, based on this antiau-

thoritarian ideal;(3) a view of human nature that justifies the hope of significant progress to-

ward the ideal; and(4) a strategy for change, involving immediate institution of noncoercive,

nonauthotitarian and decentralist alternatives.7

Dewey’s Critique of Existing Society and His Vision of the Good Society

In his most systematic work on the state, The Public and Its Problems(1927), Dewey attempts a generic and empirical approach to the question,“What Is the State?” Disavowing the utility of a series of traditional doctrines,

190 Chapter Eight

Page 220: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

he argues that associated action—a universal trait—has consequences, some-times confined to those who directly share in the transaction and sometimesnot. When those “indirectly and seriously affected . . . form a group,” we canspeak of “The Public” (PP: 35). And when this public “is organized and madeeffective by means of representatives” who “care for” its special interests,then and in so far, we may speak of the state.

This analysis, made rich by Dewey’s knack for illustration and sometimesbrilliant but not immediately relevant asides, is both important and badly mis-leading. The analysis is important because it allows us to see that “no two ages or places is there the same public” (33). Moreover, it allows us to see that State and Society are not the same, that in states, there are governments—agencies that “represent” individuals in society. Finally, it al-lows us to see that “a state” is only as good as its public and that there is no“model pattern” which makes a state a good or true state (45).

But the analysis is seriously misleading insofar as it leads us to think ofstates as universal entities. Dewey’s concern, as a long footnote makes clear,is with functions, not structures. He is thus quite ready to admit that “thestate” is a “very modern institution.” Yet, he insists, “all history, or almost all,records the exercise of analogous functions” (65–66, note 7). In a sense, ofcourse, this is true. The idea that “special agencies and measures must beformed” if “extensive and enduring consequences” are to be “attended to” has ageneral applicability—depending crucially on what is concretely meant by “spe-cial agencies and measures.” But as Dewey sees, these “special agencies” andthe public that they “represent” are open to an almost infinite range of possibil-ities. For better or for worse, “state,” “government,” even “public” is very mod-ern terms with very definite modern connotations. Moreover, while we use theterm “state” to refer indiscriminately to any sort of political body, from primitiveclan organizations to poleis, to the Roman Empire, the word “state” properly de-notes what are very modern political bodies. However great are the differencesbetween (modern) states, between, e.g., the Absolutist State which emerged inthe seventeenth century, contemporary capitalist or communist states, liberaldemocratic states and totalitarian states, all of them are states in the quite clearsense that they are legally defined entities claiming sovereignty and a monopolyof legitimate force. Each circumscribes an extended territory and a very largeand heterogeneous population. Each has a centralized organizational apparatusengaged in continuous administration and having both the “authority” and, es-pecially in this century, the ability to dramatically affect the conditions of life ofits population—for better or for worse.

Now Dewey recognized this. Not every association is a state or even hasstate-like characteristics. At one extreme are associations “which are too nar-row and restricted in size to give rise to a public” (39). “Immediate contiguity,

John Dewey: Anarchism and the Political State 191

Page 221: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

face to face relationships, has consequences which generate a community ofinterests, a sharing of values, too direct and vital to occasion a need for polit-ical organization” (39). Indeed, within a community, “the state is an imperti-nence” (41).

“Villages and neighborhoods shade imperceptibly into a political public”(43) and there may or may not be agencies which are specifically its instru-ment. Kropotkin could still refer to the Medieval commune and Dewey to theearly New England town. Further along the continuum of historical associa-tions is perhaps the polis of the ancient world, where as Dewey says, “muchof the intimacy of the vivid and prompt personal touch of the family endureswhile there has been added the transforming aspiration of a varied, freer,fuller life, whose issues are so momentous that in comparison the life of theneighborhood is parochial and that of the household dull” (44). Indeed, it washardly an accident that the idea of democracy was an invention of the polisworld (Manicas, 1989, Chapters 1 and 2).

Still of a very different sort, we can identify “empires due to conquestwhere political rule exists only in forced levies of taxes and soldiers; and inwhich, though the word state may be used, the characteristic signs of a pub-lic are notable for their absence” (43–44). Finally, as Dewey argues, butwhich needs emphasis, “for long periods of human history . . . the state ishardly more than a shadow thrown upon the family and neighborhood by re-mote personages. . . . It rules but it does not regulate. . . . The intimate and fa-miliar propinquity group is not a social unity within an exclusive whole. It is,for almost all purposes, society itself” (41–42).

These points are not peripheral to comprehending the problem of The Pub-lic and Its Problems, even if Dewey’s effort to treat the state genericallytempts us to treat them as historical asides. As Kropotkin and others in the an-archist tradition often argued, the state as we understand it, is a very modernphenomenon. And there is nothing necessary about it. Moreover, as Deweyand the anarchists saw, the development of the modern state meant also theemergence of an entirely new organization for war, the obliteration of com-munity and the suffocation of the personal and the intimate. Both were con-cerned to address the questions and to offer analyses and programs in theseterms. But this need not be a nostalgic irrelevancy—unless, of course, we un-critically accept the framework assumptions of the modern state and then pro-ceed to political inquiry.

Methodologically, Dewey was committed to a fully historical and contex-tual mode of inquiry and he recognized that the problems of contemporarypolitical arrangements were not those of the past. Nor accordingly would pastsolutions suffice. At the global level, Dewey deeply appreciated the problemof the modern state. In Reconstruction in Philosophy (1920), he extended a

192 Chapter Eight

Page 222: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

central theme of German Philosophy and Politics, and diagnosed the failureof “many schools of thought, varying even more widely in respect to methodand conclusion, [yet] agreed upon the final consummating position of thestate. . . . They do not question the unique and supreme position of the Statein the social hierarchy.” Dewey concluded: “Indeed, that conception has hard-ened into unquestionable dogma under the title of sovereignty” (RIP: 201).But perhaps he put matters most graphically in his 1927 essay, “Nationalismand Its Fruits.” He there wrote:

Patriotism, National Honor, National Interests and National Sovereignty are thefour foundation stones upon which the structure of the National State is erected.It is no wonder that the windows of such a building are closed to the light ofheaven; that its inmates are fear, jealously, suspicion, and that War issues regu-larly from its portals (C&E, II: 803).

Except for explicit anarchist thought—and even then, not all of it—no onesaw more clearly than Dewey that for the modern age, the State was not partof the solution, but was, instead, an essential part of the problem.

The structural dynamics of interstate relations were not his only concern,however, for there were effects on the relations within states. This is a funda-mental concern of The Public and Its Problems, in particular as regards themost progressive form of the modern state—the Democratic State. Although itis often overlooked—or downplayed—Dewey had no illusions about it. As inmany other places, he sharply distinguishes democracy as an idea or ideal anddemocracy as a mode of government. On Dewey’s analysis, all states have gov-ernments and all governments “represent” some public. But there are differentinstitutional arrangements by which governments exist and “represent” somepublic. For Dewey, then, political democracy is “a specified practice in select-ing officials and regulating their conduct as officials” (PP: 82). Dewey arguedthat political democracy emerged at a specific period in the development ofthe modern state and that “it emerged as a kind of net consequence of a vastmultitude of responsive adjustments to a vast number of situations . . .” (84).Indeed, in no sense did Dewey succumb to the mystifying rationalizations ofliberal democratic political theory—to the idea, e.g., that democratic institu-tions function so as to implement something called “the will of the people.”He said:

Instead of individuals who in the privacy of their consciousness make choiceswhich are carried into effect by personal volition, there are citizens who havethe blessed opportunity to vote for a ticket of men mostly unknown to them, andwhich is made up by an under-cover machine in a caucus whose operations con-stitute a kind of political predestination (120).

John Dewey: Anarchism and the Political State 193

Page 223: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Political democracy, or better, modern political democracy, is a statist formand, no doubt, it has consequential merits. But as Dewey writes, it is a means,not for realizing the idea of democracy, but “to counteract the forces that haveso largely determined the possession of rule by accidental and irrelevant fac-tors and . . . to counteract the tendency to employ political power to serve pri-vate “instead of public ends” (83). For Dewey, the “full reality” of politicaldemocracy was not that painted by patriotic publicists, nor did it meet thegoals, limited as they were, which had brought it into existence. His indict-ment was severe: “In a word, the new forms of combined action due to themodern economic regime control present policies, much as dynastic interestscontrolled those of two centuries ago. They affect thinking and desire morethan did the interests which formerly moved the state” (108).

Even more in the spirit of Marx and left anarchism, Dewey observed thatthe fusion of political and economic liberalism, the attainment of politicalrights and guarantees of private property which liberal democracy represents,had “emancipated the classes whose special interest they represented, ratherthan human beings impartially” (270). The text continues:

The notion that men are equally free to act if only the same legal arrangementsapply equally to all—irrespective of differences in education, in command ofcapital, and the control of the social environment which is furnished by the in-stitution of property—is a pure absurdity, as facts have demonstrated (271).

Dewey’s analysis of the Democratic State is radical and called for radical so-lutions. In his terms, the problem was not with the instruments of the public,but with the public itself. The public was “inchoate and unorganized,” “lost,”“eclipsed,” “confused,” and “bewildered.” This theme, expressed in manydifferent ways and in many different places is at the basis of his radical cri-tique of the political state.

In Individualism Old and New, he attacked the ideology of individualism,repeating earlier indictments of its mythological character, and he spoke of“the lost individual,” lost because while persons “are now caught up in vastcomplex of associations, there is no harmonious and coherent reflection of theimport of these connections into the imaginative and emotional outlook onlife.” Blunted, if not impossible, is “the give and take of participation, of asharing and significance of the integrating factors.” Instead, we have con-formity, “a name for the absence of vital interplay; the arrest and benumbingof communication” (ION: 85–86).

Moreover, here as in other places, Dewey attributes our “rapacious nation-alism” to a situation in which “corporateness has gone so far as to detach in-dividuals from their old local ties and allegiances but not far enough to givethem a new center and order of life.” While “modern industry, technology and

194 Chapter Eight

Page 224: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

commerce have created modern nations in their external form” and “armiesand navies exist to protect commerce, to make secure the control of raw ma-terials, and to command markets, . . . the balked demand for genuine cooper-ativeness and reciprocal solidarity in daily life finds an outlet in nationalisticsentiment.” Finally, “if the simple duties of peace do not establish a commonlife, the emotions are mobilized in the service of a war that will supply itstemporary simulation” (61–62). Indeed, the windows of our building are alsoclosed to the light of heaven.

In Freedom and Culture (1939), he spoke of a kind of “molluscan organi-zation, soft individuals within and a hard constrictive shell without” (F&C:160). In this text, the problem is put in terms of “culture”: “The problem is toknow what kind of culture is so free in itself that it conceives and begets po-litical freedom as its accompaniment and consequence” (6). Dewey is clearthat present culture militates against such a consequence and that “the situa-tion calls emphatic attention to the need for face-to-face associations, whoseinteractions with one another may offset if not control the dread impersonal-ity of the sweep of present forces” (159). Nevertheless, as John McDermotthas rightly noted, “Dewey expresses deep reservations about the externalsigns of progress, whether of material or intellectual accomplishment” (Mc-Dermott, 1973: 679). McDermott calls our attention to the following:

Schooling in literacy is no substitute for the dispositions, which were formerlyprovided by direct experiences of an educative quality. The void created by lackof relevant personal experiences combines with the confusion produced by im-pact of multitudes of unrelated incidents to create attitudes which are responsiveto organized propaganda, hammering in day after day the same few and rela-tively simple beliefs asseverated to be “truths” essential to national welfare(F&C: 46).

But this problem, the problem of the public, was not for Dewey to be reducedto that of private property and to the domination of politics by “the moderneconomic regime.” To be sure, “the philosophers of ‘individualism’ predictedtruly” when “they asserted that the main business of government is to makeproperty interests secure” (PP: 108–9). Nevertheless, “economic determin-ism” was not the whole story for Dewey. On the other hand, the problem wasnot to be solved either by changes in the organization of government. “Theproblem lies deeper,” he wrote. “The search for the conditions under whichthe public may find and express itself is necessarily precedent to any funda-mental change in the machinery” (146).

What, then, are the conditions that need to be brought into existence to re-discover the public? According to Dewey, it is simply democracy. But we mustrepeat, this doesn’t mean, “that the evils can be remedied by introducing more

John Dewey: Anarchism and the Political State 195

Page 225: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

machinery of the same kind . . . or by refining and perfecting that machinery”(144). It means democracy “in its generic social sense” (147). Democracy hererefers to the idea of democracy, the idea of democracy as community. Whatneeds to be done is to identify the conditions of community and to bring theminto existence.

The identification of the idea of democracy and the idea of community maybe Dewey’s most characteristic doctrine. He seems to have arrived at it earlyand to have never abandoned it. And he gave it a clear and special meaning.

Democracy and Education (1916) gives one of the better statements of de-mocracy as “more than a form of government” and as “primarily a mode ofassociated living, of conjoint communicated experience” (D&E: 87). Deweypoints out “we cannot set up, out of our heads, something we regard as anideal society” (83). The problem, rather, “is to extract the desirable traits offorms of community life which actually exist. . . .” From two such traits,Dewey derives “a standard”: “How numerous and varied are the interestswhich are consciously shared?” and “How full and free is the interplay withother forms of association?” (83).8

These themes are more fully developed in The Public and Its Problemswhere Dewey reasserts that “regarded as an ideal, democracy is not an alter-native to other principles of associated life. It is the idea of community itself”(PP: 148). Dewey is here at pains to emphasize the task of knowledge andparticipation in the constitution of the democratic community. The complex-ities and scope of indirect consequence had destroyed communities. Althoughas individuals we are interdependent, community exists only when “the con-sequences of combined action are perceived and become an object of desireand effort” (151). It is then that “a distinctive share in mutual action is con-sciously asserted and claimed” (152). It is then that “I” can become “We.”Moreover, as Rousseau had already seen, “interdependence provides just thesituation which makes it possible and worthwhile for the stronger and ablerto exploit others for their own ends to keep others in a state of subjectionwhere they can be utilized as animated tools” (155).

But if as Dewey reads Rousseau, the solution is “a return to the conditionof independence based on isolation,” then asserts Dewey, “it was hardly seri-ously meant” (155). “The only possible solution” is nevertheless indicated. Itis “the perfecting of the means and ways of communication of meanings sothat genuinely shared interest in the consequences of interdependent activitiesmay inform desire and effort and thereby direct action” (155). This too candefine the idea of democracy:

Wherever there is conjoint activity whose consequences are appreciated as goodby all singular persons who take part in it, and where the realization of the good

196 Chapter Eight

Page 226: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

is such as to effect an energetic desire and effort to sustain it in being just be-cause it is a good shared by all, there is insofar a community. The clear con-sciousness of a communal life, in all its implications, constitutes the idea of de-mocracy (149).

Community requires communication and it requires knowledge, but crucially,the kind of knowledge which is “the prime condition of a democratically or-ganized public is a kind of knowledge and insight which does not yet exist”(166). For Dewey, such knowledge is a knowledge which is shared, whichfunds experience with common meanings, transforms needs and wants intomutually understood goals and which thereby consciously directs conjoint ac-tivity. As it is, knowledge is merely technique: “knowledge goes relativelybut little further than that of the competent skilled operator who manages amachine. It suffices to employ the conditions that are before him. Skill en-ables him to turn the flux of events this way or that in his neighborhood. Itgives him no control of the flux” (166).

Dewey fully recognized, both as ideal and as possibility, that the idea of re-turning to some barricaded and provincially defined context was miscon-ceived. On the other hand, he persistently demanded that the basic and fun-damental locus of life had to be the neighborly community. He asserted:

In its deepest and richest sense a community must always remain a matter offace-to-face intercourse. . . . The Great Community, in the sense of free and fullcommunication is conceivable. But it can never possess all the qualities, whichmark a local community. It will do its final work in ordering the relations andenriching the experience of local associations (211).


Whatever the future may have in store, one thing is certain. Unless local com-munity life can be restored the public cannot adequately resolve its most urgentproblem: to find and identify itself (216).

For Dewey, deliberative participation in conjoint activity, the shared commu-nication of goals and outcomes of that activity, the communication of mean-ings that that presupposes, is generalizable to the Great Community, but in-evitably, at an increasing degree of abstraction and dilution. If “in its deepestand richest sense a community must always remain a matter of face-to-faceintercourse”—for there to be genuinely deliberative participation, immediaterecognition of shared meanings, and concrete satisfaction of purposes con-sciously aimed at—then as one moves away from the local community,“community” becomes increasingly shallow and more watery. On the other

John Dewey: Anarchism and the Political State 197

Page 227: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

hand, at every increasingly inclusive level there must be some “ordering” andsome sharing.

This, however, is the anarchist image of the good society. Thus Martin Buber:

The collectivity is not a warm, friendly gathering but a great link-up of eco-nomic and political forces inimical to the play of romantic fancies, only under-standable in terms of quantity, expressing itself in actions and effects—a thingwhich an individual has to belong to with no intimacies of any kind but all thetime conscious of his energetic contribution . . .

An organic commonwealth—and only such commonwealths can join to-gether to form a shapely and articulated race of (persons)—will never build it-self up out of individuals but only out of small and ever small communities: anation is a community to the degree that it is community of communities.9

There remains, however, a legitimate question and ambivalence in Dewey,even given that his image of the Great Community and his criticism of exist-ing societies is profoundly anarchistic. It is the question, whether and in whatsense, idea of government may be still relevant? There are two questions.First, what constitutes an anarchist answer to this question? Second, whatseems to be Dewey’s answer?

Different anarchists have given different sorts of analyses of the relevantissues, but it may be that the main tradition of anarchist thought is best de-scribed not as “anarchy,” but as Buber put it, as “anocracy” (�������́�)—“not absence of government but absence of domination.” This cuts twoways, meaning not only that nongovernmental forms of domination are to berejected, but also that nondominating “government” may be tolerated, in-deed required. Anarchists are antistate insofar as we keep in mind that thestate is a particular kind of political entity that because of its nature con-stricts and disallows democracy as a mode of life. Its institutions are inher-ently structures of domination. But “government” is consistent with anar-chist principles if by government one means roughly what the Greeks andRousseau had in mind; namely “a commission” or an “employment” whichserves—now to use Dewey’s extremely useful language—active and articu-lated publics. An articulated public could still use, might very well need,“agencies” in this sense. Government, a modern word which in this contextmust now be stripped of its modern connotations will not rule, and the hold-ers of “office” will not be rulers—if by that one means that they will not bein a position to legitimately dominate those they “represent.” Indeed, as ar-gued in Chapter 11, below, and extensively developed in my War and De-mocracy (1989), these ideas were clearly articulated by many writers duringthe so-called crisis period in the United States.

198 Chapter Eight

Page 228: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

This position can be reposed in terms of the question of the legitimacy oflaw and government.10 Dewey’s stance in this regard is strikingly similar tothe one advanced by William Godwin in his classic Enquiry Concerning Po-litical Justice (1793), perhaps the first systematic effort at anarchist theory.

