Siblings Under the Skin : Feminism, Social Justice and Analytic Philosophy 9781935790112, 9781888570694

Citation preview

Introduction

i

Siblings Under the Skin: Feminism, Social Justice and Analytic Philosophy Edited by Sharyn Clough

A volume in the series Critical Studies in the Humanities General Editor, Victor E. Taylor

The Davies Group, Publishers

Aurora, Colorado

ii

Siblings Under the Skin

Siblings Under the Skin. Copyright © 2003, Sharyn Clough. All rights reserved. No part of the contents of this book may be reproduced, stored in an information retrieval system, or transcribed, in any form or by any means — electronic, mechanical, photocopying, recording, or otherwise — without the express written permission of the publisher, and the holder of copyright. Submit all inquiries and requests to The Davies Group, Publishers PO Box 440140 Aurora, Colorado, 80044-0140 USA.

Library of Congress Cataloging-in-Publication Data Siblings under the skin : feminism, social justice, and analytic philosophy / edited by Sharyn Clough. p. cm. -- (Critical studies in the humanities) Includes bibliographical references. ISBN 1-888570-69-5 (pbk. : alk. paper) 1. Feminist theory. 2. Logical positivism. 3. Analysis (Philosophy) I. Clough, Sharyn, 1965- II. Series. HQ1190 .S455 2002 305.42'01--dc21 2002152848

Cover photo “Children Playing, c. 1910” appears courtesy of the Museum of Garden History, London, England.

Printed in the United States of America Published 2003. The Davies Group Publishers, Aurora CO 80044-0140 1234567890

Introduction

iii

Contents

Acknowledgments

iv

Introduction Siblings under the skin, Sharyn Clough

3

Historical Context Politics and the Wissenschaftliche Weltanschauung: Neurath and Reichenbach on science and socialism, Steve Gimbel 13 Conversations A question of evidence, Lynn Hankinson Nelson Epistemology naturalized, W.V. Quine 65

41

A hasty retreat from evidence: The recalcitrance of relativism in feminist epistemology, Sharyn Clough 85 A coherence theory of truth, Donald Davidson 117 Charity and ideology: The field linguist as social critic, Bjørn Ramberg 141 A nice derangement of epitaphs, Donald Davidson

161

Lost in logical space? Wilfrid Sellars and feminist empiricism, Edrie Sobstyl 181 More on givenness and explanatory coherence, Wilfrid Sellars

211

Where values interact with science, Hugh Lacey 229 On the sources of knowledge and of ignorance, Karl Popper

265

Hempelian insights for feminism, Heather Douglas 283 Science and human values, Carl Hempel 307

iv

Siblings Under the Skin

For Phil Tumminia who gets things done

Introduction

v

Acknowledgements

A number of these essays have appeared previously and I am grateful for permission to reprint them. “A Question of Evidence” by Lynn Hankinson Nelson was first published by Indiana University Press in Hypatia: A Journal of Feminist Philosophy (1993, vol. 8 [02]: 172-189), and is reprinted with permission. Quine’s essay “Epistemology Naturalized,” is from his collection Ontological Relativity and Other Essays (1969, pp. 69-90). The copyright is held by Columbia University Press, and the essay is reprinted by permission of the publisher. “A Hasty Retreat from Evidence: The Recalcitrance of Relativism in Feminist Epistemology,” was first published by Indiana University Press in Hypatia: A Journal of Feminist Philosophy (1998, vol. 13 [04]: 88111), and is reprinted with permission. “A Coherence Theory of Truth and Knowledge,” and “A Nice Derangement of Epitaphs” appeared originally in Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, edited by Ernest LePore (1986, Oxford: Basil Blackwell). The “Afterthoughts” to “A Coherence Theory” appeared in Reading Rorty: Critical Responses to Philosophy and the Mirror of Nature (And Beyond) edited by Alan Malachowski (1991, Oxford: Basil Blackwell). A special thank you to Donald Davidson who holds the copyright for these essays and generously allowed me to reprint them here. “Charity and Ideology: The Field Linguist as Social Critic,” by Bjørn Ramberg, appeared originally in Dialogue: The Canadian Philosophical Review (1988, vol. 27: 637-651) and is reprinted with the permission of the publishers. The copyright for Popper’s “On the Sources of Knowledge and Ignorance,” is held by The British Academy (1961). The essay forms the introduction to Popper’s collection Conjectures and Refutations, but appeared originally in Proceedings of the British Academy (1960, Vol. XLVI). It is reproduced here by permission.

vi

Siblings Under the Skin

Wilfrid Sellar’s “More on Givenness and Explanatory Coherence,” was originally published in Justification and Knowledge, edited by George Pappas (1979, pp. 169-182), and appears by kind permission from Kluwer Academic Publishers. “Science and Human Values,” by Carl Hempel was included in his volume Aspects of Scientific Explanation (1965) but appeared originally in Social Control in a Free Society, edited by Robert Spiller (1960) and is reprinted by permission of the publishers, University of Pennsylvania Press. I would also like to thank Laurie Lahey for her invaluable work as editorial assistant and Heather Douglas, Hugh Lacey, Steve Gimbel, and Edrie Sobstyl for their original contributions to the collection. Alison Wylie, Rich Campbell, Raphael Sassower, and Joe Rouse were also there right from the beginning of the project, providing guidance and encouragement.Finally I’d like to thank Victor Taylor who helped get the collection started and Jonathan Kaplan who helped get it finished.

Introduction

Introduction

1

2

Siblings Under the Skin

Introduction

3

Siblings Under the Skin Sharyn Clough

Western feminists and other philosophers concerned with social justice have begun to move beyond research programs that are defined over and against our analytic inheritance. Building from a necessarily critical foundation, we have begun to construct positive theories that make clear our debts to the analytic tradition. The debts are not merely methodological. Historically, the analytic tradition has had its own concerns for social justice. Witness the tireless political work of Russell, for example. Within the Vienna Circle, the search for objective criteria for truth and rational grounds for demarcating science from non-science was at least partly a critical response to European fascism and anti-semitism, especially as these fueled Nazi persecution of Jewish scientists. Not surprisingly, the methodological issues of analytic philosophy that have proven the most fruitful for feminist appropriation concern epistemology and the philosophy of science. Analytic philosophy of language and mind are also becoming increasingly attractive for feminist investigators and other philosophers concerned with challenging the political status quo. Indeed, issues in epistemology and science are often tightly connected, conceptually, with issues of language, and language with mind. Analytic philosophers such as Neurath, Carnap, Hempel, Popper, Sellars, Quine, and Davidson ranged easily over many of these topic areas, as do the feminist and other progressive thinkers who have used the analytic tradition as a springboard for further work. It is in this shared spirit of social change and methodological focus that the current volume brings together the writings of contemporary western feminist philosophers, and other progressive writers, alongside the classic texts of the analytic tradition that gave them inspiration. In Siblings Under the Skin: Feminism, Social Justice and Analytic Philosophy, the conversation begins with an exchange between feminist philosopher Lynn Hankinson Nelson and W. V. Quine. Paired with

4

Siblings Under the Skin

Quine’s “Epistemology Naturalized,” Nelson’s essay “A Question of Evidence” examines how we might think of the relationship between politics and scientific evidence. The question how to give appropriate attention to empirical evidence without at the same time lapsing into an untenable foundationalism was a going concern for Quine and feminists who use his work. Quine’s “Epistemology Naturalized” begins with a highly technical history of the problem of justifying scientific knowledge in terms of the truths of logic. In the latter half of the essay he beings to articulate naturalism as a way to solve the problem of justification, particularly by identifying the epistemic role of observation sentences. Nelson’s essay helps illuminate the more technical aspects of Quine’s program by applying his theories of holism, naturalism and empiricism to feminist science studies. Nelson makes use of Quine’s work to argue that feminists need an adequate account of the role of empirical evidence— a theme that reappears in this collection in my exchange with Donald Davidson and Edrie Sobstyl’s exchange with Wilfrid Sellars. In “A Hasty Retreat from Evidence: The Recalcitrance of Relativism in Feminist Epistemology” I argue that while feminists working in epistemology and science have made important contributions to the deconstruction of the traditional Cartesian model, some elements of the model are retained. This retention helps explain the remnants of global scepticism and its relativist variants that appear in many feminist studies of science, especially in those cases where we have tried to articulate the relationship between empirical evidence and our political values. I suggest that we make use of Davidson’s philosophy of language to more fully displace scepticism and relativism. Like his teacher Quine, Davidson provides a holistic framework that makes clear the empirical content in even the most abstract political claims. I argue that this content can be evaluated, allowing us to combat scepticism and relativism by showing that some political claims (e.g., feminist claims) are better supported by empirical evidence than are others. My essay is paired with Davidson’s “A Coherence Theory of Truth.” In this essay, Davidson analyses the justificatory role of empirical evidence. While supportive of much of Quine’s naturalism, Davidson takes Quine to task for leaving sceptical space between our beliefs

Introduction

5

and the empirical sensations that cause them. Reminding us that our awareness of any given sensation is itself another belief, Davidson produces a coherence theory of belief, without at the same time collapsing into a relativist view that uses coherence to define truth. (That his is not a coherence theory of truth conflicts, of course, with the title of his essay —a conflict he acknowledges in an important section appended to the essay, titled “After Thoughts”). Bjørn Ramberg’s “Charity and Ideology: The Field Linguist as Social Critic” responds to a question that arises for those making political use of Davidsonian semantics. If claims are to be judged relative to their position in a holistic web of empirical belief, this presumes that we have access, ideally anyway, to the truth conditions that give those claims meaning. But how then do we make sense of ideological deception? What do we do when the truth conditions are hidden or ambiguous? This is a puzzle for feminists and others who work to undermine the ideology of the status quo. Ramberg suggests a compelling solution, based on Davidson’s “A Nice Derangement of Epitaphs.” In “A Nice Derangement” Davidson argues that language is not essentially convention driven, that, in fact, we can understand each other even when we make use of idiosyncratic locutions that fly in the face of established linguistic conventions. Ramberg shows that it is precisely this slippage between established conventions that gives ideologically-charged claims the appearance of legitimacy. While this problem appears to conflict with Davidson’s views about the holistic relation between beliefs and evidence, Ramberg argues, that quite the opposite is the case—if we are vigilant radical interpreters, we can make use of holism to better identify where ideologically-problematic claims have come loose from established claims to evidence. In an original essay inspired by the work of Wilfrid Sellars, Edrie Sobstyl returns to the question how feminists can speak about the evidential justification of their beliefs without being foundationalists about evidence. Sobstyl’s essay “Lost in Logical Space? Wilfrid Sellars and Feminist Empiricism” is paired with Sellars’s “More on Givenness and Explanatory Coherence.” In the latter essay, Sellars reviews his attack on the idea that empirical sensations provide epistemic foundations for scientific theories—an idea he calls “The Myth of the Given.” However, he renews his commitment to realism—the

6

Siblings Under the Skin

world from which those sensations emanates is real and causally efficacious. As with Quine’s work, Sellars’s essay is technically dense. Sobstyl’s response to Sellars is helpful for teasing out the larger philosophical points that relate Sellars’s positive views on science, theories, and epistemic agents. In addition to her discussion of the justificatory role of evidence, Sobstyl makes use of Sellars’s analysis of agency to expand upon the feminist idea that individual knowers can only be understood in the context of the larger community of which they are a part (cf. Nelson). The final pairings speak to the role of values in science policy, featuring original essays by Hugh Lacey and Heather Douglas, alongside essays by the analytic philosophers Karl Popper and Carl Hempel. In “Where Values Interact with Science” Hugh Lacey explores the role of political values in agricultural science in Brazil. He borrows from Popper’s work, including the essay “On the Sources of Knowledge and Ignorance” to prescribe criteria for evaluating values in science that still leaves room for a robust concept of objectivity. “On the Sources of Knowledge and Ignorance” is the introductory essay to Popper’s famous collection Conjectures and Refutations. While the companion essay, “Science: Conjectures and Refutations” has been reprinted numerous times, “On the Sources of Knowledge” is less known and well-worth examining. Here, Popper acknowledges that the most important source of our knowledge is tradition—a point often observed by those who study the role of values in science. Popper asks: How do past traditions affect present scientific observations? How can we mitigate the effects of tradition? His answers provide theorists like Lacey a starting point for politically-engaged science policy critique. Heather Douglas penned “Hempelian Insights for Feminism” after she discovered a little-discussed essay by Hempel called “Science and Human Values.” In her essay, Douglas examines the relationship between political values and the acceptance of various levels of error in scientific results—what Hempel calls “inductive risk.” Appearing in Hempel’s classic collection Aspects of Scientific Explanation and other Essays in the Philosophy of Science, the essay “Science and Human Values” has received little attention until now. In “Science and Human Values” Hempel applies his characteristically

Introduction

7

clear-headed, analytic style to a discussion of the “valuational ‘presuppositions’ of science” and the effects these presuppositions have on the rules of hypothesis confirmation, and, perhaps more, importantly, on the rules of hypothesis acceptance. The essays in this collection provide a small sample of the ongoing conversation between progressive philosophers and the analytic tradition. The first sustained conversations between feminist theory and classics in western philosophy, more generally, were published in the Re-Reading the Canon series (Pennsylvania State University Press). Many texts in this series are relevant to an examination of the strengths that analytic philosophy can bring to those concerned with social justice (e.g., Re-Reading the Canon: Feminist Interpretations of Quine, edited by Lynn Hankinson Nelson and Jack Nelson, forthcoming). The formation of the Society for Analytical Feminism spoke to a similar desire to call into question the necessarily adversarial relationship between the analytic tradition and the political goals of feminist theory. The Society website provides a clear exposition of the relationship between feminism and the analytic tradition and also includes a bibliography that is helpful for identifying the work in areas such as ethics and political theory not represented in Siblings Under the Skin (see Ann Cudd, “Analytic Feminism: One Woman’s Attempt at a Definition,” www.ukans.edu/~acudd/safhomepage). For those interested in further study of the links between progressive philosophy and analytic philosophy of science and epistemology, other useful sources include Raphael Sassower’s book Cultural Collisions: Postmodern Technoscience which shows important similarities in the work of Karl Popper and feminist theorist Donna Haraway (Sassower 1995). Among the many helpful chapters of Joe Rouse’s Engaging Science, is his essay “Against Representation: Davidsonian Semantics and Cultural Studies of Science,” which provides a clear analysis of Davidson’s anti-representationalism and illustrates how such analysis can aid progressive cultural criticism of science (Rouse 1996). Lynn Hankinson Nelson’s book Who Knows: From Quine to a Feminist Empiricism, provides the most comprehensive articulation of feminist work on Quine (Nelson 1990), joined more recently by Richmond Campbell’s Illusions of Paradox: A Feminist Epistemology Naturalized (1998). Alison Wylie’s work on the gender politics affecting archaeology makes good

8

Siblings Under the Skin

use of two analytic-minded philosophers of science, Dudley Shapere and Ian Hacking (Wylie 1996). Peg O’Connor’s book Oppression and Responsibility (forthcoming), Cressida Heyes’ Line Drawings (2001) and Wendy Lee-Lampshire’s essay “Decisions of Identity: Feminist Subjects and Grammars of Sexuality” (1995) each make use of the philosophy of the later Wittgenstein to support feminist and anti-racist theorizing. Returning to the current collection, Siblings Under the Skin begins with an original essay by Steve Gimbel that explores the historical relationship between the analytic tradition and social activism. Gimbel focuses on the political and philosophical relationship between Reichenbach, Neurath, and the sociologist Frederich Tönnies. A number of philosophers have begun to study the social and political context within which the early analytic philosophers were engaged (e.g., Cat, Cartwright and Chang 1996, Giere 1996). As this collection and these studies indicate, the relation between the analytic tradition, feminism and social justice more generally is not as distant as was once thought.

Introduction

9

References Campbell, Richmond. 1998. Illusions of paradox: A feminist epistemology naturalized. New York: Rowman and Littlefield. Cat, Jordi et al. 1996. Otto Neurath: Politics and the unity of science. In The disunity of science: Boundaries, contexts and power, edited by Peter Galison and David Stump. Stanford: Stanford University Press. Cudd, Ann. Analytic feminism: One woman’s attempt at a definition. ukans.edu/~acudd/safhomepage.htm. Giere, Ronald. 1996. The feminism question in the philosophy of science. In L. H. Nelson and J. Nelson (eds.) Feminism, science, and the philosophy of science. London: Kluwer. Heyes, Cressida. 2000. Philosophical Investigations (in a feminist voice). In Line drawings: Defining women through feminist practice. Ithaca: Cornell. Lee-Lampshire, Wendy. 1995. Decisions of identity: Feminist subjects and grammars of sexuality. Hypatia 10(4): 32-45. Nelson, Lynn Hankinson. 1990. Who knows: From Quine to a feminist empiricism. Philadelphia: Temple University Press. Nelson, Lynn Hankinson and Jack Nelson, eds. (forthcoming). Re-Reading the Canon: Feminist Interpretations of Quine. University Park: Penn State University Press. O’Connor, Peg. (forthcoming). Oppression and responsibility: A Wittgensteinian approach to social practices and moral theory. University Park: Penn State University Press. Sassower, Raphael. 1995. A feminist engagement: Popper and Haraway. In Cultural collisions: Postmodern technoscience. New York: Routledge. Rouse, Joseph. 1996. Against representation: Davidsonian semantics and cultural studies of science. In Engaging science: How to understand its practices philosophically. Ithaca: Cornell University Press. Wylie, Alison. 1996. The constitution of archaeological evidence: Gender politics and science. In The disunity of science: Boundaries, contexts and power, edited by Peter Galison and David Stump. Stanford: Stanford University Press.

10

Siblings Under the Skin

Historical Context

Historical Context

11

12

Siblings Under the Skin

Historical Context

13

Politics and the Wissenschaftliche Weltanschauung: Neurath and Reichenbach on science and socialism Steve Gimbel

Introduction In some contemporary philosophical circles, the argument for the irrelevance of analytic philosophy is the simple assertion: “positivism is dead”. “Positivism” in this assertion is generally taken to refer to the weltanschauung laid out by A. J. Ayer in Language, Truth, and Logic and believed to be dogmatically and univocally asserted by the members of the Vienna Circle, the Berlin Gesellschaft für empirische Philosophie, and a small band of confederates in the United States and England. Positivism, on this view, is a single, static position to be accepted or rejected and the renunciation of any subset of its doctrines (often taken to number two, following Quine) is to place oneself outside of the analytic tradition. The unsoundness of one or both of the foundational epistemic “dogmas” is frequently taken to have been conclusively shown by Quine, Thomas Kuhn, or the Critical Theorists among others. Apart from these epistemological attacks are the political attacks launched from the socialist left that positivism is a “conservative doctrine necessarily committed to existing social institutions and to a technocratic conception of politics” (O’Neill 1995: 29), and that its adherents stand opposed to the liberation movements that arose in the second half of the 20th century. In grounding philosophical analysis in logic and science, positivism is charged with being inherently antihumanist, anti-feminist, and anti-pluralist. While these attacks have come most notably from the Frankfurt School, e.g., Marcuse (1964) and Habermas (1972), similar charges come from orthodox Marxists and the New Left. Since the late 1980’s, thanks to scholars like J. Alberto Coffa, Alan Richardson, and Ronald Giere, an industry has arisen performing careful historical analyses of the work of the members of theVienna Circle

14

Siblings Under the Skin

and the logical empiricist movement more generally. The result of this work has been largely to debunk the above caricature of early analytic philosophy. The logical empiricists, in fact, were far from unified. Careful consideration reveals astute criticisms among the major figures in the movement and wide points of contention. The breadth and intricacy of the discussions attempting to settle these differences, or to bridge seemingly divergent viewpoints. provide insights that show these views to be far from moribund. While the search for a criterion of cognitive significance has rightly been called off, and a revival of 1930s-style positivism will find few if any adherents, the subtleties embedded in the views of these philosophers are being rediscovered and found to contain significant insights for the contemporary philosophical context. This holds not only for the epistemological positions advanced by the logical empiricists, but also for their socio-political views. The charges of political conservatism are far off the mark. Most of the logical empiricists were themselves socialists, and their socio-political views are best understood as at least consistent with, if not arising from, the epistemological/scientific stance towards the world as a whole that they asserted. The logical empiricist line not only was not intended to oppose the central notions of human liberation, but was in fact explicitly and mindfully designed to assert them. Indeed, positivism can only be truly understood in the social context in which it arose, a context which included both strong opposition to the rising tide of the right-wing clerical and national socialist movements and great hopes for the instantiation of socialist and social democratic institutions when the usually stolid cultures of German-speaking Europe became briefly plastic in the aftermath of World War I. This line of discussion is most clear when it comes to cases. While the exploits of Bertrand Russell, the most famous socially active founder of analytic philosophy, are legion the details of the lives of Hans Reichenbach and Otto Neurath, central figures in the blossoming logical empiricist movement in Berlin and Vienna, respectively, are less well known, but equally persuasive. Reichenbach was the center of the Berlin Gesellschaft and cofounder and co-editor with Rudolf Carnap of Erkenntnis, the journal of logical empiricism. His non-tenured position at the University in Berlin was opposed by the philosophy faculty at least in part because as

Historical Context

15

a student he had been the leader of the Socialists Student Party Berlin, but was awarded on the political muscle of Albert Einstein who saw a philosophical ally in his former student. Neurath was a foundational figure in the Vienna Circle. Karl Popper held that it was Neurath “perhaps more than anyone else, who was instrumental in turning it into the ‘Vienna Circle’ (Popper 1973: 54).” Carnap referred to Neurath as “the big locomotive” of the Circle’s worldwide influence. A sociologist by trade, Neurath turned down a prestigious offer to join Max Weber at Heidelberg University in order to become the president of the Central Economic Administration, charged with establishing a planned, socialist economy in the newly formed Bavarian republic. When the Administration was overthrown by the clerical Dollfuss government, Neurath was imprisoned. He was able to return to Austria only on the testimony of Weber and Otto Bauer. In Reichenbach and Neurath, we have two figures who are central to the history of analytic philosophy and who are not merely possessing of socialist sympathies, but who were activists and leaders in the liberation fights of their day. Most importantly, in both cases, their ethical and socio-political theories cannot be understood apart from their broader epistemological concerns. For feminists, green theorists, and others interested in justice or liberation to write off the works of these figures, or the logical empiricists more generally, is to do a disservice to all. There is much in the analytic tradition that is useful for contemporary theorists of social justice, as so many of the essays in this volume make clear.

*** I. The dismissal of positivism is often coupled with the charge that positivism is the very pinnacle (or depth) of the enlightenment project. This is a charge that the positivists not only gladly accepted, but explicitly cultivated. The positivists sought to take the new tools, offered particularly by advances in logic that arose at the end of the 19th and beginning of the 20th centuries, and use them to further the enlightenment ideals.

16

Siblings Under the Skin

It is frequently overlooked, by those for whom “enlightenment” and “conservative” are used interchangeably, that the Enlightenment itself was born a liberation movement. The Enlightenment was a radical attempt to wrest complete control, both political and intellectual, from the religious and aristocratic institutions of the day and turn it over to the people. The cornerstone of the enlightenment ideal is the view that the ability to acquire genuine knowledge of the world is independent of personal virtue or social position. Popes and kings, Bishops and Knights, have no special access to genuine knowledge. What matters is the correct employment of natural reason, and that is, in principle, within the grasp of any normal person (Giere 1996a: 6).

The universe was taken to speak freely to all; we need not rely on powerful social institutions for our ears, we need only to listen carefully. Such a statement is not merely a claim of scientific methodology, but a political strike of the most radical sort. The very possibility of science was a threat to the system that perpetuated social and economic inequality. So it was also seen by the positivists. Science was considered to be a reaction against, indeed as the opposition to, the conservative alliance of the bourgeoisie and clerical groups that served to preserve their power and privileges. The spirit of science led to involvement in the world and a drawing of attention toward injustice and inequity rather than toward the sort of metaphysical speculation that was often used to manipulate the working class into supporting the institutions that oppressed them. For the positivists, the proper use of science is to provide possibilities to society, to tell us what can be done. All of us in the Vienna Circle took a strong interest in the political events in our country, in Europe, and in the world…I think that nearly all of us shared the following three views as a matter of course, which hardly needed any discussion. The first is the view that man has no supernatural protectors or enemies and that therefore whatever can be done to improve life is the task of man himself. Second, we had the conviction that mankind is able to change the condition of life in such a way that many of

Historical Context

17

the sufferings of today may be avoided and that the external and internal situation of life for the individual, the community, and finally for humanity will be essentially improved. The third is the view that all deliberate action presupposes knowledge of the world, that the scientific method is the best method of acquiring knowledge and that therefore science must be regarded as one of the most valuable instruments for the improvement of life (Carnap 1963: 82-3).

As scientific truth is to be independent of nation, culture, and gender, the adoption of a general scientific stance towards the world would lead to international cooperation, an end to nationalism and anti-Semitism, and a universal spirit of democracy. “The Scientific Conception of the World,” the pamphlet produced by Neurath, Carnap, and Hans Hahn that served as the founding document of the Vienna Circle, ends with such a sentiment: We witness the spirit of the scientific world-conception in growing measure the forms of personal and public life, in education, upbringing, architecture, and the shaping of economic and social life according to rational principles. The scientific world-conception serves life and life receives it (Carnap, Hahn, Neurath 1929: 33940).

Indeed, it was not merely the positivists who saw their scientific world-conception as a force for broad social change, but also those resisting this change. [I]t was also seen as a social movement by its political enemies ‘the right-wing Austro-German nationalists, the reactionary Catholic clerical establishment in Vienna and in the university, the protoNazis and anti-Semites’ who saw the Vienna Circle as ‘red’ and ‘jewish,’ and therefore as a social and political threat and not only an intellectual one (Wartofsky 1982: 86).

These political enemies not only vilified the positivists’ scientific world-conception, but also their science. With Einstein’s theories disparaged in Nazi propaganda for being “Jewish science,” the act

18

Siblings Under the Skin

of logical analysis must have been conceived as a political strike against evil. To remove this socio-political context from an understanding of logical empiricism is to misunderstand logical empiricism. There was from its inception a liberationist spirit to the movement and we see this instantiated in the social lives of Reichenbach and Neurath. Hans Reichenbach was born in Hamburg in 1891 to a protestant mother and a Jewish father. From a young age, Reichenbach was affiliated with the German counter-culture. In high school he joined the Wandervogel movement. Wandervogel was a loosely organized “revolt by withdrawal” against the mainstream means of socializing youth in Germany (Landauer 1978: 26). Bands of youth sought to escape to nature, hiking and camping for many days at a time, repudiating all aspects of “civilized society,” and rejecting all authority. The style of dress, the playing of guitars, the exaltation of wanderlust and a life of simplicity with the land leads to a more than superficial comparison to the development of the American counter-culture of the 1960s. While attending university in Berlin and Munich, Reichenbach became a leader in the Freie Studentschaft movement. Pre-World War I German universities were dominated in the student, faculty, and administrative contexts by reactionary, socially exclusive, anti-Semitic fraternal organizations called Korporationen. Non-incorporated students, or Finken, were second-class citizens in the academic community. The Freie Studentschaft was created in part as a solidarity movement and as a self-help mechanism for these unincorporated students seeking health insurance, less expensive access to education and books, and housing agencies for the student population. Over time, it began to acquire a more definite leftist political stance pressing for a democratization of the universities. Among the aims of the Freie Studentschaft was the fight against the Lex Arons, which forbade Social Democrats to hold academic positions; a striving for greater pluralism with respect to the choices of academic courses and majors; juster disciplinary procedures on campus; freedom of assembly and opportunity; non-discrimination with respect to political views, race, creed, and nationality (Reichenbach 1978: 95-6).

Historical Context

19

Upon his return from service as a radio engineer on the Russian front during the First World War, Reichenbach matriculated at the University of Berlin where he was one of a small handful of students to attend Einstein’s first seminar on relativity theory. This was the time of the German revolution and the rise in power of the Social Democrats. In its wake, Reichenbach formulated the platform for and was made the first chairman of the Socialist Student Party Berlin, which was designed to ally itself with the Social Democrats in socializing the university. Reichenbach conceived the purpose of the SSPB to change the structure of the university from one of training wealthy youth for a life of civil service perpetuating oppressive capitalist institutions, to one that would provide middle class and proletarian youth with the knowledge necessary to maintain a socialist society. In its foundational document are found demands for freedoms of academic speech, assembly and the “selection of lecturers and students without regard to their membership in any class, party, church, sect, race, sex, or state”; a call for an early form of affirmative action that would create “legal forms that will eliminate the predominance of particular groups” in the student body; student input on curricular and governance issues; and demands for “new chairs in hitherto suppressed and neglected subjects (Reichenbach 1918b: 133-5).” Holding a Ph.D. at the time, Reichenbach offered non-affiliated courses at the University in the philosophy of socialism and brought in a number of lecturers including Ernst Cassirer, Eduard Bernstein, Karl Kautsky, a leading figure in the Social Democratic movement, and Rudolf Hilferding, who was offered a central position in the Bavarian socialist government. This activity would return to haunt Reichenbach when he sought to move from his position of Associate Professor at the Technische Hochschule at Stuttgart to the faculty at the University at Berlin. Reichenbach’s appointment was roundly opposed by the faculty in part because of his political activities. His repudiation by the philosophy faculty in particular forced his sponsors, most notably Einstein, Planck, and von Laue “left leaning physicists, all” to create a non-tenured position in the physics department to teach “die erkenntnistheoretischen Grundlagen der Physik.”

20

Siblings Under the Skin

It was here that Reichenbach would organize the Gesellschaft für empirische Philosophie that would later change its name to Gesellschaft für wissenschaftliche Philosophie following a suggestion from David Hilbert. Included in the group were Walter Dubislav, Kurt Grelling, Kurt Lewin, and C. G. Hempel. Reichenbach loosely aligned the group with the Vienna Circle, corresponding with Moritz Schlick, Carnap, and Neurath. There were deeply shared sympathies and endeavors between the groups, most notably the founding of the journal Erkenntnis, although the concerns of the Berlin group tended more towards the analysis of specific scientific results and less towards the general pronouncements on the nature of science as a whole that one found coming from Vienna. The Berlin group was also more open to the inclusion of social/political issues within its official business, receiving talks from left-wing theorists and activists such as Karl Korsch. In the spring of 1933, the Nazi purge of the universities included Reichenbach. He fled to Turkey where he stayed until emigrating to the United States, living out his life as a professor at the University of California at Los Angeles. Otto Neurath was born in 1882. His parents were well educated, his mother being the daughter of a Viennese lawyer and his father a largely self-educated political economist who came from a poor Jewish family. Contrary to his father’s wishes, Neurath too studied political economy. He received his doctorate from the University in Berlin for a dissertation on conceptions of commerce in ancient Egypt. His interests in economics lay in connecting standard economic indicators to real measures of the standard of living in populations. Economics is not to be a dispassionate science, but a tool to be used in bringing about increased quality in housing, nutrition, education, entertainment, and health for all people. His historical studies brought with them an objection to the standard methodology in economics, which considered free market economies to be the normal state and all other economic orders unimportant perturbations. In particular, he noticed that in a state of war the government places extreme restrictions on the freedom of the market in order to maximize production of goods needed by the country for the sake of its military interests. A planned economy often successfully supersedes the free market in order to accomplish a socially

Historical Context

21

determined goal. In considering such planned wartime economies, Neurath noticed that profit motive is replaced by concerns over maximizing productivity and that the monetary economy is replaced by a barter-based economy in kind. If this transformation is possible during wartime and is by and large successful in bringing about a society able to carry out its external goals of military conquest, should not such a transformation also be possible in order to carry out the internal goal of maximizing the standard of living of the members of the society? Suppose a society had the will to fully enter into a “war on poverty” or a “war on human discontentment”? Such consideration led Neurath to become a committed socialist. The key to bringing about the improvement of life for all members of a society, according to Neurath, is to put in place a planned, money-less economy. But such a radical change will not be possible given the entrenchment of traditional capitalist institutions. The opportunity to put new systems in place comes from catastrophic events. Wars or natural disasters afford temporary opportunities and it is for this reason that we see such systems arising from them. A complete replacement of economic institutions would require a catastrophic event of historic proportions in order to completely uproot all traditional institutions. Such a situation would be furnished, Neurath surmised before the event, by a world war. His prediction was accurate. On the heels of World War I came the German revolution. With the economy in shambles and the leftists of the former opposition in control, the stage was set for a socialist reshaping of the German economy. Neurath, well connected with the new Social Democratic elite, began giving impassioned addresses to workers’ and farmers’ councils throughout Bavaria and Saxony outlining his vision of a planned socialist economy. It was a vision that placed less stress on seizure of means of production by the state, and more on the organization and uniting of production and distribution means through a central planning body that would have to answer to a hierarchy of democratic councils representing all interests. The plan for the economy could not be derived a priori, but must come from a democratic sentiment—a socialist economy that would last requires active cooperation of those it encompasses.