Godwin began with a dialectical criticism of liberal political philosophy,and especially with the familiar idea that obedience to and the authority ofgovernment derive from contract. Finding unsurmountable difficulties inthis theory, Godwin shifts to a utilitarian ground and finds that justice isthe key. Three kinds of authority are distinguished: to one’s judgment, tospecialized knowledge, and to sanctions. On this view, then, if a rule isjust, it should be complied with—but because one can see that the rule isjust, not because of some mythological contract. Similarly, for Dewey, thequestion, “Why should the will of rulers have more authority than that ofothers?” and “Why should the latter submit?” are spurious questions, the“dialectical consequence of . . . theories . . . which define the state in termsof an antecedent causation . . . ” (PP: 53). Dewey was quite correct in hold-ing that it was these sorts of theories that dominated modern politicalthought. But in rejecting the very formulation of the question, Dewey camepreciously close to the anarchist Godwin.

Thus, “the regulations and laws of the State are misconceived when theyare viewed as commands” (53). Commands presuppose a commander. If so,we can then ask the question, what gives the commander the right to com-mand? What grounds my duty to him? For Dewey, however, laws are but in-struments: “the institution of conditions under which persons make theirarrangements with one another” (54). They are “a means of doing for a per-son what otherwise only his own foresight, if thoroughly reasonable, coulddo” (56). Evidently, on this view, as Godwin had also insisted, rules are goodor bad only insofar as they are means for doing what reasonable people woulddo and for assisting them in getting those things done. Their justificationneeds nothing else. Indeed, it is the spurious theories of law and the statewhich lead us to look elsewhere and which, ultimately, cause us to blindly fol-low rules which are not so justified.l1

“Anarchism,” a transliteration from the Greek, means literally “without aruler” and because in some contexts, in the absence of a ruler—a commanderor someone to give orders—there is disorder, anarchy can also denote chaos.Anarchists do not, of course, assume that this must be so and believe thatideally at least, individuals can be self-governing. Indeed, they believe thatit is primarily the mystified complexity of the state, as Godwin puts it, “thecraft and mystery of governing,” “the pernicious notion of an extensive ter-ritory,” “the dreams of glory, empire and national greatness” which prohibitsuch self-governance. Even worse, it is the myth of popular sovereignty that,

John Dewey: Anarchism and the Political State 199

Page 229: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

paradoxically, has so successfully propelled the mystification of the state.Compare here Godwin and Dewey. First Godwin:

Too much stress has undoubtedly been laid upon the idea, as of a grand andmagnificent spectacle, of a nation deciding for itself upon some great publicprinciple, and of the highest magistracy yielding its claim when the generalvoice has pronounced (Godwin, 1971: 115).

And Dewey:

The familiar eulogies of the spectacle of “free men” going to the polls to deter-mine by their personal volitions the political forms under which they live is aspecimen of the tendency to take whatever is readily seen as the full reality ofthe situation (PP: 101).

Dewey could no doubt agree with Godwin, that in his society, like Godwin’s,too many revere too many established institutions and bad laws, that “as su-pernatural matters have progressively been left high and dry . . . the actualityof religious taboos has more and more gathered about secular institutions, es-pecially those connected with the nationalist state” (PP: 170). Indeed, “if‘holy’ means that which is not to be approached nor touched, save with cere-monial precautions and by specially anointed officials, then such things areholy in contemporary political life” (170).

David Wieck has perceptively observed that “the values which Deweyhoped to realize in a democracy . . . are realizable only in something ap-proaching anarchy.” But he may be correct in saying that “about decentralism. . . Dewey hadn’t paid heed to Kropotkin.”12 Still, we may wonder. As earlyas 1918, Dewey wrote that “if we are to have a world safe for democracy anda world in which democracy is safely anchored, the solution will be in the di-rection of a federated world government and a variety of freely experiment-ing and freely cooperating self-governing local, cultural and industrialgroups” (C&E, II: 559–60). Forty years and two world wars after he hadcalled for “more radical thinking” about the sovereign state, Dewey wrote (inhis 1946 Afterword to The Public and Its Problems) that “the State is a myth.”He there offered “as a working principle,” “the idea of Federation as distinctfrom both isolation and imperial rule” (255). Dewey did not, it seems, getclear about what this might mean concretely. Nevertheless, if we take therestoration of the local community as his point of departure, his vision is in-deed powerful:

Territorial states and political boundaries will persist; but they will not be barri-ers which impoverish experience by cutting man off from his fellows; they willnot be hard and fast divisions whereby external separation is converted into in-

200 Chapter Eight

Page 230: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ner jealousy, fear, suspicion and hostility. Competition will continue, but it willbe less rivalry for acquisition of material goods, and more emulation of localgroups to enrich direct experience with appreciatively enjoyed intellectual andartistic wealth (PP: 217).

And the material basis for such a Great Community is within reach:

If the technological age can provide mankind with a firm and general basis ofmaterial security, it will be absorbed into a human age (217).

Buber noted that “the socialist idea points of necessity, even in Marx andLenin, to the organic construction of a new society out of little societies in-wardly bound together by a common life and common work and their associ-ations” (Buber, 1949: 99). But how much more is it true that the main andmost distinctive themes in Dewey’s social philosophy point—of necessity—to such a vision?

In the next part, I will seek to reinforce the foregoing claim by arguing thatDewey, along with the anarchists, parted ways with the Marxist-Leninists for ap-proximately the same reasons and with approximately the same conclusions.


Dewey’s social and political philosophy is close to anarchism as regards hisview of social change. Both sharply contrast with Marxism in rejecting theidea that a “social revolution” could be made by the few for the many and inrejecting the idea that the “proletariat” must be the agent of an insurrectionaryrevolution.

The anarchist, as Dewey, does not deny the existence of class division insociety and both affirm that a good society could not be class divided. But itwas a mistake, on both of their views, to suppose that progressive changecould have but one agent or that any one agent of change could be sufficient.As Hook rightly pointed out, Dewey spoke of “class struggles in their pluralform.”13 Anarchists tended to speak more vaguely of “the people” or “themasses.” Marxists will insist, of course, that this offers a dubious politics.

The matter of insurrection is more complicated. In the first place, some an-archists did believe in insurrection and in the use of such violence as had toattend insurrection.14 Bakunin saw this in apocalyptic terms; Kropotkin be-lieved it to be unavoidable, but hoped that the violence could be kept to aminimum. Tolstoi, at opposite poles from Bakunin, was consistently pacifist.But more important, anarchists tended to be undoctrinaire about the problemsof revolutionary change, to orient programs to specific contexts and to

John Dewey: Anarchism and the Political State 201

Page 231: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

emphasize pedagogic means. These emphases are, of course, wholly congen-ial to Dewey.

The emphasis on pedagogic means was implied by the most characteristiccriticism by anarchists of Marxian politics. For the anarchist, there could beno separation of the revolution process from the revolutionary goal. Thus,Alexander Berkman:

It is only by growing to a true realization of their present position, by visualiz-ing their possibilities and powers, by learning unity and cooperation, and prac-ticing them, that the masses can attain freedom.15

Or Gustav Landauer:

One can throw away a chair or destroy a pane of glass; but those are idle talk-ers and credulous idolaters of words who regard the state as such a thing or as afetish that one can smash in order to destroy it. The state is a condition, a cer-tain relationship among human beings, a mode of behavior between (people);we destroy it by contracting other relationships, by behaving differently towardone another. . . . We are the state, and we shall continue to be the state until wehave created institutions that form a real community and society of (persons).16

All this could be Dewey. But especially noticeable are Dewey’s arguments,which parallel and often supplement and enrich the anarchist position.

Most anarchists have not been so naive as to suppose that persons are nat-urally good. Nor have they based their hopes or, more important, their pro-grams on such a postulate—even if, unfortunately, they are too frequentlyread that way. Some, for example, Landauer, were influenced by romanticismand especially by Nietzsche. Others, for example, Kropotkin, were influencedby the naturalism of Darwin. While I cannot here assess these different viewsof human nature, there is little doubt that Dewey was deeply concerned withthe question and that his approach has considerable force.

Dewey did not think that human beings were naturally good, naturally in-telligent, or naturally free, since as is well known, on his view, human im-pulses and capacities were always realized socially. This meant that institu-tional arrangements were decisive: “Social arrangements are means ofcreating individuals” (RIP: 194).

But at the same time, Dewey did not find himself caught up in what hecalled “a vicious circle.” For him, individuals, beginning from where theywere, could change themselves as they change society. As Arthur Lothsteinhas rightly pointed out, Dewey dropped the self-enclosed metaphor of the cir-cle for the dynamic and open-ended metaphor of the spiral. Dewey wrote, forexample:

202 Chapter Eight

Page 232: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

We are not caught in a circle; we traverse a spiral in which social customs gen-erate some consciousness of interdependencies, and this consciousness is em-bodied in acts which in improving the environment generate new perceptions ofsocial ties, and so on forever (HNC, in McDermott: 721).

This was possible since, in Dewey’s terms, habits and customs could be de-liberatively transformed. “Habits,” he argued, were the “mainspring of humanaction, and habits are formed for the most part under the influence of the cus-toms of the group” (PP: 119). We are never “habitless,” to be sure, but cus-toms are nothing but the social “grooves” which are the result of previous ha-bituations. At the social-psychological level, habits can be altered throughacting differently; and because custom and social structure are themselves in-carnate in the repeated and multiplied acts of persons, acting differentlychanges them too. Indeed, Dewey well sees that, in Giddens’s terms, actingdifferently creates new conditions as it changes our perceptions, beliefs, anddesires. And this means also that we need pertinent knowledge of the condi-tions and consequences of our actions and we need to be clear about ourgoals. (See Chapter 4.)

For Dewey, the problem was essentially pedagogic. Purposive and pro-gressive change in society, however, has to be directed and there must be def-inite goals in mind. Dewey was perfectly clear about this as regards educa-tion: “The conception of education as a social process and function has nodefinite meaning, until we define the kind of society we have in mind” (D&E:97). And in Human Nature and Conduct as in many, many other places,Dewey put great emphasis on educating the young:

the cold fact of the situation is that the chief means of continuous, graded eco-nomical improvement and social rectification lies in utilizing the opportunitiesof educating the young to modify prevailing types of thought and desire. Theyoung are not as yet as subject to the full impact of established customs (HNC:127).17

Dewey was optimistic in his assessment that the school could be “the chiefmeans” of social rectification. But on his own premises, the school—like theexperimental anarchist community—was not and could not be independentand disconnected from the large society. It is inevitable, accordingly, that itwould tend to reproduce the habits and ideas of the larger society, since es-pecially in the case of the schools, they depended for their existence on insti-tutions interested explicitly in maintaining the status quo.

But the school was not the only place where changes could be wrought. In-deed, in terms of Dewey’s theory, since all “habits” and all “customs” weresustained by repeated activities, any of them could be changed by changing

John Dewey: Anarchism and the Political State 203

Page 233: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

these acts that sustain them. Nevertheless, if the alteration of activities andthence of habits and customs was not to issue in chaos—into merely a break-down of the prevailing order of things, it had to be directed and unified, andas with the schools, there had to be definite goals in mind. There are but twoalternatives. Either one imposed the change on individuals or one took ad-vantage of opportunities to encourage and develop tendencies on the part ofthose affected to make the changes themselves. The former route, of course,is the method of revolutionary vanguard parties; the latter is the method of de-mocracy. Indeed, knowledge is liberating insofar as understanding how weparticipate in maintaining oppressive conditions gives us reasons to changethose conditions.

It is possible to discern in Dewey’s writings at least three “powerful objec-tions to the strategy of imposed change.” First, and very generally, for Dewey,if intelligence is to be brought to bear on progressive social change, not onlymust we consider goals, but as well, we must consider the particular condi-tions and particular context and assess the complex possible consequences ofpossible strategies for change. Imposing change may produce the desired out-come, but even if it does, it doesn’t produce just that outcome. As Deweysummarized the point:

Doctrines, whether proceeding from Mussolini or Marx, which assume that be-cause certain ends are desirable therefore those ends and nothing else will resultfrom the use of force to attain them is but another example of the limitations puton intelligence by any absolutist theory (PM: 139).

But Emma Goldman—with some of the same “doctrines” clearly in mind,would seem to concur heartily:

Anarchism is not . . . a theory of the future to be realized through divine inspi-ration. It is a living force in the affairs of life, constantly creating new condi-tions. The methods of anarchism therefore do not comprise an ironclad programto be carried out under all circumstances. Methods must grow out of the eco-nomic needs of each place and clime, and of the intellectual and temperamentalrequirements of the individual. . . . Anarchism does not stand for military drilland uniformity; it does, however, stand for the spirit of revolt, in whatever form,against everything that hinders human growth (Goldman, 1969).

A second and related objection regards the difficulty—if not impossibility—of all-at-once, totalist, attempts at change. Dewey concluded:

The revolutionary radical . . . overlooks the force of ingrained habits. He is right,in my opinion, about the infinite plasticity of human nature. But he is wrong inthinking that patterns of desire, belief and purpose do not have a force compa-

204 Chapter Eight

Page 234: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

rable to the momentum of physical objects. . . . Habit, not original human na-ture, keeps things moving most of the time (PM: 190).

Imposed change cannot be sustained because habits cannot be dramaticallyaltered. That is why revolutionary societies tend to revert to older ways of ac-tivity, and worse, to reproduce the old structures but in new institutionalforms. Similarly, anarchists write of the “preparation” for the revolution toanarchistic society and emphasize the pedagogic problem of changing people—a problem not solvable by “divine inspiration,” by violence and au-thoritarian tactics. Such changes will take time and if they are to be sustained,must be deeply rooted. This is suggested by the previous text quoted fromEmma Goldman; it reoccurs with a different emphasis in this earlier text, al-most in the language of Dewey:

The true criterion of the practical . . . is not whether (some scheme) can keep in-tact the wrong or foolish; rather is it whether the scheme has vitality enough toleave the stagnant waters of the old, and build as well as sustain, new life (Gold-man, 1969: 9).

I am suggesting here, of course, that anarchists do not have a utopic concep-tion of social change, that they realize full well, that their ideal could notcome into existence by means of some totalist transformation, as Goldmanput it, through “divine inspiration.” The idea that anarchists must reject any-thing short of their ideal as unjustifiable and therefore deserving of immedi-ate destruction is not anarchism but nihilism. And this means that the new so-cial forms, new habits and customs will be but painfully and slowly evolved.Daniel Guerin, a contemporary French anarchist writer notes:

Proudhon, in the midst of the 1848 Revolution, wisely thought that it wouldhave been asking too much of his artisans to go, immediately, all the way to ‘an-archy.’ In default of his maximum program, he sketched out a minimum liber-tarian program: progressive reduction in the power of the State, parallel devel-opment of the power of the people from below. . . . It seems to be the more orless conscious purpose of many contemporary socialists to seek out such a pro-gram (Guerin, 1970: 550)

The anarchist, like Dewey, can have a vision of the good society without be-ing lacking in programs. And if Dewey and the anarchists are correct, it is notthe doctrinaire revolutionaries who are “practical,” but as Goldman suggests,those who seek programs which are vital enough “to leave the stagnant wa-ters of the old, and build as well as sustain new life.”

It is hardly assumed here that the commitment of the anarchist tradition—andof Dewey—to a revolutionary process consistent with genuine democracy

John Dewey: Anarchism and the Political State 205

Page 235: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

settles any of the difficult questions which will still need to be asked, or thatDewey and particular anarchist writers would necessarily or even likely agreeon particular plans of action. Indeed, there is a very great difference betweenDewey’s emphasis on the methods of political democracy and the anarchist em-phasis on what is called “direct action.” However, as April Carter has correctlypointed out, while direct action “must be distinguished from constitutional andparliamentary styles of activity on the one hand, and from guerilla warfare onthe other,” not only do forms of direct action shade into parliamentary styles,as, e.g., in sit-ins, strikes, and “civil disobedience,” but as well, they may bebest construed as a kind of crude and creative form of direct democracy. Inso-far, accordingly, direct action may be entirely consistent with, indeed, a logicalimplication of, Dewey’s social philosophy. As anarchists have often argued,such tactics are essentially pedagogic because they are vehicles by which per-sons learn and practice democratic participation (Carter, 1973).

Moreover, there may be considerable disagreement on whether the state isitself to be used and if so how, whether, e.g., as some anarchists have argued,one should entirely reject the vote, or political parties, whether one shouldseek alternative and parallel forms or whether it is possible to effect progres-sive change through existing structures.18

Nor finally, need there be agreement on the prospects and probability ofchange toward the ideal of democracy. Perhaps Dewey should here have thelast word:

The foundation of democracy is faith in the capacities of human nature; faith inhuman intelligence and the power of pooled and cooperative experience. It isnot because these things are complete but that if given a show they will growand be able to generate progressively the wisdom needed to guide collectiveagain (D&EA: 402).


1. C. Wright Mills, Sociology and Pragmatism (New York: Oxford UniversityPress, 1969): 443–44, and more recently, Charles Frankel, “John Dewey’s Social Phi-losophy,” New Studies in the Philosophy of John Dewey (Hanover, N.H.: UniversityPress of New England, 1977) found Dewey’s thought to be nostalgic. Alphonso I.Damico, Individualism and Community, The Social and Political Thought of JohnDewey (Gainesville: University Presses of Florida, 1978) discusses some of the liter-ature relevant to “pluralistic democracy” and some of the problems regarding the re-lation of Dewey’s theory of inquiry to his social philosophy. See also Mills on thispoint: 318–20 and Chapters 20 and 21. George Novack in Pragmatism versus Marx-ism (New York: Pathfinder Press, 1975) cites Maurice Cornforth as one example ofthe view that “pragmatism, particularly in the form which Dewey has given it, is the

206 Chapter Eight

Page 236: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

philosophy of American imperialism” (275). Novack rejects this (silly) view eventhough he finds that Sidney Hook’s uneven political career, including his defense ofAmerican imperialism, stems from pragmatism’s “promiscuousness” (82). It wasHook, of course, who attributed to Dewey “the best elements of Marx’s thought”(Reason, Social Myths and Democracy [New York, Harper Torchbooks, 1966]: 132).

2. References to Dewey’s writings will be indicated with abbreviated titles andpage numbers within parentheses. See below for abbreviations of Dewey texts cited.

3. Quotations from Baldwin are from Paul Avrich, The Modern School Move-ment: Anarchism and Education in the United States (Princeton, N.J.: Princeton Uni-versity Press, 1980: 38) from an interview with Baldwin.

4. Sidney Hook, John Dewey: An Intellectual Portrait (New York: John Day, 1939).See also Arthur Lothstein’s excellent dissertation, “From Privacy to Praxis: The Case forJohn Dewey as a Radical Social Philosopher,” (New York: New York University Press,1979, Chapter 4).

5. For a discussion of Dewey’s response to the trial and execution, see GeorgeDykhuizen, The Life and Mind of John Dewey, edited by Jo Ann Boydston (Carbon-Dale: Southern Illinois Press, 1973).

6. See Frank A. Warren, An Alternative Vision (Bloomington: Indiana UniversityPress, 1974). For Dewey’s analysis of the need for a third party, see Bingham andRodman (eds.), Challenge of the New Deal (Nashville: Falcon Press, 1934) whichreprints Dewey’s important Common Sense essay, “Imperative Need for New RadicalParty.” See James Campbell’s very useful account, in his dissertation, “Pragmatismand Reform: Social Reconstruction in the Thought of John Dewey and George Her-bert Mead,” (New York: SUNY, 1979).

7. The foregoing criteria for an anarchist social philosophy are quoted from JohnP. Clark, “What Is Anarchism?” in Anarchism: Nomos XIX, edited by J. R. Pennockand J. W. Chapman (New York: New York University Press, 1978): 13.

8. In this book, Dewey identifies the dire consequences of Identifying “the civicfunction” of education with The State, and concludes that the very idea of nationalsovereignty gives rise to a contradiction “between the wider sphere of associated andmutually helpful social life and the narrow sphere of exclusive and hence potentiallyhostile pursuits and purposes” (97).

9. Martin Buber, Paths in Utopia (Boston: Beacon Press, 1949): 42. This indis-pensable little book may be the best single treatment of the dilemmas of the statist/anarchist tensions, theoretical and practical, in the radical tradition. Buber and Deweyhave very much in common and, perhaps, not surprisingly; Buber’s anarchism isoverlooked.

10. I have discussed the question of the legitimate state in The Death of the State(New York: G. P. Putnam’s Sons, 1974), Chapter 2.

11. On this view, the problem of “civil disobedience” is misconceived, as Godwin,Thoreau—and Dewey, show. That is, whether in any given case, one should or shouldnot comply with a law depends upon its justness and the consequences of complying ornot. Cf. Dewey’s little essay, “Conscience and Compulsion” (in Characters and Events).

One should also note that Godwin, as most anarchist writers, conceives of the prob-lem of coercive sanctions in straightforward consequentialist terms and argues that

John Dewey: Anarchism and the Political State 207

Page 237: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

only at the limit of the anarchist ideal are they unjustifiable. Cf., for example, God-win, 1971: Book VII.

12. The first text is from David Wieck, “Anarchist Justice,” in Nomos XIX: 235,the second is from his review of Paul Goodman’s Drawing the Line, Telos, No. 35(Spring 1978).

13. Lothstein carefully examines the pertinent literature in his “From Privacy toPraxis,” Chapter IV. He rescues Dewey from Dewey’s “right-wing epigones” and re-sponds with force to some of Dewey’s more mechanically minded left critics. ButLothstein is himself critical of Dewey.

14. This is as good a place as any to comment on the unfortunate association of an-archism with terrorism. Terrorism must be distinguished from violence as such. Un-derstood as the idea that acts of violence against individuals or groups: assassinations,bombings, kidnappings, etc., are means of revolutionary change, terrorism has beenrejected by almost all anarchist writers, yet it is true that one can identify a period ofterrorist activity in several national histories. On this see, James Joll, The Anarchists(New York: Grosset and Dunlap, 1964) and Paul Avrich, The Russian Anarchists(Princeton, N.J.: Princeton University Press, 1967). The identification of anarchismwith terrorism has, of course, had enormous consequences and, no doubt, goes someway toward explaining the discrediting of anarchism. For a perceptive account, seeEmma Goldman, “The Psychology of Political Violence” (1910), in Anarchism andOther Essays (New York: Dover, 1969). Dewey, we should recall, argued that “theonly question which can be raised about the justification of force is that of compara-tive efficiency and economy of use” and “what is justly objected to as violence or un-due coercion is a reliance upon wasteful and destructful means of accomplishing re-sults” (“Force and Coercion” (1916) in Characters and Events (II: 789).

15. Alexander Berkman, What Is a Communist Anarchist? with an Introduction byPaul Avrich (New York: Dover, 1972), originally published as New and After: TheABC of Communist Anarchism (1929). This is a very clear exposition of many of thekey points of difference between anarchism and (the prevailing) Marxism.

16. Quoted by Eugene Lunn, Prophet of Community: The Romantic Socialism ofGustav Landauer (Berkeley: University of California Press, 1975: 226, translatedfrom “Schwache Stattsminner, Schwacheres Volk,” (Der Sozialist June, 1910).

17. American anarchists also put enormous emphasis on educating the young. TheSpaniard, Francisco Ferrar, was a more direct influence on early twentieth-century ef-forts, but the pertinence of Dewey’s view were fully recognized. For discussion, seeespecially, Paul Avrich, The Modern School Movement. It is interesting to notice thatDewey was not especially interested in educational experiments conducted within an-archist colonies, although he did visit the Stelton Colony and school in Stelton, N.J.On the other hand, many anarchists had little confidence in this route. Berkman said,“I myself . . . have little faith in colonies. You cannot build the new society that way”(quoted from Avrich: 306).

18. Much recent radical theory has come to the conclusion that in the advancedcapitalist states where liberal democracies exist, the only strategy to be pursued—consistent with a genuine socialism—requires an answer to this question:

208 Chapter Eight

Page 238: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

How is it possible radically to transform the State in such a manner that the extension anddeepening of political freedoms and the institutions of representative democracy . . . arecombined with the unfurling of forms of direct democracy and the mushrooming of self-management bodies? (Nicos Poulantzas, State, Power and Socialism [London: Verso,1980]).


GPP: German Philosophy and Politics (New York: Henry Holt, 1915).D&E: Democracy and Education (New York: Free Press, 1966, Enlarged Edition).RIP: Reconstruction in Philosophy (Boston: Beacon Press, 1957).HNC: Human Nature and Conduct (New York: Henry Holt, 1922).PP: The Public and Its Problems (Chicago: Swallow Press, 1954).PF: “Philosophies of Freedom,” in R. J. Bernstein (ed.), Experience, Nature and Free-

dom (Indianapolis, Ind.: Bobbs-Merrill, 1960).C&E: Characters and Events. Two Volumes, edited by Joseph Ratner (New York:

Henry Holt, 1929)LSA: Liberalism and Social Action (New York: Capricorn, 1963).ION: Individualism Old and New (New York: Capricorn, 1962).F&C: Freedom and Culture (New York: Capricorn, 1963).D&EA: “Democracy and Educational Administration,” in Joseph Ratner (ed.), Intel-

ligence and the Modern World (New York: Modern Library, 1939),PM: Problems of Men (New York: Philosophical Library, 1946).

John Dewey: Anarchism and the Political State 209

Page 239: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 240: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


If we begin with the political calamities of the last one hundred years and add,as at least part consequence of these, the upheavals in philosophy, literature,art and science, we can appreciate the present attractiveness of a political phi-losophy without foundations: There is no truth; only an endless “conversa-tion” in a self-sufficient linguistic realm which is totally disconnected fromany extra-linguistic reality—if such there be. Because God is dead, “humannature” has no content, and history is meaningless, the dream of creating anew kind of human society—the dream of utopian and revolutionary modernpolitics—is instead a nightmare. There is no knowable, objective, definable,transmittable common good; there are only “interests,” not to be judged, stillless to be accommodated. There is no responsible politics which is not impo-tent: Either we irresponsibly offer “the masses” ungrounded hope or, more re-sponsibly, we reject the quest for “glittering triumph,” perhaps even im-provement, and settle for “the far more modest, though indispensable,concern to prevent ‘catastrophes.’” On this view of things, the belief that“everything is possible seems to have proved only that everything can be de-stroyed,” that efforts “to escape from the grimness of the present into nostal-gia for a still intact past, or into the anticipated oblivion of a better future, arevain.”1 With Camus, anguishing over Algeria, we may wonder how one caneven write, knowing that what is said might provide an alibi for a Pol Pot ora terrorist willing to throw—or drop—a bomb.

The alternatives would seem to be these: Either one knows what is goodand true, or one does not. If one does, then, must not one act on that knowl-edge, even if, finally, it turns out that one is wrong? On the other hand, if onedoes not know what is good and true, if perhaps there is no good and true,then, must we not be unwilling to act in the name of “the people,” or history,

Chapter Nine

Philosophy and Politics: A HistoricalApproach to Marx and Dewey

Page 241: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

or of our ideals? Instead, is it not the case that our commitment must be butto keep the conversation going?2

As “liberal,” the antifoundationalist alternative has an appeal. Indeed, it isjust this that makes it a useful counterrevolutionary ideology. But the issue ismisformulated. It is not that there is no basis for human solidarity—even ifthat basis is not to be discovered but created, or that there is no “truth,” evenif like solidarity, it too is a social product. Nor is violence, as such, the prob-lem. Like inaction, it is sometimes justified and sometimes not justified. For,unnoticed by the antifoundationalist is the possibility of a politics that needsno foundations, a politics which does not guarantee success and does not pre-suppose that some person or party has a truth not shared by those in whosename they.act. Put in other terms, the problem of modern politics is less thelack of “foundations” and more the absence of a genuinely democratic poli-tics, a politics which aims at the creation of communities by the active par-ticipation of interdependent individuals, a politics in which “interests” be-come shared goods, a politics which insists that truth can only be our truth. Inwhat follows, I argue that Marx and Dewey offered versions of this. On thisview, we are “sculptors” or crafters, but we lack blueprints. On this view, weare actors in history, but we can—and must—write our own scripts.


Marx had some clear and definite ideas about democracy, ideas which remainunrealized, but which cannot be dismissed as utopian or as youthful extrava-gances or cynical subterfuge. Marx consistently held that participatory democracy was the goal of revolutionary transformation, that what we callmodern democracy, though a form of alienated politics, was genuinely pro-gressive, and finally—and most critically—that there could. be no separationof revolutionary means from revolutionary ends. While the issues are con-tentious, I want to suggest that opposing views depend largely on an ahistor-ical reading of important texts and events. Critical in this regard is the per-sistent tendency to construe Marx’s politics in terms of a misconstrual of thedifferences between Marx and the anarchists and to be anachronistic as re-gards later debates between revolutionary Marxists and social democrats.3


Characteristically, Marx’s point of departure was criticism of Hegel, butwhile he rejected much in Hegel, he also found much of value. Hegel saw that

212 Chapter Nine

Page 242: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

the French Revolution had raised the problem of sovereignty in a critical wayand that it was this problem with which “history is now occupied, and whosesolution it has to work out in the future” (Hegel, 1956: 452). The problem wascritical, because in the fully developed modern state the people could not besovereign. On Hegel’s view of the matter, “‘the sovereignty of the people’ isone of the confused notions based on the wild idea of the ‘people.’ Takenwithout its monarch and the articulation of the whole . . . the people is a form-less mass and no longer a state” (Hegel, 1952: 279).

In the fully developed modern state, there was a bifurcation of civil soci-ety and the state. Individuals live private lives and relate anonymously. It wasthus that, unless articulated, the people are a “formless mass.” On the otherhand, the government, the king, parliament, the bureaucracy and the policebecame the mode by which “the whole” was articulated and expressed. Fail-ing to grasp the full force of the American “solution” to the problem of sov-ereignty, Hegel opted for a reactionary solution, a constitutional monarchy.

Marx agreed fully with the Hegelian analysis of the bifurcation of civil so-ciety and the state, but he saw also that the Americans had, in a remarkableway, already solved the problem that Hegel believed had still to be solved.That is, Marx saw that the fully realized modern state would be a democraticstate. The solution, however, required that in the fully realized modern state,the alienation of individuals would be fully realized and at the same time fullyobscured. In the democratic state, every adult is a “citizen” with full civil andpolitical rights. Moreover, in virtue of the mechanisms of representative gov-ernment, the people are “sovereign.” But for Marx the reality was otherwise:Each citizen is “an imagined member of an imagined sovereignty, divested ofhis actual individual life and endowed with an unactual universality.” In thedemocratic state, “liberators reduce citizenship, the political community, to amere means for preserving [the] so-called rights of man.” “But this meansman in his uncivilized and unsocial aspect, in his fortuitous existence and justas he is, corrupted by the entire organization of our society, lost and alienatedfrom himself, oppressed by inhuman relations and elements” (Easton andGuddat, 1967: 225–26, 231).

To say that in the democratic state individuals are “uncivilized” and“unsocial” and that they have “fortuitous existences” is not to say that in-dividuals are barbaric, or nasty or motivated by greed—though this mayalso be true. It is to say rather that they live private lives, that they are iso-lated, that they relate anonymously, that their situations are “accidental”like their sex or race, and that while their powers are social, they are notsocially realized. This is the result of “the entire organization of society,’’but in particular of the market structure of bourgeois society and the alien-ating structures of the modern state. As Thomas points out, “the state

Philosophy and Politics: A Historical Approach to Marx and Dewey 213

Page 243: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

becomes a fetishistic personification of political potential, very much asthe concept of capital designates the separation between the conditions oflabor and the producer. Both are the members of society’s own real forceset up against them, opposed to them, and out of control” (Thomas, 1980:196). The people are sovereign but they have no power over the conditionsof their lives. As sovereigns in an illusory community, they are in realitycontrolled by “inhuman relations.” A number of critical implications flowfrom Marx’s analysis.

First, the problem for Marx is first and last political, of what has to be doneand to happen if people are to gain control of the circumstances—now alien-ated—which structure their lives. This problem was not to be solved “econo-mistically” or by perfecting the instrumentalities of the democratic state. Thisdistinguished Marx’s view from, e.g., Proudhon on the one hand, and on theother, from republicans. Second, to achieve the goal is to overcome the dual-ity of civil society and the state, and this means, as Marx writes, “in true de-mocracy, the political state disappears (untergeht).” This view, of course, hasits Rousseauian intimations and suggests the critical point of comparison tonineteenth-century anarchisms and social democrats. But third, in contrast toanarchisms which share with Marx the idea that state power must be broken,if this was to be achieved, it had to be achieved by an agency which did notreflect the alienated relations of private property. As he wrote, it was the workof “a class with radical chains, a class in civil society that is not of civil soci-ety, a class that is the dissolution of all classes, a sphere of society having auniversal character because of its universal suffering” (Easton and Guddat,1967: 263–64).

These ideas lead directly to the 1848 Communist Manifesto where, recur-ring to the Aristotelian lineage of the idea that democracy was class rule bythe poor, Marx and Engels write “the first step in the revolution by the work-ing class, is to raise the proletariat to the position of ruling class, to establishdemocracy.” In this democracy, to be sure, “the political state” has not disap-peared, for as a statist form, this democracy was still a “dictatorship,” albeita dictatorship of the majority, the proletariat, against the minority—the own-ers of the means of production. It was thus that it was but “a first step in therevolution of the working class” (McClellan, 1977: 237).4

But it was the Paris Commune of 1871 which seems to have given Marxa paradigm for what might be, a paradigm which, prefigured in actuality,has earlier intimations of “true democracy.” As he said, the democracy ofthe Commune was a “historically new creation” and “the glorious harbin-ger of a new society.”5 Still, it is vital to be clear about what Marx saw in itand why also that, for him, it was a premature and finally unwise act ofheroism.

214 Chapter Nine

Page 244: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


For us, “democracy” refers to statist electoral politics and “anarchism” referseither to a silly utopianism or to terrorist politics. If in the nineteenth century,these terms had not yet been thoroughly appropriated by the enemies of ruleby the demos, it was already the case, as Marx noted, that “it was generallythe fate of completely new historical creations to be mistaken for the coun-terpart of older and even defunct forms of social life, to which they bear a cer-tain likeness” (Marx, 1971: 73). In what sense, then, was the Commune “acompletely new historical creation”?

On the one hand, it was anarchist in the sense that it broke state power. Ashe wrote: “The unity of the nation was not to be broken, but, on the contrary,to be organized by the communal Constitution and to become a reality by thedestruction of the State power which claimed to be the embodiment of thatunity independent of, and superior to, the nation itself, from which it was buta parasitic excrescence” (73). But it was not anarchist in the sense of Stirner,Proudhon, or Bakunin. Marx took these writers seriously—with due cause—but because, on his view, they lacked an adequate understanding of politicaleconomy, they were mistaken both as regards their vision of a good societyand as regards the means of attaining it. Putting the matter as briefly as pos-sible, for Marx, anarchists were inverted statists. Since on their view the statewas the problem, once rid of it, all would be well. Because for them there waslittle point in discriminating between forms of the state, transformative activ-ity, whether the antirevolutionary activity of Proudhon or the conspiratorialactivity of Bakunin, had to wash its hands of the state. As Proudhon said, “toindulge in politics is to wash one’s hand in dung” (quoted by Thomas, 1980:184).

Marx saw the matter very differently. His view of the Commune gives us astart in seeing how. In the first place, the Commune was

a thoroughly expansive political form, while all previous forms of governmenthad been emphatically repressive. Its true secret was this. It was essentially aworking-class government, the product of the struggle of the producing againstthe appropriating class . . . (Marx, 1971: 75).

We must be clear what this means. The Commune had fashioned the firstgovernment—the word must be used gingerly—which aimed at realizingfull control over the circumstances of life by ordinary citizens. It was inthis sense “expansive” in contrast to those forms that took for granted the conditions of ordinary life. Even the best case, for example, the de-mocracy of ancient Athens, took these for granted. Previous democracies,

Philosophy and Politics: A Historical Approach to Marx and Dewey 215

Page 245: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

like nondemocracies, were “repressive” in that they aimed but to replacethe rule of one class for another—without altering the alienating conditionswhich called for class rule in the first place. For example, in the ancientpolis, “politics” regarded the struggle between rich and poor over deci-sions of law and war, but the poor were not social revolutionaries in thesense that they either did or could aim at reconstituting society. This idea,which was owed to the French Revolution, regarded the perception, nowfamiliar largely through the thought of Hegel and of Marx that human his-tory was radically unlike natural history in being the product of human ac-tivity. For Marx this meant that

Communism differs from all previous movements in that it overturns the basisof all earlier relations of production and intercourse, and for the first time con-sciously treats all natural premises as the creatures of hitherto existing men,strips them of their natural character and subjects them to the power of unitedindividuals (McClellan, 1977: 179).

Political democracy was “a great step forward” because it acknowledged “thesovereign people,” but it was the final form of emancipation only “within theprevailing order of things.”

Indeed, it was just for these reasons that the Commune could not have suc-ceeded: “Apart from the fact that this was merely the rising of a city underexceptional circumstances, the majority of the Commune was in no way so-cialist nor could it be” (Marx, 1971: 293). The “exceptional circumstances”which produced the Commune left the full machinery of the repressiveFrench State in place, and insofar, the Communards faced formidable odds.But in addition, because the Communards were not organized, politically ac-tive workers, their political capacities were undeveloped.

The problem was not that the Communards lacked revolutionary con-sciousness, for they surely knew how to die on the barricades, nor was theproblem economistic, regarding their incapacity at the existing stage of eco-nomic development to conquer scarcity, but that an alienated citizenry was inno position to reabsorb their alienated social powers. They were still isolated,“private” persons who, as not yet thoroughly interdependent, could not or-ganize themselves so as to realize fully the powers that they had. As Edwardssays: “the Communards belonged more to the past tradition of Paris revolu-tionaries than presaging the industrial struggles of the future” (Edwards,1971: 360). As an alliance of artisans, of workers in craft industries, of pettybourgeois shopkeepers and traders, it was impossible for them to overcome“the contradiction between public and private life, between general and par-ticular interests.” It was thus that “with a modicum of common sense . . . [theCommune] could have reached a compromise with Versailles useful to the

216 Chapter Nine

Page 246: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

whole mass of the people the only thing that could have been reached at thetime” (Marx, 1971: 293).

The anarchists saw and celebrated the fact that the Commune was not madeby “the proletariat” and was instead an alliance of “the people,” and they sawand celebrated the spontaneity and disorganization that was a critical mark ofthe Commune. But, for Marx, their failure to see that unless the Communardswere to have a long period for self-education in self-rule—an impossibility inthe circumstances of Civil War—the Commune had to fail.

A second point of comparison between Marx’s understanding of the Com-mune and anarchist thought regards the question of government. For us, ei-ther there is a government or there is anarchy—no authority, no coercive or-ganization. The idea was surely reinforced by a great deal of anarchistpolemics—especially in the nineteenth century, but the confusion is deeper,depending on the eighteenth-century identification of rule with government.