22

Siblings Under the Skin

The plan, or at least Neurath’s presentation of it, was well received and Neurath was offered the chance to head the central planning office of the newly formed Bavarian republic. In a difficult decision, Neurath turned down a prestigious offer of a lectureship with Max Weber at Heidelberg University to accept his position instantiating a socialist economy. Neurath’s Central Economic Administration worked through the tumult of the short-lived leftist control of Bavaria. He strove to implement his plan under the Hoffman republican government and the Bavarian soviet republic under Landauer and then under Leviné.1 Neurath saw economic and political matters as independent and felt that as long as economic plans had popular support, they could be put in place independent of political structure. This view of the independence of political and economic function saved Neurath’s life. When the second soviet republic fell to fascist forces, Neurath was arrested and in danger of execution. Witnesses at his trial included Max Weber and Otto Bauer. Throughout the trial, Neurath was portrayed as a devoted civil servant who acted to serve the people and not any particular government. He was convicted of assisting high treason and sentenced to one and a half years imprisonment, but, on the intervention of Bauer, was deported to Austria before serving his entire term. Returning to Vienna, after a brief time in Czechoslovakia setting up an institute to train factory councilors to organize factory councils for the German Trade Union Association, Neurath organized and became the general secretary of the Austrian Co-operative Housing and Allotment Association, arguing that capitalism bred substandard housing for workers that, in turn, played a significant causal role in illness, alcoholism, and a general diminished quality of life. Associated with the housing association was the adult education movement to which Neurath was deeply committed—the more educated the proletariat, the less likely it is to be easily dominated. These interests further collided when Neurath was asked to put together an exhibition about the Housing Association to inform the public of its work. He quickly came to see the power of such open displays in bringing knowledge to the public. The temporary exhibition transformed into a permanent Museum for Housing

Historical Context

23

and Urban Development in Vienna and ultimately to the Museum of Economy and Society. Neurath strove to change the notion of a museum from a storehouse for ancient or odd artifacts into a means of popular education. He developed a pictorial means of conveying statistical data that would require no mathematical training. This ISOTYPE picture language was designed to demystify social facts and bring them to the public where they could enter popular discourse and affectthe forming of rational social opinions. So effective were these ISOTYPE displays that Neurath’s group was commissioned to build exhibitions in both Moscow and Chicago. With the rise of Nazism, Neurath’s series of high profile socialist positions placed him in danger. Indeed, Cartwright et. al. conjecture that when Hitler proclaimed that Communists and Jews were dragging Germany to social, moral, and economic annihilation, Neurath was one of those to whom Hitler was referring (Cartwright, et. al. 1996). When the right-wing clerical government of Dollfuss seized power in Austria, opening the door to Hitler, Neurath escaped to the Hague. From here he would have to flee once again, landing, ultimately, in England.

*** II The existence of strong activist engagement on the part of philosophers like Reichenbach and Neurath provides reason to suspect that their broader philosophical work ought not be dismissed as antagonistic to those with an interest in questions of liberation and justice. But making such a connection requires further that these figures (1) possessed well-formulated considerations of ethical and socio-politico-economic issues, and (2) that these considerations were germane to their more general philosophical interests. In this section, the first of these points will be addressed. Cartwright et. al. point out correctly that Neurath and Reichenbach have a common view about the foundations of ethics. They, like others in the logical empiricist movement, denied the cognitive content of moral statements unless they be mere statements of personal preference. Unfortunately,

24

Siblings Under the Skin

this usually gets translated into the claim that the positivists were naïve subjectivists. A full understanding of the connection between the epistemology and socialist views of Reichenbach and Neurath requires a replacing of this caricature. When we consider one of the largely undiscussed influences upon the logical empiricists, sociologist Ferdinand Tönnies, a more interesting picture develops. If we are to group the positivists with any of the contemporary ethical camps, it is with care ethicists like Nel Noddings. While the ethics of care may not be entirely well-received within the academic feminist community, placing the positivists in line with any contemporary ethical discussions about gender radically repositions their views. Ferdinand Tönnies’ seminal work Gemeinschaft und Gesellschaft was an attempt to synthesize the romantic and rationalist approaches to sociology. His main goal was the explication of what he saw as the two contrasting archetypes of human relations: gemeinschaftliche relations are organic, spring naturally from the will, and are amorphous in their internal structure, whereas gesellschaftliche relations are mechanical, constructed, and well-defined by conventional rules. Using these concepts, Tönnies sought to understand the historical evolution of human relations that has resulted from industrialization and how such an understanding may be used to better the lot of people in the future. Tönnies illustrates the Gemeinschaft/Gesellschaft distinction using an analogy with mechanics. We may formulate scientific principles, e. g., equations of motion, to govern the behavior of bodies. But the practical application of these equations requires that we not deal with the actual interaction of the two objects themselves. We must replace the properties of the real objects with idealizations. We deal with point masses that do not differ in size or shape or that are perfectly solid or perfectly elastic. Our laws work perfectly for our ideal entities, but only as approximations for real objects. Of course, in small cases, approximations will be undesirable. When we have but a few objects to consider, we do not want to reduce them to less than they are. In the case of a large ensemble, however, such idealization is necessary. If we are incapable of dealing with each individually, we must establish a fair way of dealing with all. As such we do not attach value judgments to Gemeinschaft and Gesellschaft. They

Historical Context

25

are different ways of relating that are appropriate to different relations. Indeed, we do not encounter relations that are purely in one category or the other; rather, all relations are some combination of the two with one type dominant. The relationships that are most typical of Gemeinschaft fall into three categories. First is the relation between mother and child, which is “most deeply rooted in liking or pure instinct” and implies long duration. Second is the sexual relationship that begins instinctively, but does not imply long duration and thereby requires an act of will to avoid becoming a case of one-sided subjugation of the weaker member. Third is the relation between siblings, in which an act of will is required for the forming of the relationship. Less intimate relationships, e.g., kinship, neighborhood membership, and friendship may be understood in terms of this type. The relationships of Gesellschaft are those between people who are removed from one another, “separate in spite of all unifying factors.” When one encounters a strangers the lack of connection does not allow one to relate in an organic manner. Such relations must be regulated. This stranger is to be approached just as the last and the next will be. In place of the person who is the stranger we put a generalized approximation of a person and respond in a well-metered socially predetermined fashion. Business relations stand as the quintessential example of such relations. In the move from an agrarian to a mercantile economy, Tönnies notes that the default means of relating has shifted towards Gesellschaft. He points to Adam Smith’s remark that “Every man...becomes in some measure a merchant.” An effective capitalist economy requires a universal set of individual-independent rules to be set in place and be obeyed by all in the fashion of Thomas Hobbes. In the conception of Gesellschaft, the original or natural relations of human beings to each other must be excluded. The possibility of a relation in the Gesellschaft assumes no more than a multitude of mere persons who are capable of delivering something, and consequently of promising something. Gesellschaft as a totality to which a system of conventional rules applies is limitless; it constantly breaks through its chance and real boundaries. In Gesellschaft every person strives for that which is to his own

26

Siblings Under the Skin

advantage and he affirms the actions of others only insofar and as long as they can further his interest. Before and outside of convention and also before and outside of each special contract, the relation of all to all may therefore be conceived as potential hostility or latent war (Tönnies 1887: 77).

Competition becomes universal and what used to be a communal possession whose value was in its usage becomes an individually owned entity whose value derives from speculation. This social transformation was not a mere intellectual curiosity for Tönnies, indeed it was a reality that played itself out in stark terms before his own eyes. In his formative years, Tönnies observed his home province of Schleswig-Holstein converted from a rural culture to one of mechanization and commercialization. He saw the rootedness of the peasant in the earth replaced by the mercantile concern for profit. The Gemeinschaft/Gesellschaft distinction is used to account for the change in the nature of the institutions and relationships that affect the real lives of real people. According to Tönnies, the industrialization did not merely bring different institutions, it brought different types of institutions, new types geared for new and different jobs. But just as industrialization forced a reshaping of social institutions and affected the nature of human relations, so too may additional economic changes. Tönnies sees in socialism the possibility of tipping the social balance back towards Gemeinschaft. In a later revision, Tönnies added, After the terrible disruption which the capitalistic system of Gesellschaft has undergone, it has still more ruthlessly used its destructive forces. In view of these phenomena, the cry for ‘Gemeinschaft’ has become more vocal, often with explicit (or in the case of British Guild socialism, tacit) reference to this book. This cry deserves more credit, the less it voices a Messianic hope in the ‘spirit’ alone. For the spirit as a separate entity is real only in ‘ghost’ magic. To attain reality, it must incarnate itself in a living principle capable of development. Such principle is found in the idea of co-operative production, if and when it is capable of protecting itself against relapsing into mere business (Tönnies 1887: 1).

Historical Context

27

Socialism may restore Gemeinschaft, but it requires activism and must not be left up to the historical determinism of some passages in Marx. We can have no doubt that both Reichenbach and Neurath were both well acquainted with Tönnies’ work. Reichenbach’s plans to socialize the university in (1918a) are explicitly and unapologetically modeled upon Tönnies’ book, using Gemeinschaft and Gesellschaft as his central concepts roughly in the way explicated by Tönnies. Neurath met Tönnies at a summer academy in 1903 and corresponded with him intensely for several years. Neurath’s non-reductive notion of the unity of science, considering artificial individuals as legitimate entities for inclusion in scientific laws comes directly out of Tönnies.2 One place where Tönnies’ influence on Reichenbach and Neurath is profound but largely undiscussed, however, is in the position that both take on the foundation of ethics. The similarity between Neurath and Reichenbach on ethical foundations is noted by Cartwright, et. al.: Neurath expects people to care about human misery, and he hopes to educate them to do so where they do not. In the end, he thinks that without the blinders of metaphysics, enough people will do so that it will be possible to get up the new socialist economy. But he never makes the claim that one should do so because it is right or good. The fully planned economy will eliminate crises and that is what we aim to do. It is not necessary to tack on the idle addendum ‘and crises are wrong.’ Neurath’s attitude here reflects his repudiation of the entire Kantian framework, in which he resembled Reichenbach. For both Neurath and Reichenbach, the idea was repellent that we should prefer people who try to redistribute coal because it is their duty rather than because doing so alleviates human misery (1996: 248).

Both Neurath and Reichenbach hold that moral propositions contain only trivial cognitive content, yet both hold strong views about the way people ought to approach ethical decision making. It is Tönnies’ notions of Gemeinschaft and Gesellschaft that fills in the gap between the seemingly inconsistent rejection of the meaningfulness of ethical statements and the concern for human misery that is the root of ethical action for Reichenbach and Neurath.

28

Siblings Under the Skin

Both adhere to the standard positivist line that sentences with cognitive content are sentences for which one may in principle hold evidence for or against. In the case of ethical statements, i.e., statements that proclaim the moral rightness, permissibility, or wrongness of an act, either in type or token, two classes of analysis are possible. The first explication considers ethical utterances to be “a declarative statement about the volitional state of the speaker” (Reichenbach 1959: 194). When we assert “Setting small children on fire is morally unacceptable,” we are asserting a proposition about our desire that people not engage in such activities. In this way, ethical theories do contain cognitive content and are, therefore, meaningful. It is this interpretation that often leads to the positivists being portrayed as subjectivists. If all that an ethical statement means is a statement of one’s preferences and if, as seems obvious, people will differ in the preferences held, then, the line goes, the positivists must be arguing that each one of us defines his or her own sense of moral rightness and wrongness. But this line neglects the subsequent and more important move that is made. Reichenbach, for example, is well aware that the meaning derived from this parsing is not the ethical content of which we intended to speak. “[T]he volitional attitude of the speaker appears merely as an addition, interesting from a personal or psychological point of view, but dispensable for the purpose of the discussion” (Reichenbach 1959: 196). The important point for the current discussion is that when one enters into ethical discourse, the discussion has a purpose. We need to determine how we ought to act and the first explication does not account for this. The second explication is that ethical utterances only have the appearance of declarative statements. “Pushing old ladies under moving busses is morally wrong” seems to assert a moral fact about the act of pushing old ladies under busses. This pretense, however, is false. The statement is not a declarative statement at all, but an imperative. The statement of an act’s wrongness is equivalent to the instruction not do it. Ethical statements of the form “X is right/ permissible/wrong” appear declarative, but are in fact disguised commands “You must/you may/you may not perform X.”

Historical Context

29

Imperatives, however, being neither true nor false, will lack cognitive content. As such, ethical utterances do not possess cognitive meaning. The modern analysis of knowledge makes a cognitive ethics impossible: knowledge does not include any normative parts and therefore does not lend itself to an interpretation of ethics. The ethico-cognitive parallelism renders ethics a bad service: if it could be carried through, if virtue was knowledge, ethical rules would be deprived of their imperative character (Reichenbach 1951: 277).

This sentiment leads to the misunderstanding that ethical utterances were considered dispensable by the positivists. If these thinkers are proclaiming ethical propositions to be meaningless, the line goes, then surely they are suggesting that we ought to do without them. Again, positivist ethics supposedly reduces to subjectivism. But this interpretation is simply false. Ethical statements are treated very seriously by the logical empiricists both in the particular and the general. As Reichenbach notes: “it is a misunderstanding of the nature of moral directives to conclude that if ethics is not objectively demonstrable, everybody may do what he wants” (Reichenbach 1951: 287). How do we account for this seeming contradiction of the discussion of proclaimed meaningless statements? The answer is that while ethical utterances are devoid of cognitive content, they do possess instrumental meaning. We must make decisions about how to intend to act and these are real decisions with real consequences even if the result of moral deliberation is a sentence without a truth-value. Moral deliberation is to be taken seriously, but must receive a different sort of explication. The key to understanding the nature of ethical discussion for the logical empiricists is to keep in mind that logical empiricism must be understood as a reaction to Kant. Just as the logical empiricist line on the nature of space is a reaction to Kant’s transcendental aesthetic, so too this discussion is a reaction to Kant’s ethics. The categorical imperative imposes a fixed, mechanical ethic based upon metaphysics. If we eliminate the metaphysics, as the positivists were wont to do,

30

Siblings Under the Skin

what is left are rules established by mere convention. The idea of rigid, mechanistic, conventional rules governing relationships is nothing more than Tönnies’ concept of Gesellschaft. Kantian ethics are a set of commands derived from metaphysics and capitalism designed to keep the masses within the system. To anti-metaphysical socialists like the logical empiricists, the debunking of such a system is both a logical and a political act. For the logical empiricists, human behavior—the only means by which human suffering will be lessened—is not to be governed by Gesellschaft; rather, human action ought to be based upon Gemeinschaft. This is why Reichenbach and Neurath would find it repellant that one act out of duty rather than in order to alleviate human misery. We ought to act to avoid the suffering which is a part of the “external and internal situation of life for the individual, the community, and finally for humanity” (Carnap 1963: 83). The “scientific humanism” that Carnap attributes to the positivists generally is the result of replacing of the Kantian Gesellschaft with Gemeinschaft. Reichenbach traces the feeling of obligation we experience from moral imperatives to a sense of the common will. [T]hese volitions are imposed upon us by the social group to which we belong, in other words, they are originally group volitions. This origin accounts for the superpersonal dignity and the feeling of subordination with which we make the moral decision (Reichenbach 1951: 285).

We see moral decisions made not for the satisfaction of personal desires, but in terms of the goals of the community within which the agent is embedded. [C]riminals have an ethics of their own class; within their class they do not steal or kill, but they oppose their class to the larger class of what we call a civilized society and disregard all moral obligations with respect to this wider class. Students of a high-school class may regard their class as a group opposed to the teacher and find it in their moral right to deceive and harass him. Conversely, there are teachers who are highly esteemed by the students and are seldom deceived; such a teacher has succeeded in making the

Historical Context

31

students incorporate him into their group. The working class has an ethics of its own; so has the class of big capitalists, or the aristocracy of countries that have not yet eliminated the remnants of feudalism (Reichenbach 1951: 286-7).

This notion of decisions by group volition ought to be understood as broader than a mere social relativism. It is not that each society may randomly choose what to require of people, rather moral imperatives stem from the goals of the community, from the source of unity that underlies the Gemeinschaft. We find Tönnies giving the same sort of stress to group will in his explication of Gemeinschaft. [T]he theory of Gemeinschaft starts from the assumption of perfect unity of human wills as a natural condition which is preserved in spite of actual separation...everything that conforms to this conception of the Gemeinschaft relationship is to be considered as the proper and real will of those bound together (Tönnies 1887: 47).

The basis of moral deliberation is to consider the other and the community as an entity whose overall welfare one ought to be concerned about. This stance has reappeared in the ethical literature. Nel Noddings’ explication of ethical caring, for example, may be seen as a direct extension of the project of replacing a Gesellschaft-based ethic with a Gemeinschaft-based ethic. Noddings and other care-based theorists point to the rigidity and mechanical nature of Kantian ethics as the source of their divergence from real-life applicability. We must move away from rule-based, contractual business models towards those based upon the model of the caring mother. While Noddings characterizes the current state of ethical consideration as embodying “the language of the father”, at the expense of the voice of the mother, the Tönnies-influenced logical empiricists would characteristic it as embodying the language of the capitalist at the expense of the voice of the community.

***

32

Siblings Under the Skin

III In the face of this discussion, the question naturally arises why such an analysis is necessary. Why has the interplay between science and left-wing politics in the work of the logical empiricists been hidden or denied? How could it be that these figures have been branded conservatives? There appear to be several factors explaining the contemporary caricature of logical empiricism. The first concerns American political history. When purged from the German-speaking universities, the logical empiricists, mostly socialist and many of them Jewish, sought refuge from Nazism in the west. Many, e.g., Reichenbach, Carnap, Hempel, Bergmann, and Feigl emigrated to the United States. The political climate of the U.S. was deeply affected by the end of the Second World War, which was accompanied by the onset of McCarthyism.3 The socialist and Jewish philosophers of the Vienna Circle left a society that persecuted them for their views and backgrounds, but the host countries that promised them relief proved instead to be similarly restrictive of their freedoms. If the positivists had made their socio-political views prominent, their welcome may have been rescinded. Similarly, in the midst of the Cold War, the aid of German-accented scientists like Werner von Braun encouraged the view that German speakers who discussed science would be eager to prove themselves useful to the American military-industrial complex. By placing the analytical, scientific ends of their work as a public face, the logical empiricists sought political safety. The second factor concerns the nature of philosophical progress. Philosophy does not, by and large, build upwards, establishing new and bigger results. Rather, the march of philosophy is often to increased depths, questioning and overturning the presuppositions of the previous generation. Even those, like Philip Frank (1949), who shared the Enlightenment picture, acknowledged that “in every period a new Enlightenment is required” in order to keep the presuppositions of one generation from becoming too entrenched and thereby oppressive. In fact, Neurath himself wrote, “It will be interesting when we new critics of language will be criticized with the tools we have devised” (Neurath, in Uebel 1996. p.109). The next wave of critiques, both analytic and

Historical Context

33

continental, took aim at the logical empiricists and in this game of going one deeper, the radicals of the past were destined to become the conservatives of the present. But this does not entirely account for the contemporary misrepresentation. Even if contemporary critiques were to go beyond logical empiricism, why not do so with the reverence that Isaac Newton gave his predecessors? Why not argue that the positivists did not see as far as we, but that is only because we stand upon their shoulders? Why ignore or vilify the socio-political views of these committed leftists? The traditionally-held distaste for history within analytic philosophy itself may provide additional clues. The extensive, careful study of history for history’s sake, was, until recently, seen by many analytic philosophers as a non-philosophical endeavor.4 Importance is placed on what is found in analysis and once an analysis has been shown faulty, revisiting it is considered a waste of time. As such, detailed discussions of the foundational works of the positivists and the broader historical context of their creation have been largely neglected. But this is changing. As the work of Richardson, Cartwright, Giere, et. al. shows, there has been an upsurge in interest in the historical foundations of logical empiricism, focusing largely on the connection between various historical factors and the development of empiricist epistemologies. It is very exciting to see the essays collected here in Siblings Under the Skin acknowledge and celebrate the connection between the analytic movement and its historically entrenched commitment to social justice.

34

Siblings Under the Skin

Notes 1. 2. 3. 4.

See Cartwright, Cat, Fleck, Uebel (1996, pp. 50-2) for a detailed discussion. See chapter 2 of Cartwright, Cat, Fleck, Uebel (1996) for a detailed discussion of Tönnies’ effects upon Neurath. Detailed discussions of the effects of McCarthyism on the logical empiricists are given in McCumber (1996) and Giere (1996b). E.g., Carnap’s discussion of the problems of “historical neutralism” in American university philosophy departments (Carnap 1963).

References Carnap, R. 1963. Intellectual Autobiography. In P. A. Schillp (ed.) The philosophy of Rudolf Carnap. LaSalle: Open Court. Carnap, R, H. Hahn & O. Neurath. 1929. The scientific conception of the world. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. Cartwright, N. & J. Cat. 1996. Neurath against method. In R. N. Giere and A. Richardson (eds.) Origins of logical empiricism. Minnesota Studies in the Philosophy of Science, vol. XVI, Minneapolis: University of Minnesota. Cartwright, N., J. Cat, L. Fleck, & T. E. Uebel. 1996. Otto Neurath: Philosophy between science and politics. New York: Cambridge. Frank, P. 1949. Modern science and its philosophy. Cambridge: Harvard University. Giere, R. N. 1996a. The feminism question in philosophy of science. In L. H. Nelson and J. Nelson (eds.) Feminism, science, and the philosophy of science. London: Kluwer. ——. 1996b. From wissenschaftliche Philosophie to philosophy of science. In R. N. Giere and A. Richardson (eds.) Origins of logical empiricism. Minnesota Studies in the Philosophy of Science, vol. XVI, Minneapolis: University of Minnesota. Habermas, J. 1972. Knowledge and human interests. London: Heinemann, 1972. Jacobs, S. & K. Otto. 1990. Otto Neurath: Marxist member of the Vienna Circle. Auslegung, vol. 16, no.2, pp.175-189.

Historical Context

35

Jørgenson, J. 1951. The development of logical empiricism. Chicago: University of Chicago. Lakenbacher, E. 1973. Memories of Otto Neurath. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. Landauer, C. 1978. Memories of Hans Reichenbach. In M. Reichenbach and R. S. Cohen (eds.) Hans Reichenbach: Selected Writings, 1909-1953. Boston: D. Reidel. Marcuse, H. 1964. One dimensional man. London: Routledge and Kegan Paul. McCumber, J. 1996. Time in the ditch: American philosophy and the McCarthy era. Diacritics, vol. 26, no. 1, pp.33-49. Neurath, O. 1936. Individual sciences, unified science, and pseudorationalism. In R. Cohen and M. Neurath (eds.) Otto Neurath: Philosophical papers. Dordrecht: D. Reidel. ——. 1932. Protocol statements. In R. Cohen and M. Neurath (eds.) Otto Neurath: Philosophical papers. Dordrecht: D. Reidel. ——. 1920. Experiences of socialization in Bavaria. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. ——. 1919. Through war economy to economy in kind. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. ——. 1913. The theory of war economy as a separate discipline. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. ——.1912, The problem of the maximum pleasure. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. Noddings, N. 1984. Caring: A feminine approach to ethics and moral education. Berkeley: University of California. O’Neill, J. 1995. In partial praise of a positivist: The work of Otto Neurath. Radical Philosophy, vol. 74, pp. 29-38. Popper, K. 1973. Memories of Otto Neurath. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. Reichenbach, H. 1918a. Socializing the University. In M. Reichenbach and R. S. Cohen (eds.) Hans Reichenbach: Selected writings, 1909-1953. Boston: D. Reidel. ——. 1918b. Platform of the Socialist Students’ Party. In M. Reichenbach and R. S. Cohen (eds.) Hans Reichenbach: Selected writings, 1909-1953. Boston: D. Reidel. ——. 1918c. Report of the Socialist Student Party, Berlin. In M. Reichenbach and R. S. Cohen (eds.) Hans Reichenbach: Selected writings, 1909-1953. Boston: D. Reidel. ——. 1920. The theory of relativity and a priori knowledge. Berkeley: University of California.

36

Siblings Under the Skin

——. 1951. The rise of scientific philosophy. Berkeley: University of California. ——. 1959. On the explication of ethical utterances. In M. Reichenbach (ed.) Modern philosophy of science. New York: Humanities Press. ——. 1978. Student years: An introductory note. In M. Reichenbach & R. S. Cohen (eds.) Hans Reichenbach: Selected Writings, 1909-1953. Boston: D. Reidel. Schumann, W. 1973. Memories of Otto Neurath. In M. Neurath and R. S. Cohen (eds.) Empiricism and sociology. Dordrecht: Reidel. Tönnies, F. 1887. Community and society. East Lansing: Harper Torch Books. Uebel, T. E. 1999. Otto Neurath, the Vienna Circle, and the Austrian Tradition. In A. O’Hara (ed.) German philosophy since Kant. New York: Cambridge. ——. 1996. The enlightenment ambition of epistemic utopianism: Otto Neurath’s theory of science in historical perspective. In R. N. Giere and A. Richardson (eds.) Origins of logical empiricism. Minnesota Studies in the Philosophy of Science, vol. XVI, Minneapolis: University of Minnesota. Wartofsky, M. W. 1982. Positivism and politics: The Vienna Circle as a social movement. Grazer-Philosophische-Studien, vol. 82, no. 16/17, pp. 79102. Weber, M. 1904. ‘Objectivity’ in social sciences. In E. Shils and H. Finch (ed.) The methodology of the social sciences. New York: Free Press.

Historical Context

37

Biographical Note Steven Gimbel (b. 1968) is an assistant professor of philosophy at Gettysburg College where his research focuses mainly upon philosophy of science and the history of analytic philosophy and early 20th century physics and mathematics. He has published articles in Philosophy of Science, Erkenntnis, the Journal of Applied Philosophy, and Studies in Practical Philosophy. The path that led to Steve’s interest in the political philosophies of Reichenbach and Neurath is somewhat circuitous. Beginning with questions about the relation between epistemology and the history of relativity, Steve’s interest focused on Reichenbach’s early writings on geometric conventionalism. Finding formal similarities in Reichenbach’s understanding of Einstein’s argumentation and his early attempts to modify Kantian epistemology, Steve examined other aspects of Reichenbach’s writings, such as his work in ethics and political philosophy, in search of further structural parallels. The unmistakable influence of Ferdinand Tönnies on Reichenbach’s social picture led Steve to examine the impact of Tönnies on the early analytics more generally. Tönnies’ personal and professional acquaintance with the socialist philosopher and economist Otto Neurath showed Neurath to be an important part of the story and helped Steve better articulate the relationship between the philosophy of the members of the Vienna Circle and their concerns for social justice.

38

Siblings Under the Skin

Nelson and Quine

Conversations

39

40

Siblings Under the Skin

Nelson and Quine

41

A Question of Evidence Lynn Hankinson Nelson

Truth and belief, W. V. Quine once noted, are sticky concepts; they stick to each other (Quine 1981a, 38). Evidence is equally sticky. Consider how strange things become when we are asked to unstick belief and evidence: We sit and think, but do we sit and believe? The White Queen, indeed, professed to do so: “When I was your age, I always did it for half-an-hour a day. Why, sometimes I’ve believed as many as six impossible things before breakfast” (Quine 1987, 19).

“But it will be agreed,” Quine continues, “that the White Queen was atypical.” Decisions to believe independently of any evidence “real or imagined,” he notes, “stretch the term ‘belief ’ beyond belief ” (Quine 1987, 19). Belief is (or ought to be) constrained by available evidence. But it is not similarly constrained by truth. We do not, without further explanation, understand someone’s claim to believe what is obviously inconsistent with available evidence. But we do understand belief of what is plausible, supported by available evidence, but in fact false. In particular cases, it is easier to unstick belief from truth than to unstick belief from evidence (or at least from the absence of counterevidence). “Intellectual progress,” John Dewey claimed, “usually occurs when we abandon questions and both of the alternatives they assume—an abandonment that results from their decreasing vitality and a change of urgent interest.” We move on, he argued, not because we answer questions but because we get over them (Dewey 1910, 313). If Tarski and Davidson are right, “What is truth?” is such a question (though “What is true?” is not). Theories of truth, insofar as they invite talk of correspondence or facts, do not advance our understanding of the world. But the same cannot be said of evidence. It is a substantial concept and belief remains well stuck to it.

42

Siblings Under the Skin

Indeed, any claim to forgo both the concept and the use of evidence is patently disingenuous. No claim or theory worth our attention comes unconnected to evidence. With feminist communities, including but by no means limited to those currently engaged in debates about epistemology, what Christine Di Stefano calls “the falsely innocent indifference” that may be appropriately attributed to postmoderns would be at least paradoxical (Di Stefano 1990, 77). For feminists, the nature of evidence, and hence a theory of evidence, is of the essence. But I stress that it is evidence, and theories thereof, of which I am speaking, not truth. I suspect that some considerable part of the misunderstanding among feminists involved in discussions of epistemology can be traced to confusions between evidence and truth. Philosophical epistemology, a central feature of the background against which feminist epistemologies and debates about epistemology have developed, has created a litany of mutually implicated alternatives—epistemology or sociology of knowledge, modernism or postmodernism, objectivism or relativism—as well as the questions and categories that present these as authentic and exhaustive options. There is an increasing awareness in feminist discussions that it is such alternatives and the questions that prompt them that we need to get over (for recent discussions, see Nicholson 1990; Alcoff and Potter 1992). To advance that project, I outline here a pragmatic account of evidence. My claim is that the account allows us to underwrite the insights and implications of various feminist epistemologies, and at least some of the theories and projects with which these epistemologies are interdependent, and to abandon some of the questions underlying the litany of dichotomies listed above. “Pragmatism,” as Richard Rorty notes, “is a vague, ambiguous, and overworked word” (Rorty 1982, 160). My account of evidence is pragmatic in the sense that on that view there are no pretheoretic notions of evidence, no standards or methods laid down prior to the business of constructing theories to explain and predict what we experience. Rather, it is within our various epistemological projects—in daily life, in science, in philosophy, in politics—that notions and standards of evidence emerge, concomitantly with the unfolding of those projects. The point, I will argue, holds for all of our notions of evidence—from our more general notions (that all of it is, in the end, sensory, for example,

Nelson and Quine

43

or, alternatively, that there are innate ideas), to notions that are more context-specific (e.g., that instrument readings constitute evidence for events on the subatomic level or that “imperfections” are evidence of evolution). In explicating the account, I make use of Quine’s arguments for holism and for the radical interdependence of epistemology and other current theories, extending these in light of issues raised in and by feminist science criticism. There is precedent, of course, for using the term “pragmatism” to describe Quine’s views; Quine himself uses the term, as have feminists who have discussed his work (e.g., Addelson 1992; Harding 1986). I have also used “empiricist” to describe his views (Nelson 1990), as he himself does.1 The connection between the two is naturalism: Quine’s pragmatism is the result of his rejection of foundationalism, and of his commitment to the empiricist thesis that in the end the evidence for our various theories and claims is sensory—a thesis that comes, as he puts it, “after physics, physiology, and psychology, not before” (Quine 1966a, 212).2 Defining science broadly as inclusive of all efforts to explain and predict experience, Quine argues that “the scientific system, ontology and all, is a conceptual bridge of our own making, linking sensory stimulation to sensory stimulation” (Quine 1981b, 20).3 So stated, the view is compatible with any social constructivist account of knowledge. But naturalism precludes a wholesale skepticism about evidence. Hence Quine continues: But I also [have an] unswerving belief in external things—people, nerve endings, sticks, stones… Now how is all this robust realism to be reconciled with the barren scene that I have just been depicting? The answer is naturalism, the recognition that it is within science itself, and not in some prior philosophy, that reality is to be identified and described (Quine 1981b, 21).

That naturalism underwrites Quine’s argument that epistemology is radically interdependent with other current knowledge and undertakings (Quine 1960, l966a, 1969): Epistemology…is not logically prior somehow to common sense or to the refined common sense which is science; it is part rather of the overall scientific enterprise, an enterprise which Neurath

44

Siblings Under the Skin

has likened to that of rebuilding a ship while staying afloat in it (Quine 1966b, 240).

Elsewhere I have argued that several of Quine’s positions support important implications of feminist science critiques and epistemologies: that knowledge is socially constructed and constrained by evidence; that canons of evidence, metaphysical commitments, and epistemology are interdependent; and, indeed, that all of our ways of organizing and explaining experience share a radical interdependence (Nelson 1990). Here I make explicit the account of evidence at work in these positions and further naturalize it in light of knowledge provided by feminist scholarship. What will emerge is a view of evidence more inclusive than what Quine envisioned (but also, my arguments indicate, “naturally” extended) and commensurate with two further implications of feminist scholarship and activism: that social relations of gender, race, class, and culture are epistemologically significant4 and that the primary epistemological agents are multiple, overlapping, and evolving communities. What makes the account I will offer promising for feminists are three shifts it incorporates. One is the abandonment of any pretense to a view from nowhere, as well as the fantasy, as Susan Bordo recently characterized a view implicit in some postmodern arguments, of being everywhere (Bordo 1990, 136); this view of evidence and the arguments I will advance and rehearse in support of it insist that we are always “here.” A second is the abandonment of the assumption that evidence is something only individuals can gather or have; the account indicates that evidence is fundamentally communal (see Nelson 1992). Third, and relatedly, the account abandons the assumption that specific theories and claims develop or face experience in isolation, insisting instead on an inclusive and dynamic system of evidential relations. Recognizing the tenacity of the categories, questions, and assumptions that have underwritten the litany of dichotomies noted earlier, I begin to make the case for a pragmatic or naturalized account of evidence by briefly discussing two other familiar and contested notions, truth and epistemology.