In the ancient polis, there was, strictly speaking, no government; there wasrule by one, few or many. Political power was unmediated. In the modernstate, however, there are always governments, the executive, parliament, bu-reaucracies and the police, and they always claim to “represent” the “gov-erned.” It is easy then to suppose that the middle ground between self-ruleand rule by others is modern political democracy, “representative govern-ment.” But for Marx (and Dewey, as we shall see), this was not the only al-ternative. The institutional novelty of the Commune was in just this.

There would be “functionaries” of the people, “agents” in the strictest ofsenses, and these would be under strict “instructions” from those whom theyrepresent. These functionaries would not be as in bourgeois democracymerely “authorized” to rule those who elect them, but would be as ambassa-dors or military commanders, “responsible” and “revocable” at the pleasureof those who elected them. “Sovereignty,” as in Rousseau, would not bealienated. Accordingly, it would not be illusory. As Marx wrote:

Instead of deciding once in three or six years which member of the ruling classwas to misrepresent the people in Parliament, universal suffrage was to serve thepeople, constituted in Communes, as individual suffrage serves every other em-ployer in the search for the workmen and managers in his business. And it iswell-known that companies, like individuals, in matters of real business gener-ally know how to put the right man in the right place, and, if they once make amistake, to redress it promptly. On the other hand, nothing could be more for-eign to the spirit of the Commune than to supersede universal suffrage by hier-archic investiture (Marx, 1971: 73).6

It is hard to know how to classify this arrangement. Is it a government withpower but no authority or with authority but no power? It would “manage”

Philosophy and Politics: A Historical Approach to Marx and Dewey 217

Page 247: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

but it would be, like the police and the courts—which are not to be abolished—“elective, responsible, and revocable.” We need to emphasize here that the re-jection of bourgeois democracy did not involve the rejection of its patentlymost critical democratic features: free elections, free communication, etc.Marx never questioned their indispensability. His criticism, like Dewey’s,was that democracy required something more, not something less. For Marx,it meant a form of real participation consistent with government.

We should note also that Marx’s analogy is nearly perfect. A “company”could be “operated” by “workmen and managers,” and no one supposes thatthe owner of the company has lost his “sovereignty.” It was but elitist propa-ganda to believe that the same principles could not apply to a “commune.”The only real question was whether a commune, like a company, could agreeon the goals of the association. This problem, of course, relates to the firstpoint regarding the question of whether the citzenry is or is not alienated.

Finally, as already suggested, we must not assume that a full-fledged de-mocracy would be totalitarian, that “the people” ruling themselves wouldtrample “personal freedom.” As Aristotle and Madison both saw, this as-sumption was warranted where “personal freedom” meant “the rights of prop-erty,” freedom of the exploiters to exploit the exploited. Insofar as the Com-mune was “expansive” and not “repressive,” it would be different fromprevious democracies. It is true, of course, that Marx did mock liberal con-stitutional theory and did not himself pay heed to the institutional problemsof true democracy, but it is clear that he profoundly valued “personal free-dom,” saw, rightly, that “only in community do the means exist for every in-dividual to cultivate his talents in all directions,” and assumed, optimistically,that individuals in a community would act so that, in contrast to previousforms, all would realize “personal freedom.” The absence of concern to insti-tutional detail here stems from his commitment to a democratic politics, to hisrepeated contention that no one could write scripts for others—still less forfuture others who will need to solve just those problems which they have.7

It is also clear that if “the unity of the nation” was not to be broken, somesort of federation was involved, and while this was not to be a “federation ofsmall States, as dreamt of by Montesquieu and the Girondins,” it is not clearwhat Marx has in mind. There are, from Marx’s point of view, two possibleobjections to “small States.” First, if they are states, then they still embody analienated politics. Second, if they are small, “the unity of the nation is bro-ken.” This second problem is critical, but it is hard to say whether, as he sug-gests, “the nation” is the smallest unit for social production in the modernworld, or whether, perhaps as part of this, if we are to think of moving pro-gressively toward the future and be realistic, we need, in a world of aggran-dizing nation-states, to think in terms of nations?

218 Chapter Nine

Page 248: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

The problem is resolvable, however, if we return Marx’s discussion to its con-text. Marx’s conclusion, that had the Commune shown “a modicum of common-sense,” it would have tried to reach “a compromise with Versailles,” powerfullyreinforces a host of evidence that Marx’s politics were gradualist. This means, inthis context, that the best that one could have hoped for, in these circumstances,was the best possible compromise consistent with the continuing existence of theFrench State. To suppose that the Communards could have mapped out and re-alized the future is the worst kind of utopian thinking.


It may be doubted that Marx’s politics were gradualist. But a gradualist poli-tics is not necessarily “reformist” nor is it necessarily antirevolutionary. It isa politics that seeks to realize what is at the time realizable. Marx surelywanted society to be revolutionized and he surely also believed that, at theright moment, a revolution would occur; but there is strong evidence thatMarx was never naive about “the right moment” and that, in contrast to anarchist-inspired politics, he was always perfectly prepared to work withinthe state—if it was a liberal democratic state.8

Already against Stirner, he had argued, “it is only in the mind of the ideol-ogist that [the ‘will’ to abolish competition and with it the state and law]arises before conditions have developed far enough to make its productionpossible” (quoted by Thomas, 1980: 343). And to emphasize, the conditionsreferred to are political, regarding the political capacities of the people whoseactivities had sustained “competition” and “law and the state.” But perhapsthe clearest statement of his position is his speech, given upon resigning fromthe Central Council of the Communist League:

The minority have substituted the dogmatic spirit for the critical, the idealist in-terpretation of events for the materialist. Simple willpower, instead of the truerelations of things, has become the motive force of the revolution. While we sayto the working people, “You will have to go through fifteen, twenty-five yearsof civil wars, and wars between nations not only to change existing conditionsbut to change yourselves and make yourselves worthy of political power,” youon the other hand, say “We ought to get power at once, or else give up the fight.”. . . Just as the democrats make a fetish of the word ‘people’ you make one ofthe word “proletariat.” Like them, you substitute revolutionary phrases for rev-olutionary action (quoted by Thomas: 331).

We can be reminded that, writing in 1852, he held that “the carrying of universal suffrage in England would . . . be a far more socialist measure than

Philosophy and Politics: A Historical Approach to Marx and Dewey 219

Page 249: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

anything that has been honoured with that name on the continent.” He con-tinued, optimistically, that “its inevitable result . . . [would be] the politicalsupremacy of the working class” (345). And his mind had not changed, afterthe Commune, when in 1871, he wrote that:

The ultimate object of the political movement of the working class is . . . theconquest of political power for this class; and this naturally requires that the or-ganization of the working class, an organization which arises from its economicstruggles, should previously reach a certain level of development. On the otherhand, however, every movement in which the working class as a class confrontsthe ruling classes and tries to constrain them by pressure from without is a po-litical movement. For instance, the attempt by strikes, etc., in a particular fac-tory or even in a particular trade to compel individual capitalists to reduce theworking day, is a purely economic movement. On the other hand, the movementto force through an eight-hour, etc., law is a political movement. And in thisway, out of the separate economic movements of the workers grows up every-where a political movement, that is to say, a class movement, with the object ofenforcing its interests in a general form, in a form possessing general, sociallycoercive force. While these movements presuppose a certain degree of previousorganization, they are in turn equally a means of developing this organization(quoted by Thomas, 347).9


Events did not proceed, however, as Marx had thought they would. While, onthe one hand, industrialization in England and then in Germany did promotean increasingly politicized labor movement, radicals were by no means ableto agree on either strategy or tactics. During his lifetime, Marx had had somesuccess in negotiating the differences and in maintaining the extraordinarilydiverse elements of the International on a course of gradualist change. But theevents which led to the effective demise of the First International (in 1872,formally in 1876 in Philadelphia) were critical in the subsequent Internation-alist movement, in Marxist politics, and in our retrospective understanding ofMarx’s politics.

Oversimplifying a very complicated story, instead of being, as Thomasputs it, “a form of doctrine having some vague and, as far as Marx was con-cerned, irksome appeal,” anarchism became a movement “having a consider-able, and widespread, appeal across national boundaries” (Thomas: 249). Be-cause Marx wanted not merely that there should be a revolution, but that it bethe right sort of revolution, he fought the anarchists tooth and nail. He did notfight them, if the foregoing is correct, because he believed in the state, or in

220 Chapter Nine

Page 250: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

centralized authority, or worse, in a revolution made by revolutionaries in thename of the people. For Marx, the First International and the Paris Communedid in fact prefigure the future, yet the repression that the Paris Communeprovoked, itself a critical factor in the demise of the International, and Marx’srole in this collapse, had serious consequences. Instead of being a pluralist,ideologically heterogeneous vehicle for the transformation of the interna-tional order, subsequent Internationals were monolithic and ideologicallydoctrinaire. Indeed, Marx’s machinations versus Bakunin became the alibi forthis and at the same time confirmed as “prophetic” the anarchist suspicionthat Marx and Marxism were authoritarian, offering but a new version of ab-solutist politics. Finally, failing to grasp what the Commune was and why itfailed, instead of being for radicals a premature glimmer into a historicallynovel form of society, for some, because it lacked “centralization and author-ity,” the Commune became an apocalyptic and hopelessly degenerative fit ofrevolutionary madness. For others, it was the very model of revolution, “ofthe spontaneity of the masses.” For many anarchists, the Commune provedthat revolution did not require a “vanguard working class,” still less, “orga-nization”; or, quite oppositely, as in Bakunin and the Blanquists, it showed theneed for a conspiratorial revolutionary party to provide the match whichwould light the fire of revolution.

But events continued to undermine Marx’s hopes. The destruction of In-ternationalism, of social democracy in Germany10 and the collapse of TsaristRussia, all sparked by the World War, completed that redefinition of Marxistpolitics that had begun with the violent end of the Commune. After the Bol-shevik Revolution, Marxist struggles to transform society will rapidly col-lapse into two poles, between a Blanquist-style demand for a revolutionaryconspiratorial vanguard of the working class aimed at “smashing the state”and reformist social democracy, aimed not at a gradualist transformation ofthe political conditions for revolution, but at winning economic concessionsfrom the capitalists.11

By the turn of the century, it had already become clear to Marxists thatwhile “economic development” was creating the conditions for “socialist pro-duction,” it was not creating a revolutionary consciousness among the work-ers. Writing in his enormously influential—and often misunderstood—“WhatIs to be Done?” (1902), Lenin noted that “‘everyone agrees’ that it is neces-sary to develop the political consciousness of the working class.” “The ques-tion was,” he continued, “how is that to be done and what is required to doit?” For Lenin, “economism”—the effort to win economic concessions—andterrorism “have one common root, subservience to spontaniety.” For Lenin,both views presumed historical inevitability and were thus apolitical. Whileas with Marx and the social democrats, “the working class” remained the

Philosophy and Politics: A Historical Approach to Marx and Dewey 221

Page 251: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

center of a mass-based social revolution, Lenin insisted that “we must ‘goamong all classes of the population’ as theoreticians, as propagandists, as ag-itators and as organizers”; and while he protested against Blanquism andagainst “confining the political struggle to conspiracy,” he also insisted thatthis did not deny “the need for a strong revolutionary organization”—if theworkers were to be politicized (Lenin, 1966: 112–19). But it was easy to readthis tract in the light of the Bolshevik Revolution and to hold that it had al-ready set out the principles of socialist revolution, not as a movement ofworkers, nor of workers allied with other classes, but of an authoritarian organization of dedicated professional revolutionaries, a minority acting for“the workers.”12

It is exactly here, then, where we can begin our analysis of Dewey’s ver-sion of a democratic politics. As we shall see, there are both strong parallelsand some critical differences.


Dewey was born in 1859, the year that Marx published the Critique of Polit-ical Economy, but Dewey did not turn explicitly to political philosophy untilhis 1915 German Philosophy and Politics. Moreover, while it is well-knownthat Dewey was a critic of Marxism, especially in his writings beginning inthe 1930s, Dewey could not at that time have read Marx’s critique of Hegel’sphilosophy of state and the tracts on alienation, nor what we call the Grun-drisse, nor even the critically important German Ideology. For Dewey, Marx-ism was philosophically what may be called “Second International Marx-ism,” a variant that was powerfully influenced by the monist philosophy ofhistory of Georgi Plekhanov. Politically, Marxism was, by that time, definedlargely by Lenin—by then understood as the promoter of the “vanguardParty.” Dewey’s philosophical roots, of course, like Marx’s, trace to Hegel—a fact of some importance; but his political sensitivities were shaped not bythe revolutions of 1848 or the Paris Commune, but by New England localismand an understanding of this which in critical ways was Jeffersonian.

In German Philosophy and Politics, Dewey traced the philosophic basis ofpatriotic statism in Germany and concluded, “the present situation presentsthe spectacle of the breakdown of the whole philosophy of Nationalism, po-litical, racial and cultural.” Dewey rejected the sufficiency of “arbitration,treaties, international judicial councils, schemes of international disarma-ment, peace funds and peace movements.” He called for “more radical think-ing” of the problem (Dewey, 1915: 130). The problem of statist politics was

222 Chapter Nine

Page 252: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

also present in Democracy and Education, published the next year. He asked:“Is it possible for an educational system to be conducted by a national stateand yet the full social ends of the educative process not be restricted, con-strained, and corrupted?” (97). Dewey thought that the answer could be yes,if, to be sure, we were talking about education “in and for a democratic soci-ety.” In Reconstruction in Philosophy (1920), he argued that despite wide dif-ferences, political philosophies were “agreed upon the final consummatingposition of the state” and he concluded that assumptions regarding the“unique and supreme position of the State in the social hierarchy” had solid-ified into dogma (201). Just as Marx had supposed that capitalist modes of in-tercourse would destroy national boundaries, make for international proletar-ian solidarity, and politicize workers, Dewey hoped that these forces wouldpropel democracy as a way of life.13

But the critical eleven years that separated Democracy and Education andThe Public and Its Problems forced Dewey to different conclusions. Not onlyhad he developed a critique of democracy as “a form of government,” but alsoit was now clear to him that democracy, as a way of life was not being fos-tered by the new interdependencies and the new capacities of technologicalsociety. On the contrary, democracy as a mode of associated living was beingprofoundly undermined by these same forces.

In terms familiar to the anarchism of Gustave Landauer and Martin Buber,Dewey argued that democracy was not “an alternative to other principles ofassociated life. It is the idea of community itself.” “Within a community, thestate is an impertinence.”

Turning his attention to democracy as a form of government, Dewey ar-gued that at best, political democracy “represents an effort . . . to counteractforces that so largely determined the possession of rule by accidental and ir-relevant factors, and . . . an effort to counteract the tendency to employ polit-ical power to serve private instead of public ends.” But political democracyhad failed even to realize these limited goals. “In a word,” he concluded, “thenew forms of combined action due to the modern economic regime controlspresent policies, much as dynastic interests controlled those of two centuriesago. They affect thinking and desire more than did the interests which for-merly moved the state” (Dewey, 1954).14

The analysis compares easily to Marx, but especially insofar as it sug-gests that the constraints on policies, as on “thinking and desire,” are struc-turally rooted. Accordingly, the problem of constituting democracy as away of life had no easy solutions and regarded the incapacity of interde-pendent people even to perceive the consequences of “combined action,”still less to be able to perceive shared goods and to act on them: What,then, was to be done?

Philosophy and Politics: A Historical Approach to Marx and Dewey 223

Page 253: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


Dewey visited the Soviet Union in 1928, the year after he published The Publicand Its Problems. In no sense was Dewey an ideological anti-Marxist. He hadassessed the Bolshevik revolution as “an experiment to discover whether the fa-miliar democratic ideals—familiar in words at least—will not be most com-pletely realized in a social regime based on voluntary co-operation, on conjointworkers’ control and management of industry.” Dewey saw, it is clear, that so-cialism was inconceivable without democracy and that, as the foregoing sug-gests, democracy in its complete sense demanded socialism. Nevertheless, hisanalysis of the democratic state made it plain to him, though not to the Marxists,that in the United States, at least, proletarian revolution was not on the historicalagenda. For him, the critical issue was the very idea of class.

A great deal of what Dewey wrote during this period sounded like a Marxianclass analysis. In addition to what has already been noted, consider Individual-ism Old and New. In that critical text, Dewey argued that the issue that Marx hadraised, “the relation of the economic structure to political operations—is one thatactively persists.” “Indeed,” he continues, “it forms the only basis of present po-litical questions.” In the pages that follow, Dewey gives an account of the crisisthat could have come from Capital, Vol. III. He writes:

There are now, it is estimated, eight billions of surplus savings a year, and theamount is increasing. Where is this capital to find its outlet? Diversion into thestock market gives temporary relief, but the resulting inflation is a “cure” whichcreates a new disease. If it goes into the expansion of industrial plants, how longwill it be before they, too, “overproduce” (Dewey, 1962: 85–86).

There is in this text even a clear reference to the upshot of the Marxian labortheory of value:

That the total earnings of eight million wage workers should be only four timesthe amount of what the income-tax returns frankly call the “unearned” incomeof . . . eleven thousand millionaires goes almost without notice (109).

Perhaps even more Marxist sounding is his claim that “large and basic eco-nomic currents cannot be ignored for any length of time, and they are work-ing in one direction.” Indeed, “economic determinism is now a fact, not a the-ory.” His account concludes with a text that could have been written byEngels:

There is a difference and a choice between a blind, chaotic and unplanned de-terminism, issuing from business conducted for pecuniary profit—the anarchy

224 Chapter Nine

Page 254: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

of capitalist production—and the determination of socially planned and ordereddevelopment. It is the difference and the choice between a socialism that is pub-lic and one that is capitalistic (119).

Finally, Dewey also sees the relation between this theoretically informedanalysis and the problem of the “lost individual.” “We live,” he writes, “po-litically from hand to mouth.”

The various expressions of public control . . . have taken place sporadically andin response to the pressure of distressed groups so large that their voting powerdemanded attention. They have been improvised to meet special occasions.They have not been adopted as parts of any general social policy (114).

It is clear enough .why this is the case. Under present arrangements, “finan-cial and industrial power, corporately organized, can deflect economic conse-quences away from the advantage of the many to serve the privilege of thefew.” The political parties themselves, the ostensible vehicles of mobilizationfor change, “have been eager accomplices in maintaining confusion and un-reality” (114).

This analysis, as pertinent today as when it was offered, is not untypical ofDewey. It suggests that Dewey’s understanding of the political possibilities ofdemocratic politics in capitalist America was anything but naive, and that, inimportant ways, it was close to Marxism. But for all this, there is evidencethat he sometimes lost sight of the critical issue. An instance is the bookwhich some have taken to be one of Dewey’s more radical political tracts,Liberalism and Social Action, written in 1935. As is well known, Dewey thereinsisted that

Liberalism must become radical, meaning by “radical” perception of the neces-sity of thorough-going changes in the set-up of institutions and correspondingactivity to bring changes to pass (62).

Dewey emphatically rejected reform that dealt with “this abuse and now thatwithout having a social goal based on an inclusive plan,” but he was less clearwhat that goal and plan was. One thing was clear: Dewey rejected Marxism,but especially “the idea of a struggle between classes, culminating in openand violent warfare as being the method for production of radical socialchange” (78).

Dewey had a clear and adequate instrumentalist view of violence. In an-other place he had written that “what is justly objected to as violence or un-due coercion is a reliance upon wasteful or destructful means of accomplish-ing results” (Dewey, 1929, Vol. II: 785). In Liberalism and Social Action,

Philosophy and Politics: A Historical Approach to Marx and Dewey 225

Page 255: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

similarly, it was not violence as such that was the issue. He recognized, withthe Marxists, that “force, rather than intelligence is built into the proceduresof the existing social system,” and that even free expression will be tolerated“as long as it does not seem to menace in any way the status quo of society”(63, 64).27 When it does, he wrote, the state will be quick to use official vio-lence in the name of protecting “the general welfare.” Dewey had learnedfrom his experience with the Palmer raids and the tragedy of Sacco andVanzetti.

What then was his objection to the class war notion of the Marxists?Dewey might have argued, though he did not, that the Marxist analysis wassubstantially correct, but that for good historical reasons, the idea of strugglebetween classes, culminating in open warfare, had to be rejected. The argu-ment for this conclusion would be complicated, but it would be fully consis-tent with Dewey’s own analysis of the democratic state.

In agreement with the anarchists of his day, for Dewey, “the workers” werenot to be agents of social change. It was not that there was no oppression andinequality in America, that workers were not exploited, nor that they werehappy with their lot. One did not hear their “angry voices,” Dewey wrote, butthat was not because they were drowned out by “shouts of eagerness for ad-venturous opportunity.” Rather, “the murmurs of discontent are drowned” by“the murmurs of lost opportunities, along with the din of machinery, motorcars and speakeasies” (1962: 78–79). The metaphor, suggestive of the muchlater writings of Marcuse or Foucault, was employed in the context ofDewey’s brilliant analysis of America individualism. It was not roast beef, but“repressive needs,” “normalization,” and “atomization” which had disinte-grated class-consciousness.

As I argued, Marx knew that the struggle would be long and hard, but hecould not have anticipated the fantastic flexibility of capitalism in the liberaldemocracies, and especially in America, the fragmenting effects of race andethnicity. In America, then, “workers” had become a politically useless cate-gory. But if so, then, “class struggle” was, at its worse, a slogan for assuringthe faithful or, at its best, an abstraction at a different level of analysis.

The issue is complicated, but I must be brief. In Marxism, classes are notdefined by a set of empirically given characteristics, e.g., income, social sta-tus, or occupation. Rather, “class” is a theoretical concept, grounded in thecentral concept of mode of production. Marx’s Capital provides a theoreticaland abstract account of the capitalist mode of production. Abstractly consid-ered, there are but two classes, the owners of the means of production and the producers of surplus value, the proletariat. This analysis, the core of any Marxism, provides an understanding of capitalism as a mode of produc-tion. It shows what problems need to be solved if it is to be reproduced—

226 Chapter Nine

Page 256: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

including thus an explanation of why capitalist relations need to be mystified,the famous “fetishism of commodities.” It shows that capitalism is “irra-tional,” that it is subject to periodic crises, and it gives an argument for so-cialism by showing that as long as the means of production are not jointlycontrolled, there is no way to end domination and alienation—including thealienated politics of the modern state.

But of course the real world is not just a mode of production. It is com-prised of societies with modes of production. Like all others, societies with acapitalist mode of production have state structures, churches, gender andracial conflicts, schools, and mass media. They have housewives, “profes-sionals,” civil servants, and all sorts of “workers” who are not proletarians—defined as producers of surplus value. But this means that Marx’s projectioninto the future of the effects of the capitalist mode of production could wellbe wrong—as, indeed, it was.

In the nineteenth century, it was still possible to keep things simple byidentifying the growing class of industrial workers with a growing and in-creasingly organized proletariat, to suppose, as the Manifesto had it, that “thesmall tradespeople, shopkeepers, and retired tradesmen generally, the handi-craftsmen and peasants—all these sink gradually into the proletariat.” “Theworkers,” now “the immense majority,” now organized, now pauperized,would become a revolutionary class. But close as that had come to beingprophetic, just before World War I, that time had now passed. More generally,then, since Marxist revolutionary class politics was predicated on the as-sumption of global capitalist transformation, of increasing polarization andimmiseration, and on the consequent development of the political capacitiesof organized workers, and since these had not obtained as Marx had hoped,Marxists might well have abandoned the idea of proletarian revolution. Theproblem, then would not be to find or make a revolutionary class, but as Hin-dess writes, “to mobilize effective support around socialist objectives out ofthe forces, struggles and ideologies operative in particular societies.”15

The Marxists of the interwar period did not, of course, see this; nor, giventhat their Marxism was the monocausal Marxism of the Second International,was it surprising that Dewey would reject Marxist analysis.

He observed: “according to the Marxians . . . the economic foundations ofsociety consist of two things, the forces of production on one side and, on theother side, the social relations of production.” Further, for Marxians, scien-tific technology is part of the forces of production. It is dynamic while the so-cial relations are static; they “lag behind.” Dewey here was ready to admitthat “what was happening socially is the result of the combination of thesetwo factors,” and thus it would seem that here, as above, Dewey had fully ap-propriated the extremely influential Preface account of what came to be

Philosophy and Politics: A Historical Approach to Marx and Dewey 227

Page 257: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

called “historical materialism.”16 But Dewey then insisted that it was but “atruism” to call this combination “capitalism” and to say that “capitalism is the‘cause’ of all the important changes that have occurred.” On his view,

Colossal increase in productivity, the bringing together of men in cities and largefactories, the elimination of distance, the accumulation of capital, fixed and liquid—these things would have come about, at a certain stage, no matter whatthe established institutional system. They are the consequences of the newmeans of technological production (1963: 81).17

The text is a startlingly clear expression of technological determinism, but ifindeed, for “orthodox” Marxists, “technology” produces changes in the rela-tions of production and thus explains the emergence of capitalism, Deweysaw what they did not, that once one holds that technology directly definesthe labor process and, through this, the wider social relations, “historical ma-terialism” entails that there need be little real difference between “capitalism”and “socialism”!18

Dewey’s concrete approach should have put him on guard. While he oftensuccumbs to the high abstraction, “industrial society,” he seems also to haveseen that the logic and consequences of the accumulation of capital was a fun-damental cause of the way changes occurred in the West, of the particular ap-plication of technologies and the particular distribution of wealth and re-sources, that had capitalism been other than what it is—and here we areindebted to Marx—technological production could surely have been differ-ent. Putting the matter as briefly as possible, insofar as the relations of privateproperty define the accumulation of capital, the state is preferably liberal.This means not just that private and public are bifurcated but that governmentwill be predictably limited in addressing problems thrown up by the processof capital accumulation. At the very least, it must be constrained to activitiesconsistent with the maintenance of the system of private accumulation.

Dewey’s claim that “the release of productivity is the product of coopera-tively organized intelligence” is correct. As Marxists point out, production issocialized in capitalism. Moreover, if one wants the productivity associatedwith industrial societies, there is no alternative to that. But Dewey’s idea that“coercion and oppression on a large scale exist” because “of the perpetuationof old institutions and patterns not touched by scientific method” is patentlyfallacious. Indeed, in the text already quoted from Individualism Old and New,he had it right: “There is a difference and a choice between a blind, chaotic andunplanned determinism, issuing from business conducted for pecuniary profit,and the determination of a socially planned and ordered development,” be-tween “a socialism that is public and one that is capitalistic.” This difference,of course, is exactly the extension of political democracy to the economy, the

228 Chapter Nine

Page 258: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

elimination of the contradiction between socialized production and private ap-propriation, and finally, as Marx insisted and Dewey surely seemed also to see,the reappropriation of social powers by united individuals.

Once Dewey lost touch with the root of the problem, he could no longer of-fer plausible solutions. In Liberalism and Social Action, he offered:

The question is whether force or intelligence is to be the method upon which weconsistently rely and to whose promotion we devote our energies. Insistence thatthe use of force is inevitable limits the use of available intelligence. . . . There isan undoubted objective clash of interests between finance-capitalism that con-trols the means of production and whose profit is served by maintaining relativescarcity, and idle workers and hungry consumers. But what generates violentstrife is failure to bring the conflict into the light of intelligence, where the con-flicting interests can be adjudicated on behalf of the interests of the great ma-jority (1963: 79, 80).

The argument is a bad argument for at least three reasons. First, Dewey’s ab-solutist either/or, either force or intelligence, is unwarranted. No serious rev-olutionary, not Marx, not Lenin, not even Bakunin, so tied his hands in theway that Dewey suggests, even if, for them, violence was inevitable. Onewould have supposed that Dewey’s fine understanding of the use of violenceby the state in defense of the status quo would have led him to the conclusionthat as regards radical social change, some violence would, at some point, benecessary.

Second, whatever the difficulties of a Marxian analysis, Marxists were notso foolish as to suppose that the lions, the finance capitalists, would sit downwith the lambs and “adjudicate” away their privileged power. The “objectiveclash of interests” which Dewey rightly acknowledged was neither temporarynor negotiable. Rooted in the capitalist system as such, it left the partieslocked “in a death clutch.”19

Third, Dewey here presupposes that publics exist, for it is only then, as hehere implies, that “cooperative intelligence” can be a mode of social recon-struction. Immediately after he condemns Marxists for “a rigid logic,” hesays: “The ‘experimentalist’ is one who would see to it that the method de-pended upon by all in some degree in a democratic community can be fol-lowed through to completion” (80, my emphasis).

It will not be easy to explain Dewey’s continuing optimism that creative in-telligence can be effective even where it so patently lacks institutions. It is eas-ier to explain his decisive turn against Marxism. By 1928 at least, Dewey hadgiven up on the Socialist Party.20 By this time, the Soviets had already severelyabused, perhaps irreparably, the idea of “socialism.” They would, in the yearscoming, disillusion still more. As noted, Dewey’s notion of Marxism was

Philosophy and Politics: A Historical Approach to Marx and Dewey 229

Page 259: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

essentially the Marxism of the Second International, a Marxism flawed inmore ways than one. In his 1939 Freedom and Culture, Dewey attacked Marx-ism as unscientific on grounds that it had “a monistic block-universe theory ofsocial causation.” The monistic theory of history was a disaster: Social causa-tion was plural and reciprocal. As I have suggested in this essay, the actualcourse of history is the cluttered product of contingencies that no theory couldassimilate.21 On the other hand, Dewey—like the Marxists—had never him-self been clear on the causal questions regarding capitalism, industrial societyand the modern state. If anything, he shared with them a tendency toward tech-nological and economic determinism. Similarly, he was fully correct to chargethat no one was less scientific than the “scientific” Soviet Marxists: “Scientificmethod in operating with working hypotheses instead of with fixed and final Truth is not forced to have an Inner Council to declare just what is theTruth . . .” (97). Finally, Marxists had all too often argued that capitalism wasthe only evil and that therefore, once rid of it, all would be lovely. One wouldhave thought that the thirties proved otherwise.

Still, by 1939, Dewey had definitely shifted his emphasis—if not worse.After reminding his readers that he had “from time to time pointed out theharmful consequences the present regime of industry and finance had uponthe reality of democratic ends and methods,” and that he had “nothing to re-tract,” he went on to say that the Marxists were wrong in holding that “gov-ernment in the so-called democratic states is only the organ of a capitalistclass.” Now if this meant—as by then Marxists had supposed—that the statecould not be used by revolutionary socialists, then Dewey was surely correct.But as I have argued, Marx would have agreed with Dewey here. Yet Deweyseems to mean more than this:

the effect of constant criticism of governmental action; of more than one politi-cal party in formulating rival policies; of frequent elections; of the discussionand public education that attend majority rule, and above all the fact that polit-ical action is but one factor in the interplay of a number of cultural factors, havea value that critics of partial democracy have not realized (94).

Though admittedly “partial,” he was now prepared to defend what he took tobe a characteristic American “looseness of cohesion and indefiniteness in di-rection of action.”

We take for granted the action of a number of diverse factors in producing anysocial result. There are temporary waves of insistence upon this and that partic-ular measure and aim. But there is enough democracy so that in time any onetendency gets averaged up in interplay with other tendencies. An average pre-sents qualities that are open to easy criticism. But as compared with the fanati-

230 Chapter Nine

Page 260: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

cism generated by monistic ideas when they are put into operation, the averag-ing of tendencies a movement toward the mean, is an achievement of splendor(94–95).

But from the fact that what happens in history is the product of complex mul-tiple factors working in unpredictable ways, it did not follow that American“pluralist” politics generated some splendid “movement toward the mean,”that conflicting “interests” somehow get “averaged up,” that “the equilibriumin social affairs” was desirable. Indeed, how “pluralist” could a politics bewhen, as he had argued in 1927, the public was lost? Finally, it was true that“political action is but one factor,” but does not this mean, as in 1929 he alsohad seen, that in a capitalist society, this left a free hand for “financial and in-dustrial power, corporately organized”?

Dewey could accept “the criticism that much of our political democracy ismore formal than substantial, provided,” he now insisted, “it is placed in con-trast with totalitarian political control.” To be sure, one does not need a verygood society to compare well with Nazism and Stalinism. One might arguehere that the despair of politics, so characteristic of our day, had by then in-fected Dewey and that, like Sidney Hook and later pragmatists, he was nowprepared to celebrate bourgeois democracy. But this would be most unfair toDewey.

Not only was he unflinching in his rejection of the “new kind of Stoicism”which had gripped postwar Europe, but he was unflinching in his recognitionof the profoundly troubled situation and in his commitment to the idea thatthings could be made better—a great deal better. As Lothstein writes:

The central point for Dewey was that while suffering and setback suffused thetotal human endeavor, it was neither daemonic or unremitting. Rather sufferingand celebration . . . were experimental correlatives, happiness supervening upontheir conjoint and dialectical origination. Although nested in a radically contin-gent and indeterminate world, our situation, Dewey argued, was not that of Sisy-phus or Job, our fate sealed by divine fiat or historical obsolescence. He saw usinstead as freewheeling sculptors of meaning in a world bereft of ultimate guar-antees, but open to experimental improvement.22

Nor, despite his unflinching optimism, was it the case that Marx had ever-expected miracles. There are, he once remarked, “no ready-made utopias tointroduce par decret du people, and if so, then surely there were none to beinstituted by “a vanguard party” wreaking death in the name of the people. Ifwe are to be emancipated, we need “to work out our own emancipation” and“to pass through long struggles, through a series of historic processes, trans-forming circumstances and [people].”

Philosophy and Politics: A Historical Approach to Marx and Dewey 231

Page 261: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

This returns us to the question of “foundations.” Whatever his blind spots,Dewey’s was a politics that needed no foundations. While Marxists did finda foundation for politics in an eschatological philosophy of history, I have ar-gued that this is much less clear as regards Marx. But, however that may be,what is now needed, it seems to me, is to renew the possibilities of democraticpolitics which acknowledges the insights of Marx but yet strips Marxism ofthe idea that history is on the side of emancipation. We need, that is, to com-bine the best in Marx and Dewey.

The public, lost and eclipsed, has not been found. For those of us living ina democratic state, finding it is the primary imperative. But how to do this?Dewey’s answer might go as follows: Try, by taking advantage of any oppor-tunity that presents itself, to bring into existence publics; try to give direct ex-perience and educative quality by informing it; try to create from our atom-ized relations incipient communities which can be fostered and enlarged, andtry to do this by identifying common goods which can call for active supportand participation. Of course, this is not to say much, even if, as I think, it istrue and important. Still, armed with a Marxist understanding of what is hap-pening to us and why, it may be possible to take advantage of opportunitiesand to try, as Dewey offered, to build some incipient but progressively grow-ing democratic publics.


1. Cf. Norman Jacobson, Pride and Solace: The Functions and Limits of PoliticalTheory : (Berkeley, University of California Press, 1978). Jacobson’s book is perhapsthe most systematic effort to examine the implications of a foundation-less politics,but his moral is equivocal. See my review, “The Crisis of Contemporary PoliticalTheory,” Interpretation, Vol. 9 (September 1981). The texts quoted are from Arendt,as quoted by Jacobson, Chapter V, passim.

2. An antifoundationalist politics need not reflect despair. Rorty suggests a ver-sion when he writes that “we should be more willing than we are to celebrate bour-geois capitalist society as the best polity actualized so far, while regretting that itis irrelevant to most of the problems of most of the population of the planet,”“Method, Social Science, Social Hope,” in Consequences of Pragmatism (Min-neapolis: University of Minnesota Press, 1982: 210). But of course, ideologicalcertitude is an obvious feature of those American policies which, in the pursuit oftriumph in what can only be called a Holy War, are as limitless in their means asany which Orwell, Camus or Arendt condemned. Indeed, “bourgeois capitalist so-ciety” is not “irrelevant” to most of the problems of most of the population of theplanet exactly because it is a large part of the problem of these peoples—whetherthe societies are capitalist “miracles,” e.g., Korea, or “socialist” disasters, e.g.,Nicaragua.

232 Chapter Nine

Page 262: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

3. An excellent contextual reading of Marx’s politics is Alan Gilbert’s Marx’sPolitics: Communists and Citizens (New Brunswick, N.J.: Rutgers University Press).Gilbert, following a path marked by Michael Harrington (in his Socialism, New York:Bantam, 1973), shows that Marx persistently altered his political strategies in the lightof experience and that he was no economic determinist, inflexibly committed to patformulas—unlike most of his later epigones. Paul Thomas’s Karl Marx and the An-archists (London: Routledge and Kegan Paul, 1980) is an indispensable account ofMarx’s relations, ideologically and politically, to nineteenth-century anarchism. Seealso his “Alienated Politics,” in Terence Ball et al. (eds.), After Marx: Cambridge,Cambridge University Press, 1984). Martin Buber’s Paths in Utopia (Boston: Beacon,1949) remains valuable. Barry Hindess offers a crisp account of the critical debatesbetween Lenin, Kautsky, and Bernstein. Unfortunately, he does not discuss Rosa Lux-emburg, who was perhaps closest to Marx on the critical issues. See “Marxism andPolitical Democracy,” in Alan Hunt (ed.), Marxism and Democracy (London:Lawrence and Wishart, 1980). On Luxemburg, see Norman Geras, The Legacy ofRosa Luxemburg (London, NLB, 1976).

4. This is not to say that there are not difficulties and ambiguities in Marx’s writ-ings on the critical issues. An excellent treatment is Frederic L. Bender, “The Ambi-guities of Marx’s Concepts of ‘Proletarian Dictatorship’ and ‘Transition to Commu-nism’”, History of Political Thought, Vol. II (November 1981). See also, Harrington:54–60; Gilbert, Chapter VIII.

5. Karl Marx, “The Civil War in France,” in On the Paris Commune (Moscow:Progress, 1971): 97. As Bender points out, Engels confirmed that for them, the Com-mune was a new type of polity. In an 1875 letter to Bebel, Engels wrote: “The wholetalk about the state should be dropped [from our party’s statements] especially sincethe Commune . . . was no longer a state in the proper sense of the word [because itwas a state in-the-process of dissolving]. . . . We would therefore propose to replacestate everywhere by Gemeinwesen, a good old German word which can very wellconvey the meaning of the French word ‘commune’” (cited by Bender: 549).

6. For an extended development of these ideas in the American Confederation pe-riod, see my War and Democracy (Oxford: Basil Blackwell, 1989), Part V, and “TheForeclosure of Democracy in America,” History of Political Thought, Vol. 9 (Spring1988).