***

Nelson and Quine

45

I. Truth and epistemology The most basic point to note about truth is that exploring the possibility of a feminist theory thereof may be a viable future project, but it is a different project from the present one in which I explore the questions of what warrants beliefs and claims and what warrants our distinguishing claims as “more” or as “less” warranted. As I see things, these are questions about evidence, nor about truth. But perhaps there is more that should be said. I don’t suggest that feminists (or anyone else) “give up” on truth. I assume that, unlike the White Queen (when we are not being disingenuous), we believe to be true what we say to be true and that “What is true?” remains a viable, indeed a crucial, question. But at present I know of only two ways to maintain more than a “disappearance” or “redundancy” theory of truth (the view, as Quine puts things, that “truth is immanent” (Quine 1981b, 21). One is to define truth in terms of internal consistency, to subscribe to a coherence theory of truth. While I will argue that coherence with a current body of knowledge and practices is one of the constraints on specific theories and projects, it is not a sufficient criterion. Our collective experience indicates that nature is a point of resistance with which we need to contend, and the hypothesis that there is a world that constrains what it is reasonable to believe is woven through most of our theories because it makes the most sense of what we experience. The second option, which gives the world its due, defines truth in terms of correspondence, in terms of some notion of “fitting” or “being true to” the facts. But correspondence, and theories of truth that incorporate some version of it, are also beset with problems. Not the least of these is that correspondence presumes a dualism of “organizing schemes and that which is waiting to be organized” (Davidson 1974) and the latter presumes a view from nowhere and that there is ‘‘something’’ by virtue of which claims are ‘‘true.’’ It may be that some other approach will be found to save the intuition underlying theories of truth. Sandra Harding’s recent suggestion that we can and should aim at less partial and perverse theories (Harding 1990) seems to me on the mark, but not to require a theory of truth to underwrite it. I think it can be underwritten with

46

Siblings Under the Skin

an adequate account of evidence, and I regard naturalism as one of the requirements for adequacy. An important premise underlying my project is that epistemology, like other notions with which it is deeply connected—including evidence, truth, and knowers—is a contested notion and, like these others, a dynamic one. History attests to the fact that epistemologies emerge, evolve, and fade in relation to new undertakings, experience, and knowledge: e.g., in modern thought, in positivism, in Marxism, and, more recently, in feminism and in the variety of responses to the demise of foundationalism, which include sociology of knowledge, naturalized epistemologies, and postmodern arguments against epistemology. This historical process has implications for current debates about feminist epistemology. One implication is that epistemology was not defined once and for all by Cartesian or any other epistemology, and hence we are currently faced with a number of interesting, interconnected, and open questions, including “Who knows?” “How is knowledge generated?” and “What is the nature and strength of evidence available to us?” Put another way, we would need an argument that did not rely on the tenets of Cartesian epistemology (or some other) to be warranted in concluding that epistemologies are necessarily wedded to the existence of “one truth,” to abstract individualism, or to foundationalism or that they are by their “very nature” both partial and universalizing. Lacking such an argument, the questions I have noted remain open (see Nelson 1992; Nicholson 1990; Alcoff and Potter 1992). A second implication is that epistemologies, theories about how knowledge is constructed and about the evidence available for and used in such constructions, are deeply connected to other current theories and undertakings. For those who have abandoned foundationalism with its promise of “extratheoretic” groundings and standards, there are several ways to construe that relationship and its implications. One response is to construe epistemology as an allegedly “purely descriptive” enterprise, as those working in the strong program in sociology of knowledge propose, a proposal that at least acknowledges that epistemology is dynamic.5 Another response is to abandon epistemology altogether, as postmoderns advocate; my earlier comments suggest that the proposal assumes a view of epistemol-

Nelson and Quine

47

ogy undermined by its history. A third is to “naturalize” epistemology, as Quine and some feminists advocate: to acknowledge that an epistemology is radically interdependent with other knowledge and undertakings and should be pursued as such and that the justification of beliefs and theories (if it occurs) will be on the basis of their coherence with experience rather than by appeal to a current epistemology.6 A fourth response is to try to have it both ways: to acknowledge (at least implicitly) the interdependence of epistemologies and (other) current knowledge and projects but to attempt to salvage a justificatory role for epistemology (Harding 1986), perhaps by embracing the circularity involved and arguing that one circular system is better than others (Jaggar 1983). The account of evidence I will advocate supports the third option, that we naturalize epistemology. It indicates that the grounds appropriate to judging theories, claims, and methods are broader than a favored epistemology (although appeals to epistemology will also be made): that they encompass experience, as well as other knowledge and undertakings with which, according to this account of evidence, epistemology shares a radical interdependence. Hence, a naturalized view of evidence allows us to retain the prescriptive aspects of epistemology without appeal to alleged “first principles” or foundations. The most direct support for the claims made in this section awaits the account of evidence in the next. And perhaps the best way to conclude these introductory remarks is to make clear some pragmatic suspicions concerning traditional epistemologies and theories of truth by paraphrasing Rorty’s description of William James and Dewey. So long as we see Quine and other pragmatists as having a “theory of truth,” we get them wrong; so long as we see pragmatists as having a “theory of knowledge” in the sense of a “first knowledge”—in the sense of a theory to justify other theories, methods, and practices—we get them wrong. In either case, as Rorty says of James and Dewey, we ignore their “criticisms of the assumption that there ought to be theories about such matters” (Rorty 1982, 160). But evidence is another matter.

***

48

Siblings Under the Skin

II. Evidence We have learned, or perhaps relearned, much about evidence in the last four decades.7 We now recognize that what we say and believe about the world, the social worlds and natural worlds within which we function and of which we are part, far exceeds all the evidence we have or ever will have. There is “slack,” to use Quine’s term, between all of our theories and the evidence we have for them (Quine 1966b, 241). Put another way, it is compatible with our collective experience that we will eventually abandon our current theories (though not, of course, all at once) for theories that are commensurate with much of our experience to date but incompatible with our present theories (Longino 1990; Nelson 1992). We have also learned that indefinitely many theories might equally well organize and explain what we experience, that we are not warranted in assuming there is a unique, true theory of nature awaiting discovery. As I have made the point, it is commensurate with our collective experience that an alternative theory of nature that did not include Boyle’s law (or for that matter any “law”), a theory of nature that organized things differently, might equally well explain and predict what we experience (Nelson 1992; Potter 1992). Minimally, then, there is no one “most probable” account of the world, and hence we need some further argument to warrant the assumption that there is (nonetheless) one true account. Quine makes the argument this way: Even if we bypass such troubles [as are raised by assuming a final organon of scientific method]…we have no reason to suppose that man’s [sic] surface irritations even unto eternity admit of any one systematization that is scientifically better or simpler than all possible others. It seems likelier, if only on account of symmetries or dualities, that countless alternatives would be tied for first place. Scientific method is the way to truth, but it affords even in principle no unique definition of truth (Quine 1960, 23).

I have argued that another lesson of the last forty years is that there is nothing in our collective experience to warrant the assumption that our sensory organs are sufficiently refined to discriminate a “best” theory or “most probable” theory (if there is such a thing) from alternative candidates (Nelson 1992). It is commensurate with that experience and

Nelson and Quine

49

with our knowledge (e.g., in empirical psychology and evolutionary epistemology) that our sensory organs are refined to a degree that (so far) they enable us to survive by organizing and predicting relevant future experience. But there is nothing to warrant the inference that they are adequate to the task of encompassing all that goes on. Indeed, it is an implication of current evolutionary theory that our sensory organs are the product of “jury-rigging” and probably only one of the possible functional results (Gould 1989). Feminist philosophers of science have made use of the first two lessons (and I am suggesting we also make use of the third) to remind us that provided we remain committed to taking evidence seriously, the door is not closed on any question (although not all, of course, are worthy of pursuit), including the question of whether gender or politics could have anything to do with “good” or with “serious” science. They have also used them in the service of more specific arguments concerning the role of background assumptions and values in theory acceptance, in their evaluations of particular theories and research programs, and in their assessments of the emphases of traditional philosophy of science (see, e.g., Longino 1990; Nelson 1990; Potter 1992). The final lesson I note here, a lesson that, while making use of the others, contributes important—and in some ways unique—insights into the nature of evidence, has resulted from feminist participation and interest in knowledge building. I focus again on feminist science criticism. It is an implication of that body of criticism—not only of its findings but of its existence and evolution as well—that the experience and knowledge we bring to bear on the theorizing we do in science will include that shaped by the social relations of gender, race, and class that characterize our society. Nor, many of us have argued, can we take the lesson of that body of criticism to mean that stricter methodological controls are needed to “filter out” these factors and relationships, which surely are present in feminist science criticism (Harding 1986, 1991; Longino 1990; Nelson 1990; Seigfried 1990; Tuana 1989). Rather, we need at the very least more authentic views of the relationships between knowledge and social and political relations than were possible given the categories (or lack thereof ) of much of traditional epistemology.8

50

Siblings Under the Skin

One can, of course, view these lessons as “the bad news.” The first three underscore limitations on the evidence available to us; the fourth broadens the factors relevant to our knowledge and undertakings (including science) to encompass social relations, politics, values, and other factors long regarded as a threat to objectivity, if not the very antitheses of evidence. But I suggest that the inclusiveness appropriately understood is not to be feared. Full support for my suggestion awaits the account of evidence, but both are foreshadowed in the arguments so far rehearsed. Each of these arguments makes use of things we know (the “we,” of course, requires specification),9 and each presumes that knowledge (e.g., in empirical psychology, evolutionary theory, philosophy of science, and feminist science criticism) and collective experience (“collective,” of course, also requires specification) constitute evidence for our views about evidence. In other words, the limitations and the inclusiveness are immanent and do nothing to further the case for skepticism or relativism. The alternative to acknowledging these lessons provides no comfort. Holding on to the view that there is one most probable theory, or that knowledge will someday be complete, or that our sensory organs are sufficiently refined to encompass all that happens, or that there are real boundaries between “serious” knowledge and the social and political relations that characterize our society would be, from here, at best an article of faith. As such, it would be no more warranted, no more defensible, than any other article of faith.10 More to the point, we do not need to settle for an article of faith; there is a view that allows for evidence and reasonable belief without derivability from unshakable foundations, without certitude. The view accepts coherence, and with it, explanatory power, as a measure of reasonableness. I begin to explicate it using a description of research in high-energy physics offered by physicist Leon Lederman. Lederman begins by noting that “no subatomic particle is ever observed directly”: Two particles collide and spew debris and new matter inside the accelerator. Physicists infer the existence of new particles from the fact that they collide with other particles which leave electromagnetic tracks in a $65 million detector. Think of a bus that drives

Nelson and Quine

51

by your house every day. One afternoon while you’re at work the bus collides with a Subaru. The bumper flies off the Subaru and hits your mailbox, which is hurled through your window. When you come home, you look at the pattern of shattered glass and say, “Hmmm, a Subaru.” That’s not unlike what high-energy physicists do for a living (Lederman 1989, 43).

We can use Lederman’s analogy in making a first start at answering the question, “What is the evidence for subatomic particles and for claims about their behavior?” Underwriting or supporting the view that electromagnetic tracks and debris are evidence of new subatomic particles, there is, most obviously, a body of theory and accepted practices in which subatomic particles figure directly, as well as other theories and methodological commitments in physics, mathematics, and technology. Lederman’s analogy also reveals a connection between a theory that posits subatomic particles and so-called commonsense knowledge of macroscopic objects and events, for it is by virtue of the latter—in this case, instrument readings indicating the presence of debris in a $65 million particle detector, and the analogy drawn between such debris and what happens when macroscopic objects collide—that the evidence for, including the explanatory power of, subatomic particles becomes apparent.11 Lederman is also clearly presuming broad metaphysical and methodological commitments, including the views that there are objects and events that are not “directly” observable that explain more systematically what happens on the macroscopic level and that particular macroscopic events (instrument readings, for example) are evidence for these. My claim is that the theories, methods, and commitments that have emerged as underwriting claims about subatomic particles, some within science and some of a broader reach, constitute a large part of the evidence for those claims (see also Nelson 1992). That is, the terms “underwrite” and “support” serve to identify essential relationships, for the “system” sketched above, to use a Quinean metaphor, is akin to an arch, each “piece” supporting, serving as evidence for, and being supported by others (Quine 1960, 11). Knowledge about subatomic particles and claims about their behavior are not isolable from a larger system of theories, practices, and standards of evidence. Hence there is no discrete piece of evidence that warrants them. The evidence includes

52

Siblings Under the Skin

an extensive body of scientific theories, technologies, methodologies, standards, and practices; knowledge and experience of macroscopic objects and events and the standards we use to identify these; and broad metaphysical and methodological commitments, including those noted above.12 This is only part of the story, of course, for internal consistency is not an adequate criterion for reasonable beliefs and viable explanations. Part of the evidence for physicists’ current knowledge about subatomic particles is experience—theories that posit subatomic particles both organize and are compatible with our experiences—and such theories have explanatory power—they allow us to explain and to predict some of what happens. Our consideration of Lederman’s statement also has implications for “experience”: namely, that it can not be afforded a foundational status.13 The sensory experiences that are currently possible and viewed as relevant to claims about subatomic particles are shaped and mediated by a historically specific system of theory and accepted practices—a system that also constitutes part of the evidence for such particles. I was once asked if this view of evidence was appropriate to “nontheoretical” entities. The short answer to the question is that there are no such entities. Part of the evidence for mailboxes, Subarus, and instrument readings is provided by the larger systems of theories and practices in which such objects figure; and part of the evidence is provided by the fact that theories that include such objects (as well as political parties and social movements) help us to organize, explain, and predict some of what happens. Our abilities as individuals to recognize such objects depend as deeply on communal theories and practices as do those of the physicist to recognize electromagnetic tracks, for it is such theories and practices that organize our sensory experiences into coherent and recoverable accounts (Nelson 1990, 1992). Quine once combined these several arguments. Noting that “any defense of the molecular doctrine has to do with its indirect bearing on observable reality” and that, from the point of view of molecular theory, physical objects are “just posits” that help us to organize what we experience, he argued that “we should do well to conclude: such, then, at bottom, is what evidence is, both for ordinary bodies and for molecules” (Quine 1966b, 233-38). The conclusion yields a natural-

Nelson and Quine

53

ized and holistic account of evidence: standards of evidence, standards that determine what we recognize and countenance as evidence, emerge within the processes through which we generate knowledge. Our understanding of evidence emerges concomitantly with our success in understanding those very processes.

*** III. A naturalized account of evidence I turn next to an example from feminist science criticism as a way of further developing this account of evidence and to show that it is compatible with what I previously cited as implications of feminist scholarship: that knowledge is both socially constructed and constrained by evidence and that social relations, including gender, race, and class, are epistemologically significant. I will also briefly address its implications for the agents of epistemology. The example is biologist Ruth Bleier’s criticism of a research project in neuroendocrinology devoted to finding a hormonal basis for sex differences in hemispheric lateralization (Bleier 1984, 80-109). A fundamental problem with research into a hormonal basis for sex-differentiated lateralization, Bleier argued, is that the central assumption underlying it—the assumption that there are sex differences in lateralization—is not itself warranted. The assumption, she noted, is in fact borrowed from a research project in reproductive neuroendocrinology that has attempted to establish such differences but has never succeeded in doing so (Bleier 1984, 84). Bleier cited additional and fundamental problems underlying research into sex-differentiated lateralization: there is no evidence for a correlation between hemispheric lateralization and cognitive abilities (e.g., visuospatial abilities) that does not rely on circular reasoning; the alleged sex differences in cognitive abilities to be explained by lateralization have not been established (indeed, they remain a matter of considerable debate within empirical psychology); the arguments used to support the hypothesis of sex-differentiated lateralization rely on circular reasoning and androcentric assumptions; and two of the more widely discussed studies that use lateralization to explain “women’s lack of success in science and math-

54

Siblings Under the Skin

ematics” contradict each other about its cause, with one claiming that it is due to a lesser degree of lateralization and the other citing a greater degree of lateralization (Bleier 1984, 80-109 passim). At the conclusion of an extensive analysis, Bleier attributed to sexism the flaws inherent in such research, the failure to recognize or acknowledge them, and the proliferation of studies into a biological basis for sex-differentiated cognitive abilities. What the studies are in fact trying to explain are the differences that exist between the sexes in status, privilege, or power within known industrial, patriarchal systems, and they do this through attempting to scientifically establish biologically determined, sex-differentiated cognitive [characteristics]... that would make women’s subordinate position inevitable (Bleier 1984, 109).

I will not evaluate the viability of the two research programs mentioned or Bleir’s claims about the specific flaws underlying or incorporated in them.14 Rather, I will use the issues raised in her critique to support the account of evidence argued for in the preceding section and to naturalize it further in light of feminist science criticism, including Bleier’s arguments. Note, first, that even for those who reject a holistic account of evidence, the borrowing of hypotheses is not a mark of flawed science. To the contrary, synthesizing hypotheses and results is commonly regarded as a mark of good science. (But, of course, presenting results as if the hypothesis one has borrowed has been confirmed when it has not been, as Bleier documented in this and other cases, is a mark of flawed science.) Second, while historical studies suggest clear connections between research into sex differences and social and political context (and, as I argue below, constitute evidence supporting feminist suspicions about such research), the evidence for the viability of individual research projects into sex differences and for the hypotheses at work in these projects (e.g., that hormones are related to sex-differentiated fetal brain organization) encompasses more than androcentric and sexist assumptions (see Longino 1990). Some of the questions, hypotheses, and models Bleier cited as unwarranted do appear to be so when considered in isolation—the assumption, for example, of a causal relation between hormones and sex-differentiated lateraliza-

Nelson and Quine

55

tion, in the absence of results establishing (or even lending credence to) sex-differentiated lateralization. But, as Bleier’s critique and those of other feminist critics indicate, these questions and hypotheses are strongly connected to other working hypotheses and research programs that contribute to the apparent plausibility of a hormonal basis for sex differences in lateralization. These include research in empirical psychology into possible correlations between lateralization and cognitive abilities, research in reproductive endocrinology linking androgens to sex differences in fetal rat brains, an extensive literature claiming that sex differences in mathematical abilities have been established, as well as the research in neuroendocrinology that Bleier noted, devoted (at least in part) to establishing sex-differentiated lateralization (Bleier 1984; Longino 1990; Nelson 1990). Now a hundred research projects, interconnected and even interdependent at various levels, do not necessarily add up to one larger good research project, particularly if all incorporate wrongheaded assumptions, and I concur that there is evidence which indicates that research into a biological basis of sex differences in cognitive abilities is wrongheaded. But we need a framework broader than individual projects or hypotheses to encompass the evidence that actually supports the research Bleier and other feminists have criticized. Equally important, we need such a framework to arrive at an adequate account of the evidence that supports feminist criticisms of these projects and of the larger enterprise of attempting to discover or establish sex differences in cognitive abilities and/or a biological basis for those differences. That framework, I suggest, is provided by a naturalized view of evidence that does not erect artificial boundaries between politics and science: that recognizes a broad system of theories and practices, including those of science (construed narrowly), “common sense,” and politics, as constituting part of the evidence for individual research projects into sex differences—and for feminist criticism of them. This view is naturalistic, for the expansion of evidence to include politics has emerged concomitantly with feminist experience and knowledge. Construing evidence broadly and naturalistically, we can point to the following as evidence supporting feminist claims that research into sex-differentiated lateralization is unwarranted: the historical and current fascination within the sci-

56

Siblings Under the Skin

ences to study (or establish) sex differences, and the fact that such interest intensifies during public debates about “women’s position”; a staggering list of proposed explanations for women’s inferior cognitive abilities; the fact that although explanations of women’s alleged inferiority have often contradicted each other, the “results” have remained consistent; the androcentric (and often sloppy) assumptions incorporated within research questions (one of Bleier’s examples is the research question, “Why are males masculine?”); and the fact that research into a biological basis for sex-differentiated cognitive abilities persists even as other current research, as well as the gains being made by women in mathematics and the sciences, undermines its rationale, a point to which I return below. In other words, part of the evidence underwriting our suspicions about this research is a history that would be laughable if the results were not so tragic, a history with strong parallels to the present; and part of the evidence consists of other knowledge and developments—which, in fact, are available to those engaging in the research. Included in the last category (but by no means exhaustive of it) are research into postnatal neurobiological growth that indicates that the development of the structures necessary to and enabling cognitive functioning requires a sociolinguistic environment—and hence like other aspects of neurobiological development, is not driven solely by biology; psychological and sociological research documenting both the role of environment in the development of cognitive skills and differences in the environments” of girls and boys relevant to such development; and the fact that girls are beginning to close the gap in (what are arguably gender-biased) math achievement tests and other areas, a gap the research in question is purportedly designed to explain (see, for example, Bleier 1984; Longino 1990; Nelson 1990). It is in light of such evidence—some emerging within specific sciences and some within broader contexts, including history and politics—as well as in view of the flaws Bleier and other scientists have documented that we are warranted in concluding that androcentric and sexist assumptions are factors, indeed, that they are functioning as evidence, in current research into sex-differentiated lateralization.

***

Nelson and Quine

57

IV. Conclusion It is a consequence of the several arguments advanced and rehearsed here that there is no simple formula for distinguishing viable theories or research programs from nonviable ones, that such judgments require extensive, multifaceted, and always incomplete evaluation (see Longino 1990; Nelson 1992; Seigfried 1990). When we acknowledge this, we raise the level of frustration of those who want feminists to answer “How could a theory about electrons be related to gender or politics?” in twenty-five words or less; but then were we to switch roles to ask a physicist to defend the theory of subatomic particles or a philosopher to defend modal logic, given the same restriction, it would become clear that we are, to vary Neurath’s metaphor, in the same boat. More importantly, when we acknowledge this, we block the all-too-facile conclusion that the lesson provided by feminist scholarship is (what we all knew all along) that politics is inherently distorting of research and is itself a mark of “bad science.” Politics, including an intense interest in gender and other social relations, is neither overcome nor absent in feminist science criticism. To deny its relevance, we would need to assume either that feminists (for some reason other than their politics) just happen to be better scientists (more careful observers, more creative theoreticians, more logically adept, etc.) or that many scientists have deliberately distorted their accounts of humans and other species to construct androcentric theories (Harding 1986; Longino 1990; Nelson 1990). My argument is that the important lesson to take from the last forty years is that individual theories neither develop nor face experience in isolation; that the evidence available, relevant, and appropriate includes theories, assumptions, projects, and values of a broad reach including politics; and that the evidence (when it is interesting) is neither arbitrary nor unable to be evaluated. These are among the more farreaching implications of feminist science criticism, and they are central features of the view of evidence I am advocating. It is a consequence of this view of evidence and an implication of the analysis of research into sex differences in cognitive abilities undertaken in the previous section that a naturalized feminist epistemology is a viable enterprise. In constructing accounts of how knowledge is generated and of the evidence available and used in such construc-

58

Siblings Under the Skin

tions, feminist epistemologists can and should appeal to, among other things, feminist experience and knowledge, acknowledging the radical interdependence between epistemology and other knowledge and undertakings. The relationship is better characterized as a spiral than as circular, for in attempting to understand the knowledge and standards accepted by another community or those of our own communities, we may find that we need to reconstruct the experiences of those involved (e.g., neuroendocrinologists) to make the most overall sense and/or to alter the assumptions with which we began—to abandon some of the standards of evidence and/or revise our views about what we know. Both kinds of reconstruction have, of course, characterized feminist scholarship to date. My arguments also indicate the need for an additional reconstruction relevant to current debates about feminist epistemology and to the naturalized epistemology I am advocating. It concerns who it is who knows, and I touch on it briefly here (see also Nelson 1990, 1992). On the view of evidence I have outlined, claims to know are shaped and made possible, and ultimately subject to, community criteria—the standards and knowledge constructed by communities within which we come to know (see Addelson 1992; Harding 1991; Longino 1990). As I have put the point, “acceptable answers to the question ‘Who knows?’ include ‘Many of us,’ ‘All of us,’ ‘Everyone’…but only very problematically, ‘Only me’” (Nelson 1990, 255-56). My claims to know are subject to the knowledge and standards constructed by the various communities of which I am a member; indeed, I have the ability to know only because there are such communities, and both my communities and I will judge my claims by reference to communal standards and knowledge. Hence the primary epistemological agents are groups—or, as the foregoing suggests, communities.15 This claim might be understood in two ways. One is that what comes to be recognized (and some of it “certified”) as knowledge (recognized by our friends in the first case; cited in footnotes, published in refereed journals, printed in boldface in textbooks, in the second) is the result of negotiations between, consensus achieved by, or other activities engaged in by individuals who, as individuals, know in some logically or empirically “prior” sense.

Nelson and Quine

59

But evidence is not an individual matter. And so the claim is stronger, the “we” a primitive: the collaborators, the consensus achievers, and, in more general terms, the agents who generate knowledge are communities, not individuals. Individuals do know, of course; but your knowing or mine depends on our knowing, for some “we.” More to the point, you or I can only know what we know (or could), for some “we.” Hence the “agents” in the current debate concerning research into sex-differentiated lateralization, the appropriate focus of epistemologists, are dynamic communities, including at least neuroendocrinology, feminist communities, and their overlap (consider feminist endocrinologists), as well as the larger social and political community of which these are subcommunities. The reconstructions that have resulted from feminist scholarship, including those undertaken and proposed here, are compatible with and supported by a view of epistemology and other current knowledge and undertakings as mutually supporting, without benefit or need of pretheoretic or extrasystematic foundations. Neurath has likened science to a boat, which, if we are to rebuild it, we must rebuild plank by plank while staying afloat in it. … Our boat stays afloat because at each alteration we keep the bulk of it intact as a going concern (Quine 1960, 3-4).

A feminist, naturalized account of evidence is also a going concern.

60

Siblings Under the Skin

Notes 1. 2. 3. 4.

5. 6. 7. 8. 9.

10. 11. 12. 13.

It is not clear, however, that those of us who attribute pragmatism or empiricism to Quine are using the same understandings of either. See, e.g., Harding (1986); Longino (1990); Nelson (1990); and Tuana (1992). As Charlene Haddock Seigfried points out, Dewey’s instrumentalism or pragmatism “was often simply known as his naturalism” (private correspondence). See, e.g., Kennedy (1970). Ironically, Quine also wants to maintain a boundary between politics and science. If we adopt the holism and naturalism he advocates and I pursue here, no such boundary can be drawn. See also Nelson (1990). I make use here of Harding’s insight that gender is a social relation (Harding 1991). Seigfried points out that the original pragmatists, in particular Mead and Dewey, recognized race, class, and culture—and, to a lesser extent, gender—as epistemologically significant and that “the extension of gender [follows logically]” from their recognition of other social relations (private correspondence). The qualification reflects my view that all such accounts make at least implicit use of current knowledge and assumptions. See, for example, Addelson (1992); Duran (1991); and Nelson (1990, 1992). This paragraph and the next two repeat arguments made in Nelson (1992). Similar arguments are found in Code (1991); Duran (1991); Harding (1991); Longino (1990); Nelson (1990); and Alcoff and Potter (1992). It will become clear that 1 view the homogeneous “we” of some mainstream epistemology as no less a fiction than the abstract “every man” of traditional epistemology. See also Code (1991); Harding (1991); and Nelson (1992). Elsewhere I argue that such articles of faith may be a consequence of a commitment to epistemological individualism. Nelson (1992). I qualify “commonsense” knowledge to remind us that it is as dynamic and theoretical (and sometimes as contested) as “scientific” knowledge. See also Nelson (1990). For an alternative and comprehensive account of empiricism, as well as arguments against the holism I am advocating, see Longino (1990). Appeals to experience can become vacuous unless the notion of what constitutes experience is further specified. Here and elsewhere I rely on Quine’s notion of experience as the firings of sensory receptors (Nelson 1990, 1992). But as those discussions and this indicate, we do not experience those firings; we experience the world.

Nelson and Quine

61

14. Bleier’s analysis was more extensive than my summary indicates. See also Longino (1990); Nelson (1990); and Tuana (1989). 15. Seigfried notes that “the community of knowledge of feminist epistemology has historical roots in C. S. Peirce’s community of scholars, which Dewey extended to actual communities” (private correspondence).

References Addelson, Kathryn Pyne. 1992. Knower/doers and their moral problems. In Feminist epistemologies. See Alcoff and Potter 1992. Alcoff, Linda and Elizabeth Potter, eds. 1992. Feminist epistemologies. New York and London: Routledge. Al-Hibri, Azizah Y. and Margaret A. Simons, eds. 1990. Hypatia reborn: Essays in feminist philosophy. Bloomington: Indiana University Press. Bleier, Ruth. 1984. Science and gender. New York: Pergamon Press. Bordo, Susan. 1990. Feminism, postmodernism, and gender-skepticism. In Feminism/postmodernism. See Nicholson 1990. Boydsron, Jo Ann, ed. 1970. Guide to the works of John Dewey. Carbondale: Southern Illinois University Press. Code, Lorraine. 1991. What can she know? Feminist theory and the construction of knowledge. Ithaca, NY: Cornell University Press. Davidson, Donald. 1974. On the very idea of a conceptual scheme. Proceedings and Addresses of the American Philosophical Association 47: 5-20. Dewey, John. [1910] 1970. The influence of Darwin on philosophy. Reprinted in Darwin: A Norton critical edition, ed. Philip Appleman. New York and London: W. W. Norton. Di Stefano, Christine. 1990. Dilemmas of difference. In Feminism/ postmodernism. See Nicholson 1990. Duran, Jane. 1991. Toward a feminist epistemology. Totowa, NJ: Rowman and Littlefield. Gould, Stephen J. 1989. Wonderful life: The Burgess shale and the nature of history. New York and London: W. W. Norton. Harding, Sandra. 1986. The science question in feminism. Ithaca, NY: Cornell University Press. ——. 1990. Feminism, science, and the anti-Enlightenment critiques. In Feminism/postmodernism. See Nicholson 1990.

62

Siblings Under the Skin

——. 1991. Whose science? Whose knowledge? Thinking from women’s lives. Ithaca, NY: Cornell University Press. Jaggar, Alison M. 1983. Feminist politics and human nature. Totowa, NJ: Rowman and Allenfeld. Kennedy, Gail. 1970. Dewey’s logic and theory of knowledge. In Guide to the works of John Dewey. See Boydston 1970. Lederman, Leon. 1989. An interview with Leon Lederman. Omni (Oct): 4652. Longino, Helen. 1990. Science as social knowledge: Values and objectivity in scientific inquiry. Princeton, NJ: Princeton University Press. Nelson, Lynn Hankinson. 1990. Who knows: From Quine to a feminist empiricism. Philadelphia: Temple University Press. ——. 1992. Epistemological communities. In Feminist epistemologies. See Alcoff and Porter 1992. Nicholson, Linda J., ed. 1990. Feminism/postmodernism. New York and London: Routledge. Potter, Elizabeth. 1992. Gender and epistemic negotiations. In Feminist epistemologies. See Alcoff and Potter 1992. Quine, W. V. 1960. Word and object. Cambridge: MIT Press. ——. 1966a. On mental entities. In The ways of paradox and other essays. New York Random House. ——. 1966b. Posits and reality. In The ways of paradox and other essays. New York: Random House. ——. 1969. Epistemology naturalized. In Ontological relativity and other essays. New York: Columbia University Press. ——. 1981a. On the very idea of a third dogma. In Theories and things. Cambridge: Harvard University Press. ——. 1981b. Things and their place in theories. In Theories and things. Cambridge: Harvard University Press. ——. 1987. Quiddities: An intermittently philosophical dictionary. Cambridge, MA, and London: Belknap Press. Rorty, Richard. 1982. Consequences of pragmatism. Minneapolis: University of Minnesota Press. Seigfried, Charlene Haddock. [1985] 1990. Second sex: Second thoughts. In Hypatia-reborn: Essays in feminist philosophy. See Al-Hibri and Simons 1990. Tarski, Alfred. 1956. Logic, semantics, metamathematics. Oxford: Clarendon Press. Tuana, Nancy, ed. 1989. Feminism and science. Bloomington: Indiana University Press. ——. 1992. The radical future of feminist empiricism. Hypatia 7(1): 100-14.

Nelson and Quine

63

Biographical note Lynn Hankinson Nelson was drawn to philosophy at an early age. Although very good at math, her father, a mathematician, told her she was more likely to be a philosopher. This prediction was prompted by a call from Lynn’s second grade teacher who complained that Lynn kept asking “but what are numbers? They’re not the numerals after all.” Lynn was inspired to study Quine’s philosophy of science by her then professor, now husband, Jack Nelson, and by Susan Haack’s Philosophy of Logics (that her philosophy of logic came later to be paired with feminist philosophy is unlikely to please Professor Haack, but then that’s part of the fun). Lynn published her first book Who Knows: From Quine to a Feminist Empiricism (Temple, 1990) before meeting Quine in person. She treasures the hand-written note she received from him about the book. Quintessentially Quine, he wrote “You have me dead to rights,” (she was thrilled!) “But,” he continued: “the idea of a feminist epistemology still has an aura of unreality to me. I must read on.” She has since met Quine several times and once she and Jack had a lovely dinner with him. Here she had been forewarned to keep the conversation away from philosophy (a topic he reserved for formal discussion) and toward Quine’s more favored dinner topics—travel and languages. To this day, Lynn is unsure whether she managed to dissuade Quine from his charitable assumption that, given their last names, both she and Jack were fluent in Swedish. Lynn has co-authored a monograph with Jack, On Quine (Wadsworth, 2000)—it was a labor of love and the work she’s most proud of. They were thrilled when Quine’s son, Doug Boynton Quine, read the manuscript to his father just months before Quine died. Doug reported that his father thoroughly enjoyed it. Lynn and Jack are currently working on Reconstituting Empiricism: The Project and Legacy of W.V. Quine.

64

Siblings Under the Skin

Nelson and Quine

65

Epistemology Naturalized W. V. Quine

Epistemology is concerned with the foundations of science. Conceived thus broadly, epistemology includes the study of the foundations of mathematics as one of its departments. Specialists at the turn of the century thought that their efforts in this particular department were achieving notable success: mathematics seemed to reduce altogether to logic. In a more recent perspective this reduction is seen to be better describable as a reduction to logic and set theory. This correction is a disappointment epistemologically, since the firmness and obviousness that we associate with logic cannot be claimed for set theory. But still the success achieved in the foundations of mathematics remains exemplary by comparative standards, and we can illuminate the rest of epistemology somewhat by drawing parallels to this department. Studies in the foundations of mathematics divide symmetrically into two sorts, conceptual and doctrinal. The conceptual studies are concerned with meaning, the doctrinal with truth. The conceptual studies are concerned with clarifying concepts by defining them, some in terms of others. The doctrinal studies are concerned with establishing laws by proving them, some on the basis of others. Ideally the obscurer concepts would be defined in terms of the clearer ones so as to maximize clarity, and the less obvious laws would be proved from the more obvious ones so as to maximize certainty. Ideally the definitions would generate all the concepts from clear and distinct ideas, and the proofs would generate all the theorems from self-evident truths. The two ideals are linked. For, if you define all the concepts by use of some favored subset of them, you thereby show how to translate all theorems into these favored terms. The clearer these terms are, the likelier it is that the truths couched in them will be obviously true, or derivable from obvious truths. If in particular the concepts of mathematics were all reducible to the clear terms of logic, then all the truths of mathematics would go over into truths of logic; and surely

66

Siblings Under the Skin

the truths of logic are all obvious or at least potentially obvious, i.e., derivable from obvious truths by individually obvious steps. This particular outcome is in fact denied us, however, since mathematics reduces only to set theory and not to logic proper. Such reduction still enhances clarity, but only because of the interrelations that emerge and not because the end terms of the analysis are clearer than others. As for the end truths, the axioms of set theory, these have less obviousness and certainty to recommend them than do most of the mathematical theorems that we would derive from them. Moreover, we know from Gödel’s work that no consistent axiom system can cover mathematics even when we renounce self-evidence. Reduction in the foundations of mathematics remains mathematically and philosophically fascinating, but it does not do what the epistemologist would like of it: it does not reveal the ground of mathematical knowledge, it does not show how mathematical certainty is possible. Still there remains a helpful thought, regarding epistemology generally, in that duality of structure which was especially conspicuous in the foundations of mathematics. I refer to the bifurcation into a theory of concepts, or meaning, and a theory of doctrine, or truth; for this applies to the epistemology of natural knowledge no less than to the foundations of mathematics. The parallel is as follows. Just as mathematics is to be reduced to logic, or logic and set theory, so natural knowledge is to be based somehow on sense experience. This means explaining the notion of body in sensory terms; here is the conceptual side. And it means justifying our knowledge of truths of nature in sensory terms; here is the doctrinal side of the bifurcation. Hume pondered the epistemology of natural knowledge on both sides of the bifurcation, the conceptual and the doctrinal. His handling of the conceptual side of the problem, the explanation of body in sensory terms, was bold and simple: he identified bodies outright with the sense impressions. If common sense distinguishes between the material apple and our sense impressions of it on the ground that the apple is one and enduring while the impressions are many and fleeting, then, Hume held, so much the worse for common sense; the notion of its being the same apple on one occasion and another is a vulgar confusion. Nearly a century after Hume’s Treatise, the same view of bodies was espoused by the early American philosopher Alexander Bryan Johnson

Nelson and Quine

67

(Johnson, [1836] 1947). “The word ‘iron’ names an associated sight and feel,” Johnson wrote. What then of the doctrinal side, the justification of our knowledge of truths about nature? Here, Hume despaired. By his identification of bodies with impressions he did succeed in construing some singular statements about bodies as indubitable truths, yes; as truths about impressions, directly known. But general statements, also singular statements about the future, gained no increment of certainty by being construed as about impressions. On the doctrinal side, I do not see that we are farther along today than where Hume left us. The Humean predicament is the human predicament. But on the conceptual side there has been progress. There the crucial step forward was made already before Alexander Bryan Johnson’s day, although Johnson did not emulate it. It was made by Bentham in his theory of fictions. Bentham’s step was the recognition of contextual definition, or what he called paraphrasis. He recognized that to explain a term we do not need to specify an object for it to refer to, nor even specify a synonymous word or phrase; we need only show, by whatever means, how to translate all the whole sentences in which the term is to be used. Hume’s and Johnson’s desperate measure of identifying bodies with impressions ceased to be the only conceivable way of making sense of talk of bodies, even granted that impressions were the only reality. One could undertake to explain talk of bodies in terms of talk of impressions by translating one’s whole sentences about bodies into whole sentences about impressions, without equating the bodies themselves to anything at all. This idea of contextual definition, or recognition of the sentence as the primary vehicle of meaning, was indispensable to the ensuing developments in the foundations of mathematics. It was explicit in Frege, and it attained its full flower in Russell’s doctrine of singular descriptions as incomplete symbols. Contextual definition was one of two resorts that could be expected to have a liberating effect upon the conceptual side of the epistemology of natural knowledge. The other is resort to the resources of set theory as auxiliary concepts. The epistemologist who is willing to eke out his austere ontology of sense impressions with these set-theoretic auxiliaries is suddenly rich: he has not just his impressions to play with, but sets of

68

Siblings Under the Skin

them, and sets of sets, and so on up. Constructions in the foundations of mathematics have shown that such set-theoretic aids are a powerful addition; after all, the entire glossary of concepts of classical mathematics is constructible from them. Thus equipped, our epistemologist may not need either to identify bodies with impressions or to settle for contextual definition; he may hope to find in some subtle construction of sets upon sets of sense impressions a category of objects enjoying just the formula properties that he wants for bodies. The two resorts are very unequal in epistemological status. Contextual definition is unassailable. Sentences that have been given meaning as wholes are undeniably meaningful, and the use they make of their component terms is therefore meaningful, regardless of whether any translations are offered for those terms in isolation. Surely Hume and A. B. Johnson would have used contextual definition with pleasure if they had thought of it. Recourse to sets, on the other hand, is a drastic ontological move, a retreat from the austere ontology of impressions. There are philosophers who would rather settle for bodies outright than accept all these sets, which amount, after all, to the whole abstract ontology of mathematics. This issue has not always been clear, however, owing to deceptive hints of continuity between elementary logic and set theory. This is why mathematics was once believed to reduce to logic, that is, to an innocent and unquestionable logic, and to inherit these qualities. And this is probably why Russell was content to resort to sets as well as to contextual definition when in Our Knowledge of the External World and elsewhere he addressed himself to the epistemology of natural knowledge, on its conceptual side. To account for the external world as a logical construct of sense data—such, in Russell’s terms, was the program. It was Carnap, in his Der logische Aufbau der Welt of 1928, who came nearest to executing it. This was the conceptual side of epistemology; what of the doctrinal? There the Humean predicament remained unaltered. Carnap’s constructions, if carried successfully to completion, would have enabled us to translate all sentences about the world into terms of sense data, or observation, plus logic and set theory. But the mere fact that a sentence is couched in terms of observation, logic, and set theory does not mean that it can be proved from observation sentences by logic and

Nelson and Quine

69

set theory. The most modest of generalizations about observable traits will cover more cases than its utterer can have had occasion actually to observe. The hopelessness of grounding natural science upon immediate experience in a firmly logical way was acknowledged. The Cartesian quest for certainty had been the remote motivation of epistemology, both on its conceptual and its doctrinal side; but that quest was seen as a lost cause. To endow the truths of nature with the full authority of immediate experience was as forlorn a hope as hoping to endow the truths of mathematics with the potential obviousness of elementary logic. What then could have motivated Carnap’s heroic efforts on the conceptual side of epistemology, when hope of certainty on the doctrinal side was abandoned? There were two good reasons still. One was that such constructions could be expected to elicit and clarify the sensory evidence for science, even if the inferential steps between sensory evidence and scientific doctrine must fall short of certainty. The other reason was that such constructions would deepen our understanding of our discourse about the world, even apart from questions of evidence; it would make all cognitive discourse as clear as observation terms and logic and, I must regretfully add, set theory. It was sad for epistemologists, Hume and others, to have to acquiesce in the impossibility of strictly deriving the science of the external world from sensory evidence. Two cardinal tenets of empiricism remained unassailable, however, and so remain to this day. One is that whatever evidence there is for science is sensory evidence. The other, to which I shall recur, is that all inculcation of meanings of words must rest ultimately on sensory evidence. Hence the continuing attractiveness of the idea of a logischer Aufbau in which the sensory content of discourse would stand forth explicitly. If Carnap had successfully carried such a construction through, how could he have told whether it was the right one? The question would have had no point. He was seeking what he called a rational reconstruction. Any construction of physicalistic discourse in terms of sense experience, logic, and set theory would have been seen as satisfactory if it made the physicalistic discourse come out right. If there is one way there are many, but any would be a great achievement. But why all this creative reconstruction, all this make-believe? The stimulation of his sensory receptors is all the evidence anybody has had

70

Siblings Under the Skin

to go on, ultimately, in arriving at his picture of the world. Why not just see how this construction really proceeds? Why not settle for psychology? Such a surrender of the epistemological burden to psychology is a move that was disallowed in earlier times as circular reasoning. If the epistemologist’s goal is validation of the grounds of empirical science, he defeats his purpose by using psychology or other empirical science in the validation. However, such scruples against circularity have little point once we have stopped dreaming of deducing science from observations. If we are out simply to understand the link between observation and science, we are well advised to use any available information, including that provided by the very science whose link with observation we are seeking to understand. But there remains a different reason, unconnected with fears of circularity, for still favoring creative reconstruction. We should like to be able to translate science into logic and observation terms and set theory. This would be a great epistemological achievement, for it would show all the rest of the concepts of science to be theoretically superfluous. It would legitimize them—to whatever degree the concepts of set theory, logic, and observation are themselves legitimate—by showing that everything done with the one apparatus could in principle be done with the other. If psychology itself could deliver a truly translational reduction of this kind, we should welcome it; but certainly it cannot, for certainly we did not grow up learning definitions of physicalistic language in terms of a prior language of set theory, logic, and observation. Here, then, would be good reason for persisting in a rational reconstruction: we want to establish the essential innocence of physical concepts, by showing them to be theoretically dispensable. The fact is, though, that the construction which Carnap outlined in Der logische Aufbau der Welt does not give translational reduction either. It would not even if the outline were filled in. The crucial point comes where Carnap is explaining how to assign sense qualities to positions in physical space and time. These assignments are to be made in such a way as to fulfill, as well as possible, certain desiderata which he states, and with growth of experience the assignments are to be revised to suit. This plan, however illuminating, does not offer any key to translating the sentences of science into terms of observation, logic, and set theory.

Nelson and Quine

71

We must despair of any such reduction. Carnap had despaired of it by 1936, when, in “Testability and meaning,” (1936; 1937) he introduced so-called reduction forms of a type weaker than definition. Definitions had shown always how to translate sentences into equivalent sentences. Contextual definition of a term showed how to translate sentences containing the term into equivalent sentences lacking the term. Reduction forms of Carnap’s liberalized kind, on the other hand, do not in general give equivalences; they give implications. They explain a new term, if only partially, by specifying some sentences which are implied by sentences containing the term, and other sentences which imply sentences containing the term. It is tempting to suppose that the countenancing of reduction forms in this liberal sense is just one further step of liberalization comparable to the earlier one, taken by Bentham, of countenancing contextual definition. The former and sterner kind of rational reconstruction might have been represented as a fictitious history in which we imagined our ancestors introducing the terms of physicalistic discourse on a phenomenalistic and set-theoretic basis by a succession of contextual definitions. The new and more liberal kind of rational reconstruction is a fictitious history in which we imagine our ancestors introducing those terms by a succession rather of reduction forms of the weaker sort. This, however, is a wrong comparison. The fact is rather that the former and sterner kind of rational reconstruction, where definition reigned, embodied no fictitious history at all. It was nothing more nor less than a set of directions—or would have been, if successful—for accomplishing everything in terms of phenomena and set theory that we now accomplish in terms of bodies. It would have been a true reduction by translation, a legitimation by elimination. Definire est eliminare. Rational reconstruction by Carnap’s later and looser reduction forms does none of this. To relax the demand for definition, and settle for a kind of reduction that does not eliminate, is to renounce the last remaining advantage that we supposed rational reconstruction to have over straight psychology; namely, the advantage of translational reduction. If all we hope for is a reconstruction that links science to experience in explicit ways short of translation, then it would seem more sensible to settle for psychology. Better to discover how science is in fact

72

Siblings Under the Skin

developed and learned than to fabricate a fictitious structure to a similar effect. The empiricist made one major concession when he despaired of deducing the truths of nature from sensory evidence. In despairing now even of translating those truths into terms of observation and logicomathematical auxiliaries, he makes another major concession. For suppose we hold, with the old empiricist Peirce, that the very meaning of a statement consists in the difference its truth would make to possible experience. Might we not formulate, in a chapter-length sentence in observational language, all the difference that the truth of a given statement might make to experience, and might we not then take all this as the translation? Even if the difference that the truth of the statement would make to experience ramifies indefinitely, we might still hope to embrace it all in the logical implications of our chapter-length formulation, just as we can axiomatize an infinity of theorems. In giving up hope of such translation, then, the empiricist is conceding that the empirical meanings of typical statements about the external world are inaccessible and ineffable. How is this inaccessibility to be explained? Simply on the ground that the experiential implications of a typical statement about bodies are too complex for finite axiomatization, however lengthy? No; I have a different explanation. It is that the typical statement about bodies has no fund of experiential implications it can call its own. A substantial mass of theory, taken together, will commonly have experiential implications; this is how we make verifiable predictions. We may not be able to explain why we arrive at theories which make successful predictions, but we do arrive at such theories. Sometimes also an experience implied by a theory fails to come off; and then, ideally, we declare the theory false. But the failure falsifies only a block of theory as a whole, a conjunction of many statements. The failure shows that one or more of those statements is false, but it does not show which. The predicted experiences, true and false, are not implied by any one of the component statements of the theory rather than another. The component statements simply do not have empirical meanings, by Peirce’s standard; but a sufficiently inclusive portion of theory does. If we can aspire to a sort of logischer Aufbau der Welt at all, it must be to one in which the texts slated for translation into observa-

Nelson and Quine

73

tional and logico-mathematical terms are mostly broad theories taken as wholes, rather than just terms or short sentences. The translation of a theory would be a ponderous axiomatization of all the experiential difference that the truth of the theory would make. It would be a queer translation, for it would translate the whole but none of the parts. We might better speak in such a case not of translation but simply of observational evidence for theories and we may, following Peirce, still fairly call this the empirical meaning of the theories. These considerations raise a philosophical question even about ordinary unphilosophical translation, such as from English into Arunta or Chinese. For, if the English sentences of a theory have their meaning only together as a body, then we can justify their translation into Arunta only together as a body. There will be no justification for pairing off the component English sentences with component Arunta sentences, except as these correlations make the translation of the theory as a whole come out right. Any translations of the English sentences into Arunta sentences will be as correct as any other, so long as the net empirical implications of the theory as a whole are preserved in translation. But it is to be expected that many different ways of translating the component sentences, essentially different individually, would deliver the same empirical implications for the theory as a whole; deviations in the translation of one component sentence could be compensated for in the translation of another component sentence. Insofar, there can be no ground for saying which of two glaringly unlike translations of individual sentences is right. For an uncritical mentalist, no such indeterminacy threatens. Every term and every sentence is a label attached to an idea, simple or complex, which is stored in the mind. When on the other hand we take a verification theory of meaning seriously, the indeterminacy would appear to be inescapable. The Vienna Circle espoused a verification theory of meaning but did not take it seriously enough. If we recognize with Peirce that the meaning of a sentence turns purely on what would count as evidence for its truth, and if we recognize with Duhem that theoretical sentences have their evidence not as single sentences but only as larger blocks of theory, then the indeterminacy of translation of theoretical sentences is the natural conclusion. And most sentences, apart from observation sentences, are theoretical. This conclusion,

74

Siblings Under the Skin

conversely, once it is embraced, seals the fate of any general notion of propositional meaning or, for that matter, state of affairs. Should the unwelcomeness of the conclusion persuade us to abandon the verification theory of meaning? Certainly not. The sort of meaning that is basic to translation, and to the learning of one’s own language, is necessarily empirical meaning and nothing more. A child learns his first words and sentences by hearing and using them in the presence of appropriate stimuli. These must be external stimuli, for they must act both on the child and on the speaker from whom he is learning. Language is socially inculcated and controlled; the inculcation and control turn strictly on the keying of sentences to shared stimulation. Internal factors may vary ad libitum without prejudice to communication as long as the keying of language to external stimuli is undisturbed. Surely one has no choice but to be an empiricist so far as one’s theory of linguistic meaning is concerned. What I have said of infant learning applies equally to the linguist’s learning of a new language in the field. If the linguist does not lean on related languages for which there are previously accepted translation practices, then obviously he has no data but the concomitances of native utterance and observable stimulus situation. No wonder there is indeterminacy of translation—for of course only a small fraction of our utterances report concurrent external stimulation. Granted, the linguist will end up with unequivocal translations of everything; but only by making many arbitrary choices—arbitrary even though unconscious— along the way. Arbitrary? By this I mean that different choices could still have made everything come out right that is susceptible in principle to any kind of check. Let me link up, in a different order, some of the points I have made. The crucial consideration behind my argument for the indeterminacy of translation was that a statement about the world does not always or usually have a separable fund of empirical consequences that it can call its own. That consideration served also to account for the impossibility of an epistemological reduction of the sort where every sentence is equated to a sentence in observational and logico-mathematical terms. And the impossibility of that sort of epistemological reduction dissipated the last advantage that rational reconstruction seemed to have over psychology.

Nelson and Quine

75

Philosophers have rightly despaired of translating everything into observational and logico-mathematical terms. They have despaired of this even when they have not recognized, as the reason for this irreducibility, that the statements largely do not have their private bundles of empirical consequences. And some philosophers have seen in this irreducibility the bankruptcy of epistemology. Carnap and the other logical positivists of the Vienna Circle had already pressed the term “metaphysics” into pejorative use, as connoting meaninglessness; and the term “epistemology” was next. Wittgenstein and his followers, mainly at Oxford, found a residual philosophical vocation in therapy: in curing philosophers of the delusion that there were epistemological problems. But I think that at this point it may be more useful to say rather that epistemology still goes on, though in a new setting and a clarified status. Epistemology, or something like it, simply falls into place as a chapter of psychology and hence of natural science. It studies a natural phenomenon, viz., a physical human subject. This human subject is accorded a certain experimentally controlled input—certain patterns of irradiation in assorted frequencies, for instance—and in the fullness of time the subject delivers as output a description of the three-dimensional external world and its history. The relation between the meager input and the torrential output is a relation that we are prompted to study for somewhat the same reasons that always prompted epistemology; namely, in order to see how evidence relates to theory, and in what ways one’s theory of nature transcends any available evidence. Such a study could still include, even, something like the old rational reconstruction, to whatever degree such reconstruction is practicable; for imaginative constructions can afford hints of actual psychological processes, in much the way that mechanical simulations can. But a conspicuous difference between old epistemology and the epistemological enterprise in this new psychological setting is that we can now make free use of empirical psychology. The old epistemology aspired to contain, in a sense, natural science; it would construct it somehow from sense data. Epistemology in its new setting, conversely, is contained in natural science, as a chapter of psychology. But the old containment remains valid too, in

76

Siblings Under the Skin

its way. We are studying how the human subject of our study posits bodies and projects his physics from his data, and we appreciate that our position in the world is just like his. Our very epistemological enterprise, therefore, and the psychology wherein it is a component chapter, and the whole of natural science wherein psychology is a component book—all this is our own construction or projection from stimulations like those we were meting out to our epistemological subject. There is thus reciprocal containment, though containment in different senses: epistemology in natural science and natural science in epistemology. This interplay is reminiscent again of the old threat of circularity, but it is all right now that we have stopped dreaming of deducing science from sense data. We are after an understanding of science as an institution or process in the world, and we do not intend that understanding to be any better than the science which is its object. This attitude is indeed one that Neurath was already urging in Vienna Circle days, with his parable of the mariner who has to rebuild his boat while staying afloat in it. One effect of seeing epistemology in a psychological setting is that it resolves a stubborn old enigma of epistemological priority. Our retinas are irradiated in two dimensions, yet we see things as three-dimensional without conscious inference. Which is to count as observation—the unconscious two-dimensional reception or the conscious three-dimensional apprehension? In the old epistemological context the conscious form had priority, for we were out to justify our knowledge of the external world by rational reconstruction, and that demands awareness. Awareness ceased to be demanded when we gave up trying to justify our knowledge of the external world by rational reconstruction. What to count as observation now can be settled in terms of the stimulation of sensory receptors, let consciousness fall where it may. The Gestalt psychologists’ challenge to sensory atomism, which seemed so relevant to epistemology forty years ago, is likewise deactivated. Regardless of whether sensory atoms or Gestalten are what favor the forefront of our consciousness, it is simply the stimulations of our sensory receptors that are best looked upon as the input to our cognitive mechanism. Old paradoxes about unconscious data and inference,

Nelson and Quine

77

old problems about chains of inference that would have to be completed too quickly—these no longer matter. In the old anti-psychologistic days the question of epistemological priority was moot. What is epistemologically prior to what? Are Gestalten prior to sensory atoms because they are noticed, or should we favor sensory atoms on some more subtle ground? Now that we are permitted to appeal to physical stimulation, the problem dissolves; A is epistemologically prior to B if A is causally nearer than B to the sensory receptors. Or, what is in some ways better, just talk explicitly in terms of causal proximity to sensory receptors and drop the talk of epistemological priority. Around 1932 there was debate in the Vienna Circle over what to count as observation sentences, or Protokollsätze (Carnap and Neurath, 1932). One position was that they had the form of reports of sense impressions. Another was that they were statements of an elementary sort about the external world, e.g., “A red cube is standing on the table.” Another, Neurath’s, was that they had the form of reports of relations between percipients and external things: “Otto now sees a red cube on the table.” The worst of it was that there seemed to be no objective way of settling the matter: no way of making real sense of the question. Let us now try to view the matter unreservedly in the context of the external world. Vaguely speaking, what we want of observation sentences is that they be the ones in closest causal proximity to the sensory receptors. But how is such proximity to be gauged? The idea may be rephrased this way: observation sentences are sentences which, as we learn language, are most strongly conditioned to concurrent sensory stimulation rather than to stored collateral information. Thus let us imagine a sentence queried for our verdict as to whether it is true or false; queried for our assent or dissent. Then the sentence is an observation sentence if our verdict depends only on the sensory stimulation present at the time. But a verdict cannot depend on present stimulation to the exclusion of stored information. The very fact of our having learned the language evinces much storing of information, and of information without which we should be in no position to give verdicts on sentences however observational. Evidently then we must relax our definition of observation sentence to read thus: a sentence is an observation

78

Siblings Under the Skin

sentence if all verdicts on it depend on present sensory stimulation and on no stored information beyond what goes into understanding the sentence. This formulation raises another problem: how are we to distinguish between information that goes into understanding a sentence and information that goes beyond? This is the problem of distinguishing between analytic truth, which issues from the mere meanings of words, and synthetic truth, which depends on more than meanings. Now I have long maintained that this distinction is illusory. There is one step toward such a distinction, however, which does make sense: a sentence that is true by mere meanings of words should be expected, at least if it is simple, to be subscribed to by all fluent speakers in the community. Perhaps the controversial notion of analyticity can be dispensed with, in our definition of observation sentence in favor of this straightforward attribute of community-wide acceptance. This attribute is of course no explication of analyticity. The community would agree that there have been black dogs, yet none who talk of analyticity would call this analytic. My rejection of the analyticity notion just means drawing no line between what goes into the mere understanding of the sentences of a language and what else the community sees eye-to-eye on. I doubt that an objective distinction can be made between meaning and such collateral information as is community-wide. Turning back then to our task of defining observation sentences, we get this: an observation sentence is one on which all speakers of the language give the same verdict when given the same concurrent stimulation. To put the point negatively, an observation sentence is one that is not sensitive to differences in past experience within the speech community. This formulation accords perfectly with the traditional role of the observation sentence as the court of appeal of scientific theories. For by our definition the observation sentences are the sentences on which all members of the community will agree under uniform stimulation. And what is the criterion of membership in the same community? Simply general fluency of dialogue. This criterion admits of degrees, and indeed we may usefully take the community more narrowly for some studies than for others. What count as observation sentences

Nelson and Quine

79

for a community of specialists would not always so count for a larger community. There is generally no subjectivity in the phrasing of observation sentences, as we are now conceiving them; they will usually be about bodies. Since the distinguishing trait of an observation sentence is intersubjective agreement under agreeing stimulation, a corporeal subject matter is likelier than not. The old tendency to associate observation sentences with a subjective sensory subject matter is rather an irony when we reflect that observation sentences are also meant to be the intersubjective tribunal of scientific hypotheses. The old tendency was due to the drive to base science on something firmer and prior in the subject’s experience; but we dropped that project. The dislodging of epistemology from its old status of first philosophy loosed a wave, we saw, of epistemological nihilism. This mood is reflected somewhat in the tendency of Polányi, Kuhn, and the late Russell Hanson to belittle the role of evidence and to accentuate cultural relativism. Hanson ventured even to discredit the idea of observation, arguing that so-called observations vary from observer to observer with the amount of knowledge that the observers bring with them. The veteran physicist looks at some apparatus and sees an x-ray tube. The neophyte, looking at the same place, observes rather “a glass and metal instrument replete with wires, reflectors, screws, lamps, and pushbuttons” (Hanson 1966). One man’s observation is another man’s closed book or flight of fancy. The notion of observation as the impartial and objective source of evidence for science is bankrupt. Now my answer to the x-ray example was already hinted a little while back: what counts as an observation sentence varies with the width of community considered. But we can also always get an absolute standard by taking in all speakers of the language, or most.1 It is ironical that philosophers, finding the old epistemology untenable as a whole, should react by repudiating a part which has only now moved into clear focus. Clarification of the notion of observation sentence is a good thing, for the notion is fundamental in two connections. These two correspond to the duality that I remarked upon early in this lecture: the duality between concept and doctrine, between knowing what a sentence means and knowing whether it is true. The observation sentence

80

Siblings Under the Skin

is basic to both enterprises. Its relation to doctrine, to our knowledge of what is true, is very much the traditional one: observation sentences are the repository of evidence for scientific hypotheses. Its relation to meaning is fundamental too, since observation sentences are the ones we are in a position to learn to understand first, both as children and as field linguists. For observation sentences are precisely the ones that we can correlate with observable circumstances of the occasion of utterance or assent, independently of variations in the past histories of individual informants. They afford the only entry to a language. The observation sentence is the cornerstone of semantics. For it is, as we just saw, fundamental to the learning of meaning. Also, it is where meaning is firmest. Sentences higher up in theories have no empirical consequences they can call their own; they confront the tribunal of sensory evidence only in more or less inclusive aggregates. The observation sentence, situated at the sensory periphery of the body scientific, is the minimal verifiable aggregate; it has an empirical content all its own and wears it on its sleeve. The predicament of the indeterminacy of translation has little bearing on observation sentences. The equating of an observation sentence of our language to an observation sentence of another language is mostly a matter of empirical generalization; it is a matter of identity between the range of stimulations that would prompt assent to the one sentence and the range of stimulations that would prompt assent to the other.2 It is no shock to the preconceptions of old Vienna to say that epistemology now becomes semantics. For epistemology remains centered as always on evidence, and meaning remains centered as always on verification; and evidence is verification. What is likelier to shock preconceptions is that meaning, once we get beyond observation sentences, ceases in general to have any clear applicability to single sentences; also that epistemology merges with psychology, as well as with linguistics. This rubbing out of boundaries could contribute to progress, it seems to me, in philosophically interesting inquiries of a scientific nature. One possible area is perceptual norms. Consider, to begin with, the linguistic phenomenon of phonemes. We form the habit, in hearing the myriad variations of spoken sounds, of treating each as an approximation to one or another of a limited number of

Nelson and Quine

81

norms—around thirty altogether—constituting so to speak a spoken alphabet. All speech in our language can be treated in practice as sequences of just those thirty elements, thus rectifying small deviations. Now outside the realm of language also there is probably only a rather limited alphabet of perceptual norms altogether, toward which we tend unconsciously to rectify all perceptions. These, if experimentally identified, could be taken as epistemological building blocks, the working elements of experience. They might prove in part to be culturally variable, as phonemes are, and in part universal. Again there is the area that the psychologist Donald T. Campbell calls evolutionary epistemology (Campbell 1959). In this area there is work by Hüseyin Yilmaz, who shows how some structural traits of color perception could have been predicted from survival value (Yilmaz, 1962; 1967). And a more emphatically epistemological topic that evolution helps to clarify is induction, now that we are allowing epistemology the resources of natural science.3

82

Siblings Under the Skin

Notes 1.

2. 3.

This qualification allows for occasional deviants such as the insane or the blind. Alternatively, such cases might be excluded by adjusting the level of fluency of dialogue whereby we define sameness of language. (For prompting this note and influencing the development of this paper also in more substantial ways I am indebted to Burton Dreben.) Cf. Quine, Word and Object (1960: 31-46; 68). See Quine, “Natural Kinds,” in Ontological Relativity and other Essays (1969).

References Campbell, D.T. 1959. Methodological suggestions from a comparative psychology of knowledge processes. Inquiry 2: 152-182. Carnap, R. 1936. Testability and meaning. Philosophy of Science 3: 419-471; 4: 1–40. Carnap, R. and Otto Neurath. 1932. Debate: Protokollsätze. Erkenntnis 3: 204-228. Hanson, N. R. 1966. Observation and interpretation. In S. Morgenbesser, ed., Philosophy of science today. New York: Basic Books. Johnson, A. B. [1836] 1947. A treatise on language. Berkeley. Quine, W.V. 1969. Ontological relativity and other essays. New York: Columbia University Press. ——. 1960. Word and object. Cambridge: MIT Press. Yilmaz, Hüseyin. 1962. On color vision and a new approach to general perception. In E. E. Bernard and M. R. Kare, eds., Biological prototypes and synthetic systems. New York: Plenum. ——. 1967. Perceptual invariance and the psychophysical law. Perception and Psychophysics 2: 533-538.

Nelson and Quine

83

Biographical note W.V. Quine was born in 1908 in Akron, Ohio. At the time of his death, Dec. 25, 2000, he was the Edgar Pierce Professor of Philosophy, Emeritus, at Harvard University. Quine attended Oberlin College as an undergraduate and completed his Ph.D. at Harvard under the direction of C.I. Lewis. Quine was among the most pre-eminent philosophers of the 20th century. He set the agenda for British-American philosophy from about 1950 through at least the 1980s. Philosophers working during that period, on any of the many topics Quine has touched upon, recognized a need to engage Quine’s views. Those who cling firmly to theories of language that countenance talk of the meaning of individual sentences feel obliged to offer a response to Quine’s rejection of the analytic/synthetic distinction and the notion of meaning as something distinct from empirical content. Those who hold some form of correspondence theory of truth and/or a commitment to the reality of macro, micro, or atomic or subatomic objects independent of the conceptual scheme we weave, often feel the need to address Quine’s arguments for holism and pragmatism. And those who hold firmly to the idea of the stability and transparency of language and meaning, are often compelled to address Quine’s arguments for the indeterminacy of translation. Quine’s legacy can be seen in the work of any number of his students, including Donald Davidson. A number of volumes have been devoted to exchanges between Quine and his students, admirers and critics, notably, Perspectives on Quine, edited by Robert Barrett and Roger Gibson (1990, Blackwell).