7. Of course, insofar as he ignored the very real dangers of usurpation of powerand violation of individual rights, Marx was, as Bender notes, to this extent “respon-sible” for later vanguard interpretations.

8. Harrington, Thomas, and Gilbert each provides ample evidence on this criticalpoint.

9. See M. Levin, “Marx and Working-Class Consciousness,” History of PoliticalThought, Vol. I (Autumn, 1980).

10. Cf. Carl E. Schorske, German Social Democracy, 1905–1917: The Develop-ment of the Great Schism (Cambridge, Mass., Harvard University Press), 1955 andmy War and Democracy, Chapters 11 and 12.

11. See F. Claudin, “Democracy and Dictatorship in Lenin and Kautsky,” New LeftReview, Vol. 107 (Nov./Dec. 1977).

Philosophy and Politics: A Historical Approach to Marx and Dewey 233

Page 263: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

12. This is hardly the place to survey the literature on Lenin and Leninism. Myviews are influenced by Roy Medvedev, Leninism and Western Socialism (London,NLB, 1981); Stephen F. Cohen, Bukharin and the Bolshevik Revolution (New York:Knopf, 1974), Moshe Lewin, Lenin’s Last Struggle (New York: Vintage, 1970). Seealso my War and Democracy, Chapter 11.

13. Compare, of course, Marx and Engels: “National differences and antagonismsbetween peoples are daily more and more vanishing, owing to the developments ofthe bourgeoisie, to freedom of commerce, to the world-market, to uniformity in themode of production and in the conditions of life corresponding thereto” (CommunistManifesto, in McClellan: 235).

14. The most extensive treatment of Dewey as a “vestibular” anarchist, “in theAmerican grain,” is Arthur Lothstein’s excellent “From Privacy to Praxis: The Casefor John Dewey as a Radical American Philosopher,’” PhD Dissertation, NYU, 1979.See also his “Salving from the Dross: John Dewey’s AnarchoCommunalism,” ThePhilosophical Forum, Vol. 10 (Fall 1978).

15. Hindess argues that the series of critical debates between Kautsky, Lenin, andBernstein, from 1891 to World War I, are “variations on a single theme,” viz., whereto locate the boundaries “for non-economic, non-class determinants of political lifeand stop it from getting out of hand” (37). Thus, while none of these writers was sim-ply class-reductionist and while even Bernstein does not break completely with theconception that the economy is ultimately determining, they differ enormously onwhat and how much of what is political is not determined by the economy. But onHindess’s view, the debate between them is irresolvable because “there is no one gen-eral mechanism of connection between politics and the economy that is characteristicof capitalism as such—or for that matter, of particular historical phases of its devel-opment” (41). In other words, as in Marx’s own political practice, political questionsmust always be posed concretely, considering the particular details of the particularsociety under consideration. A “revisionist” politics becomes plausible, then, at thepoint where, in the liberal-democratic state, socialism is no longer primarily a classissue.

16. The phrase “historical materialism” is not used by Marx at all. Engels first em-ployed the term “materialist conception of history” in a review of Marx, Contributionto the Critique of Political Economy that has the famous Preface that became the au-thority for Second International versions of “historical materialism.” For a recent de-fense of this view, see G. A. Cohen, Karl Marx’s Theory of History: A Defense (NewYork, Oxford University Press) and the critique by Andrew Levine and Eric OlinWright, “Rationality and the Class Struggle,” New Left Review, Vol. 123 (Sept./Oct.1980).

17. Liberalism and Social Action: 81. In Individualism Old and New, Dewey hadchastized Marx for reasoning “too much from psychological economic premises” anddepending “too little upon technological causes” (102).

18. See Phillip Corrigan, Harvie Ramsay, and Derek Sayer, Socialist Constructionand Marxist Theory: Bolshevism and Its Critique (New York: Monthly Review,1978); Marc Rakovski, Towards an East European Marxism (London: Allison andBusby, 1978).

234 Chapter Nine

Page 264: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

19. The expression is C. Wright Mills’s. Mills made a similar critique of Dewey inhis Sociology and Pragmatism (New York: Oxford, 1966).

20. See Dewey, “The Need for a New Party,” “Who Might Make a New Party?”and “Politics for a New Party,” New Republic, Vol. 66 (1931); “The Future of Radi-cal Political Action,” Nation, Vol. 136 (1933); “The Imperative Need for a New Rad-ical Party.” Commonsense, Vol. II (1933). For a general account of the Socialist Partyand its relation to the Dewey-led League for Independent Political Action, see FrankA. Warren, An Alternative Vision: The Socialist Party in the 1930s (Bloomington: In-diana University Press, 1974), esp. Chapter V.

21. See my A Realist Philosophy of the Social Sciences (Cambridge: CambridgeUniversity Press, 2006).

22. Arthur Lothstein, “From Privacy to Praxis,” p. 80. Lothstein points out that thecriticism of “a new kind of Stoicism,” was made a month after Dewey’s eighty-eighthbirthday, in 1947. Dewey argued that on this view, “existence reduces pretty well towhat the individual can make out of it on his own hook,” and added, “I think they arereactions of people who are scared and have not the guts to face life” (ibid., p. 60f.,quoting from a letter to William Daniels, “Letters of John Dewey to Robert V.Daniels, 1946–50,” Journal of the History of Ideas, Vol. XX (October–December1959).

Philosophy and Politics: A Historical Approach to Marx and Dewey 235

Page 265: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 266: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


The problem of justice continues to be a topic of lively debate. The continu-ing struggle over civil rights, our anguish over the war in Iraq and the so-called “war” on terror, our continuing unease over the economy, immigration,and the environment, skepticism regarding the system of criminal justice, anddoubts about our educational institutions identify the main historical forcesand problems which underlie the current discussion. The symptoms and is-sues cover a very wide range: attacks on affirmative action, death penaltylegislation and attacks on the rights of the accused, profoundly exacerbatedby the war on terror, and, propelled by free market ideology, tax revolt andcutbacks in education and social services.

At the level of social theory, John Rawls’s A Theory of Justice, publishedin 1971, generated a veritable industry. His difficult and highly praised bookseems to have arrived at exactly the right time, perhaps because it offered apowerful statement and defense of what was probably mainstream liberalthinking on justice. Rawls’s theory is individualistic, but his recognition of“the least advantaged,” coupled with his attention to what he called, “pureprocedural justice,” seemed to confirm our most basic intuitions about justice.To be sure, there were immediate criticisms of some of the more vulnerablearguments, but these received a greater response from the right than from theleft. Indeed, Robert Nozick’s critical treatment in his 1974 Anarchy, State andUtopia gave him a cover of the popular and liberal New York Times Magazine.Not only was the presence there of a philosopher significant and unusual; buthe was there heralded as the most articulate of the new spokesmen of conser-vative, individualistic thinking on justice. Indeed, as regards the current stateof opinion, Nozick, seems to have conquered Rawls’s version of “New Deal”Liberalism.

Chapter Ten

John Dewey and the Problem of Justice

Page 267: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

In this context, it may be useful to consider the writings of John Dewey,America’s foremost social philosopher. And in this context, if one surveys thevoluminous writings of Dewey, writings which span over seven decades un-til his death in 1952, the first thing that one notices is the relative inattentionpaid by Dewey to the problem of justice. Altogether, there are perhaps notmore than a dozen pages of sustained discussions devoted explicitly to thetopic. These discussions are insightful and important, little gems, and in whatfollows, it will be a pleasure to appeal to them. But the first task, really themain task of this presentation, is to try to explain this apparent imbalance inDewey’s attention.

Dewey’s lack of discussion of the theory and problems of justice is not tobe explained by a failure to see problems, nor by an unwillingness to dealwith them at the theoretical or practical level. Indeed, the two most sub-stantial discussions of justice were written during two periods of acute crisisin our economy, during the 1890s and again, in 1932. Similarly, Dewey’s ac-tive involvement in a host of public matters of social and political nature aretoo well known to recite here. Several of these, e.g., the trial of Sacco andVanzetti, raised serious questions of justice.1 There are, I believe, two sets ofreasons that do explain this imbalance. They are important and give us insightinto both our problems and Dewey’s unique strengths as a social philosopher.One set of reasons specifically regards this stance. As a social philoso-pher, Dewey was a writer who aimed not to write a social philosophy—a doctrine—but who aimed rather to show how we must try to seek solution ofour social problems. The other set regards the very idea of justice. It seems tome that Dewey, for various reasons to be detailed, sought to displace justiceas the central concept of social philosophy. However, for these same reasons,he found himself using the term less and less, until ultimately he abandonedit altogether. These two sets of reasons are definitely related, but it may nev-ertheless be desirable to treat them more or less separately, taking the idea ofjustice first.


It is possible to show that there are two dominating conceptions of justice inWestern civilization. The first had its home and only full articulation in the an-cient Greek polis.2 Plato and Aristotle, of course, develop it with sophisticationand vigor. The other concept of justice was also first formulated in the ancientGreek polis, in Periclean Athens. It is associated with the names of Democritusand Protagoras. But this idea of justice did not come into its own until the mod-ern period, with Hobbes and Locke, Kant, Hume, and nineteenth-century liberal

238 Chapter Ten

Page 268: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

philosophy. We should call it the liberal concept of justice. It remains the domi-nating concept in the West, and Rawls and his more conservative critics, despitedifferences, stand in this tradition.

The liberal concept of justice favors atomistic metaphors and voluntary re-lations, for example, the contract; it is conventionalistic, arguing that justiceand political society are “artifacts” deliberately and rationally constructed; itis legalistic, emphasizing formal and procedural justice; it employs marketnotions of distributive justice, presupposes scarcity and finally, it is “harshand hard.” As Hume put it, justice is the “mean virtue.”

By contrast, the Platonic—Aristotelian notion of justice favors organicmetaphors and conceive of human relations and political society as “natural”;it presupposes natural inequalities, emphasizes morality instead of law, andthinks of justice very widely. As Aristotle said, it is “the whole of virtue.”

While this is not the place to develop this radical contrast, an illustration ofeach concept may be helpful if only to fix our ideas.

Plato’s Republic aimed to answer the question: What is justice? For him, itwill be remembered, justice is a condition of the soul, a “psychic harmony,”a prerequisite of just acts and, crucially, of well-being and happiness (eudai-monia) itself. Parallel with this analysis, justice is a condition of the polis thatis itself thought of as an organic unity. Each element, “class,” or person, hasa task (ergon) which, when performed well, contributes to the well-being ofthe totality. The four virtues, Wisdom, Courage, Temperance, and Justice areeach defined as functions of the social and psychic unities of society and self.

At the very opposite pole is Thomas Hobbes. Where men are in “the natu-ral condition,” for Hobbes, there is neither justice nor injustice. The consen-sually introduced mechanism of impersonal law that constitutes political so-ciety also constitutes the very possibility of justice. Once done, natural equityand justice is replaced by one principle, “performance of the convenant.” Thewhole apparatus of customary rights and privileges is similarly reduced: “Toeach according to the agreements he has made.”

To be sure, Hobbes’s theory was too extreme, too tough-minded. And nodoubt, subsequent versions of “contract’“ theory from Locke to Rawls re-sponded with corrections and additions. Nevertheless, there is a clearly iden-tifiable tradition that must be sharply separated from the earlier one.

Where then does Dewey stand as regards these two concepts and traditionsof justice? In the last analysis, Dewey could not accept either, even if he didpull strongly toward the Greek.

In his Syllabus of 1884, Dewey noted “in many respects, the discussionsof virtue in Plato and Aristotle are still unequaled.” Indeed, following them,if in his own novel fashion, he argued that that “courage, temperance andwisdom denote simply phases of every moral act” and that “the name is

John Dewey and the Problem of Justice 239

Page 269: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

given according to the phase which, in a given case, happens to be domi-nant.” “Justice, then,” argues Dewey,

is the name for the deed in its entirety. . . . It is not another virtue, it is the sys-tem of virtue, the organized doing: whose organic members are wisdom, the willto know; courage, the impulse to reach; control, the acquired power to do. (EW,Vol. 4: 363, 357).

As this text shows, Dewey is very, very Greek in holding that justice is thevirtue of a unity “organically” related, even if at the same time, Dewey re-jected the “faculty” psychology which is generally imputed to Greek moralphilosophy. He is Greek, too, in couching virtue in terms of self-developmentand self-realization, even if for him in contrast to the Greeks, the underlyingnotion of human nature is open, dynamic, and changing; not closed, static,and fixed.3 Dewey and the Greeks agreed that persons were doers, exerting,developing, and enjoying human powers and capacities and that the concernfor “realization” of these powers should be at the center of a moral and socialphilosophy.4 But if so, then “justice” could not be reduced to “obedience tolaw” or to “just desert.” His text continues:

the current distinction between justice as penal, and justice as concrete recogni-tion of positive merit by the share awarded an agent . . . is far too rigid. . . . Un-consciously there is smuggled in the assumption that worth is static; that what aman has done is somehow complete in itself, and serves to indicate his merit,and therefore, the way he should be treated. Service is taken as some thing ren-dered, not as a function . . . (EW, Vol. 4: 359).

The idea that “worth” is static and that “deserts” and “entitlements” are likecommodities, exchangeable as equivalents for “things” exactly characterizesthe market conception of justice, the dominant mode of modern thinking onthe subject. Dewey struggles in this text, as in others, to drive home the lim-iting and incomplete nature of this framework for justice. He writes:

When it was said that the ordinary concept of desert concealed a momentous as-sumption, it was meant that the whole dualism of justice and love is involved.If justice be conceived as mere return to an individual of what he has done, ifhis deed, in other words, be separated from his vital, developing self, and if,therefore, the “equivalent return” ignore the profound and persistent presence ofself-hood in the deed, then it is true that justice is narrow in its sphere, harsh inform, requiring to be supplemented by another virtue of larger outlook and freerplay—Grace. But if justice be the returning to a man of the equivalent of hisdeed, and if, in truth, the sole thing which equates the deed is self, then quiteotherwise. Love is justice brought to self-consciousness; justice with a full, in-

240 Chapter Ten

Page 270: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

stead of a partial standard of value; justice with a dynamic, instead of a static,scale of equivalency (361).

Our “ordinary” sense of justice is narrow, is harsh. Recognizing, though am-bivalently and sometimes incoherently, that persons are dynamic selves relat-ing humanly to one another and to the world, we think of justice as requiringsupplementation, by mercy, by kindness, by love. But surely that misleads aswell. It is not either justice or love. It is not justice or charity—from caritas,“love.” This accepts the dualism and allows us to paste over injustice withgratuities. But how to overcome this dualism? Dewey had it right: Once wereject the idea that “deeds” are “things” to be exchanged for equivalents, weundermine the dualism, for then, it is possible to link the deed with the self.But Dewey remains contaminated by the market theory of justice. Signifi-cantly, he puts the matter conditionally: if justice be the returning to a personof the equivalent of his deed. And he lets us muse as to whether he wouldhave preferred to deny the hypothesis altogether. But if we are to think of jus-tice in these terms, and as moderns, it is hard to see how to do otherwise, thenfor Dewey, at least at this point, let us try not to disconnect the deed from thevital, changing self.

Dewey will return to these themes just twice more, in the Ethics written col-laboratively with James Tufts, published originally in 1908 and then in substan-tial revision in 1932. As in the earlier Syllabus, Dewey focuses on the notion thatjustice is “hard” and “harsh” and he develops another dimension of this attitude.Here, he argues that it comes from identifying justice “with the working of somefixed and abstract law . . . as if man was made for law, not law for man” (MW,Vol. 5: 373). Although pursuing this idea systematically would take us directlyinto Dewey’s problem-centered and inquiry-oriented style of social philosophy,we should pause here if only briefly to emphasize the pervasiveness of the no-tion, as it bears on the problem of justice.

Dewey clearly saw that alongside the market conceptualization of justicewas another that derived ultimately from Kant. It put heavy emphasis on dutyand obligation and its most austere and rigorous form is captured by the Latin,Fiat justitia, ruat coelum, “let justice be done, let the heavens fall.” Deweytook this phrase as the title for a brief, popular essay written for the New Re-public in 1971. Rejecting the legalism and formalism which so typically char-acterizes “moral” discussions of justice, whether of war and international re-lations, as in this case, or of race or sex, Dewey identified such ethics as“resolutely irrelevant to the circumstances of action and the conditions oflife” (1929: 592). In another and earlier essay, entitled “Nature and Reason inLaw” (1914), Dewey pregnantly characterized “the chief working differencebetween moral philosophies in their application to law.” It was, he argued,

John Dewey and the Problem of Justice 241

Page 271: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

that “some of them seek for an antecedent principle by which to decide; whileothers recommend the consideration of the specific consequences that flowfrom treating a specific situation this way or that, using the antecedent mate-rials and rules as guides of intellectual analysis but not as norms of decision”(795). This text, we might note here, could well be the point of departure foran extended analysis of the current debate over affirmative action programsas, of course, it compresses an entire potential legal philosophy.

For Dewey, this methodological inversion explained in part the limitingand narrowing conception of justice. And in the 1908 discussion, Deweyagain calls, optimistically, for “the transformation of the conception of justiceso that it joins hands with love and sympathy” (MW, Vol. 5: 373–74).

But one can hardly be heartened by these remarks. The problem seems in-escapable. The liberal notion of justice was liberating insofar as it made menindiscriminately subject to impersonal law and insofar as it broke the basis of“privilege” based on hereditary status, but Dewey saw early on that the lib-eral notion was far too narrow, too rigid. So he struggled to enlarge it, to rem-edy its partiality, to supplement it. And if we grant that love and sympathy arethe requisite supplementations, one may legitimately wonder how good amerely just society or merely just person would be? For the Greek, this couldnot be a question. With their notion of justice, the just man and the just soci-ety had to be good. And, of course, if love and sympathy are the requisite sup-plementations to our ordinary sense of justice, then one may legitimatelywonder how we are to proceed. Indeed, the deeper Dewey looked into theproblems and issues of liberal society, the more disjointed became the effortto transform and widen liberal justice.


Dewey was never sanguine regarding the mechanisms of distribution in lib-eral society. Still less was he mystified by the rhetoric of the current theories.This may be brought out by considering Tufts’s contribution to the collabora-tive 1908 edition of their Ethics. Dewey must surely have endorsed the perti-nent pages. Indeed, in the revision of 1932, it seems that Dewey himselfwrote the crucial Chapter 21 that restated the issues and reaffirmed their ear-lier stance. We may look first at the earlier version, noting well the early dateof the text.

Characteristically, the locus of the critique is “individualism.” They beginby arguing that if we take a purely “formal” view and make “formal freedomof contract the only criterion, then any price is fair which both parties agreeto” (MW, Vol. 5: 475). This position, characteristic of Hobbes, and of classi-

242 Chapter Ten

Page 272: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

cal and neoclassical political economy, is substantially the position argued forby Nozick in the recent work cited earlier. Although the argument cannot bedeveloped here, Nozick’s “entitlement” theory, while very much enriched bydetail and the sophistication of modern decision theory, is, I would argue,subject quite precisely to Tufts’s criticism. It is this: If the exchange is to befair, the parties to the bargain must be equal. “But in a large part of the ex-changes of business and services, the two parties are not equal.”5 In otherterms, where some must accept the conditions of the contract, “formal free-dom of contract” is not a sufficient condition. In his 1932 statement of thistheory, Dewey characterizes it even more economically. Its motto is to each“what he can get through his ability, his shrewdness, his advantageous eco-nomic position due to inherited wealth and every other factor that adds to hisbargaining power . . .” Dewey rightly notes that “this is the existing methodunder capitalism” (Dewey and Tufts, 1912: 454), as today, owners of baseballteams, school boards, and the AMA tend to forget.