84

Siblings Under the Skin

Clough and Davidson

85

A Hasty Retreat From Evidence: The Recalcitrance of Relativism in Feminist Epistemology Sharyn Clough

In this essay I contribute to an ongoing debate about the nature of relativism and its presence in feminist critiques of traditional epistemology. I argue that an unnecessary level of relativism enters into some of the epistemological writings of three influential feminist theorists: Helen Longino, Evelyn Fox Keller, and Sandra Harding. However, my diagnosis of relativism is not meant to support the antifeminist views of some traditional epistemologists (i.e., those who argue that feminists introduce relativism into what would otherwise be a world of objective truth-telling).1 I take the strength of most feminist writings on epistemology to be the suggestion that we should dismantle the entire traditional epistemological project, including both objectivism and relativism. In this essay I highlight those areas where the dismantling is not yet complete. Beginning with an analysis of Longino’s “Can There be a Feminist Science?” (1987) and Science as Social Knowledge (1990), and proceeding to three essays from Keller’s collection Secrets of Life, Secrets of Death (1992a; 1992b; 1992c), I argue that each of these writings employs a similarly relativist use of the “underdetermination thesis.” This is the thesis, often associated with W.V. Quine, that every scientific theory is underdetermined by the evidence brought forward in its support, i.e., theoretically, any particular piece of evidence can be used to support an infinite number of theories. Conversely, for any theory that fits the available evidence, there may be another theory that fits the same evidence equally well (Quine 1981, 28-29). The relativist applications of the underdetermination thesis involve some version of the claim that because some scientific theories are chosen over others, these choices must, ultimately, be relative to a political or cultural “worldview,” “explanatory scheme,” or “interpretive framework”; there can be no objective adjudication on the basis of how any one theory simply “corresponds” to the empirical evidence.2 I show how

86

Siblings Under the Skin

this sort of relativism involves the conceptual splitting of the empirical evidence, on the one hand, from the filter of politics or culture, on the other—a split similar to that between “content” and “scheme” critically discussed by Donald Davidson (e.g., Davidson 1984). Davidson argues that the scheme/content split is indicative of a questionable representationalist model of knowers and the world. I support his interpretation and believe it to be an important diagnostic aid for feminist discussions of epistemology and science. I include some examples from Harding’s standpoint theory (Harding 1991; 1993) to further illustrate the pervasiveness of the representationalist model within feminist science critique and the subsequent pervasiveness of relativism. Finally, I offer a pragmatist reading of Davidson’s nonrepresentationalist alternative that would make feminist concessions to relativism unnecessary. Relativism: A necessary evil? Longino, Keller, and Harding have each taught us the importance of distancing our scientific discourse from the traditional epistemological project. Their struggles have shown us, however, that the inherited epistemological legacy is not easily displaced. The result, for feminists, and others critical of the tradition, is that unless we are fully distanced from our epistemological legacy, any moves critical of one end of the epistemological continuum (typically, objectivism) will involve an endorsement, however begrudging, of the other end (typically, relativism).3 This phenomenon is manifest in each of the feminist writings referred to above. In each of their arguments, Longino, Keller, and Harding begin by providing compelling criticisms of objectivism—roughly, the view that a true theory has a one-to-one correspondence with the evidence adduced in its support. For example, in her discussion of competing anthropological theories regarding the use of ancient chipped stones (specifically, the choice between the “woman-the-gatherer” or the “man-the-hunter” theory), Longino claims that the choice cannot be based on a one-to-one correspondence between one of the theories and the evidence because each theory has been influenced by an accompanying “interpretive framework” through which the data is screened

Clough and Davidson

87

(Longino 1990, 106-11). When the data is thus screened, she argues, it can then form evidence for the theory in question. Specifically, feminist or gynocentric interpretive frameworks screen the data as evidence for the woman-the-gatherer theory and androcentric frameworks screen data as evidence for the man-the-hunter theory. In this way, each theory is underdetermined, i.e., each theory can be supported equally well (or equally poorly) by the available evidence. Further, she writes, the historical nature of the competing theories compounds the underdetermination problem, so that “not only do we not now have evidence [that would definitively support one theory over the other] but we cannot have it” (Longino 1990, 111). Theory choice, she argues, must be relative to the political values that form the interpretive frameworks because there is no such thing as value-free evidence. Evidence, construed as value-free, cannot play the deciding role in the way objectivists thought it could. Similarly, Keller speaks of “abandoning the hope for a one-to-one correspondence with the real” (Keller 1992c, 73). She uses the underdetermination thesis to argue that the “real” actually corresponds to any number of theories and that “since nature is only accessible to us through representations and since representations are necessarily structured by language (and hence, by culture), no representation can ever ‘correspond’ to reality” (1992a, 5). Finally, Harding argues that certain aspects of culture, namely the social standpoint of the theorist, filter the correspondence between any one theory and the evidence gathered in support of that theory. Harding makes the Marxist claim that one’s social standpoint will “organize and set limits” on one’s understanding of the world (Harding 1993, 54). In other words, the choice of which theories of the world we take to be true will be relative, in some way, to our social standpoint. Harding does not put the issue in terms of the underdetermination of theories by evidence, but her criticism of the objectivist view of one-to-one correspondence between theory and evidence is remarkably similar to the criticisms offered by Longino and Keller. For each of these theorists, some level of relativism is presented as a necessary evil; an evil that must be accepted by critics of objectivism and applied consistently, not only to androcentric or sexist scientific theories, but also, reflexively, to those theories offered by feminist scien-

88

Siblings Under the Skin

tists and feminist critics of science. While I appreciate these criticisms of objectivism, I believe the elements of relativism that remain give too much away to opponents of feminism and sceptics of a more general philosophical sort. Longino, for instance, admits that her criticism of the possibility of direct correspondence between theory (or hypothesis) and evidence, relativizes what counts as evidence, and that “by relativizing what counts as evidence to background beliefs or assumptions, hypothesis acceptance on the basis of evidence is also thus relativized” (Longino 1990, 61). While Longino does not abandon objectivity, her contexualized account results in a restricted notion of “objectivity by degrees” that is relative to the social dynamics of various science communities (1990, ch. 4). Keller too, concedes that in the absence of any one-toone correspondence between representations and reality, our feminist decisions between representations should be made relative to the differing political “interventions” each representation affords (Keller 1992c, 76). Similarly, because Harding argues that all beliefs are filtered through the social standpoint of the believer, she disavows the claim that the standpoints of women or feminists will produce true beliefs about reality—just less partial, less distorted ones than those produced by “anti-liberatory interests,” for example (1991, 185, 149). While these lingering elements of relativism evoke a dangerous level of self-directed scepticism, they are also unnecessary. I believe that both relativism and the objectivist theory of one-to-one correspondence remain predicated on an unnecessary, and untenable, representationalist view of language users and the world.4 The representationalist metaphor in epistemology What is it about the epistemological continuum from objectivism to relativism that leads to trouble no matter what end one approaches? Feminist epistemologists have long been struggling with this question, but I find the diagnostic work of Richard Rorty and Donald Davidson to be particularly persuasive. Rorty has suggested that part of the problem results from the representational metaphors that frame the epistemological debate (e.g., Rorty 1991, 151-61). Discussants on both sides of the debate, he explains, view beliefs, sentences or theories as repre-

Clough and Davidson

89

senting the world. The acquisition of these representations is viewed as a filtering process. Here, in Davidson’s terms, our “language scheme,” “worldview,” or culture is described as a medium through which the empirical “content,” “sense-data,” or “the facts” of the external world are filtered (Davidson 1984). Hypotheses, or more formally, theories, are then viewed as the combination or systematization of representations. Sometimes the resulting theory is said to feed back into the filtering system, so that our allegiance to the theory affects our ability to perceive new data and to form new representations. My concern is that by invoking this filtering process, the representationalist invokes a metaphysical gap between the subjective end-product of belief and the objective, external reality the belief is about, a gap between mental “inner space” and the outside world.5 The ontological task presented by the representationalist model is the identification of those normative properties would that indicate the level of interference or filtering between the representations and the bits of the world to which they refer. Typically, these normative properties are identified as relational, such as the property of correspondence. A scientific theory that has the property of correspondence is one that successfully bridges, with little or no interference, the implied metaphysical gap between our inner subjective beliefs and objective external reality. Epistemology is then a process of adjudicating between representations, based on the detection of these normative relational properties. This process used to be motivated by a quest for certainty. Feminists (among others) were right to suggest that this epistemological goal needed to be scaled down (e.g., Harding 1991). It also seemed appropriate that we supported the shift toward naturalistic accounts that replaced a priori theorising about the detection of truth and evidence (e.g., Lynn Hankinson Nelson 1990). Other changes consistent with feminist politics included the development of a more holistic approach to the subjects of epistemological theorizing, focusing on human beliefs in their natural “ecological” settings of larger theories and worldviews, rather than on the abstract “S knows that p” model (e.g., Lorraine Code 1991). However, while we have thus restrained and restructured the traditional quest for certainty, I believe that these differences reflect changes in degree more than kind. Epistemology, in both feminist and

90

Siblings Under the Skin

traditional versions, remains an attempt to specify, either a priori, or through a naturalised account of human cognition, the property of theories, sentences or beliefs that makes them true, or least partial, or maximally objective. Despite the qualifiers, the key similarity is the epistemological specification of a normative property, the detection of which will help us choose between competing knowledge claims.6 Even when this property is characterized as “justified by the empirical evidence” as some empiricists have maintained, this philosophical specification remains separate from the everyday ad hoc question whether any particular knowledge claim is justified by the evidence. Epistemology is a required response, given representationalist metaphysics. The metaphysical independence or gap between the inner subjective stuff of mind and the external objective reality makes coherent the worry that the two worlds might not be bridgeable—all of our subjective theories about external reality “might be just as they are and yet reality—and so the truth about reality—be very different” (Davidson 1990, 298). In other words, the independence is such that all of our theories would “float free” of the bits of the world they purported to describe, unless securely anchored via the normative relational properties. When one conceives of such a gap between representations and the world represented, there is always the possibility of massive error in the representations (i.e., it becomes conceivable that all of our bridgework could be completely undependable). This is the worry of global scepticism, so clearly articulated by the proto-representationalist, Descartes, in his theory of mind/body dualism. Epistemological responses to scepticism range from objectivism to relativism, with instrumentalism, or empiricism playing a moderate role somewhere between the two. A brief examination of these familiar positions, in terms of their less-familiar representationalist commitments, follows. In representationalist terms, objectivism is the view that scepticism about the truth of our representations can be defeated—that we can delineate a priori the criteria for judging whether or not our representations have the requisite relational property. In other words, the claim goes, we can tell if and when the bridgework between our subjective inner space and the objective outer reality is dependable. “Objectivity” names the prescribed approach to the detection of truth. If we are objective, if we stand apart from the filters of all our subjective theories, if

Clough and Davidson

91

we enlist the help of other objective observers similarly placed, then we can clear the bridgeway between our sensory receptors and the causal forces of the empirical data. We can tell whether a particular theory is in a correspondence relation with the external data the theory purports to describe.7 Instrumentalists and empiricists take the objectivist views to be generally coherent but false for a specified range of claims. More specifically, they disagree with the traditional objectivist view about the possibility of identifying relational properties that would hold between the theoretical or unobservable elements of any given theory and the external world the theory represents. But they typically agree with the objectivist that a normative identification of relational properties is possible, and necessary, as long as the focus is restricted to the observable or otherwise empirically accessible elements of the theory.8 Those who have moved well away from objectivism but who stay within the representationalist framework of the debate, end up with a relativism that resigns them to doubt, at a very general level, the existence of any firm causal relationships between their theories and the world. At this relativist end of the continuum, we find the tacit claim that if we are critical of the objectivist notion of relational properties linking our representations to the world, then we are left with the position that links can be made only between representations themselves. Our subjective filtering of the external world is so opaque that criteria for adjudicating between representations—criteria such as truth—can only be said to be relative to our interests, our politics, our worldviews, and not to the world. At best, we can attain “maximal” objectivity, or “least partiality.” According to relativists, the metaphysical gap between our representations and the world represented remains unbridged, or at least any bridgework we construct is irredeemably blocked by the filters of our worldviews and conceptual schemes. I believe that some version of this claim underlies the relativist use of the underdetermination thesis that theory-choice cannot be made strictly on the basis of evidence; that evidence for theory choice is screened through “interpretive,” “linguistic,” or “cultural” frameworks. In the next three sections I will highlight this sort of relativism in the writings of Longino, Keller, and Harding.

92

Siblings Under the Skin

Helen Longino’s feminist science In her essay “Can There Be a Feminist Science?” (1987) Longino previews the major themes of her book Science as Social Knowledge: Values and Objectivity in Scientific Inquiry (1990). One of these themes is the debate over the criteria for what makes feminist science “feminist.” Longino argues against some feminist accounts that equate objectivity with value-free scientific method (1987, 60). Her negative project seems to me to be largely correct. My criticisms concern her positive articulation of science (including feminist science). Here, Longino explains that objective, good science is always biased with the “contextual values” of our “interpretive frameworks,” and these, in turn, guide our observations. For example, the interpretive frameworks of theories in particle physics are necessary to guide observations of elementary particles in cloud chambers (Longino 1987, 54). The observations are not “given,” they are guided or filtered by our interpretive frameworks and prior value commitments. Much of science, Longino claims, is guided by interpretive contextual values so feminist diagnoses of bad, unobjective science as that science which is “biased by contextual values,” will not fully capture the problem (1987, 56). What we need to do, she writes, is to redefine objectivity as that which allows us to better examine the influence of these interpretive frameworks and values. She explains: We cannot restrict ourselves simply to the elimination of bias, but must expand our scope to include the detection of limiting and interpretive frameworks and the finding or construction of more appropriate frameworks. We need not, indeed should not, wait for such a framework to emerge from the data (Longino 1987, 60).

On her model, the prescribed focus for feminist work in science then becomes a search for better conceptual filters (or schemes), as distinct from, and more coherent than, the search for better evidence (or content). Note, also, the scheme/content split in her explanation that the data of the external world are “dumb” and that it is only through subjective conceptual filters that they are given voice as evidence for a particular hypothesis or theory (Longino 1990, 111). The problem here is that when one conceives of a split between an inner conceptual world of values and interpretive frameworks and

Clough and Davidson

93

an outer world of unanalyzed data one invites an unanswered (and unanswerable?) scepticism about the relationship between evidence and theory. While Longino is surely correct that evidence of elementary particles, for example, is not simply “given,” I do not think that we then have to redirect our project away from questions of evidence and toward some nonevidential investigations of interpretive frameworks. On the Davidsonian model I prescribe, interpretive frameworks are holistically of a piece with other evidential considerations. The positive implications of such holism are described below. Longino continues her argument by outlining two sorts of values that, she claims, are part of even the best science: those values that are constitutive of scientific practice; and those that affect the context in which science is practiced (1987, 54). The constitutive values govern “what constitutes acceptable scientific practice” (Longino 1987, 54). The contextual values are the background values, explanatory schemes, or political commitments that each scientist might bring to her laboratory. Against more traditional philosophers of science, and paralleling, to some extent, the work of Thomas Kuhn (1970), Longino argues that the second set of values—the contextual values—play an active role not only in what some call the context of discovery, but also in “the inner workings of scientific inquiry” or the context of justification (Longino 1987, 54). These contextual values have a similar function to that of “paradigms” in Kuhn’s writings.9 Contextual values must play a role in scientific justification and theory-choice, she argues, because we cannot choose theories on the basis of the evidence alone. Given that theories are underdetermined, theory-choice, even “objective” theory-choice, is always based on something more than evidence (Longino 1987, 54-56). Thus, Longino explains, there is “no formal basis for arguing that an inference [from data to theory that is] mediated by contextual values is thereby bad science” (1987, 55). It could be bad science, but the presence of contextual values is not the deciding factor. Indeed, she argues, the influence of contextual values in “the inner workings of science” can be part and parcel of good science as usual (1987, 56). From here she proceeds to describe feminist science and feminist science criticism—two obvious sites of contextual values at work—as good science as usual. She claims that feminist scientific practice will be good, objective science insofar as it “admits political considerations as relevant con-

94

Siblings Under the Skin

straints on reasoning, which, through their influence on reasoning and interpretation, shape content” (Longino 1987, 61). As I have suggested earlier, I believe this formulation to be unnecessarily weak and will prescribe, instead, a nonrepresentationalist approach whereby feminist political considerations are viewed as further elements of evidential reasoning, rather than as nonevidential “constraints on reasoning.” There are a number of other representationalist elements in Longino’s move toward the relativist corollary of the underdetermination thesis. For example, illustrating how the contextual values of our worldview play a role in science, Longino discusses the role of feminist and non-feminist background assumptions in the “woman-the-gatherer” versus “man-the-hunter” interpretations in anthropology and in the selection of interactionist versus linear models used to “mediate data” within sex-hormone research (Longino 1990, chs. 6-7; 1987, 58). In the latter case, the interactionist model of the influence of sex hormones highlights the two-way interaction between the presence of prenatal hormones and resulting physiological changes at the pre- and postnatal cellular and macro levels. Linear models focus on a more deterministic, one-way relationship where the prenatal presence of hormones are assigned all, or most, of the causal power, with little, or no causal attention given to the feedback from the rest of the system either pre- or postnatal. Examining the two competing hormonal theories, Longino, and her research partner Ruth Doell, found sexism and androcentric bias in many aspects of the linear theory research, though they did not find that sexism or androcentric bias affected the inferences from data to theory in any straightforward way (Doell and Longino 1988). Instead, they claim that the inferences were affected at a deeper level by prior commitments to patriarchal political ideals, which, in turn, affected commitments to the linear explanatory model. Inferences from data to theory within the level of the linear explanatory model were found to be sound. Longino describes the patriarchal contextual values, and the linear explanatory model associated with them, as screens or filters in the process of scientific justification. For example, she writes, “In the conduct of research [explanatory models] serve as background assumptions against which data are ordered, in light of which data

Clough and Davidson

95

are given status as evidence for particular hypotheses and as a context within which studies gain significance” (1990, 135). Objectionable, patriarchal politics favored commitments to the linear hormonal explanatory models, which, in turn resulted in a one-way, deterministic view of prenatal hormones in control of adult human behaviour. Longino describes the linear model as a patriarchal, hierarchical view of human behaviour that limits understanding of “human capacities for self-knowledge, self-reflection, [and] self-determination” (Longino 1987, 58). Longino prescribes the nonlinear, interactionist model instead, because self-knowledge, self-reflection, and self-determination are part of a feminist political vision or worldview (1987, 59). She then concludes that an interactionist model should be chosen by feminists, “because of explicitly political considerations” (1987, 61). Here, again, “political considerations” seem to be distinct from evidential considerations, just as the representationalist model distinguishes “scheme” from “content.” Leaving aside, for the moment, the problematic nature of the split between feminist politics and evidence, there does, indeed, seem to be evidence (however conceived) that the interactionist model is better than the linear model.10 Even Longino writes that the interactionist model “allows not only for the interaction of physiological and environmental factors but also for the interaction of these with a continuously self-modifying, self-representational (and self-organizing) central processing system”—something that the linear model cannot do (1987, 58). But, says Longino, this is not enough. “Obviously model-choice is also constrained by (what we know of ) reality, that is, by the data. But reality (what we know of it) is, I have already argued, inadequate to uniquely determine model choice” (1987, 61). My sense is that Longino’s use of the hedge “what we know of reality” is the sort of scepticism made coherent but unanswerable by the metaphysics of the representationalist model. It is the scepticism that results from conceiving of a metaphysical gap between the raw data of the world, out there waiting, and our organizing schemes primed to filter the waiting data. The organizing filters of feminism or androcentrism block unmediated knowledge of reality, serving as preconceived explanatory frameworks that organize the raw data of sex hormones, for example. In Longino’s words, again, explanatory

96

Siblings Under the Skin

models “serve as background assumptions against which data are ordered, in light of which data are given status as evidence” (1990, 135). For all Longino knows, all our theories might be floating free of the real world because they are part of our subjective representational system of explanatory frameworks or political worldviews, that filter reality. Keller’s arguments contain similar concessions to scepticism. Keller’s Secrets of Life, Secrets of Death In the introduction to her essay collection, Secrets of Life, Secrets of Death, Keller argues that objective method cannot be that which distinguishes theories based on ideology or myth from theories based on truth (1992a, 4). She describes this more critical view of objectivity, as “my ‘linguistic turn,’” which, she continues, “represents a shift from my earlier preoccupation with the frailties of description, and in one respect at least, a departure from my initial confidence in the possibility of identifying certain beliefs as ‘myth-like,’ as distinct from other beliefs that are, by implication, ‘myth-free.’ Such a notion now seems to me suspiciously reminiscent of the old demarcation between ‘truth’ and ‘ideology’, or between ‘good science’ and ‘value-laden science,’ demarcations that are themselves residues of the copy theory of truth” (Keller 1992a, 4-5). In representationalist terms, Keller no longer believes that objective method involves identifying which theories have the truth-conferring property of correspondence and which are based merely on ideology. In another essay in the collection, “Critical Silences in Scientific Discourse,” she writes of “abandoning the hope for a one-to-one correspondence with the real” (Keller 1992c, 73). However, it seems throughout these essays that her abandonment of objectivist correspondence is based on what she sees as a failure of execution and not necessarily of conception. Correspondence remains the ideal sort of relation we need to bridge the metaphysical gap between the world and us. The problem, says Keller, is that we simply cannot construct successful bridges, because they are always blocked by the influence of cultural conceptual schemes. Where Longino writes of political inclinations and explanatory frameworks, Keller writes more generally of culture—but in both cases a conceptual scheme is invoked. Keller explains further that “Since nature is only accessible to us

Clough and Davidson

97

through representations and since representations are necessarily structured by language (and hence, by culture), no representation can ever ‘correspond’ to reality” (1992a, 5). Again, for the representationalist, this criticism of objectivism inevitably leads to some version of relativism. The only option is to view our representations as filtered products of our subjective language scheme or culture. The representationalist is then faced with the question: how are we to choose between subjective representations if none has the objective, truth-conferring property of one-to-one correspondence with the external world? Employing the relativist view of the underdetermination thesis, Keller’s response is that we should choose those representations that facilitate certain “interventions.” Specifically, we should choose those interventions that best suit our feminist political goals. In the following passage Keller explains the options she believes this sort of conceptual relativism leaves for feminists: Since it is demonstrably possible to envision different kinds of representations, we need now to ask what different possibilities of change might be entailed by these different kinds of representation? For this, we need to understand the enmeshing of representing and intervening, how particular representations are already committed to particular kinds of interventions. Is there, for instance, a sense in which we might say that the program of modern genetics already has, written into its very structure, a blueprint for eugenics? Or that nuclear weapons are prebuilt into the program of nuclear physics? And if so, what kinds of theories of the natural world would enable us to act on the world differently? (Keller 1992c, 76).

In the above quotation, Keller makes the representationalist acknowledgment that subjective linguistic filters play an instrumental role in our choice of theories. However, she is still concerned to acknowledge the flip side of the representationalist scheme/content coin, namely, the role of the “nonlinguistic” realm—the objective reality the theories describe. For example, in the first essay in the collection, “Gender and Science: An Update,” Keller writes that for feminist critics who take the objective success of science seriously, the new task is to answer the question “How do ‘nature’ and ‘culture’ interact in the production

98

Siblings Under the Skin

of scientific knowledge?” (1992b, 36). But discovering how these two metaphysically distinct realms interact becomes as much of a sceptical problem for Keller as it was for Descartes. Keller is in good company when she is unable to provide a compelling answer. Despite her switch from objectivist searches for truth to instrumentalist searches for success, the representationalist elements remain, as do the scepticism and relativism. Harding and feminist standpoint theory Paralleling Keller’s and Longino’s views on the underdetermination thesis, Sandra Harding’s work on feminist standpoint theory is critical of the claim that objective method consists in detecting a one-to-one correspondence between true representations and the world (Harding 1991; 1993). But, again, in parallel with Longino and Keller, Harding does not fully dismantle the representationalist model, rather she focuses her criticism on the clarity of the correspondence relation. In “Rethinking Standpoint Epistemology: What is ‘Strong Objectivity’?” (1993), Harding explains her commitment to the general tenets of standpoint theory: The starting point of standpoint theory—and its claim that is most often misread—is that in societies stratified by race, ethnicity, class, gender, sexuality, or some other such politics shaping the very structure of a society, the activities of those at the top both organize and set limits on what persons who perform such activities can understand about themselves and the world around them. ... In contrast, the activities of those at the bottom of such social hierarchies can provide starting points for thought—for everyone’s research and scholarship—from which humans’ relations with each other and the natural world can become visible. This is because the experience and lives of marginalized peoples, as they understand them, provide particularly significant problems to be explained or research agendas (Harding 1993, 54; italics in original).

In Whose Science? Whose Knowledge? (1991) Harding argues that while it is true that every social standpoint filter “organizes and sets

Clough and Davidson

99

limits” on understanding, i.e., every filter provides only a partial representation of reality, not all social standpoints generate equally partial representations or beliefs. The social standpoints of women, or feminists with “maximally liberatory social interests,” for example, “have generated less partial and distorted beliefs than others” (Harding 1991, 144, 148). She explains: The history of science shows that research directed by maximally liberatory social interests and values tends to be better equipped to identify partial claims and distorting assumptions, even though the credibility of the scientists who do it may not be enhanced during the short run. After all, anti-liberatory interests and values are invested in the natural inferiority of just the groups of humans who, if given real equal access (not just the formally equal access that is liberalism’s goal) to public voice, would most strongly contest claims about their purported natural inferiority. Antiliberatory interests and values silence and destroy the most likely sources of evidence against their own claims. That is what makes them rational for elites (Harding 1991, 148-49).

As with Longino and Keller, Harding rightly criticizes the traditional epistemological view of objectivism. The “value-free” approach of objectivism, she argues, results in a “semi-science” that “turns away from the task of critically identifying all those broad, historical social desires, interests, and values that have shaped the agendas, contents, and results of the sciences much as they shape the rest of human affairs” (Harding 1991, 143). Harding prescribes, instead, “strong objectivity” that extends the idea of scientific research “to include systematic examination of…powerful background beliefs” thereby “maximizing objectivity” (1991, 149). However, paralleling Longino in particular, Harding continues with the representationalist metaphor by characterizing strong objectivity as the critical examination of linguistic or social filters, “the powerful background beliefs” that continually block our knowledge-seeking of the nonlinguistic, natural realm. We cannot get at objective representations so the best we can do is to search for better conceptual filters. The worry of philosophical scepticism that results from scheme/content relativism begins to appear in Harding’s work, just as it appeared in the writings of Longino and Keller.

100

Siblings Under the Skin

For example, because she acknowledges that all beliefs have a social filter, Harding disavows the claim that the standpoints of women or feminists will produce true beliefs (1991, 185, 149). While she purchases some consistency by claiming that all knowledge is somehow distorted, this sceptical claim robs her of the foundation she then needs to argue her thesis—namely that the knowledge produced from maximally liberatory social standpoints is less distorted, generally, than that produced from others. Even as Longino, Keller, and Harding rightly reject the claims that objective method involves impartial (value-, social-, culture-free) detection of one-to-one correspondence, none seems to fully critique the metaphysical gap of representationalism that correspondence sets out to bridge. They argue that our subjective conceptual schemes filter our gathering of evidence (our belief acquisitions) so that, on Longino’s and Harding’s analysis, in particular, objective method (“objectivity by degrees” for Longino; “strong objectivity” for Harding) is simply the least subjective method for judging which conceptual schemes, filters, or interpretive frameworks make for the least opaque filters between us and the world. Unfortunately, accepting the metaphysical gap between our theories and the world, while criticizing the ability of correspondence to bridge that gap, makes global scepticism a concern. All of our representations could be floating free of the world, to varying degrees. If this is the case, then, Longino, Keller, and Harding (and critics of feminism) are right, we must concede a certain amount of relativism. When feminist scientists and science commentators choose between representations that are underdetermined by evidence, our decision can be made only on the basis of our feminist political interpretive frameworks. But wait. We shouldn’t give up on the potentially decisive role of evidence just yet. While feminist scholars, and others, have shown that correspondence doesn’t bridge the metaphysical gap, relativist resignation is not our only other option. We need to more fully deconstruct the traditional epistemological project by dismantling the representationalist metaphor of the gap. So says Davidson, or so, at least, I have been hinting.

Clough and Davidson

101

A Davidsonian prescription Davidson makes a number of points against the representationalist model that informs the epistemological debates between objectivists and relativists. For example, paralleling the work of many feminist critics of epistemology, he argues against the claim that the objective detection of sensory data can be used to justify or stand as evidence for beliefs that represent those data.11 Davidson notes that for the justification process to work, we have to be aware of the detection of sense data, and this awareness is simply another belief. His argument undercuts the objectivist attempt to construe awareness of sensory data as an evidential entity that stands independent from our beliefs. It might seem, however, that in revealing the incoherence of harnessing sensations as independent evidence, Davidson has removed any justificatory scheme for our empirical beliefs. This seems to leave us with the scepticism encountered by Longino, Harding and Keller, a scepticism that Davidson’s nonrepresentationalist model is supposed to avoid. If explanations appealing to the sensory origins of our beliefs do not justify those beliefs, how do we know that we are not globally mistaken about the world? In this section I introduce Davidson’s “radical interpreter” as a heuristic device that provides a “reason for supposing most of our beliefs are true that is not a form of evidence” (Davidson [1986] 1991a, 127). It is important to make clear that the term “most” in the above quotation is not meant as a quantificational claim guaranteeing, for example, that a certain number of our beliefs must be true. Rather, Davidson uses the concept of the radical interpreter to support a philosophical claim, namely the claim that the detection of false beliefs requires that we have a background of true beliefs against which the error of the false beliefs can be measured. This latter claim undercuts the global sceptic who wants to make error a general concern, i.e., who wants to deny or question the existence of norms against which errors can be measured and detected. The “radical interpreter”—an adult interpreter faced with a completely foreign language—is an idealized concept Davidson borrows from Quine. Quine introduced the character in his explanation of how we would have to proceed to learn a completely foreign language when

102

Siblings Under the Skin

no “translation manual” is available (e.g. Quine “Ontological Relativity” 1969). I will argue that if we analyze meaning from the perspective of the radical interpreter, a whole host of traditional epistemological problems can be set aside. Davidson equips the radical interpreter with the abilities of a competent adult speaker of a language. Parachuted into the midst of a foreign land, she has general expectations about how to proceed. She has a sense of basic logical structure (i.e., she understands the implications of those elements of a language [“and,” “if…then,” etc.] that give the sentences that contain them their particular logical form). She also has the ability to discern when the speakers of the foreign language are making assertions, that is, expressing, in the form of sentences, beliefs held true (even though, in the beginning, she has no idea what those sentences mean). Davidson notes that, in order to make any progress in her new world, the radical interpreter must watch for correlations between types of sounds uttered by the native speakers and the kinds of events in their shared world that caused the utterances. In the beginning this is all she has to go on. She does not have any preconceived notion of the particular semantic role that is played by any particular noises uttered by the native speakers. Rather, at this early stage, it is the radical interpreter’s successful (accurate) identification of the environmental reference that prompted the native speakers’ noises, which provides those noises with semantic content in the first place. For example, the interpreter’s understanding of the meaning of the native speaker’s utterance “There’s the bus!” is provided by the shared causal relationship between the arrival of a bus in the visual (or aural) fields of the interpreter and the native speaker, and the native speaker’s utterance.12 The foreign noises that express basic or simple beliefs, in sentences such as “There’s the bus!” are the starting points for the radical interpreter. These basic beliefs are expressed in what Quine called “occasion sentences” (Quine 1960). Occasion sentences are so-named because their truth values change depending on precise, salient variables such as the time and place the sentences are uttered and who utters them. The truth of the sentence “There’s the bus!” for example, will depend on the presence of a bus at the time the sentence is uttered. For these

Clough and Davidson

103

“basic” beliefs expressed in occasion sentences, it is possible for the radical interpreter to make an educated guess about the truth conditions of the native utterance, because she has such immediate access to the truth values of her guesses. Quine contrasts these occasion sentences with “standing sentences,” such as “There have been some buses.” These latter sentences will be true depending on much more general variables, such as the presence of buses at any number of times prior to the occasion “There have been some buses” is uttered. What makes occasion sentences, as opposed to standing sentences, the basic entry points for the radical interpreter is not the epistemic simplicity of the terms involved in the sentences (as the empiricist might claim), but the relative ease with which a non-native speaker can guess the truth conditions of the native occasion sentences.13 The causal triangular relationship between the interpreter, the native speakers’ utterances of occasion sentences, and the objects and events in their world, requires that the interpreter assume the natives are speaking truthfully about their beliefs. Of course, while the adult language user has the ability to recognize when a native speaker is making an assertion, this recognition does not guarantee that the native speaker’s assertion is true. But, says Davidson, at the beginning, the radical interpreter must assume that the native speaker’s assertions are true. For interpretation to occur she must assume that the same relation between belief and truth holds for those she interprets, as for herself—what Davidson and Quine have called “the principle of charity.” In other words, starting with the most simple utterances such as “There’s the bus!” the radical interpreter must assume that she and the native speakers agree about what would make those utterances true (e.g., the presence of a bus). Why is this agreement necessary at the beginning when the interpreter is collecting sentences in the native language and correlating them with the sorts of environmental conditions that prompted the sentences? It is necessary, says Davidson, because in order to identify her teachers as having any beliefs, she must assume the beliefs they hold are true. Once she has established an empirical base of correlations between their sentences and hers, then she can start to make judgments of inconsistency and falsehood. Before that point, identifying her teachers’

104

Siblings Under the Skin

.

beliefs as false would deplete the empirical base from which she needs to begin her interpretative project in the first place. As one Davidson commentator explains, assigning “too much falsity among beliefs undermines the possibility of identifying beliefs at all” (Jeffrey Malpas 1992, 159). Identifying falsehoods and misconceptions, is “parasitic” on an established coordinate of shared meaning.14 We are getting closer, then, to explaining Davidson’s “anti” sceptical claim about the necessity of having true beliefs for the identification of false beliefs. It might still be unclear, however, why the existence of a “shared coordinate of meaning” between the native speaker and the radical interpreter, guarantees, in Davidson’s words, that “it cannot happen that most of our plainest beliefs about what exists in the world are false” (1991b, 195). Just because there must be agreement between the radical interpreter and the native speakers’ about the truth of basic beliefs, does not guarantee that those beliefs are, in fact, true. Davidson responds by examining the concept of truth itself. Where, he asks, do we come up with the concept of objective truth? The answer is in shared language. “Unless a language is shared there is no way to distinguish between using the language correctly and using it incorrectly; only communication with another can supply an objective check” (Davidson 1991c, 157). And communication with another can only start by assuming agreement on what makes utterances true—the principle of charity. Davidson’s apologists note that the principle of charity is unfortunately named, because it does not operate as advice that we could choose to follow or not (see Bjørn Ramberg 1989; Malpas 1992). Ramberg emphasises this point: “The principle of charity, ...offers no advice to us as interpreters, it yields no interpretational strategy. It is not a heuristic device, nor is it, accordingly, something we could get by without; it is a condition of the possibility of interpretation” (Ramberg 1989, 74, italics in original). If the principle of charity is a precursor for successful interpretation, this means that truth must be held primitive for words and sentences to be meaningful. This takes us back to the example of the radical interpreter correlating environmental circumstances with basic native utterances, e.g., “There’s the bus!” The radical interpreter has no initial preconceptions about how to link a native utterance with specific semantic content. Rather, her attention to the correct (true) reference

Clough and Davidson

105

of the native sentence is what provides her with clues to the meaning of the utterance, in the first place. The meaning of an utterance is given by its truth conditions, and not the reverse. Davidson uses these points about the radical interpreter to support his extensionalist claim that in the simplest cases of beliefs, i.e., those expressed in occasion sentences, the events and objects that cause those beliefs (the extension of the beliefs) also determine their contents, or meaning (the intension of the beliefs) (Davidson 1989a, 164; 1989b; [1986] 1991a; 1991b, 195). This means that in the simplest cases, there cannot be wholesale slippage between our understanding the meaning of a sentence and our understanding of the conditions that would make that sentence true. Davidson describes this approach to meaning further, in the following passage: “As long as we adhere to the basic intuition that in the simplest cases words and thoughts refer to what causes them, it is clear that it cannot happen that most of our plainest beliefs about what exists in the world are false. The reason is that we do not first form concepts and then discover what they apply to; rather, in the basic cases, the application determines the content of the concept” (Davidson 1991b, 195). Davidson’s extensionalist approach to meaning excludes the possibility that the speech of the radical interpreter could be, in principle, indistinguishable from her teachers and idiosyncratic with respect to meaning. In the simplest cases of beliefs expressed in occasion sentences, the meaning of her utterances is determined by their being used correctly in the presence of another speaker and the event in the world that caused the utterance. Taking a holistic approach to build from the simpler cases of beliefs, to beliefs expressed in more complex theories, any idiosyncrasies in the radical interpreter’s meaning are, in principle, available for her correction through a purely extensional examination of how she has applied her references. Somewhere along the line, any discrepancies can, in principle, be revealed. There is no subjective “inside” to her beliefs that is metaphysically separate and inaccessible from the viewpoint of the native speakers in the objective, outer world. For example, if the radical interpreter has interpreted “She’s candid,” in the native language, as “She’s rude” in her own language, the difference in meaning between the two sentences could, in principle, be revealed to her. The two words “rude” and “candid” are linked in a web-

106

Siblings Under the Skin

like fashion to different, simpler concepts, which in turn have different causes. The two utterances are correctly applied on different occasions; this is what gives them different meanings. Using the model of the radical interpreter, Davidson’s causal analysis of belief provides us less-than-radical interpreters with a presumption in favour of the truth of any particular belief. However, a presumption is not a guarantee. He cheerfully admits that the truth of each belief is up for grabs, though not all or even most of these beliefs can be up for grabs at once. It is the veridicality of beliefs generally, as understood through his causal account, that makes “meaningful disagreement” over particular beliefs possible (Davidson 1984, 196-197). Our beliefs have no content unless we have established a common convergence between ourselves, another speaker (or speakers), and a shared environmental stimulus. Occasion sentences provide the entry points for this convergence. Once we have established a pattern of successful convergence, a pattern of semantic “firmness,” then we can say of any particular belief that it is false. You have to be right about a large background of beliefs before you can critically examine the validity of particular ones. Similarly, successful communication with others indicates that you know many things about your world (Davidson 1989a; 1990, sec. III). We now have a way to explain how, in Davidson’s view, scepticism does not arise as a coherent option that needs epistemological attention. Davidson does not show that global scepticism is wrong, he simply argues that on the model of the radical interpreter, a metaphysical gap between language users and the world is unthinkable. Recall that on the representationalist view, beliefs are conceived as an “inner” non-natural, subjective representation of the outer, natural realm. In contrast, Davidson asks us to try viewing belief as the production of a triangular causal relationship between three naturalised entities, namely, ourselves, other speakers, and our shared environment. From the perspective of the radical interpreter, our ability to use language comes from direct, unmediated, causal contact with the world, which, in turn, guarantees that we have an established background of true beliefs against which our false beliefs can be measured. As Davidson writes, “communication begins where causes converge” ([1986] 1991a, 132). If we want to doubt in a wholesale, global fashion the causal etiology of our beliefs, we must also “give up language” (Ramberg 1989, 97).

Clough and Davidson

107

Davidson on underdetermination theory Applying Davidson’s model of language use, we are cautioned against the metaphysical bifurcation of inner, subjective, political reasons for scientific beliefs from external, objective, evidential reasons for these beliefs. This advice is particularly relevant for addressing the problems that Longino and Keller encountered when they prescribed a relativist version of the underdetermination thesis, and for Harding when she made similar representationalist claims about the filtering of social standpoints. Longino and Keller argued that the underdetermination of theory by evidence means that, because interpretive frameworks or cultural worldviews filter any evidence brought forward in support of a theory or hypothesis, we cannot choose between theories or hypotheses on the basis of evidence. Adjudication can only be made on the basis of our political values. However, this construal presumes the representationalist view that the “evidence” and our feminist “political values” emanate from two metaphysically separate spheres—the first from the objective, external world; the second from the subjective, internal mind (or minds). The “evidence” is construed as providing independent (objective) support for a theory, while political values are viewed as dependent and subjective. In response to this representationalist claim about the belief-independence of empirical evidence, Davidson reminds us that when we marshal empirical evidence in support of a belief or theory, we need first to be aware of the empirical evidence, and that awareness is itself another belief. In the project of marshaling epistemic justification for our individual beliefs there is no independent, “non-belief ” entity to which we can appeal. The evidence for a belief must itself be a belief. It is also important to see that both our political values and our more straightforwardly empirical commitments are beliefs of this evidential sort. On Davidson’s model even our (feminist) political beliefs must have some web-like relation to empirical evidence, if they are to have any content. There are a number of ways in which feminist political values can interact with and support the more straightforwardly empirical commitments that, together, make up our growing web of beliefs (e.g., our beliefs critical of sexism and oppression in science). For example,

108

Siblings Under the Skin

recall Longino’s particular discussion of the role of political values in choosing between competing archaeological interpretations of chipped stones. One theory, highlighting the role of the male hunter, interprets the stones as hunting tools. The other, competing theory, highlights or includes the role of the female gatherer and interprets the stones as implements for gathering and preparing edible vegetation. According to Longino, the available evidence supports both theories equally well, so the choice between the male-focused model or the more inclusive, female model must be relative to an underlying political commitment, namely to androcentrism or feminism, respectively (Longino 1990, 109). I argue, instead, that feminist political values are themselves beliefs with empirical content that can, in turn, provide good evidential reasons for rejecting the man-the-hunter interpretation. Our rejection of this interpretation does not need to be construed as relative to the non-evidence world of feminist politics. For example, feminist political analysis of past scientific practices has revealed what is by now a well-documented pattern, namely that theories of human bodies and/or behavior that ignore women’s bodies and/or behavior have proven to be inaccurate. The feminist archaeologist who disputes the man-the-hunter theory, in spite of the equivocal evidence provided by the chipped stones, still has good inductive evidence, based on her feminist political views, to support her decision. The man-the-hunter theory leaves out the role of women in the human development of technology and culture. The feminist archaeologist who chooses to interpret the chipped stones on the basis of a theory that includes or even highlights the role of female agrarian behaviour is making her choice based on past evidence that to ignore the role of women is to get the “human” story drastically wrong. Her decision is not merely relative to feminist politics, it is not based on some nonevidence belief entity brought in when all the objective evidence, independent of belief, is equivocal. Rather it is a decision well supported by inductively observed instances of past scientific errors. On my nonrepresentationalist view, then, the man-the-hunter and the woman-the-gatherer interpretations are not equally well supported by the evidence. The former is not supported by feminist analyses of past scientific practice. It is not the case that, faced with interpretations equally well supported by the “belief-independent”

Clough and Davidson

109

empirical evidence, we are forced to the inner belief world of politics to make our choice. On Davidson’s model, our empirical beliefs have no better metaphysical links than do our political beliefs to the outer, independent objective world, just as our political beliefs are no more closely related than our more straightforwardly empirical beliefs to our inner subjective world. But this is because, on Davidson’s view, there is no inner or outer world, there is no metaphysical bifurcation. There is only one world, an objective view of which can be made meaningful only by the language users who are part of it. While it is certainly possible that some of the political beliefs that make up our belief webs might be more geographically remote from the empirical beliefs at the edge of our webs, the holism of Davidson’s model indicates that the political beliefs are still connected, by some threads, to those empirical beliefs. When we examine meaning on the model of the radical interpreter, we see that changes in empirical beliefs can, and must, in principle, affect more theoretical beliefs, even if the effect is only slight. For the radical interpreter, no two theoretical beliefs can both conflict with each other in drastic ways and have the same truth conditions. Of course, even though Longino might not have found an example of two underdetermined theories or models, there still might be cases where we want to say that, from the point of view of us nonradical interpreters, two conflicting theories are equally well-supported by the empirical evidence. Here, if we are careful to construe both the “empirical” and “political” evidence in support of each theory as themselves beliefs, we might say that both types of belief can be epistemically underdetermined by their causal relationship with the external world. But, in principle, the radical interpreter must be able to identify the precise causal history of any individual belief, even if we, less-than-radical-interpreters cannot. In the sceptic’s world, the fear is that the metaphysical separation between the world and us makes coherent the worry that we are, in principle, unable to speak with confidence about the causal links between our representations and the world represented. Davidson’s point is not to offer comfort to the sceptic that her representations are indeed accurate, but to rethink the “beliefs as representations” model itself.

110

Siblings Under the Skin

He uses the radical interpreter to give life to an alternate view of the relationship between language users and the world, whereby all we have (and all we need) is an interconnected web of empirical and theoretical/ political beliefs, where for any one attribution of error, that potentially false belief must be connected sufficiently firmly to a sufficiently rich background of true beliefs before we can even identify that belief as being false about some feature of the world. To review, Davidson’s repudiation of the representationalist metaphor is a repudiation of a metaphysical gap between our representations and the world. Unlike the filtering conceptual schemes invoked in the writings of Longino, Keller, and Harding, Davidson views our language use as a guarantee of an unmediated causal relationship between most of our beliefs and the world. But, unlike the correspondence theory of objectivism, he does not use this unmediated contact to justify particular beliefs. I think we feminists should examine this option further. A nonrepresentationalist understanding of contextual values or worldviews would conceive of them, not as filters between our beliefs and some non-belief form of evidence, but as further important strands in our web of belief. When we justify particularly crucial elements of our feminist worldviews, such as our beliefs about oppression and justice, our appeals to the evidence have been well-documented and are powerfully persuasive as a result. There is no need for us to doubt the evidence of our feminist political values, as long as we conceive of such evidence as that which is provided by other beliefs in our web. This inter-belief comparison is where all justification happens, and has happened, for decades of feminist research. In this way, we can make stronger claims than those allowed by the sceptical arguments of Longino, Keller, and Harding. Our scientific theories and our beliefs about oppression and justice are not merely relative to our feminist conceptual schemes, they are justified by the evidence and they are true.

Clough and Davidson

111

Notes 1.

See Elisabeth Lloyd’s “Science and Anti-Science: Objectivity and its Real Enemies” (Lloyd 1997) for an excellent discussion of this sort of antifeminist argument. 2. Quine’s views on the scepticism underlying this relativist application are difficult to identify (see Bergström [1993] for a review of this point). 3. See the parallel analysis offered by Ilkka Niiniluoto in “The Relativism Question in Feminist Epistemology” (1997). 4. Karen Barad makes similar suggestions based on the philosophy of Bohr (Barad 1997). 5. My thanks to Bjørn Ramberg for suggesting this characterization. 6. See Linda Alcoff ’s Real Knowing for a recent endorsement of this sort of epistemological project (Alcoff 1996, 2-3). 7. A prototypical version of objectivism, as I have described it, can be found in Mind and the World Order by C. I. Lewis ([1929] 1956). He explains that “the two elements to be distinguished in knowledge are the concept, which is the product of the activity of thought, [such as the forming of an hypothesis or a theory] and the sensuously [empirically] given, which is independent of such activity” (Lewis [1929] 1956, 37). For Lewis, the “given” of experience is “what remains unaltered, no matter what our interests, no matter how we think or conceive” ([1929] 1956, 52). This is what Davidson calls the “content.” Our conceptualization of the given content is the perceptual imposition of a filter or “scheme.” 8. Bas van Fraassen’s constructive empiricism, for example, maintains that we can only have objective knowledge of the truth about the observable entities in a theory. He explains that “to accept a theory is (for us [constructive empiricists]) to believe that it is empirically adequate—that what the theory says about what is observable (by us) is true” (van Fraassen 1980, 18; italics in original). When a theory makes reference to unobservable entities we cannot have such knowledge. van Fraassen argues that theories that contain unobservables can be “empirically adequate” but not true or false as a whole (1980, 18). Many feminist philosophers, notably Nelson (1990), have begun to reexamine feminist versions of empiricism, but these approaches are not the focus of this essay. 9. Longino’s differences with Kuhn are discussed in chapter 2 of Science as Social Knowledge (Longino 1990). 10. Nelson makes the same observation about Longino on this point, but Nelson’s diagnosis is from a Quinean rather than a Davidsonian perspective (Nelson 1990, 238-39; see also Nelson 1993). Insofar as Nelson

112

Siblings Under the Skin

makes use of Quine’s nonrepresentationalist moments, I am generally in agreement with her proposals. 11. For an alternative examination of Davidson, see Alcoff ’s Real Knowing (1996). 12. Charlene Haddock Seigfried reminds me that an even more thorough-going pragmatist interpretation would involve pairing the sentence “There’s the bus!” with a whole host of public transit practices—riding the bus, smelling the bus fumes, etc. I think Davidson would appreciate the addition of these other practical, experiential details. 13. Quine goes on to distinguish a subclass of occasion sentences, the observation sentences, in order to make a number of empiricist claims that rely on a representationalist model (Quine 1960, ch. 2). Like some of the more traditional foundationalists such as Hempel (1965), Quine tries, at times, to use the empirical simplicity of occasion sentences as an epistemological grounding for claims about more complex sentences. As I will argue, Davidson’s discussion of occasion sentences has no such epistemological implications. 14. Another interpretive point suggested to me by Ramberg.

References Alcoff, Linda Martín. 1996. Real knowing: New versions of the coherence theory. Ithaca: Cornell University Press. Barad, Karen. 1997. Meeting the universe halfway: Realism and social constructivism without contradiction. In Feminism, science and the philosophy of science, ed. Lynn Hankinson Nelson and Jack Nelson. Dordrecht: Kluwer. Bergström, Lars. 1993. Quine, underdetermination, and skepticism. The Journal of Philosophy 90 (7): 331-58. Code, Lorraine. 1991. What can she know? Feminist theory and the construction of knowledge. Ithaca: Cornell University Press. Davidson, Donald. 1984. On the very idea of a conceptual scheme. In Inquiries into truth and interpretation. Oxford: Clarendon Press. ——. 1989a. The myth of the subjective. In Relativism: Interpretations and confrontations, ed. Michael Krausz. Bloomington: Indiana University Press. ——. 1989b. The conditions of thought. In The mind of Donald Davidson, ed. Johannes Brands and Wolfgang Gombocz. Grazer Philosophische Studien, no. 36. Amsterdam: Editions Rodopoi.

Clough and Davidson

113

——. 1990. The structure and content of truth. The Journal of Philosophy 87 (6): 279-328. ——. [1986] 1991a . A coherence theory of truth and knowledge. Reprinted in Reading Rorty: Critical responses to Philosophy and the mirror of nature (and beyond), ed. Alan Malachowski. Oxford: Basil Blackwell. ——. 1991b. Epistemology Naturalized. Dialectica 45 (2-3): 191-202. ——. 1991c. Three varieties of knowledge. In A.J. Ayer Memorial Essays, ed. A. Phillips Griffiths. Cambridge: Cambridge University Press. Doell, Ruth, and Helen Longino. 1988. Sex hormones and human behavior: A critique of the linear model. Journal of Homosexuality 15 (3/4): 55-79. Harding, Sandra. 1991. Whose science? Whose knowledge? Thinking from women’s lives. Ithaca: Cornell University Press. ——. 1993. Rethinking standpoint epistemology: What is “strong objectivity?” In Feminist epistemologies, ed. Linda Alcoff and Elizabeth Potter. New York: Routlege. Hempel, Carl G. 1965. Aspects of scientific explanation and other essays in the philosophy of science. New York: The Free Press. Keller, Evelyn Fox. 1992a. Introduction to Secrets of life, secrets of death: Essays on language, gender and science. New York: Routledge. ——. 1992b. Gender and science: An update. In Secrets of life, secrets of death: Essays on language, gender and science. New York: Routledge. ——. 1992c. Critical silences in scientific discourse. In Secrets of life, secrets of death: Essays on language, gender and science. New York: Routledge. Kuhn, Thomas. 1972. The structure of scientific revolutions, 2d ed. Chicago: University of Chicago Press. Lewis, C.I. [1929] 1956. Mind and the world order, republication of 1st ed. with corrections. New York: Dover Publications. Lloyd, Elisabeth. 1997. Science and anti-science: Objectivity and its real enemies. In Feminism, science and the philosophy of science, ed. Lynn Hankinson Nelson and Jack Nelson. Dordrecht: Kluwer. Longino, Helen. 1987. Can there be a feminist science? Hypatia 2 (3): 51-64. ——. 1990. Science as social knowledge: Values and objectivity in scientific inquiry. Princeton: Princeton University Press. Malpas, Jeffrey, E. 1992. Donald Davidson and the mirror of meaning. Cambridge: Cambridge University Press. Nelson, Lynn Hankinson. 1990. Who knows: From Quine to a feminist empiricism. Philadelphia: Temple University Press. ——. 1993. A question of evidence. Hypatia 2 (8): 172-189. Niiniluoto, Ilkka. 1997. The relativism question in feminist epistemology. In Feminism, science and the philosophy of science, ed. Lynn Hankinson Nelson and Jack Nelson. Dordrecht: Kluwer.

114

Siblings Under the Skin

Quine, Willard Van Orman. 1960. Word and object. Cambridge: MIT Press. ——. 1969. “Ontological relativity” and other essays. New York: Columbia University Press. ——. 1981. Empirical content. In Theories and things. Cambridge: The Belknap Press. Ramberg, Bjørn. 1989. Donald Davidson’s philosophy of language: An introduction. Oxford: Basil Blackwell. Rorty, Richard. 1991. Philosophical papers. Vol. 1, Objectivity, relativism, and truth. Cambridge: Cambridge University Press. van Fraassen, Bas. 1980. The scientific image. Oxford: Oxford University Press.

Clough and Davidson

115

Biographical note Like a number of the contributors to this volume, Sharyn Clough (b. 1965) is an ex-pat. Hailing from western Canada, Sharyn came stateside to begin her academic career teaching philosophy of science, feminist theory and symbolic logic. Responding to the challenge of teaching material that is often dry and abstract, Sharyn took up the pedagogical banner How I Stopped Worrying and Learned to Love American Pop-Culture, and reports that now even logic can be fun. She has published essays in journals such as Hypatia, Social Epistemology, and Studies in the History and Philosophy of the Biological and Biomedical Sciences. Her monograph Beyond Epistemology: A Pragmatist Prescription for Feminist Science Studies (Rowman and Littlefield, April 2003) continues the themes found in “A Hasty Retreat.” Sharyn’s interest in the work of Donald Davidson was piqued after meeting Bjørn Ramberg, a Davidson scholar, when she was in graduate school at Simon Fraser University (PhD, 1997). Her prior training in the social sciences had left her with a number of questions concerning the assumptions that informed scientific method. Sharyn was particularly interested in politicized (e.g., feminist) criticism of these assumptions, but the spectre of relativism continued to haunt the available feminist accounts. The work of Donald Davidson, through the pragmatic interpretations provided by Ramberg, provided a way out of the dilemma. While working on her dissertation, Sharyn recalls reading Lynn Hankinson Nelson’s book Who Knows: From Quine to a Feminist Empiricism with both excitement and a little bit of worry—perhaps Nelson had solved the problem already! By the end of the book, Sharyn was relieved—a few problems still needed solving (imagine!). Perhaps, Quine’s student, Davidson, could help sharpen the focus. Sharyn continues to be Nelson’s number one fan, even as she continues her attempts to convert Nelson to the Davidsonian camp. In response to reading her essay “A Hasty Retreat” Davidson wrote to Sharyn to say that in her discussion of feminist epistemology she had made “good and legitimate use” of his attack on representationalism and that “it isn’t easy for most people to take in” what he’s been up to, but that she seemed “to have got everything straight.” He noted too his pleasant surprise that she should take on as an ally an “old male fogey” like him and expressed a new respect for feminist philosophy. Thus marked for Sharyn the beginnings of the idea that exchanges between analytic philosophy and feminist theory could be fruitful, encouraging, and, wonder of wonders, reflexive.

116

Siblings Under the Skin

Clough and Davidson

117

A Coherence Theory of Truth and Knowledge Donald Davidson

In this paper I defend what may as well be called a coherence theory of truth and knowledge. The theory I defend is not in competition with a correspondence theory, but depends for its defence on an argument that purports to show that coherence yields correspondence. The importance of the theme is obvious. If coherence is a test of truth, there is a direct connection with epistemology, for we have reason to believe many of our beliefs cohere with many others, and in that case we have reason to believe many of our beliefs are true. When the beliefs are true, then the primary conditions for knowledge would seem to be satisfied. Someone might try to defend a coherence theory of truth without defending a coherence theory of knowledge, perhaps on the ground that the holder of a coherent set of beliefs might lack a reason to believe his beliefs coherent. This is not likely, but it may be that someone, though he has true beliefs, and good reasons for holding them, does not appreciate the relevance of reason to belief. Such a one may best be viewed as having knowledge he does not know he has: he thinks he is a sceptic. In a word, he is a philosopher. Setting aside aberrant cases, what brings truth and knowledge together is meaning. If meanings are given by objective truth conditions there is a question how we can know that the conditions are satisfied, for this would appear to require a confrontation between what we believe and reality; and the idea of such a confrontation is absurd. But if coherence is a test of truth, then coherence is a test for judging that objective truth conditions are satisfied, and we no longer need to explain meaning on the basis of possible confrontation. My slogan is: correspondence without confrontation. Given a correct epistemology, we can be realists in all departments. We can accept objective truth conditions as the key to meaning, a realist view of truth, and we can insist that knowledge is of an objective world independent of our thought or language.

118

Siblings Under the Skin

Since there is not, as far as I know, a theory that deserves to be called ‘the’ coherence theory, let me characterize the sort of view I want to defend. It is obvious that not every consistent set of interpreted sentences contains only true sentences, since one such set might contain just the consistent sentence S and another just the negation of S. And adding more sentences, while maintaining consistency, will not help. We can imagine endless state-descriptions — maximal consistent descriptions — which do not describe our world. My coherence theory concerns beliefs, or sentences held true by someone who understands them. I do not want to say, at this point, that every possible coherent set of beliefs is true (or contains mostly true beliefs). I shy away from this because it is so unclear what is possible. At one extreme, it might be held that the range of possible maximal sets of beliefs is as wide as the range of possible maximal sets of sentences, and then there would be no point to insisting that a defensible coherence theory concerns beliefs and not propositions or sentences. But there are other ways of conceiving what it is possible to believe which would justify saying not only that all actual coherent belief systems are largely correct but that all possible ones are also. The difference between the two notions of what it is possible to believe depends on what we suppose about the nature of belief, its interpretation, its causes, its holders and its patterns. Beliefs for me are states of people with intentions, desires, sense organs; they are states that are caused by, and cause, events inside and outside the bodies of the entertainers. But even given all these constraints, there are many things people do believe, and many more that they could. For all such cases, the coherence theory applies. Of course some beliefs are false. Much of the point of the concept of belief is the potential gap it introduces between what is held to be true and what is true. So mere coherence, no matter how strongly coherence is plausibly defined, cannot guarantee that what is believed is so. All that a coherence theory can maintain is that most of the beliefs in a coherent total set of beliefs are true. This way of stating the position can at best be taken as a hint, since there is probably no useful way to count beliefs, and so no clear meaning to the idea that most of a person’s beliefs are true. A somewhat better way to put the point is to say there is presumption in favour of the truth of a belief that coheres with a significant mass of belief. Every

Clough and Davidson

119

belief in a coherent total set of beliefs is justified in the light of this presumption, much as every intentional action taken by a rational agent (one whose choices, beliefs and desires cohere in the sense of Bayesian decision theory) is justified. So to repeat, if knowledge is justified true belief, then it would seem that all the true beliefs of a consistent believer constitute knowledge. This conclusion, though too vague and hasty to be right, contains an important core of truth, as I shall argue. Meanwhile I merely note the many problems asking for treatment: what exactly does coherence demand? How much of inductive practice should be included, how much of the true theory (if there is one) of evidential support must be in there? Since no person has a completely consistent body of convictions, coherence with which beliefs creates a presumption of truth? Some of these problems will be put in better perspective as I go along. It should be clear that I do not hope to define truth in terms of coherence and belief. Truth is beautifully transparent compared to belief and coherence, and I take it as primitive. Truth, as applied to utterances of sentences, shows the disquotational feature enshrined in Tarski’s Convention T, and that is enough to fix its domain of application. Relative to a language or a speaker, of course, so there is more to truth than Convention T; there is whatever carries over from language to language or speaker to speaker. What Convention T, and the trite sentences it declares true, like “‘Grass is green”, spoken by an English speaker, is true if and only if grass is green’, reveal is that the truth of an utterance depends on just two things: what the words as spoken mean, and how the world is arranged. There is no further relativism to a conceptual scheme, a way of viewing things, a perspective. Two interpreters, as unlike in culture, language and point of view as you please, can disagree over whether an utterance is true, but only if they differ on how things are in the world they share, or what the utterance means. I think we can draw two conclusions from these simple reflections. First, truth is correspondence with the way things are. (There is no straightforward and non-misleading way to state this; to get things right, a detour is necessary through the concept of satisfaction in terms of which truth is characterized.1) So if a coherence theory of truth is acceptable, it must be consistent with a correspondence theory. Second, a theory of knowledge that allows that we can know the truth must be

120

Siblings Under the Skin

a nonrelativized, non-internal form of realism. So if a coherence theory of knowledge is acceptable, it must be consistent with such a form of realism. My form of realism seems to be neither Hilary Putnam’s internal realism nor his metaphysical realism (Putnam 1978, 125). It is not internal realism because internal realism makes truth relative to a scheme, and this is an idea I do not think is intelligible.2 A major reason, in fact, for accepting a coherence theory is the unintelligibility of the dualism of a conceptual scheme and a ‘world’ waiting to be coped with. But my realism is certainly not Putnam’s metaphysical realism, for it is characterized by being ‘radically non-epistemic’, which implies that all our best researched and established thoughts and theories may be false. I think the independence of belief and truth requires only that each of our beliefs may be false. But of course a coherence theory cannot allow that all of them can be wrong. But why not? Perhaps it is obvious that the coherence of a belief with a substantial body of belief enhances its chance of being true, provided there is reason to suppose the body of belief is true, or largely so. But how can coherence alone supply grounds for belief? Perhaps the best we can do to justify one belief is to appeal to other beliefs. But then the outcome would seem to be that we must accept philosophical scepticism, no matter how unshaken in practice our beliefs remain. This is scepticism in one of its traditional garbs. It asks: why couldn’t all my beliefs hang together and yet be comprehensively false about the actual world? Mere recognition of the fact that it is absurd or worse to try to confront our beliefs, one by one, or as a whole, with what they are about does not answer the question nor show the question unintelligible. In short, even a mild coherence theory like mine must provide a sceptic with a reason for supposing coherent beliefs are true. The partisan of a coherence theory can’t allow assurance to come from outside the system of belief, while nothing inside can produce support except as it can be shown to rest, finally or at once, on something independently trustworthy. It is natural to distinguish coherence theories from others by reference to the question whether or not justification can or must come to an end. But this does not define the positions, it merely suggests a form the argument may take. For there are coherence theorists who hold that some beliefs can serve as the basis for the rest, while it would

Clough and Davidson

121

be possible to maintain that coherence is not enough, although giving reasons never comes to an end. What distinguishes a coherence theory is simply the claim that nothing can count as a reason for holding a belief except another belief. Its partisan rejects as unintelligible the request for a ground or source of justification of another ilk. As Rorty has put it, ‘nothing counts as justification unless by reference to what we already accept, and there is no way to get outside our beliefs and our language so as to find some test other than coherence’ (Rorty 1979, 178). About this I am, as you see, in agreement with Rorty. Where we differ, if we do, is on whether there remains a question how, given that we cannot ‘get outside our beliefs and our language so as to find some test other than coherence’, we nevertheless can have knowledge of, and talk about, an objective public world which is not of our own making. I think this question does remain, while I suspect that Rorty doesn’t think so. If this is his view, then he must think I am making a mistake in trying to answer the question. Nevertheless, here goes. It will promote matters at this point to review very hastily some of the reasons for abandoning the search for a basis for knowledge outside the scope of our beliefs. By ‘basis’ here I mean specifically an epistemological basis, a source of justification. The attempts worth taking seriously attempt to ground belief in one way or another on the testimony of the senses: sensation, perception, the given, experience, sense data, the passing show. All such theories must explain at least these two things. what, exactly, is the relation between sensation and belief that allows the first to justify the second? And why should we believe our sensations are reliable, that is, why should we trust our senses? The simplest idea is to identify certain beliefs with sensations. Thus Hume seems not to have distinguished between perceiving a green spot and perceiving that a spot is green. (An ambiguity in the word ‘idea’ was a great help here.) Other philosophers noted Hume’s confusion, but tried to attain the same results by reducing the gap between perception and judgement to zero by attempting to formulate judgments that do not go beyond stating that the perception or sensation or presentation exists (whatever that may mean). Such theories do not justify beliefs on the basis of sensations, but try to justify certain beliefs by claiming that they have exactly the same epistemic content as a sensation. There are

122

Siblings Under the Skin

two difficulties with such a view: first, if the basic beliefs do not exceed in content the corresponding sensation they cannot support any inference to an objective world; and second, there are no such beliefs. A more plausible line is to claim that we cannot be wrong about how things appear to us to be. If we believe we have a sensation, we do; this is held to be an analytic truth, or a fact about how language is used. It is difficult to explain this supposed connection between sensations and some beliefs in a way that does not invite scepticism about other minds, and in the absence of an adequate explanation, there should be a doubt about the implications of the connection for justification. But in any case, it is unclear how, on this line, sensations justify the belief in those sensations. The point is rather that such beliefs require no justification, for the existence of the belief entails the existence of the sensation, and so the existence of the belief entails its own truth. Unless something further is added, we are back to another form of coherence theory. Emphasis on sensation or perception in matters epistemological springs from the obvious thought: sensations are what connect the world and our beliefs, and they are candidates for justifiers because we often are aware of them. The trouble we have been running into is that the justification seems to depend on the awareness, which is just another belief. Let us try a bolder tack. Suppose we say that sensations themselves, verbalized or not, justify certain beliefs that go beyond what is given in sensation. So, under certain conditions, having the sensation of seeing a green light flashing may justify the belief that a green light is flashing. The problem is to see how the sensation justifies the belief. Of course, if someone has the sensation of seeing a green light flashing, it is likely, under certain circumstances, that a green light is flashing. We can say this, since we know of his sensation, but he can’t say it, since we are supposing he is justified without having to depend on believing he has the sensation. Suppose he believed he didn’t have the sensation. Would the sensation still justify him in the belief in an objective flashing green light? The relation between a sensation and a belief cannot be logical since sensations are not beliefs or other propositional attitudes. What

Clough and Davidson

123

then is the relation? The answer is, I think, obvious: the relation is causal. Sensations cause some beliefs and in this sense are the basis or ground of those beliefs. But a causal explanation of a belief does not show how or why the belief is justified. The difficulty of transmuting a cause into a reason plagues the anti-coherentist again if he tries to answer our second question: what justifies the belief that our senses do not systematically deceive us? For even if sensations justify belief in sensation, we do not yet see how they justify belief in external events and objects. Quine tells us that science tells us that ‘our only source of information about the external world is through the impact of light rays and molecules upon our sensory surfaces (Quine 1975, 68). What worries me is how to read the words ‘source’ and ‘information’. Certainly it is true that events and objects in the external world cause us to believe things about the external world, and much, if not all, of the causality takes a route through the sense organs. The notion of information, however, applies in a non-metaphorical way only to the engendered beliefs. So ‘source’ has to be read simply as ‘cause’ and ‘information’ as ‘true belief ’ or ‘knowledge’. Justification of beliefs caused by our senses is not yet in sight.3 The approach to the problem of justification we have been tracing must be wrong. We have been trying to see it this way: a person has all his beliefs about the world—that is, all his beliefs. How can he tell if they are true, or apt to be true? Only, we have been assuming, by connecting his beliefs to the world, confronting certain of his beliefs with the deliverances of the senses one by one, or perhaps confronting the totality of his beliefs with the tribunal of experience. No such confrontation makes sense, for of course we can’t get outside our skins to find out what is causing the internal happenings of which we are aware. Introducing intermediate steps or entities into the causal chain, like sensations or observations, serves only to make the epistemological problem more obvious. For if the intermediaries are merely causes, they don’t justify the beliefs they cause, while if they deliver information, they may be lying. The moral is obvious. Since we can’t swear intermediaries to truthfulness, we should allow no intermediaries between our beliefs and their objects in the world. Of course there are causal intermediaries. What we must guard against are epistemic intermediaries.