The “take-advantage-of-your-bargaining-power” theory of justice has an-other, less rude, version. It is characterized by Dewey by the motto: “To eachwhat he earns.” This theory, plausible enough on its face, due perhaps to themultitude of difficulties concealed in the notion of “earn,” does not, arguesDewey, characterize capitalist distribution. But it must be rejected in anycase, since it cannot be realized. It cannot be realized because production issocial. The point is important but often misunderstood.

In “producing”—toasters, services, skills, or knowledge—individuals em-ploy knowledge, skills, and instruments that are the legacy of previous gen-erations of workers. Moreover, and characteristically, “products”—includingknowledge—are jointly produced in the more obvious sense that they areproducts of many hands and minds. Suppose, then, we take the Gross Na-tional Product to represent the combined social product—an entirely artificialmeasure for the “product” we need to measure, but useful perhaps to fix ourideas. The earning theory of distribution, then (like its sophisticated relative,modern productivity theory), presupposes that it is possible to divide up theGNP and assign to each individual exactly what is hers or his. No part of thisis to be shared on grounds that we can’t disentangle our contribution; no partis a residue earned by past labor and no part is earned by anyone. The prob-lem here is not simply whether this division is fair, whether each receives ajust share, but whether, indeed, any coherent sense can be given to the ideathat respective contributions to the social product can be so disentangled. ForDewey, rightly, it was obvious that they could not.6

But this is not the end of the difficulties for the individualistic theories, foras Dewey and Tufts write, they suffer from a serious moral failure. Achieve-ment and failure, what one does “contribute,” or “earn” is a function of three

John Dewey and the Problem of Justice 243

Page 273: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

things: heredity, social advantage and the socially produced conditions ofknowledge and environment and, finally, individual effort. It is not a matterof individual effort alone. It is at this juncture that Rawls’s influential theorydeparts from traditional individualisms, for with Tufts and Dewey, Rawlsagrees that accidents of birth—the good fortune to be born rich and hand-some—are not in themselves morally relevant. And indeed, if so, then one canask, as does Dewey and Tufts, “If all men are accounted equal in the State,why not in wealth?” (Dewey and Tufts, 1932: 359).

It is perhaps here that the contrast in the orientation of Dewey and of Rawlsis most graphic and where, at the same time, Dewey reveals both his greateststrength as a social philosopher and perhaps, as well, his greatest weakness.

Consider Rawls’s response first. For Rawls, the family is the key problemand, short of restructuring it, natural talents and social advantages will in-evitably be rewarded. I think that it can easily be shown that Rawls gives uptoo quickly here. Indeed, as we suggest, Dewey and Tufts have a more en-couraging response. But Rawls’s originality begins at exactly this point, forhis famous “difference principle” is meant precisely to justify inequalitiesthat, however they come about, had best not be removed. His argument isquite straightforward. An egalitarian distribution would be inefficient but anefficient system need not be just. It would be just, however, if social and eco-nomic inequalities were arranged so that there was “fair equality of opportu-nity” and, crucially, so that the “least advantaged” were better off than theyotherwise would have been. If Rawls is right, something looking very muchlike our system is, in his terms, “nearly just.” To be sure, we have some wayto go in achieving “fair equality of opportunity”—notice that this still rewardsnatural ability and that, for Rawls, the family remains (and will remain) a cru-cial limitation even on this. Nevertheless, it is fair to say that, for Rawls, weare doing perhaps as well as can be expected.

Tufts and Dewey take an entirely different tack. After evaluating the extanttheories, they offer, instead of their own theory, what they call “a workingprogram.” The gist of it is contained in a short paragraph:

A man’s power is due (1) to physical heredity; (2) to social heredity . . .; and (3)to his own efforts. Individualism may properly claim this third factor. It is just to treat men unequally so far as their efforts are unequal. It is socially desirableto give as much incentive as possible to the full development of everyone’s pow-ers. But this very same reason demands that in the first two respects we treatmen as equally as possible (490).

This working program is radical since, ultimately, it means that no unequalbenefits should accrue to persons exclusively on the basis of their natural tal-ents or on the basis of socially derived advantages. But it is a “working pro-

244 Chapter Ten

Page 274: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

gram” in the sense that it leaves entirely open the means by which the idealis to be achieved. It does not demand radical revolution in order to achieve aradical restructuring of society, even if the realization of the ideal would in-volve a radical restructuring of society. And it does not insist on any particu-lar ameliorative reforms, even if there are steps that could and should betaken. And it is in this that Dewey’s greatest strength and weakness may berevealed. For it is not at all easy to judge whether or not Dewey saw how rad-ical the ideal was or how radical would the changes have to be to bring theideal into existence. In both the 1908 and the 1932 discussions, educationcharacteristically is emphasized as means. But “conditions of food, labor, andhousing” and “the importance of private property” are also identified. In thelater discussion, Dewey responded passionately to the notion that because percapita income had increased greatly—shades of Rawls!—it was “foolish” toraise the question of distribution. Indeed, in direct contrast to Rawls, he ar-gued that wealth, not income is the crucial variable: “The individuals or cor-porations that have great wealth undertake great enterprises. They control forbetter or worse the wages and living conditions of great numbers” (454).7

These same sorts of criticisms are found in many of Dewey’s writings anddemonstrate that he was keenly aware of the bearing of the system of privateproperty not only on the problem of justice, but on the problem of freedomand democracy as well.8 Yet, many commentators have found grounds for ar-guing that Dewey was naive in having unwarranted hopes for the efficacy ofeducation even as an ameliorative factor. This is probably so.9

The backlash on affirmative action and ERA, the decisive, if inevitablefailures of poverty programs and efforts to guarantee equal education for all,would indeed have been disheartening to Dewey. And as disheartening, per-haps, is the renewed enthusiasm for what are really quite worn out individu-alistic theories of justice. His own theory could be stated in a sentence, firstwritten in 1891: “What is due the self is that it be treated as self” (Dewey andTufts, 1932: 35). In the last analysis, Dewey preferred working programs overtheories. Indeed, this takes us to the final part of our account.


It is another of the commonplaces of commentators on John Dewey’s thoughtthat he was preoccupied with method, indeed sometimes to the extent that con-tent altogether seemed to dissolve.10 This is not the place to examine all the dif-ficult questions which attend this criticism, but as regards our particular prob-lem, the problem of justice, I think that it must be said that Dewey’s socialphilosophy does represent a departure from traditional social philosophies and

John Dewey and the Problem of Justice 245

Page 275: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

that this shift is perhaps best construed as an orientation which displaced theproblem of justice as a substantive theoretical problem and replaced it with anorientation which emphasized problems and ideas that connected more directlyto method and to practice.

This alternative point of departure in Dewey’s thought may be best ex-pressed in Reconstruction in Philosophy (1920) although many texts confirmthe idea. Identifying three alternative social philosophies, the individualistic,the socialist, and the organic, Dewey argued that all “suffer from a commondefect.”

They are all committed to the logic of general notions under which specific sit-uations are to be brought. . . . They are general answers supposed to have uni-versal meaning that covers and dominates all particulars. Hence, they do not as-sist inquiry. They close it. They are not instrumentalities to be employed andtested in clarifying concrete social difficulties. . . . The social philosopher,dwelling in the region of his concepts, solves problems by showing relationshipsof ideas, instead of by helping men solve problems in the concrete . . .”(Dewey,1948: 188, 192).

This is the touchstone idea of Dewey’s emphasis on “inquiry,” “experi-mentalism,” and “instrumentalism.” Within Dewey’s frame, current debate onjustice would suffer from the same defects of abstraction, from the same ir-relevance to the actual conditions of education, of work, and of association,from the same aristocratic detachment that seems presupposed by the ideathat “philosophers” can solve human problems.

Dewey’s critics on the left are also correct, however, in judging that hisexperimentalist and method-centered attitude left him vulnerable to two alternative—and at bottom inconsistent—sorts of readings.

On the one hand, Dewey’s efforts to shift the focus of social philosophyaway from “doctrine” led some to see Dewey as advocating an engineeringand scientistic conception of social philosophy and inquiry. This view, in-spired by Dewey’s repeated assertion that social questions could be treated“scientifically” and “experimentally,” meant for those readers something likethe sort of practice which presumably goes in “laboratories” manned by per-sons in white coats and constrained by canons of “efficiency” and “positive”control. On this view—technocratic and still fashionable—a new breed of“social scientists” would provide that “expert” knowledge that would speed-ily solve concrete social problems. (See the account of Lippman, Chapter 7.)

In the last analysis, this reading cannot be sustained, even if Dewey didgive ample room for misconstrual. Perhaps his willingness and openness toincorporate and encourage ideas that seemed congenial to him further con-fused matters. One might mention here his long association with A. F. Bent-

246 Chapter Ten

Page 276: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

ley and his attraction in the late 1920s to the “operationist” views of P. W.Bridgman.

On the other hand, it led others to see Dewey as committed to a kind of un-principled reformism, to a defense of patchwork suggestions as responses tothe outcropping of crisis. As argued in Chapter 9 above, Dewey was reformistin his attitude, rejecting consistently the idea that societies could be intelli-gently transformed by radical and revolutionary programs. His approach was“piecemeal,” a call to deal amelioratively with concrete and particular prob-lems. Thus, for him, Marxism was “doctrinaire” in offering “sweeping gen-eralizations” and “general solutions” to “general problems.” Dewey wassurely sensitive to the problem of the “unintended consequences” of radicalchange and to the ease with which progressive movements become appropri-ated and distorted. But it must be said as well that while his ideals were rad-ical, as we already noted, his appreciation of obstacles preventing their real-ization was probably naive. Nevertheless, his reformism was hardlyunprincipled and his shift in focus was both sound and important. Indeed, Ibelieve that there is much to be learned from him on this score.

For Dewey, the problem of justice, as the problems of freedom and de-mocracy, cannot be “solved” by “experts” or by philosophers. They couldonly be solved—if that is still the right word—by people in the everydayworld in their doings and sufferings. Dewey seems to have grasped this andthat is why, in the last analysis, the “content” of his social philosophy seemsso thin and, finally, so painfully obvious. There are, it seems to me, but twoitems in it: First, there is the idea that “the level of action fixed by embodiedintelligence is always the important thing” (Dewey, 1954: 166), and second,the idea that “democracy is a way of life, individual and social.” These cru-cially related ideas defined the limits of social philosophy. Movement in thedirection of their realization was movement toward an ideal in the only senseof ideal that Dewey allowed—“the tendency and movement of some thingwhich exists carried to his final limit” (148). As with justice, they identifieda “working program” and, crucially, a program that could be implementedonly by people in their individual and collective doings and sufferings.

This did not mean, for Dewey, that philosophy had nothing to do. Indeed,there was a great deal to be said about both ideals and about their mode of re-alization. The whole of Dewey’s extensive writings on methods of inquiryand on education, both in and out of the school, issue in the idea of “actionfixed by embodied intelligence.” As Dewey saw it, the application of “intel-ligence” to social problems meant not the application of new techniques by“experts,” however defined, nor did it reduce to the application of an-tecedently derived principles to concrete particular situations. Rather, thecanons had to be generated in inquiry and realized in practice. And this kind

John Dewey and the Problem of Justice 247

Page 277: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

of social knowledge “does not yet exist” (166). “The only possible solution,”he wrote, to the problems generated by interdependence require “the perfect-ing of the means and ways of communication of meanings so that genuineshared interest in the consequences of interdependent activities may informdesire and effort and thereby direct action” (155). And as this text suggests,this was both condition and consequence of democracy as a way of life, theother guiding ideal of Dewey’s social philosophy. Accordingly, the whole ofhis writings on democracy, community, freedom, and culture bear on this sec-ond theme. Keeping this in mind allows us, finally, to grasp the full force ofthis wonderful text: “Philosophy recovers itself when it ceases to be a devicefor dealing with the problems of philosophers and becomes a method, culti-vated by philosophers, for dealing with the problem of men” (McDermott,1973: 95).

If it is philosophers who have the task of articulating and “cultivating”these methods, it is people themselves who must employ them. Dewey wasinsufficiently radical regarding the difficulties standing in the way of trans-forming “The Great Society” into “The Great Community,” but he never fellvictim to pat solutions. He saw that we never begin anew, from scratch, fromnothing. We either1 sustain the inherited forms or we transform them, pur-posefully and intelligently, whimsically and stupidly, coercively or coopera-tively. Dewey put his faith in the possibility that action could be conjoint, pur-poseful, and intelligent. He put the matter crisply in his 1919/20 lectures inChina. Responding to the question, “Where should we start in reforming oursociety?” Dewey answered:

. . . we must start by reforming the component institutions of the society. Fami-lies, schools, local governments, the central government—all these must be re-formed, but they must be reformed by the people who constitute them, workingas individuals—in collaboration with other individuals, each accepting his ownresponsibility. . . . Social progress is neither an accident nor a miracle; it is thesum of efforts made by individuals whose actions are guided by intelligence(Dewey, 1973: 62–63).

But he also saw, as John J. McDermott has written, that if “the responsibilityis ours and ours alone,” the transformation of the processes and forms of liv-ing is, at the same time, “laced with chance” (McDermott, 1973: xxi).


1. See George Dykhuizen, The Life and Mind of John Dewey (Carbondale, IL:Southern Illinois Press, 1973): 234.

248 Chapter Ten

Page 278: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

2. See my “Two Concepts of Justice,” Journal of Chinese Philosophy, Vol. 4(1977): 99–121.

3. Cf. here, of course, Human Nature and Conduct (New York: Henry Holt,1922).

4. C. B. Macpherson has argued that J. S. Mill tried to incorporate this funda-mental feature of Greek Idealism into his liberalism, though the result was less thansatisfactory. See his Democratic Theory (Oxford: Clarendon Press, 1973).

5. Dewey gives the same argument against “classical individualism and free en-terprise” in his lectures in China. See John Dewey, Lectures in China, 1919–1920, ed-ited and translated from the Chinese by Robert W. Clopton and Tsuin-Chen Ou (Hon-olulu: University Press of Hawaii, 1973): 113.

6. It may be noted here that modern price theory employs the fiction that marginalproducts are so divisible. This makes for quite a respectable mathematical theory, use-ful as a praxiology. But it doesn’t follow that a theory of justice that assumes the fic-tion is intelligible. In this regard, Nozick’s criticism of Rawls is interesting. Rawlssees, if not clearly, that “social cooperation” does make a difference. By assuming thatmarginal products can be “disentangled,” Nozick argues that Rawls’s account of the“problem” created by social cooperation is mistaken. Nozick does show, however,that Rawls’s individualism does not square with his view of “social cooperation.”

7. Rawls obliterates the distinction between income and wealth by inattention. Hepersistently refers to “income and wealth” but never addresses the difference. Ac-cordingly, for him it would seem to be unimportant.

8. See, for example, The Public and Its Problems (Chicago: Swallow Press,1954): 101–2, 107; “Philosophies of Freedom,” in R. J. Bernstein, ed., On Experi-ence, Nature and Freedom (Indianapolis, Ind.: Bobbs-Merrill, 1960): 271; Liberalismand Social Action, excerpted in John J. McDermott, ed., The Philosophy of JohnDewey (New York: G. P. Putnam’s Sons, 1973), two volumes, 648.

9. For example, R. J. Bernstein, Praxis and Action (Philadelphia: University ofPennsylvania Press, 1971): 228; C. W. Mills, Sociology and Pragmatism (New York:Oxford University Press, 1964): 333.

10. See, for example, Charles Frankel, “Dewey’s Social Philosophy,” in New Stud-ies in the Philosophy of John Dewey, S. M. Cahn ed., (Hanover, N.H: University Pressof New England, 1977).

John Dewey and the Problem of Justice 249

Page 279: Dewey Rescuing Dewey Essays in Pragmatic Naturalism
Page 280: Dewey Rescuing Dewey Essays in Pragmatic Naturalism


One has to be impressed by the fact that America’s premier University con-tinues to provide us with people who write “brilliant,” “wise,” “elegant,” “impressive,” and “refreshing” books which it needs so desperately. As democratic theorist Jane Mansbridge put it, Harvard’s Michael Sandel’s De-mocracy’s Discontent: American in Search of a Public Philosophy (1996) “isbound to change the course of American historiography, political philosophyand legal scholarship.” George Will, not generally thought of as a theorist ofdemocracy, was equally enthusiastic. In his words, “Sandel’s book is a think-ing person’s guide to the current rethinking of the role of government inAmerica.”

Like Robert Nozick’s Anarchy, State and Utopia (1974) that made him adarling to the minimal government-free marketers who triumphed with Rea-gan and Company, Sandel’s book is aimed at John Rawls’s influential A The-ory of Justice (1971). It seems well on its way to make him a darling of thenew “republican-communitarians.” I can only hope that some evangelical de-fender of “soulcraft” does not come to capture the imagination of our deeplytroubled society.

But this gets well ahead of what I want to say. Sandel’s argument is subtleand deceptive, even if, for me at least, it is a combination of bad philosophy,bad sociology, bad history, and bad politics. I begin with the philosophy.


There is a currently fashionable dichotomy between what George Kateb called“rights-based liberalism” and “American republican-communitarianism.” It is

Chapter Eleven

Liberalism’s Discontent: America inSearch of a Past That Never Was

Page 281: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

clear enough that Rawls and Nozick (along with Flathman, Dworkin, Feinberg,Gewirth, Sen, and many others) are, despite differences, “rights-based liberals.”The other side is a much less clear bunch and might include any number of di-verse writers who have criticized liberal philosophy and promoted some versionor other of “community,” including John Dewey, Hannah Arendt, Robert PaulWolff, Charles Taylor, Roberto Mangiabera Unger, Michael Walzer, CarolGould, Hannah Pitkin, Amitai Etzioni, and some others besides. The relation todemocracy of these writers is also very diverse.

Presumably, one of the notable achievements of Sandel have clarified thisargument, to show us both the limits of “rights-based liberalism” and thatthere is a viable alternative well within the American tradition. His main dis-tinction, accordingly, is between what he calls “the procedural republic”—theversion of liberalism “by which we live” (he offers that the label was sug-gested by Judith Sklar) and “republican” theory. As regards the procedural re-public, then:

Its central idea is that government should be neutral toward the moral and reli-gious views its citizens espouse. Since people disagree about the best way tolive, governments should not affirm in law any particular vision of the good life.Instead, it should provide a framework of rights that respects persons as free andindependent selves, capable of choosing their own values and ends. Since thisliberalism asserts the priority of fair procedures over particular ends, the publiclife it informs may be called the procedural republic (Sandel, 1996: 4).

This is, of course, a fair statement of both Rawls and Nozick. It is less clearthat it abstracts correctly the prevailing public philosophy, but I pass on thathere. In sharp contrast to republican theory, as understood by Sandel, weshould emphasize that neither Rawls nor Nozick had much to say about de-mocracy. Rawls assumed some form of “representative regime” and (withMill) even defended plural voting. While democracy is not even indexed inRawls’s book, Nozick surely goes further. After acknowledging that democ-racy is the idea that “people have a right to a say in the decisions that impor-tantly affect their lives,” Nozick asserts, remarkably: “After we exclude fromconsideration the decisions which others have a right to make and the actionswhich would aggress against me, steal from me, and so on, . . . it is not clearthat there are any decisions remaining about which even to raise the question”(Nozick, 1974: 270).