124

Siblings Under the Skin

There are common views of language that encourage bad epistemology. This is no accident, of course, since theories of meaning are connected with epistemology through attempts to answer the question how one determines that a sentence is true. If knowing the meaning of a sentence (knowing how to give a correct interpretation of it) involves, or is, knowing how it could be recognized to be true, then the theory of meaning raises the same question we have been struggling with, for giving the meaning of a sentence will demand that we specify what would justify asserting it. Here the coherentist will hold that there is no use looking for a source of justification outside of other sentences held true, while the foundationalist will seek to anchor at least some words or sentences to non-verbal rocks. This view is held, I think, both by Quine and by Michael Dummett. Dummett and Quine differ, to be sure. In particular, they disagree about holism, the claim that the truth of our sentences must be tested together rather than one by one. And they disagree also, and consequently, about whether there is a useful distinction between analytic and synthetic sentences, and about whether a satisfactory theory of meaning can allow the sort of indeterminacy Quine argues for. (On all these points, I am Quine’s faithful student.) But what concerns me here is that Quine and Dummett agree on a basic principle, which is that whatever there is to meaning must be traced back somehow to experience, the given, or patterns of sensory stimulation, something intermediate between belief and the usual objects our beliefs are about. Once we take this step, we open the door to scepticism, for we must then allow that a very great many — perhaps most — of the sentences we hold to be true may in fact be false. It is ironical. Trying to make meaning accessible has made truth inaccessible. When meaning goes epistemological in this way, truth and meaning are necessarily divorced. One can, of course, arrange a shotgun wedding by redefining truth as what we are justified in asserting. But this does not marry the original mates. Take Quine’s proposal that whatever there is to the meaning (information value) of an observation sentence is determined by the patterns of sensory stimulation that would cause a speaker to assent to or dissent from the sentence. This is a marvelously ingenious way of capturing what is appealing about verificationist theories without

Clough and Davidson

125

having to talk of meanings, sense-data, or sensations; for the first time it made plausible the idea that one could, and should, do what I call the theory of meaning without need of what Quine calls meanings. But Quine’s proposal, like other forms of verificationism, makes for scepticism. For clearly a person’s sensory stimulations could be just as they are and yet the world outside very different. (Remember the brain in the vat.) Quine’s way of doing without meanings is subtle and complicated. He ties the meanings of some sentences directly to patterns of stimulation (which also constitute the evidence, Quine thinks, for assenting to the sentence), but the meanings of further sentences are determined by how they are conditioned to the original, or observation sentences. The facts of such conditioning do not permit a sharp division between sentences held true by virtue of meaning and sentences held true on the basis of observation. Quine made this point by showing that if one way of interpreting a speaker’s utterances was satisfactory, so were many others. This doctrine of the indeterminacy of translation, as Quine called it, should be viewed as neither mysterious nor threatening. It is no more mysterious than the fact that temperature can be measured in Centigrade or Fahrenheit (or any linear transformation of those numbers). And it is not threatening because the very procedure that demonstrates the degree of indeterminacy at the same time demonstrates that what is determinate is all we need. In my view, erasing the line between the analytic and synthetic saved philosophy of language as a serious subject by showing how it could be pursued without what there cannot be: determinate meanings. I now suggest also giving up the distinction between observation sentences and the rest. For the distinction between sentences belief in whose truth is justified by sensations and sentences belief in whose truth is justified only by appeal to other sentences held true is as anathema to the coherentist as the distinction between beliefs justified by sensations and beliefs justified only by appeal to further beliefs. Accordingly, I suggest we give up the idea that meaning or knowledge is grounded on something that counts as an ultimate source of evidence. No doubt meaning and knowledge depend on experience, and experience ultimately on sensation. But this is the ‘depend’ of causality, not of evidence or justification.

126

Siblings Under the Skin

I have now stated my problem as well as I can. The search for an empirical foundation for meaning or knowledge leads to scepticism, while a coherence theory seems at a loss to provide any reason for a believer to believe that his beliefs, if coherent, are true. We are caught between a false answer to the sceptic, and no answer. The dilemma is not a true one. What is needed to answer the sceptic is to show that someone with a (more or less) coherent set of beliefs has a reason to suppose his beliefs are not mistaken in the main. What we have shown is that it is absurd to look for a justifying ground for the totality of beliefs, something outside this totality which we can use to test or compare with our beliefs. The answer to our problem must then be to find a reason for supposing most of our beliefs are true that is not a form of evidence. My argument has two parts. First I urge that a correct understanding of the speech, beliefs, desires, intentions and other propositional attitudes of a person leads to the conclusion that most of a person’s beliefs must be true, and so there is a legitimate presumption that any one of them, if it coheres with most of the rest, is true. Then I go on to claim that anyone with thoughts, and so in particular anyone who wonders whether he has any reason to suppose he is generally right about the nature of his environment, must know what a belief is, and how in general beliefs are to be detected and interpreted. These being perfectly general facts we cannot fail to use when we communicate with others, or when we try to communicate with others, or even when we merely think we are communicating with others, there is a pretty strong sense in which we can be said to know that there is a presumption in favour of the overall truthfulness of anyone’s beliefs, including our own. So it is bootless for someone to ask for some further reassurance; that can only add to his stock of beliefs. All that is needed is that he recognize that belief is in its nature veridical. Belief can be seen to be veridical by considering what determines the existence and contents of a belief. Belief, like the other so-called propositional attitudes, is supervenient on facts of various sorts, behavioural, neurophysiological, biological and physical. The reason for pointing this out is not to encourage definitional or nomological reduction of psychological phenomena to something more basic, and certainly not to suggest epistemological priorities. The point is rather

Clough and Davidson

127

understanding. We gain one kind of insight into the nature of the propositional attitudes when we relate them systematically to one another and to phenomena on other levels. Since the propositional attitudes are deeply interlocked, we cannot learn the nature of one by first winning understanding of another. As interpreters, we work our way into the whole system, depending much on the pattern of interrelationships. Take for example the interdependence of belief and meaning. What a sentence means depends partly on the external circumstances that cause it to win some degree of conviction; and partly on the relations, grammatical, logical or less, that the sentence has to other sentences held true with varying degrees of conviction. Since these relations are themselves translated directly into beliefs, it is easy to see how meaning depends on belief. Belief, however, depends equally on meaning, for the only access to the fine structure and individuation of beliefs is through the sentences speakers and interpreters of speakers use to express and describe beliefs. If we want to illuminate the nature of meaning and belief, therefore, we need to start with something that assumes neither. Quine’s suggestion, which I shall essentially follow, is to take prompted assent as basic, the causal relation between assenting to a sentence and the cause of such assent. This a fair place to start the project of identifying beliefs and meanings, since a speaker’s assent to a sentence depends both on what he means by the sentence and on what he believes about the world. Yet it is possible to know that a speaker assents to a sentence without knowing either what the sentence, as spoken by him, means, or what belief is expressed by it. Equally obvious is the fact that once an interpretation has been given for a sentence assented to, a belief has been attributed. If correct theories of interpretation are not unique (do not lead to uniquely correct interpretations), the same will go for attributions of belief, of course, as tied to acquiescence in particular sentences. A speaker who wishes his words to be understood cannot systematically deceive his would-be interpreters about when he assents to sentences—that is, holds them true. As a matter of principle then, meaning, and by its connection with meaning, belief also, are open to public determination. I shall take advantage of this fact in what follows and adopt the stance of a radical interpreter when asking about the nature of belief. What a fully informed interpreter could learn about

128

Siblings Under the Skin

what a speaker means is all there is to learn; the same goes for what the speaker believes.4 The interpreter’s problem is that what he is assumed to know—the causes of assents to sentences of a speaker—is, as we have seen, the product of two things he is assumed not to know, meaning and belief. If he knew the meanings he would know the beliefs, and if he knew the beliefs expressed by sentences assented to, he would know the meanings. But how can he learn both at once, since each depends on the other? The general lines of the solution, like the problem itself, are owed to Quine. I will, however, introduce some changes into Quine’s solution, as I have into the statement of the problem. The changes are directly relevant to the issue of epistemological scepticism. I see the aim of radical interpretation (which is much, but not entirely, like Quine’s radical translation) as being to produce a Tarskistyle characterization of truth for the speaker’s language, and a theory of his beliefs. (The second follows from the first plus the presupposed knowledge of sentences held true.) This adds little to Quine’s program of translation, since translation of the speaker’s language into one’s own plus a theory of truth for one’s own language add up to a theory of truth for the speaker. But the shift to the semantic notion of truth from the syntactic notion of translation puts the formal restrictions of a theory of truth in the foreground, and emphasizes one aspect of the close relation between truth and meaning. The principle of charity plays a crucial role in Quine’s method, and an even more crucial role in my variant. In either case, the principle directs the interpreter to translate or interpret so as to read some of his own standards of truth into the pattern of sentences held true by the speaker. The point of the principle is to make the speaker intelligible, since too great deviations from consistency and correctness leave no common ground on which to judge either conformity or difference. From a formal point of view, the principle of charity helps solve the problem of the interaction of meaning and belief by restraining the degrees of freedom allowed belief while determining how to interpret words. We have no choice, Quine has urged, but to read our own logic into the thoughts of a speaker; Quine says this for the sentential calcu-

Clough and Davidson

129

lus, and I would add the same for first-order quantification theory. This leads directly to the identification of the logical constants, as well as to assigning a logical form to all sentences. Something like charity operates in the interpretation of those sentences whose causes of assent come and go with time and place: when the interpreter finds a sentence of the speaker the speaker assents to regularly under conditions he recognizes, he takes those conditions to be the truth conditions of the speaker’s sentence. This is only roughly right, as we shall see in a moment. Sentences and predicates less directly geared to easily detected goings-on can, in Quine’s cannon, be interpreted at will, given only the constraints of interconnections with sentences conditioned directly to the world. Here I would extend the principle of charity to favour interpretations that as far as possible preserve truth: I think it makes for mutual understanding, and hence for better interpretation, to interpret what the speaker accepts as true when we can. In this matter, I have less choice than Quine, because I do not see how to draw the line between observation sentences and theoretical sentences at the start. There are several reasons for this, but the one most relevant to the present topic is that this distinction is ultimately based on an epistemological consideration of a sort I have renounced: observation sentences are directly based on something like sensation — patterns of sensory stimulation — and this is an idea I have been urging leads to scepticism. Without the direct tie to sensation or stimulation, the distinction between observation sentences and others can’t be drawn on epistemologically significant grounds. The distinction between sentences whose causes to assent come and go with observable circumstances and those a speaker clings to through change remains however, and offers the possibility of interpreting the words and sentences beyond the logical. The details are not here to the point. What should be clear is that if the account I have given of how belief and meaning are related and understood by an interpreter is correct, then most of the sentences a speaker holds to be true — especially the ones he holds to most stubbornly, the ones most central to the system of his beliefs — most of these sentences are true, at least in the opinion of the interpreter. For the only, and therefore unimpeachable, method available to the interpreter automatically puts the speaker’s beliefs in accord with the

130

Siblings Under the Skin

standards of logic of the interpreter, and hence credits the speaker with plain truths of logic. Needless to say there are degrees of logical and other consistency, and perfect consistency is not to be expected. What needs emphasis is only the methodological necessity for finding consistency enough. Nor, from the interpreter’s point of view, is there any way he can discover the speaker to be largely wrong about the world. For he interprets sentences held true (which is not to be distinguished from attributing beliefs) according to the events and objects in the outside world that cause the sentence to be held true. What I take to be the important aspect of this approach is apt to be missed because the approach reverses our natural way of thinking of communication derived from situations in which understanding has already been secured. Once understanding has been secured we are able, often, to learn what a person believes quite independently of what caused him to believe it. This may lead us to the crucial, indeed fatal, conclusion that we can in general fix what someone means independently of what he believes and independently of what caused the belief. But if I am right, we can’t in general first identify beliefs and meanings and then ask what caused them. The causality plays an indispensable role in determining the content of what we say and believe. This is a fact we can be led to recognize by taking up, as we have, the interpreter’s point of view. It is an artifact of the interpreter’s correct interpretation of a person’s speech and attitudes that there is a large degree of truth and consistency in the thought and speech of an agent. But this is truth and consistency by the interpreter’s standards. Why couldn’t it happen that speaker and interpreter understand one another on the basis of shared but erroneous beliefs? This can, and no doubt often does, happen. But it cannot be the rule. For imagine for a moment an interpreter who is omniscient about the world, and about what does and would cause a speaker to assent to any sentence in his (potentially unlimited) repertoire. The omniscient interpreter, using the same method as the fallible interpreter, finds the fallible speaker largely consistent and correct. By his own standards, of course, but since these are objectively correct, the fallible speaker is seen to be largely correct and consistent by objective standards. We may also, if we want, let the omniscient interpreter turn

Clough and Davidson

131

his attention to the fallible interpreter of the fallible speaker. It turns out that the fallible interpreter can be wrong about some things, but not in general; and so he cannot share universal error with the agent he is interpreting. Once we agree to the general method of interpretation I have sketched, it becomes impossible correctly to hold that anyone could be mostly wrong about how things are. There is, as I noted above, a key difference between the method of radical interpretation I am now recommending, and Quine’s method of radical translation. The difference lies in the nature of the choice of causes that govern interpretation. Quine makes interpretation depend on patterns of sensory stimulation, while I make it depend on the external events and objects the sentence is interpreted as being about. Thus Quine’s notion of meaning is tied to sensory criteria, something he thinks that can be treated also as evidence. This leads Quine to give epistemic significance to the distinction between observation sentences and others, since observation sentences are supposed, by their direct conditioning to the senses, to have a kind of extra-linguistic justification. This is the view against which I argued in the first part of my paper, urging that sensory stimulations are indeed part of the causal chain that leads to belief, but cannot, without confusion, be considered to be evidence, or a source of justification, for the stimulated beliefs. What stands in the way of global scepticism of the senses is in my view the fact that we must, in the plainest and methodologically most basic cases, take the objects of a belief to be the causes of that belief. And what we, as interpreters, must take them to be is what they in fact are. Communication begins where causes converge: your utterance means what mine does if belief in its truth is systematically caused by the same events and objects.5 The difficulties in the way of this view are obvious, but I think they can be overcome. The method applies directly, at best, only to occasion sentences—the sentences assent to which is caused systematically by common changes in the world. Further sentences are interpreted by their conditioning to occasion sentences, and the appearance in them of words that appear also in occasion sentences. Among occasion sentences, some will vary in the credence they command not only in the face of environmental change, but also in the face of change of credence awarded related sentences. Criteria can be developed on this

132

Siblings Under the Skin

basis to distinguish degrees of observationality on internal grounds, without appeal to the concept of a basis for belief outside the circle of beliefs. Related to these problems, and easier still to grasp, is the problem of error. For even in the simplest cases it is clear that the same cause (a rabbit scampers by) may engender different beliefs in speaker and observer, and so encourage assent to sentences which cannot bear the same interpretation. It is no doubt this fact that made Quine turn from rabbits to patterns of stimulation as the key to interpretation. Just as a matter of statistics, I’m not sure how much better one approach is than the other. Is the relative frequency with which identical patterns of stimulation will touch off assent to ‘Gavagai’ and ‘Rabbit’ greater than the relative frequency with which a rabbit touches off the same two responses in speaker and interpreter? Not an easy question to test in a convincing way. But let the imagined results speak for Quine’s method. Then I must say, what I must say in any case, the problem of error cannot be met sentence by sentence, even at the simplest level. The best we can do is cope with error holistically, that is, we interpret so as to make an agent as intelligible as possible, given his actions, his utterances and his place in the world. About some things we will find him wrong, as the necessary cost of finding him elsewhere right. As a rough approximation, finding him right means identifying the causes with the objects of his beliefs, giving special weight to the simplest cases, and countenancing error where it can be best explained. Suppose I am right that an interpreter must so interpret as to make a speaker or agent largely correct about the world. How does this help the person himself who wonders what reason he has to think his beliefs are mostly true? How can he learn about the causal relations between the real world and his beliefs that lead the interpreter to interpret him as being on the right track? The answer is contained in the question. In order to doubt or wonder about the provenance of his beliefs an agent must know what belief is. This brings with it the concept of objective truth, for the notion of a belief is the notion of a state that may or may not jibe with reality. But beliefs are also identified, directly and indirectly, by their causes. What an omniscient interpreter knows, a fallible interpreter gets right enough if he understands a speaker, and this is just the complicated causal truth that makes us the believers we are, and fixes the contents of our beliefs.

Clough and Davidson

133

The agent has only to reflect on what a belief is to appreciate that most of his basic beliefs are true, and among his beliefs, those most securely held and that cohere with the main body of his beliefs are the most apt to be true. The question, how do I know my beliefs are generally true? thus answers itself, simply because beliefs are by nature generally true. Rephrased or expanded, the question becomes, how can I tell whether my beliefs, which are by their nature generally true, are generally true? All beliefs are justified in this sense: they are supported by numerous other beliefs (otherwise they wouldn’t be the beliefs they are), and have a presumption in favour of their truth. The presumption increases the larger and more significant the body of beliefs with which a belief coheres, and there being no such thing as an isolated belief, there is no belief without a presumption in its favour. In this respect, interpreter and interpreted differ. From the interpreter’s point of view, methodology enforces a general presumption of truth for the body of beliefs as a whole, but the interpreter does not need to presume each particular belief of someone else is true. The general presumption applied to others does not make them globally right, as I have emphasized, but provides the background against which to accuse them of error. But from each person’s own vantage point, there must be a graded presumption in favour of each of his own beliefs. We cannot, alas, draw the picturesque and pleasant conclusion that all true beliefs constitute knowledge. For though all of a believer’s beliefs are to some extent justified to him, some may not be justified enough, or in the right way, to constitute knowledge. The general presumption in favour of the truth of belief serves to rescue us from a standard form of skepticism by showing why it is impossible for all our beliefs to be false together. This leaves almost untouched the task of specifying the conditions of knowledge. I have not been concerned with the canons of evidential support (if such there be), but to show that all that counts as evidence or justification for a belief must come from the same totality of belief to which it belongs.

***

134

Siblings Under the Skin

Afterthoughts, 1987 The paper printed here was written for a colloquium organized by Richard Rorty for a Hegel Congress at Stuttgart in 1981. W. V. Quine and Hilary Putnam were the other participants in the colloquium. Our contributions were published in Kant oder Hegel? ( Henrich 1983). After Stuttgart the four of us had a more leisurely exchange on the same topics at the University of Heidelberg. When the Pacific Division of the American Philosophical Association met in March of 1983, Rorty read a paper titled ‘Pragmatism, Davidson, and Truth’. It was in part a comment on ‘A Coherence Theory of Truth and Knowledge’. I replied. Rorty subsequently published his paper with revisions in Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson (LePore 1986). This note continues the conversation. A few aging philosophes, which may include Quine, Putnam and Dummett, and certainly includes me, are still puzzling over the nature of truth and its connections or lack of connections with meaning and epistemology. Rorty thinks we should stop worrying; he believes philosophy has seen through or outgrown the puzzles and should turn to less heavy and more interesting matters. He is particularly impatient with me for not conceding that the old game is up because he finds in my work useful support for his enlightened stance; underneath my ‘out-dated rhetoric’ he detects the outlines of a largely correct attitude. In his paper, both early and late, Rorty urges two things: that my view of truth amounts to a rejection of both coherence and correspondence theories and should properly be classed as belonging to the pragmatist tradition, and that I should not pretend that I am answering the sceptic when I am really telling him to get lost. I pretty much concur with him on both points. In our 1983 discussion I agreed to stop calling my position either a coherence or a correspondence theory if he would give up the pragmatist theory of truth. He has done his part; he now explicitly rejects both James and Peirce on truth. I am glad to hold to my side of the bargain. If it had not already been published, I would now change the title of ‘A Coherence Theory’, and I would not describe the project as showing how ‘coherence yields correspondence’. On internal evidence alone, as Rorty points out, my view cannot be called a correspondence theory.

Clough and Davidson

135

As long ago as 1969 I argued that nothing can usefully and intelligibly be said to correspond to a sentence (Davidson [1969] 1984); and I repeated this in ‘A Coherence Theory’. I thought then that the fact that in characterizing truth for a language it is necessary to put words into relation with objects was enough to give some grip for the idea of correspondence; but this now seems to me a mistake. The mistake is in a way only a misnomer, but terminological infelicities have a way of breeding conceptual confusion, and so it is here. Correspondence theories have always been conceived as providing an explanation or analysis of truth, and this a Tarski-style theory of truth certainly does not do. I would also now reject the point generally made against correspondence theories that there is no way we could ever tell whether our sentences or beliefs correspond to reality. This criticism is at best misleading, since no one has ever explained in what such a correspondence could consist; and, worse, it is predicated on the false assumption that truth is transparently epistemic. I also regret having called my view a ‘coherence theory’. My emphasis on coherence was probably just a way of making a negative point, that ‘all that counts as evidence or justification for a belief must come from the same totality of belief to which it belongs.’ Of course this negative claim has typically led those philosophers who held it to conclude that reality and truth are constructs of thought; but it does not lead me to this conclusion, and for this reason if no other I ought not to have called my view a coherence theory. There is also a less weighty reason for not stressing coherence. Coherence is nothing but consistency. It is certainly in favour of a set of beliefs that they be consistent, but there is no chance that a person’s beliefs will not tend to be self-consistent, since beliefs are individuated in part by their logical properties; what is not largely consistent with many other beliefs cannot be identified as a belief. The main thrust of ‘A Coherence Theory’ has little to do with consistency; the important thesis for which I argue is that belief is intrinsically veridical. This is the ground on which I maintain that while truth is not an epistemic concept, neither is it wholly severed from belief (as it is in different ways by both correspondence and coherence theories). My emphasis on coherence was misplaced; calling my view a ‘theory’ was a plain blunder. In his paper Rorty stressed a minimalist attitude

136

Siblings Under the Skin

towards truth that he correctly thought we shared. It could be put this way: truth is as clear and basic a concept as we have. Tarski has given us an idea of how to apply the general concept (or try to apply it) to particular languages on the assumption that we already understand it; but of course he didn’t show how to define it in general (he proved, rather, that this couldn’t be done). Any further attempt to explain, define, analyse or explicate the concept will be empty or wrong: correspondence theories, coherence theories, pragmatist theories, theories that identify truth with warranted assertability (perhaps under ‘ideal’ or ‘optimum’ conditions), theories that ask truth to explain the success of science or serve as the ultimate outcome of science or the conversations of some elite, all such theories either add nothing to our understanding of truth or have obvious counter-examples. Why on earth should we expect to be able to reduce truth to something clearer or more fundamental? After all, the only concept Plato succeeded in defining was mud (dirt and water). Putnam’s comparison of various attempts to characterize truth with the attempts to define ‘good’ in naturalistic terms seems to me, as it does to Rorty, apt. It also seems to apply to Putnam’s identification of truth with idealized warranted assertability (Putnam 1983, p. xvii.) A theory of truth for a speaker, or group of speakers, while not a definition of the general concept of truth, does give a firm sense of what the concept is good for; it allows us to say, in a compact and clear way, what someone who understands that speaker, or those speakers, knows. Such a theory also invites the question how an interpreter could confirm its truth—a question which without the theory could not be articulated. The answer will, as I try to show in ‘A Coherence Theory’, bring out essential relations among the concepts of meaning, truth and belief. If I am right, each of these concepts requires the others, but none is subordinate to, much less definable in terms of, the others. Truth emerges not as wholly detached from belief (as a correspondence theory would make it) nor as dependent on human methods and powers of discovery (as epistemic theories of truth would make it). What saves truth from being ‘radically non-epistemic’ (in Putnam’s words) is not that truth is epistemic but that belief, through its ties with meaning, is intrinsically veridical. Finally, how about Rorty’s admonition to stop trying to answer the sceptic, and tell him to get lost? A short response would be that

Clough and Davidson

137

the sceptic has been told this again and again over the millennia and never seems to listen; like the philosopher he is, he wants an argument. To spell this out a bit: there is perhaps the suggestion in Rorty’s ‘Pragmatism, Davidson, and Truth’ that a ‘naturalistic’ approach to the problems of meaning and the propositional attitudes will automatically leave the sceptic no room for manoeuvre. This thought, whether or not it is Rorty’s, is wrong. Quine’s naturalized epistemology, because it is based on the empiricist premise that what we mean and what we think is conceptually (and not merely causally) founded on the testimony of the senses, is open to standard sceptical attack. I was much concerned in ‘A Coherence Theory’ to argue for an alternative approach to meaning and knowledge, and to show that if this alternative were right, scepticism could not get off the ground. I agree with Rorty to this extent; I did not set out to ‘refute’ the sceptic, but to give a sketch of what I think to be a correct account of the foundations of linguistic communication and its implications for truth, belief and knowledge. If one grants the correctness of this account, one can tell the sceptic to get lost. Where Rorty and I differ, if we do, is in the importance we attach to the arguments that lead to the sceptic’s undoing, and in the interest we find in the consequences for knowledge, belief, truth and meaning. Rorty wants to dwell on where the arguments have led: to a position which allows us to dismiss the sceptic’s doubts, and so to abandon the attempt to provide a general justification for knowledge claims—a justification that is neither possible nor needed. Rorty sees the history of Western philosophy as a confused and victorless battle between unintelligible scepticism and lame attempts to answer it. Epistemology from Descartes to Quine seems to me just one complex, and by no means unilluminating, chapter in the philosophical enterprise. If that chapter is coming to a close, it will be through recourse to modes of analysis and adherence to standards of clarity that have always distinguished the best philosophy, and will, with luck and enterprise, continue to do so.

138

Siblings Under the Skin

Notes 1. 2. 3.

4.

5.

See my ‘True to the Facts’, printed originally in The Journal of Philosophy, Vol. 66 (1969), pp. 216-34. See my ‘On the Very Idea of a Conceptual Scheme’ (1974: 5-20). Many other passages in Quine suggest that Quine hopes to assimilate sensory causes to evidence. In Word and Object (1960, 22) he writes that ‘surface irritations…exhaust our clues to an external world.’ In Ontological Relativity (1969, 75), we find that ‘The stimulation of his sensory receptors is all the evidence anybody has had to go on, ultimately, in arriving at his picture of the world.’ On the same page: ‘Two cardinal tenets of empiricism remain unassailable.…One is that whatever evidence there is for science is sensory evidence. The other…is that all inculcation of meanings of words, must rest ultimately on sensory evidence.’ In The Roots of Reference (1974, 37-8), Quine says ‘observations’ are basic ‘both in the support of theory and in the learning of language’, and then goes on, ‘What are observations? They are visual, auditory, tactual, olfactory. They are sensory, evidently, and thus subjective… Should we say then that the observation is not the sensation…? No…’ Quine goes on to abandon talk of observations for talk of observation sentences. But of course observation sentences, unlike observations, cannot play the role of evidence unless we have reason to believe they are true. I now think it is essential, in doing radical interpretation, to include the desires of the speaker from the start, so that the springs of action and intention, namely both belief and desire, are related to meaning. But in the present talk it is not necessary to introduce this further factor. It is clear that the causal theory of meaning has little in common with the causal theories names and objects of which speakers may well be ignorant. The chance of systematic error is thus increased. My causal theory does the reverse by connecting the cause of a belief with its object.

References Davidson, Donald. [1969] 1984. True to the facts. Reprinted in Inquiries into Truth and Interpretation. Oxford: Oxford University Press. ——. 1974. On the very idea of a conceptual scheme. Proceedings and Addresses of the American Philosophical Association. Henrich, Dieter (ed.). 1983. Kant oder Hegel? Kett-Cotta.

Clough and Davidson

139

LePore, Ernest (ed.). 1986. Truth and interpretation: Perspectives on the philosophy of Donald Davidson. Oxford: Blackwell. Putnam, Hilary. 1978. Meaning and the moral science. London: Routledge & Kegan Paul. ——. 1983. Realism and reason. Cambridge: Cambridge University Press. Quine, W.O. 1960. Word and object. Cambridge: MIT Press. ——. 1969. Ontological relativity. New York: Columbia University Press. ——. 1974. The roots of reference. New York: Open Court Publishing. Rorty, Richard. 1978. Philosophy and the mirror of nature. Princeton: Princeton University Press.

Biographical note Donald Davidson was born in 1917 in Springfield, MA. After holding appointments across the country and abroad he is now Professor Emeritus of Philosophy, University of California at Berkeley. Davidson is one of the most influential philosophers in the analytic tradition and continues to be actively engaged in the philosophy community. Like his teacher, Quine, he is an avid traveller—in great demand as a speaker, he has delivered key note addresses at conferences around the world. Davidson is known for his work in the philosophy of language and the philosophy of mind and the intersection of these areas with epistemology. A series of his collected essays have been commisioned by Oxford—beginning with Essays on Actions and Events (1980) and Inquiries into Truth and Interpretation (1984) and his work has been honored with a volume in the prestigious series The Library of Living Philosophers (Open Court, 1999).

140

Siblings Under the Skin

Ramberg and Davidson

141

Charity and Ideology: The Field Linguist as Social Critic Bjørn Ramberg

The problem I want to raise in this paper will not be a problem for anyone who does not share both the following concerns or prejudices. The first is that we should take a materialist, extensional approach to linguistic meaning, specifically, the approach that is suggested by the work of Donald Davidson. The second is that social analysis must be emancipatory, and that this requires the conceptual possibility of postulating social structures that are in some sense hidden by discourse. The problem I want to address lies in a possible conflict between these concerns, both of which I will clarify, though not justify (except in so far as we take their compatibility to be a part of their respective justifications). The problem is this: Does the methodological core of Davidson’s semantics—the principle of charity—permit interpretation informed by a concept of ideological concealment? As with many of our philosophical problems, the existence of this one is quite precarious. We can avoid avoiding it only if we move along a rather specific route, which, I am afraid, will constitute a good chunk of my paper. Briefly, here is how it will go. I will begin by making some assumptions clear, first about ideology and the meta-theoretical dilemma of the critique of ideology, and then, at greater length, about radical interpretation and charity. The careful juxtaposition of an extensional view of language and the idea of language as concealment will reveal an apparently fundamental tension between the two. Then, bringing in the concepts of convention and social practice, I will argue that if we look really carefully, there is no problem after all. This might sound like an exercise in futility, and indeed it would be, unless the argument along the way provides some novel perspectives on either semantics or the critique of ideology or both. This, of course, is my hope, since what I have to say is intended ultimately to illuminate the concept of a language.