Both Rawls and Nozick do capture certain strands in the prevailing publicphilosophy, but the differences between them are critical—and much of whatmight be in this philosophy is not captured in the least. Thus, for example,both positions are certainly too extreme for most Americans. Rawls is tooegalitarian; Nozick is too libertarian. Both Rawls and Nozick should be com-

252 Chapter Eleven

Page 282: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

plemented for their refreshing reluctance to mostly ignore the critical idea ofdemocracy. As regards most Americans, it is true to say, I think, that democ-racy is safely understood to be defined by “free” elections (period). Since nei-ther Rawls nor Nozick reject this idea, their views are safely consistent withthe prevailing public philosophy.

Sandel acknowledges that the prevailing conception of liberal political the-ory has its roots in Locke, Kant and Mill, but at the same time, he asserts thatit “is a recent arrival, a development of the last forty or fifty years.” Indeed,it replaced a rival public philosophy, republican theory. This is characterizedas follows:

Central to republican theory is the idea that liberty depends on sharing in self-government. This idea is not inconsistent with liberal freedom. Participating inpolitics can be one among the ways people choose to pursue their ends. Ac-cording to republican political theory, however, sharing in self-rule involvessomething more. It means deliberating with fellow citizens about the commongood and helping to shape the destiny of the political community. But to delib-erate well about the common good requires more than the capacity to chooseone’s ends and to respect others’ rights to do the same. It requires knowledge ofpublic affairs and also a sense of belonging a concern for the whole, a moralbond with the community whose fate is at stake. To share in self-rule thereforerequires that citizens possess, or come to acquire certain qualities of character,or civic virtue (1996: 5–6).

Sandel’s emphasis on “self-rule” makes his version of republican theory looklike, of course, a version of democracy, but surprisingly absent in this char-acterization is any critical sense of what self-rule might mean or what wouldcount as “deliberating” about the common good, or “helping to shape” thedestiny of the political community. His concern, manifestly, is “certain qual-ities of character” essential to self-rule.

Indeed, there is nothing in the book that attends to the currently profoundstructural limits on self-rule. Perhaps the best that he does on this score is toendorse Tocqueville’s potentially beautiful trivialization of democracy: Localattachments enable citizens “to practice the art of government in the smallsphere within [their] reach” (314). Their reach may, of course, be pitiablysmall. “Ideally at least, the reach extends as the sphere extends” (314). Thisextension is also efficiently discussed: Presumably, “civic capacities firstwakened in neighborhoods and town halls, churches and synagogues, tradeunions and social movements find broader expression” (314). One need nottake a radical stance as regards self-rule to wonder about this suggestion. Thatis, suppose that all decisions of major social importance are made by either private corporations or governments dominated by two parties who

Liberalism’s Discontent: America in Search of a Past That Never Was 253

Page 283: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

fundamentally share a public philosophy? There would be no institutionalmeans to make even this “broader expression” felt. As long ago as 1925, Wal-ter Lippmann pointed out that “in spite of all that has been said about twee-dledum and tweedledee,” modern democracy “was purely and simply a mat-ter of choosing whether to “support the Ins when things are going well, or “tosupport the Outs when they seem to be going badly.” (See Chapter 7.)

Moreover, as regards “knowledge of public affairs,” Lippmann saw also—before television—that public opinion was manufactured, that, to shift themetaphor, for the ordinary citizen, the problem was “to secure maps on whichtheir own need, or someone else’s need, has not sketched in the coast of Bo-hemia.” For Lippmann, even if they were so disposed, ordinary people (in-cluding representative bodies which he called “a group of blind men”) can-not get the information they need: “what is happening, why it is happening,what ought to happen.” We are, as C. W. Mills was later to argue, a mass dom-inated by mass media. In turn (as everybody knows, but cannot do anythingabout), these are dominated by money, both because they need to sell theirproducts (and ads) and because they are owned by a handful of corporations.This is, of course, but part of this story.

But if Sandel is not interested in looking at how citizens might effectivelyparticipate in governance, he is deeply interested in looking at the qualities ofcharacter, the civic virtue, which is required if citizens are to be self-ruled. In-deed, his fundamental (almost exclusive) concern (as with Plato and, more re-cently George Will), is that, contrary to liberals, government ought not to be“neutral” (even if it could) and even more, that governments have legitimateconcerns with “soulcraft,” what he elsewhere calls “the formative project.”His clearest statement of the content of “civic virtue” comes from GeorgeWill. Sandell writes:

Unlike Falwell, who sought America’s salvation in a rebirth of Christian moral-ity, Will sought to cultivate civic virtue, the “dispositions, habits and mores” onwhich free government depends. By virtue he meant “good citizenship, whoseprinciple components are moderation, social sympathy and willingness to sacri-fice private desires for public ends” (1996: 310).

Sandel, like Will, Etzioni and many other “communitarians” would seem tohave a theory of society in which problems can be solved by changing the“morals” of persons. Perhaps it is assumed that until such time that people ac-quire that civic virtue which is the prerequisite of self-rule, we need not worryabout either how people are to get the information they need, how to decon-struct maps with the coastline of Bohemia drawn on them, or how to begin toalter structures so that people can have the power to make decisions that im-portantly affect their lives. I return to this in my last section.

254 Chapter Eleven

Page 284: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

For Sandel, the “formative project” is both inescapable and carries risk(321). Indeed, “the task of forging a common citizenship among a vast anddisparate people invites more strenuous forms of soulcraft” (319). One has tobe impressed with the ease by which he deflects this “risk.”

We may agree with him that “the civic strand of freedom is not necessar-ily exclusive or coercive,” that “it can sometimes find democratic, pluralistexpression” (321, my emphasis). But first, as liberals have long insisted, thekey virtue in this regard is toleration, reveling in difference, a willingness asequals to engage disagreement and conflict. Here, as with the question of en-larging the sphere of self-rule, he seems guided by a faith similar to many ofthose who take for granted the structures of global capitalism and defend thecurrently fashionable idea of “civil society.” Sandel writes: “Instead of col-lapsing the space between persons, it fills the space with public institutionsthat gather people together in various capacities, that both separate and relatethem” (320). He offers no suggestions on how this might be possible. Norsince he takes for granted both the modern state and global capitalism, it ishardly clear what good this would do.

His notion that the state should not be morally neutral as regards “morals”and “religion” and that it has a responsibility to cultivate civic virtue leadshim, inevitably, to some strikingly conservative conclusions. It is in theseconcrete cases where we best get the flavor of the high abstraction, civicvirtue. He writes, e.g.:

What makes a religious belief worthy of respect is not its mode of acquisition—whether by choice, revelation, persuasion, or habituation—but its place in agood life or, from a political point of view, its tendency to promote the habitsand dispositions that make good citizens (1996: 66).

What are these “habits and dispositions”? Uncritical obedience? Undyingcommitment and loyalty? And what if believers have a prior obligation toGod or to one of his emissaries? Suppose that such an obligation does not pro-mote “moderation, social sympathy and a willingness to sacrifice private de-sires for public ends,” then what? The liberal can insist that such beliefs mustbe tolerated, as liberals see, a difficult enough task in itself (321). While thereis no problem in justifying laws aimed at preventing harm to others, onSandel’s principles, is it within the legitimate province of the state to suppressbeliefs which do not “promote moderation, social sympathy and a willingnessto suppress private desires for public ends.” This includes, as we shall see, theenforcement of morals as such.

Sandel notes that in Paris Adult Theater I v. Slaton, Chief Justice Burger“wrote as if embarrassed to acknowledge the moral objection to obscenity assuch,” a reluctance that presumably, undercut the coherence of his argument.

Liberalism’s Discontent: America in Search of a Past That Never Was 255

Page 285: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

Thus, “allowing the states to decide that commerce in obscenity may ‘injurethe community as a whole’ begs the question whether communal injury canconsist in an offense against shared moral standards” (1996: 77). Burgeropted for his form of argument perhaps because he had read H. L. A Hart’sdevastation of Lord Devlin. If the state can legislate for the sake of “moral-ity” as such, then it can proscribe acts merely because they are “sinful” or“wrong.” In the light of human history, what could make anyone think thatthis is morally defensible?

Perhaps, however, if his principles are not liberal, his intuitions are. Thisseems so in a number of instances. For example, he seems to think that sex-ual relations between consenting adults, heterosexual or not, should not beproscribed. But he is uncomfortable with a straightforward liberal defense ofthis, substantially that “people should be free to choose their intimate rela-tions themselves, regardless of the virtue or popularity of the practices theychoose, so long as they do not harm others” (104). He prefers, instead, the de-cision in Griswold where the court affirmed certain values and ends. It thenarticulated “the virtues that homosexual intimacy may share with heterosex-ual intimacy, as well as any distinctive virtues of its own:” While homosexu-als have no right to intimacy, “family values” justify it.

Indeed, in the same vein, Sandel is most unhappy with no-fault divorce.“By making dependence a dangerous thing, it burdens the practice of mar-riage as a community in the constitutive sense” (115). If so, why not compellife long marriage under all conditions? In any case, blaming no fault divorcefor the grim statistics he quotes is just plain bad sociology. The problem is notthat the law affirmed the “encumbered self,” but the consequence of profoundstructural problems in American society, coupled with a familiar inattentionto the rights of divorced women. If we want to support marriage, we ought,at the very least, ensure that people have jobs that pay living wages and pro-vide families and single mothers with child support, daycare, etc. Nor evencan we say that idea of no-fault was mistaken. There is nothing in that ideathat says the former partner should not be held responsible. If alimony is notawarded and child support is not enforced, then we may suspect that patri-archy is at work in our “liberal” courts and justice system.

One should not suppose that Sandel lacks arguments against liberals whoare reluctant to embrace “the formative project.” But they are frightfully leanand implausible. What matters to the liberals, he writes (many times!), is notthe ends we choose, but our capacity to choose them. It is this that is most es-sential to our personhood. What is wrong with this? Sandel offers that:

the philosophical difficulty lies in the liberal conception of citizens as freelychoosing, independent selves, unencumbered by moral or civic ties antecedent

256 Chapter Eleven

Page 286: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

to choice. This vision cannot account for a wide range of moral and political ob-ligations that we commonly recognize, such as obligations of loyalty or solidar-ity. By insisting that we are bound only by ends and roles we choose for our-selves, it denies that we can ever be claimed by ends we have not chosen—endsgiven by nature or God, for example, or by our identities as members of fami-lies, peoples, cultures, or traditions (1996: 322).

This describes the familiar moral philosophy of “individualism.” Still onewonders about both the pertinent sociology which is being assumed andwhether “liberals” are necessarily “individualist” in this sense?

Sandel continually speaks of the liberal assumption that the self is “unen-cumbered.” A self is unencumbered, presumably, if first, the person has noobligations which are not voluntarily incurred and second, and more interest-ing, persons have identities which are independent of roles, including being“members of [a] family or city or nation or people, as bearers of that history,as citizens of this republic” (1996: 14). But if so, it is not clear that it is pos-sible for a self to be “unencumbered” and if so, surely liberalism must fail.

Consider obligations that presumably cannot be accommodated. Sandelknows that liberals can and generally do acknowledge nonvoluntary “naturalobligations,” or obligations owed to persons just because they are persons.Thus, one has no right to harm others. Similarly, liberals acknowledge specialobligations of (say) a parent to their children, just because they are voluntar-ily incurred. Are there then other obligations, to the “nation” or to other “cit-izens” of the republic, and what is their status? For Sandel, “the liberal at-tempt to construe all obligation in terms of duties universally owed orobligations voluntarily incurred makes it difficult to account for civic obliga-tions and other moral and political ties that we commonly recognize” (14). Itis not clear what the difficulty is. Of course, liberals do deny that we have anatural obligation to the political community in which we happen to live. Asis well known, the problem of legitimacy (of “political right”) was “solved”in the modern period by liberals who insisted that only if citizens “consent”are they obligated. So for this form of liberal theory, the obligation to the stateis not “natural” but voluntary. Most of us who have had problems with thisposition have no problems with the idea of voluntarily incurred obligations,but with the idea citizens consent. Since as Hume had insisted, if merely liv-ing in state (“tacit consent”) is consent, then, trivially, everybody consents.

But this is not Sandel’s problem. Rather, he seems to think that because our“identities” are tied up not only with our families and roles, but also with “ournation,” obligations generated by these have presumptive moral force. It iseasy for liberals to make sense of the claim that duties of parenthood have pre-sumptive moral force. What of our “identities as Americans, or Christians—ormales or “white men”? So it is much less clear that we are to have sympathy

Liberalism’s Discontent: America in Search of a Past That Never Was 257

Page 287: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

for Robert E. Lee who opposed secession, but concluded that his obligation toVirginia was not merely of sentimental import, but had moral force (15). Pre-sumably, as in Lee’s case, this obligation could override any other obligationshe might have, including any obligation he might have to the Republic or toresist the profoundly immoral institution of slavery. This is a dangerous doc-trine: One might, not implausibly, compare here the “obligations” of an Eich-mann or a Lt. Calley, a devout Christian, a jihadist, or member of the KKK?

Moreover as regards selves, his view is stunningly “essentialist.” Onewould have thought that he would know that nations and the identities of per-sons are not “natural kinds,” but are socially constructed from biographicallyand historically available materials. This means, critically, that they arechanged by both intended and intended actions sometimes in morally pro-gressive ways, sometimes not. The institutions of slavery might well have ledRobert E. Lee to abandoned his identity as a loyal Virginian and to identifyhimself as a loyal American. The recent construction of Serbian identity hadled, of course, to ethnic cleansing, an incredibly vicious form of “the forma-tive project.”

As regards “unencumbered selves,” one suspects that Sandel is unduly un-der the influence of Rawls. One thus might hold that the persons in Rawls’sconstructed “original position” are not “encumbered selves.” But of coursethe point of that construction was precisely to deny that one’s family, ethnicgroup, gender, or roles in society were relevant to defining the principles ofjustice. Rawls’s social philosophy is cosmopolitan. Indeed, it was just herethat Rawls was at his most emancipating. Here was a liberal who showed thatthe most “liberal” society in the world could not defend the fact that it wasthe most unequal society in the world. Moreover, as Charles Beitz shows,Rawls’s theory is easily and plausibly extended to address global injustice.1

As far as I can tell, inequality is not a concern of Sandel.The idea, of course, of an “encumbered self” is implausible, but with the

exception of Kant (and possibly of Robert Nozick), who has held to it? Kantbadly confounded matters in bifurcating the phenomenal and the noumenal.Presumably our phenomenal, flesh and blood, concrete, historically locatedselves are “encumbered. Our “noumenal self,” by contrast, is “unencum-bered.” Our “autonomy,” accordingly, depended upon our capacities as “ra-tional beings” to give law to ourselves. Sandel does not take on Kant. Norshall I (even if I think that his position is a disaster). What he does instead isto speak of Kantian liberals (generally unnamed) and then to allow us to be-lieve that all liberals are committed to a Kantian ontology. One may wish herethat he had been more careful. Indeed, one of the huge difficulties in the bookis a systematic ambiguity over whether Sandel is describing the Weltan-schauung of the times and places in his book, or whether he is engaging nor-

258 Chapter Eleven

Page 288: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

mative theory. Worse, he lacks utterly a conception of ideology: the idea thatcertain philosophical theories may be articulating what are, or are to become,widely held but false beliefs, beliefs which, in fact, are in the interest of thepowerful. This ambiguity immensely contributes to the usefulness of thisbook as ideology. I return to this.

As noted, rights theorists all acknowledge nonvoluntary obligations, butsince they generally take a consequentialist position, they also insist that anyobligation may be overidden. Moreover, directly against Sandel, they can af-firm that identities are tied up with relations and roles and still insist that our personhood requires that we choose our ends—not God, not nature, notthe government. Personhood requires agency, but agency is not, as per Kant,autonomy—self-legislating—but the capacity to choose such that given anychoice, one could have done otherwise.2

Because we are always encumbered, our choices, including our choice of“ends,” are always both enabled and constrained both by our biographies andthe social situation we find ourselves in. Liberals have not, it is true, gener-ally noticed the stunning inequalities in what is enabled, nor that these in-equalities are a straightforward consequence of socially constructed race,class, and gender. Instead, they have tended to ignore enabling conditions andto think that the only constraint is legal. Of course, this is major weakness ofmost liberalism (partly addressed by Rawls). But this hardly calls for an aban-donment of the idea that choosing one’s ends is a fundamental value. Indeed,one must fear freedom—as in Plato, Durkheim, or Freud— to think otherwise(Manicas, 1974).

Similarly, just as he confounds “autonomy” and “agency,” Sandel trades onthe idea that the liberal is committed to what he calls the voluntarist concep-tion of freedom. He writes, for example: “The voluntarist conception of free-dom that inspires this liberalism holds out a liberating vision, a promise ofagency that could be realized even under conditions of concentrated power”(Sandel, 1996: 278).

Again, agency is not the issue. Even if the choices are all grim, the capac-ity to choose is the mark of agency. But Sandel is right that “voluntarist” or“contractual” freedom is a central piece of liberal ideology. Nozick is thesurely best case. For him a choice is free if and only if it is voluntary. A choiceis involuntary if and only if it is coerced, and as above, coercion is the threator use of violence, including legal coercion exercised by the state. Coercivesocial relations just don’t count. So, neither do the immense inequalities offreedom that result from one’s position in prevailing social relations. ForNozick, if people have equal rights, then they are equally free (period).

Insofar as he recognizes “positive” rights, or rights that oblige others (in-cluding especially the state) to do something to realize one’s rights, Rawls

Liberalism’s Discontent: America in Search of a Past That Never Was 259

Page 289: Dewey Rescuing Dewey Essays in Pragmatic Naturalism

would seem to see that freedom is very unequally distributed. Thus, rights demanded by Fair Equality of Opportunity, for example, the right to a goodeducation, or those demanded by the Difference Principle, e.g., the right to in-come greater than the worst off in any other system of distribution, implynonlegal constraints on freedom. But still trapped in liberal ideology, he alsoasserts:

The inability to take advantage of one’s rights and opportunities as a result ofpoverty or ignorance, and a lack of means generally, is sometimes countedamong the constraints definitive of liberty. I shall not, however, say this, butrather I shall think of these things as affecting the worth of liberty, the value ofthe rights that the first principle allows” (1971: 204).

But not all liberals are so whimsical about freedom. Perhaps the best (intu-itively sensible, philosophically sound) definition of freedom was offered byJoel Feinberg (1973). On this view, freedom is a capacity to do something,have something or be someone. But capacities are defined by enabling factorssuch as competencies and resources, and persons are constrained by hin-drances and obstacles that prevent them from doing, having, or being. Onecan choose to sleep under the bridge, but cannot choose to sleep at the Ritz-Carlton if one lacks the required money. The idea of absolute freedom is bothincoherent and undesirable. But there are people with practically no freedomin exactly the sense that ignorance and poverty have disenabled them. Theleast of their problems is the coercive power of the state.

As, for example, Dewey well documented (Chapter 10), there are manyproblems with variant versions of liberal political philosophy; but the ideathat there is a value in choos