142

Siblings Under the Skin

I What do I mean by ideology? The concept can be-and is-used in all kinds of ways. Perhaps I ought to say, more guardedly, that we use “ideology” as a token for very different concepts. What I need now is a working definition, which you should regard as stipulative. To that end, I will begin by borrowing one of Anthony Giddens’ many fine formulations. Ideologies, he writes, are “structures of signification mobilized to legitimate the sectional interests of dominant groups, i.e., to legitimate exploitative domination” (Giddens, 1981, 61). Every word in this definition, with the exception of the preposition and the definite article, is assigned an explicit, specific and important location in Giddens’ theoretical framework for sociology, rendering the definition rich in substantive content. I will avoid any attempt at glossing Giddens’ theory, however, just because I do not want a very rich concept of ideology. I do not want to restrict the relevance of the argument to any specific sociological theory. For present purposes, the key point brought out by Giddens is that ideology serves as legitimation, of social structures that are both the medium and outcome of exploitative practices. Beyond this, I will simply beg two questions, questions that are closely related and in which it is otherwise very easy to get tangled up. Firstly, I will take it that not any legitimation is ideological. To expose the perceived legitimation of a practice (or whatever) as ideological, is to expose that practice as lacking legitimation. (Though it is conceivable that non-ideological legitimation may be found; in such cases, successful critique of ideology leads not so much to a change in the practice as to a clarification of its real purpose.) Secondly, I will assume that the so-called symmetry-thesis of the sociology of knowledge is false. This thesis, which emerges from sociology, anthropology and post-positivist philosophy of science, amounts to the claim that the truth or falsehood 1 of a belief is irrelevant to an explanation of why it is held. If this were true, the concepts of truth and ideology would be logically independent, and we would have no reason not to treat any legitimation as ideological. For the sake of the present argument, I will just dismiss the possibility that all legitimation is ideological. As a result of these additional stipulations, we are left with a concept of ideology as inherently both pernicious and deceptive. It is a systematic

Ramberg and Davidson

143

misrepresentation, distortion or concealment of social reality. Clearly, the social theorist who conceives of ideology in this manner, is thereby committed to some notion of truth which is not merely internal to any particular “structure of signification” (to use Giddens’ expression), form of life, or conceptual scheme. Hence, the primary concern of social theorists of this ilk has been to stave off cognitive relativism in its various forms. The theorist attempting to reveal the ideological nature of the legitimation of a particular social structure, must have not just a different interpretation of what is being expressed, but a better one. In diagnosing signification as ideologically distorted, she must have some normative theoretical backdrop against which this distortion can become visible. At the same time, she has some knowledge of history, and is leery of the hypostatization of the contingent mores and attributes of the particular spatio-temporal point at which she happens to feel at home. She does not want to resort to the cognitive imperialism to which cultural relativism arose as a radical response in the first place. To perform her emancipatory task, the theorist needs to preserve both the possibility of systematically different representations of the same reality and the possibility of a normative comparison of such representations. Accordingly, the Social Critic of my paper perceives herself to be caught between an untenable foundationalism in the form of universal tenets of truth or reason on the one side, and on the other an insidious form of laissez-faire conversationalism which leaves systematically concealed structures of domination sitting comfortably beneath a veil of well-bred tolerance. This is a tight squeeze to be in. Proclaiming it to be an old-fashioned squeeze, one we have “moved beyond”, does not appear to me to make it less tight. By raising the problem posed by the principle of charity, am I, then, simply dispensing another philosophical kick at someone who is already on the ground? No. The underlying ambition of this effort is not to create further theoretical difficulties for an already beleaguered position. It is, on the contrary, to hint at a possible solution to the social critic’s dilemma. On that note, I turn to Davidson’s semantic theory.

***

144

Siblings Under the Skin

II Davidson makes two requirements of a theory of meaning for a language L. It must give an interpretation of the sentences of L, and it must be testable. That is to say, a theory of meaning for L is a theoretical description of what we must know to understand what a speaker of L says, a theoretical description, moreover, that can be constructed without presupposing knowledge of the linguistic concepts of L. It will be extensional, holistic, and empirical. The following exposition of Davidson’s approach is an elaboration of each of these three features. A theory of meaning is extensional, on Davidson’s view, because the only primitive concept in the theory is the concept of truth. “Meanings” and “senses” carry no explanatory weight. A theory of meaning for L is given, roughly, by a Tarskian theory of truth for L. What is a Tarskian theory of truth? It is, first of all, a semantic theory. It is an answer to the question of what we mean by calling a given sentence of L true, not to the question of whether a given sentence of L is true. On the basis of a finite number of primitive bits of language and logical operations, the theory produces the truth-conditions of any possible sentence of L as its theorems, but does not touch on the epistemological question of when and whether these conditions do obtain. Because truth, as Davidson says, is beautifully transparent, a semantic theory of truth-in-L is an interpretation of L. Fortunately, it is not necessary to go into the mechanics of such a theory here, except to say that it must, naturally, have a finite axiomatic base of semantic primitives and hence, since its domain of application is potentially infinite, must contain recursive axioms. The theory is a formalization of the effects that words have on the truth conditions of sentences. This, indeed, is what the meanings of words amount to, on Davidson’s view: meanings are nothing but abstractions of the systematic effects of words on the truth conditions of the sentences in which they occur. It is easy to see that such a theory must also be holistic. The basis for the abstraction of the truth-functional significance of a semantic primitive, a word, can only be the totality of a language. The theory gives the meanings of words and sentences of L in so far as it assigns them a specific location and function in the semantic pattern that is

Ramberg and Davidson

145

L. It is the holistic element of Davidson’s theory that makes his extensional approach plausible. Let me illustrate. One might object that the truth conditions of a sentence cannot give its meaning, since sentences clearly can be co-extensional without being synonymous. So for example, there is no way to distinguish extensionally between the following hypothetical theorems (or T-sentences) of a theory of truth in English for German: [TI] “‘Der Schnee ist weiss’ is true if and only if snow is white.” [T2] “‘Der Schnee ist weiss’ is true if and only if grass is green.” Both T-sentences are true, but clearly only the former can plausibly be taken to give the meaning of the German sentence “Der Schnee ist weiss”. But we should not be misled by this fact to think that we cannot account for meaning extensionally. On Davidson’s holistic theory, we rule out [T2], not by a question-begging appeal to the meaning of the sentences, but by showing the impossibility of [T2] occurring as a theorem of a theory which abstracts the truth-functional role of the atomic parts of “Der Schnee ist weiss” from the indefinite number of German sentences in which they occur. The real significance of a T-sentence lies in its derivation, not in the trivial truth it iterates. This derivation is constrained by the fact that the theory must yield theorems with sentences matching the truth value of, in this case, not just “Der Schnee ist weiss”, but the truth value of every other German sentence about snow or the color white. The trivial nature (to English-German bilinguals) of the theorems, the T-sentences, attests to the truth of the theory itself. But the theory is far from trivial, embodying, as it does, the semantic structure of German, i.e., everything we need to know to understand German speakers. But even if we grant that meaning is really a matter of truth, and is given extensionally, how is such a theory constructed? How do we arrive at a theory of truth for L? This is to ask for the empirical content of a theory of meaning. And it is in the elucidation of this question that Davidson brings in the notion of radical interpretation. Imagine a situation where someone is attempting to translate into her own language a language L that is completely new to her and unrelated to anything she has ever heard or heard of, spoken by a group of people about whom she knows nothing. This person is the theoreti-

146

Siblings Under the Skin

cal Field Linguist, an abstract model of linguistic competence. She is placed in a context where any possibility of unacknowledged semantic information is defined away. Any interpretation of L-speakers she comes up with is based only on the sounds she hears them utter and what she sees them do. What Davidson adds to W. V. Quine’s account of the task of the field linguist, is the idea that the hypotheses ventured by the field linguist on the basis of her observations take the form of T-sentences, sentences like “‘s’ is true-in-L if and only if p”. Noticing that L-speakers occasionally utter, to use Quine’s example, “gavagai”, the field linguist tries to formulate a sentence that states as specifically as possible the combination of features that uniquely characterizes these occasions. She concludes that [T3] “Gavagai” is true-in-L if and only if there are rabbits hopping about in the speaker’s visual field. Collecting a good number of such hypotheses, the field linguist cannot help but discern patterns; she will individuate words, assign them structural roles, and so on. This will allow her to form testable hypotheses, theorems of her nascent theory, that she can try out on the L-speakers: will they utter the required ss under the circumstances p as predicted by the theory? Now whatever “gavagai” means, whatever the speaker intends to assert, the occasion of assertion can only count as evidence for [T3] on the assumption that “gavagai” is true or properly used. Once we admit the possibility that the speaker of L might falsely assert “gavagai” (whatever it means), the conjunct on the right side of [T3J is no longer supported by that utterance, because we would be giving up the idea that the observable features of this particular occasion of assertion of “gavagai” has any relation to the truth-conditions of “gavagai”. The field linguist cannot begin to formulate her theory unless she assumes that the L-speakers’ attitude of holding sentences true are appropriate on her sample occasions. But what if an L-speaker is only pretending to hold a sentence true, or actually does so but is mistaken in her belief? Surely we do not want to rule out mistakes or lies? No, but the point is that the only possible incentive the field linguist could have for attributing error or deceit is that a speaker’s utterance of “gavagai” conflicts with her inductively

Ramberg and Davidson

147

acquired T-sentences. And these she could only have formulated by treating the native speakers’ previous utterances of “gavagai” or other expressions in which any structural elements of “gavagai” occur as true. Certainly the field linguist will be exposed to error and deception, and against a body of strongly supported T-sentences she would probably recognize most of them. But should she be tempted to regard utterances of the subjects of interpretation as systematically false, she would merely be depriving her own theory of empirical bite, by cutting it off from its evidential base. Let me put the point this way. Let us say our field linguist encounters a conflict of this kind between two T-sentences. On the one hand, there is the T-sentence she formulates by generalizing from her observations of L-speakers’ utterances of a particular expressions and the contextual frames of these utterances as described by p. On the other hand, there is the T-sentence that she deduces from her rudimentary theory of truth for L as a prediction of the truth-conditions q of s. She then has the two T-sentences [T4] “s” is true if and only if p. [T5] “s” is true if and only if q. where p and q are assumed to be contraries. Schematically, there are three possible explanations of this conflict. The epistemic judgment of the field linguist herself might be mistaken. That is, the features of occasions of utterance of s that the field linguist identifies as p might not actually be features of these occasions. Secondly, the epistemic judgment of L-speakers might be mistaken. That is, the sentence s might be false under the conditions correctly identified by the field linguist as the conditions of the L-speakers’ assertion of s. In either of these cases, which are cases of epistemic error, the observation-based T-sentence [T4] would be false. Thirdly, the theory might be deficient. If this is the case, [T5] is false: The conditions q are not, contrary to the theorem, the truth-conditions of s. The first of these possibilities we can leave out. Not because an interpreter is never mistaken about how things are, but because she cannot attribute error to herself without changing her mind and so reformulate the observation-based T-sentence, [T4J, by altering p. If she chooses the second option, attributing epistemic error to the L -

148

Siblings Under the Skin

speaker, she preserves both her theory and her view of how things are, but at the price of disqualifying some of her own observations of the linguistic behavior of L-speakers as evidence for her theory. By declaring false [T4], the T-sentence which generalizes her observations with respect to the assertion of s, she in effect chooses to discount potential evidence. The third option, to blame the anomaly on the theory itself, is to declare false [T5], the T-sentence that falls deductively out of the theory. This is what is counselled by the so-called principle of charity. It is simply the application to theories of interpretation of the general principle of all theory construction that empirical content be maximized. Taking this route is to say that the meaning of the utterance in question was not understood, because the structural significance of its parts were misconstrued by the theory. We might be able to debug our theory by changing, for instance, the reference of a word and so both resolve the anomaly and preserve our evidential base. The result will be not just a new understanding of the utterance s, but an adjustment in our understanding of L as a whole, because every sentence in which the redefined word occurs will be directly affected. As long as such theoretical tampering results in the incorporation of more observation-based T-sentences into our evidential base, we should indeed tamper. Because the larger our evidential base, the more precisely we will be able to specify the truth-functional significance of the semantic primitives. That is to say, the more true assertions we attribute to L-speakers, the more clearly will we understand what the words of L mean.

*** III What the radical interpretation model of linguistic communication seems to imply, is that if we want to know what a speaker means, we have no choice but to treat her as on the whole a speaker of truth. It is important to realize that this is not a credo of some kind, it is not an expression of faith in the sincerity or epistemic acumen of speakers. For just as we have no choice, if we want to make sense of what oth-

Ramberg and Davidson

149

ers say, but to regard them as on the whole speaking the truth, so we have no choice but to regard ourselves as largely speakers of truth. The principle of charity is no over-estimation of our cognitive and communicative capacities. It has, rather, a somewhat deflationary effect on our self-esteem insofar as it implies that we cannot lie even if we try. Cannot lie very much, that is, just as we cannot be dramatically and romantically mistaken about how things are. Massive error, just like massive deception, presupposes an epistemic, representational relation between language and the world and a corresponding reification of the meanings of words. The unfortunately named principle of charity appears charitable only against the background provided by 2 this conception. Once we come to see the relation between language and world as a causal semantic relation, the principle of charity just becomes the expression of the naturalistic view of what it is for words to mean that runs through Davidson’s work. Can we conclude, then, that Davidson takes linguistic competence to be the knowledge of a language L as rationalized by a Tarskian theory of truth for L? I am afraid not. For what is it we know when we know a language? What is this L? Throughout I have been talking as if the identity of L was given. Though she does not know L, the linguist at least knows that L is the language to be cracked. But once the full import of Davidson’s holistic empirical strategy has soaked in, we will see that the only possible criterion of identity for languages is the successful application of one and the same theory of truth. For Davidson, there is no way in which a language can be given independently of the truththeory that determines its structure. But can we simply identify the concept of a language with the L in the formula “true-in-L” that is the core of a Tarskian theory of truth? The concept of a language as given by true in-L was the board on which Davidson stood while developing a theory of meaning without meanings. Now it turns out that this board is itself in need of replacement because it is a reification. For we never apply the exact same theory to any two speakers nor even to any one speaker at different points in time. And this has lead Davidson to reject, first, the idea that convention plays any essential role in linguistic communication and, more recently, also the idea that the concept of a language can be given theoretical content.3

150

Siblings Under the Skin

His reasons are fairly straightforward. Convention (on, e.g., David Lewis’ analysis) is a certain kind of conformity to a regularity. But if we all speak our different idiolects, there is nothing for this regularity to consist in. Nor do speakers’ actual usage have to conform to our expectations of their usage as a condition for our understanding what is meant: As field linguists, we understand each other without having to make any assumptions about whether this or that language is spoken, which is to say that no conventions need be adhered to. The radical interpretation-model rationalizes our linguistic competence not, as Davidson probably originally had thought, in terms of Tarskian theories of truth for languages. The essential competence turns out to be the ability to continuously form and reform such theories by interpreting assertions as, on the whole, true. The knowledge rationalized by any one given theory—the knowledge of a language, if we like—is fleeting, and, beyond the particular occasion of communicative exchange to which it is applied, utterly without theoretical worth. Linguistic competence, on Davidson’s view, is essentially dynamic, and no more subject to theoretical rationalization than theory-formation generally. What Davidson shows is that communication does not depend, as we naturally tend to suppose, on our speaking the same language. Radical interpretation will in principle succeed as long as we regard speakers as speaking the truth, continuously forming our theories of meaning on that assumption. The question is, where does this leave the social critic? She wants to talk about systematic linguistic distortion. She wants to analyze discourse , texts, etc., and make explicit the truth conditions of the assertions that in her view are actually being made, and then show that these conditions do not obtain. That is, she brings out the hidden meaning of discourse, and then exposes these concealed assertions as systematically false. But how can she justify her claim that these presumed hidden assertions are being made at all? The fact that her interpretation yields not just the occasional deception or error, but massive distortion, is on the view developed here simply evidence against her interpretation. She is sinning against the principle of charity, which is not, I must stress again, a recommendation of a particular interpretational strategy but a necessary condition of a theory of meaning having empirical content at all. The dilemma is this: If linguistic communication hinges essentially on our treating assertions largely as true—not on their belonging to some

Ramberg and Davidson

151

linguistic structure, like a language, given somehow independently of the particular occasions of assertion—then to say that we understand the assertions embedded in a discourse to be systematically false is a contradiction in terms. Can this contradiction be resolved? I think so. Let me begin with a brief analysis of the argument between Davidson and Michael Dummett on the concept of a language.

*** IV Dummett disagrees with Davidson on most of the central issues in contemporary analytic semantics (with the significance of Frege’s work being a notable exception). He disagrees, specifically, with Davidson’s dismissal from semantics of the concept of a language. A language, Dummett objects, “is an existing pattern of communicative speech: it is not a theory, but a phenomenon” (Dummett, 1986, 467). Sure, we all have our own evolving idiolects, and no one idiolect perfectly matches a language. But idiolects can be related in ways which make it plausible to claim that they are idiolects of the same language. This language, as an existing pattern, is a social practice, and as such is specifiable in terms of the conventions that constitute it. The fact that no one idiolect embodies all the conventions at all times, does not make any difference: These conventions are inductive generalizations over a number of speakers or idiolects, and are no less real because the pure language constituted by them is an idealization. None of this is actually inconsistent with Davidson’s views. He does not deny that there are conventions involved in communication, only that there necessarily and in principle must be. In practice, there certainly are regularities of idiolects that make it natural to group them into natural languages. What Davidson doubts, is that the concept of a language in this sense can throw any theoretical light on linguistic meaning. And here is where he and Dummett differ. For Dummett takes it that linguistic communication does essentially involve conventions, while Davidson speaks of rule-governed repetition as “a usual, though contingent, feature” (Davidson, 1982, 280) of such communication.

152

Siblings Under the Skin

There is a shared premise in this disagreement, namely the idea that if the concept of a language has any theoretically interesting content, it must come from a rationalization of linguistic competence. So, from the fact that languages are convention-governed, Dummett concludes that linguistic meaning is essentially conventional. And from the fact that communication does not, as the radical interpretationmodel shows, presuppose convention, Davidson concludes that there is nothing for a language to be. My suggestion is that we drop the shared premise of these inferences. That way, we can without contradiction claim both that languages are convention-governed social practices and that a theory of meaning conceived as a rationalization of a speaker’s linguistic competence does not involve the knowledge of conventions. In short, whether you understand what someone says and whether you know what language is being spoken are theoretically independent questions. In practice, they are of course wholly interdependent. This is because interpretation normally proceeds by way of the theoretical shortcuts which are by definition excluded from the repertoire of the radical interpreter. These shortcuts can include a host of beliefs (psychological, biological, sociological, historical, etc.), beliefs that we rely on to isolate the salient features of contexts of utterances when we are engaged in not-so-radical interpretation. But most commonly, these shortcuts are just the conventions of a language. To such an extent do we rely on these that Dummett thinks Davidson misrepresents matters when he claims that the radical interpretation-model is a model of all linguistic communication, domestic as well as foreign. We do not, Dummett thinks, interpret domestic utterances. This is true, insofar as we do not very often go about as field linguists collecting the truth conditions of the utterances of our neighbors in order to construct a theory of truth for their language. But Dummett misses Davidson’s point. It is that while the meanings or truth conditions of sentences of a natural language may well be assumed in most contexts to be conventionally given, and so not in practice determined by interpretation, the only way to theoretically specify what these sentences mean is to construct a theory of truth. And to this end, conventions make no contribution. But is not this just what Davidson was claiming when he said that conventions are a common but contingent feature of linguistic commu-

Ramberg and Davidson

153

nication? Is anything added to Davidson’s account? Yes. What is added by the divorcing of the concept of a language from that of linguistic competence is the possibility of reinstating the concept of a language as theoretically significant. And this is a good thing, because the notion of a language as an existing convention-governed practice is necessary, as both Dummett and others have pointed out, to explain, for instance, our propensity to correct ourselves when we are made aware of having misused an expression, even in cases where we are actually understood in spite of our linguistic error. There are, in fact, any number of things we need the concept of a language to account for. But the one that is of particular interest here, is the phenomenon of ideology, or systematic distortion. If, as I have suggested, we detach the concept of a language as a social practice, in which conventions govern the use of expressions, from the semantic concept of meaning, we can resolve the conflict between ideology and charity. Or so I will claim. Here is how I propose to do it. What, in practice, governs our use of expressions are the conventions of our language. What are these conventions? They are, among other things, the conventional determinations of the truth-conditions of sentences. As Dummett claims, we do not, usually, anyway, interpret assertions made in our own language. In that sense, the notion of radical interpretation is not true to our linguistic practice. But if we want to make theoretically explicit the meaning of sentences of our own language—if we ask, how do we know we are speaking the same language—then radical interpretation as homophonic interpretation comes into play. But in a normal speech situation, this will not happen, for here conventions, not the constructing of a truth-theory, determine what truth-conditions we attach to utterances. The point is this: Insofar as we are speakers of a language, the truth-conditions of the sentences of that language are conventionally taken for granted. But linguistic meaning does not essentially involve conventions. There is, then, no reason to suppose a priori that the conventionally attached truth-conditions of sentences of a language and the truth-theoretically determined truth-conditions of the assertions that speakers make, necessarily coincide. My thesis is that a certain kind of discrepancy between the conventionally taken-for-granted truth-conditions of the sentences of a language and the truth-conditions of the

154

Siblings Under the Skin

assertions we use those sentences to make is ideologically determined. In other words, the conventions of a language as a communicative social practice, may conceal the actual truth-conditions of what is being asserted. Since the actual truth-conditions of our assertions, and so the meaning of what we say, are given only by a truth-theory, we may be duped by ideology to the extent that we do not ordinarily understand utterances by subjecting them to radical interpretation. Ideology conceals, then, as, in effect, a false theory of meaning embedded in the conventions that govern a language as an existing structure. Such false theories of meaning would appear to be plausible candidates for the “structures of signification” that Giddens speaks of, structures that may, as he says, be “mobilized to legitimate the sectional interests of dominant groups” (Giddens, 1981, 61, see 638 above). How can a false theory of meaning serve an ideological purpose? It does so by trading on the dialectical relation between a speaker’s perception of how things are, and her perception of the meaning of a description of it. As an example, let us imagine the following description of the social system of the Republic of South Africa: “South Africa is a country where each population group is allowed to realize its own distinct character without interference, a country in which the natural and systematic differences between groups of people are recognized and acted upon, in such a way that every individual, no matter what her particular racial background might be, is treated fairly and in a manner which is tailored to meet her natural capacities, needs and desires.” Without the concept of a language, this little text would provide no interesting problem of semantics. The interpreter, the reading or listening subject, being true to her method, might very well regard these sentences as true, but, to the extent that she is familiar with life in South Africa, she would not understand by them what we as speakers of English take them to mean (and so not, presumably, what they would be intended to mean). But this leaves unanswerable not just the question of why lies are pernicious, but also of how such lies are even possible to effect. If, on the other hand, the subject assumes implicitly that the description is given in conventional English, she can, if she is familiar with life in South Africa, respond in at least two ways. She can,

Ramberg and Davidson

155

first of all, dismiss it as false. Again, there is no problem of semantics here—but then there is no problem of ideology, either. A false statement is recognized and treated as a false statement, and that is that. The problem of ideology arises only when the subject is unable to write the description off as a lie, yet is in a situation where the relation between how she perceives things to be and the conventional truth-conditions of the description is problematic. Such a situation may arise, for instance, when the description is backed by authority of one form or another. In these circumstances, a three-way tension emerges between the subject’s perception of the world, her perception of the conventions of her language, and her perception of the meaning of the description, a tension that does not immediately resolve itself in the view that the description is false. And it is precisely by taking advantage of this tension that ideological distortions can do their work. What gives way in the victim of ideological mystification is not her belief in the truth-value of what she hears, but the other two elements in the triangle; her immediate understanding of the conventions of the language she thinks she speaks, and her perception of social reality. Ideology works by undermining, but not destroying, the subject’s linguistic grip on the world. If the subject is deprived entirely of familiar meanings in familiar circumstances, meanings anchored by solid conventions, that is, if she loses her language completely then the conventions of her language are useless to the ideologue. The subject could not be linguistically mislead; she would, in a sense, be free to interpret the meaning of the description entirely in the light of what she observed. Of course, she might still be misled, since no one is a perfect observer, but she could not be misled by ideology, since there would be no linguistic conventions through which it could work. On the other hand, if the South African ideologues did not to some extent succeed in undermining the subject’s grasp of the conventions of the language she speaks, the ideological effect could not be achieved. The distortions in the description above would merely rise to the surface of language as a grotesque lies about South Africa, rather than maintain a political potent subterranean existence, guiding and subverting linguistic understanding. When it is successful, ideology literally consumes a language. Were there such a thing as a static language, it would not be susceptible to

156

Siblings Under the Skin

the workings of ideological distortions, because the effect is produced precisely in this continuous consumption, which is possible only because languages change over time. Ideology both uses and undermines the present conventions of a language, and thereby gradually alters the conventional meanings of the words, or rather, the truth-conditions of sentences. These altered truth-conditions must in turn be exploited and thus undermined, so that a clear theory of meaning is blocked and the tension in the semantic understanding of the victims of ideology is continuously maintained. The effecting of ideological legitimation is a dynamic process, a never completed balancing act. It works through the temporal viscosity of a language, trading equally the historical plasticity of linguistic conventions on the one hand, and the resistance to change of the linguistic understanding of the individual subject on the other. This means, naturally, that the critique of ideology, too, is a never-ending labour, a continuous struggle to clarify meaning, to recapture the efficacy of language. And it is the fact that linguistic meaning does not essentially involve conventions that theoretically enables us to penetrate the concealment effected by ideologically-loaded linguistic conventions. To the extent which is perhaps not a very great extent, that we are able to suspend the conventions of a language, at least within a particular domain, and place ourselves in a context of radical interpretation, to that extent are we able to identify the actual truth-conditions of what is being asserted. This is why the project of finding new ways of talking is central to the critique of ideology, and here the semantic counter-innovation of poetry and prose is an integral part. If conventions were essential to a theoretical model of linguistic communication, this hermeneutical effort would be in principle impossible. On the other hand, if a language as a concrete social practice were not convention governed, there could be no ideology to expose. The critique of ideology is a matter not of attributing systematically false assertions to speakers, but of formulating and exposing as systematically distorted the theory of truth by which victims of ideology normally understand what is being asserted. If ideology is a false theory of meaning, ideologically subverted conventions of language distort not merely as a lie distorts what is true. To get truth-conditions wrong is to get meaning wrong; it is to blur the meaning of what is said. Therefore,

Ramberg and Davidson

157

ideology distorts by obfuscating the meaning of what is said. We might say, quite literally, that ideology works by concealing what language it is that is actually being spoken. More specifically, it perpetuates semantic ambiguities and unclarities structured with precision to conceal contradictions such as that between our sense of what is fair and the notion that apartheid is fair. The answer to this ideological concealment of contradictions is to make explicit the incompatibility of the elements in the triangle of tension formed by ideologically distorted meaning, the conventional roots of the subject’s linguistic understanding, and the subject’s perception of the world. By doing this, the critique of ideology attempts to recover the subject’s ability to interpret ideological distortions simply as lies, and to resurrect her ability to perceive actual contradictions between ideological depictions and the social reality she lives in. So for example, the description of South Africa offered above has ideological potency only with respect to a subject whose linguistic understanding remains rooted in the conventional meaning of words like “fair”, “natural”, “character”, “desire”, “need”, and so on, while at the same time being sufficiently blunted or subverted to accept the use of those words in the description and to accept the description as true. How this subtle form of destruction, of both language and perception, is achieved is not directly relevant to the present semantic point, though the concrete critique of ideology, practical radical interpretation, will presuppose an understanding of the kinds of techniques of linguistic manipulation, indoctrination and subversion that perpetuate ideological legitimation. To conclude: It turns out not only that charity and ideology are compatible concepts, but, importantly, that radical interpretation provides a model for the critique of ideology. Based on a concept of truth that is not in internal to any particular scheme, but is a precondition for any linguistic understanding, this model does not give rise to the problems of cognitive relativism. It construes the concealment effected by ideological structures of signification as a matter both of systematically false assertions and of distortions of meaning, by making the clarity of what is meant dependent on the truth of what is asserted. The locus of ideological distortion is the conventions that govern the social practice that constitutes a language. Its modus is the exploitation of the plastic-

158

Siblings Under the Skin

ity of the subject’s perception of these conventions. Ideology can in, principle be exposed and subjected to critique precisely insofar as there is available a model of linguistic meaning on which linguistic communication is not theoretically convention-bound. This is the project of the social critic, who takes on the task of uncovering the ideological subversion of the conventions of a language in radical interpretation. By assuming the attitude of the field linguist to what she thought was her own language, the social critic works to reappropriate language as a vehicle of truth.

Ramberg and Davidson

159

Notes 1.

2. 3.

For expositions of the symmetry thesis, see for instance David Barnes and Barry Bloor, “Relativism, Rationalism and the Sociology of Knowledge”, in Martin Hollis and Steven Lukes, eds., Rationality and Relativism (Oxford: Blackwell, 1982); Mary Hesse, “The Strong Thesis of Sociology of Science”, in her Revolutions and Reconstructions in the Philosophy of Science (Bloomington, IN: Indiana University Press, 1980). Davidson now speaks of a principle of rational accommodation, rather than a principle of charity. See “Communication and Convention”, in Davidson (1982); see also Davidson (1986).

References Davidson, D. 1982. Inquiries into truth and interpretation. Oxford: Clarendon Press. ——. 1986. A nice derangement of epitaphs. In E. LePore, ed., Truth and interpretation. Oxford: Blackweli. Dummett, M. 1986. Comments on Davidson and Hacking. In E. LePore, ed., Truth and interpretation. Oxford: Blackwell. Giddens, A., 1981. A contemporary critique of historical materialism, vol. 1 London: MacMillan.

Biographical note Bjørn Ramberg was born in Oslo (Norway) in 1961. He was an undergraduate student at the University of Oslo (1979-82), and received his MA (1983)and PhD (1987) in Philosophy from Queen’s University at Kingston. He was a faculty member at Simon Fraser University from 1991 until 2000. In 1997 he moved with his family back to Norway, and is currently a member of the philosophy department at the University of Oslo. An inveterate metaphilosopher, Ramberg has published papers on the work of Donald Davidson, Richard Rorty, Daniel Dennett, and Hans-Georg Gadamer. Bjørn’s interest in the work of Davidson began in graduate school, which eventually led him to write Donald Davidson’s Philosophy of Language: An

160

Siblings Under the Skin

Introduction (Blackwell, 1989), a book enthusiastically praised by Davidson. Bjørn and he have engaged in fruitful philosophical exchange ever since. Most recently, Bjørn was asked to write an essay for the Davidson volume of the Library of Living Philosophers (Open Court, 1999). The essay is titled “The Significance of Charity,” and Davidson’s warm response is precisely the sort that nurtures the philosophical team-work documented in Siblings Under the Skin. Of Bjørn’s essay, Davidson writes: “This is the kind of criticism that makes doing philosophy worth the trouble… Bjørn Ramberg… has shown me a better way to put together some of the strands in my thinking, and gently suggested what I would do well to leave out.” When not writing philosophy, Bjørn and his dog Thea take to the wilds of Norway for winter hikes—apparently, even Canada did not have enough snow for him. While he misses the many colleagues and friends he and his family have met while abroad, he’s happy to be home.

Ramberg and Davidson

161

A Nice Derangement of Epitaphs Donald Davidson

Goodman Ace wrote radio sitcoms. According to Mark Singer, Ace often talked the way he wrote: Rather than take for granite that Ace talks straight, a listener must be on guard for an occasional entre nous and me…or a long face no see. In a roustabout way, he will maneuver until he selects the ideal phrase for the situation, hitting the nail right on the thumb. The careful conversationalist might try to mix it up with him in a baffle of wits. In quest of this pinochle of success, I have often wrecked my brain for a clowning achievement, but Ace’s chickens always come home to roast. From time to time, Ace will, in a jerksome way, monotonize the conversation with witticisms too humorous to mention. It’s high noon someone beat him at his own game, but I have never done it: cross my eyes and hope to die, he always wins thumbs down.1

I quote at length because philosophers have tended to neglect or play down the sort of language-use this passage illustrates. For example, Jonathan Bennett writes, I doubt if I have ever been present when a speaker did something like shouting ‘Water!’ as a warning of fire, knowing what ‘Water!’ means and knowing that his hearers also knew, but thinking that they would expect him to give to ‘Water!’ the normal meaning of ‘Fire!’2

Bennett adds that, ‘Although such things could happen, they seldom do.’ I think such things happen all the time; in fact, if the conditions are generalized in a natural way, the phenomenon is ubiquitous. Singer’s examples are special in several ways. A malapropism does not have to be amusing or surprising. It does not have to be based on a cliché, and of course it does not have to be intentional. There need be

162

Siblings Under the Skin

no play on words, no hint of deliberate pun. We may smile at someone who says ‘Lead the way and we’ll precede’, or, with Archie Bunker, ‘We need a few laughs to break up the monogamy’, because he has said something that, given the usual meanings of the words, is ridiculous or fun. But the humor is adventitious. Ace’s malaprops generally make some sort of sense when the words are taken in the standard way, as in ‘Familiarity breeds attempt’, or ‘We’re all cremated equal’, but this is not essential (‘the pinochle of success’). What is interesting is the fact that in all these cases the hearer has no trouble understanding the speaker in the way the speaker intends. It is easy enough to explain this feat on the hearer’s part: the hearer realizes that the ‘standard’ interpretation cannot be the intended interpretation; through ignorance, inadvertence, or design the speaker has used a word similar in sound to the word that would have ‘correctly’ expressed his meaning. The absurdity or inappropriateness of what the speaker would have meant had his words been taken in the ‘standard’ way alerts the hearer to trickery or error; the similarity in sound tips him off to the right interpretation. Of course there are many other ways the hearer might catch on; similarity of sound is not essential to the malaprop. Nor for that matter does the general case require that the speaker use a real word: most of ‘The Jabberwocky’ is intelligible on first hearing. It seems unimportant, so far as understanding is concerned, who makes a mistake, or whether there is one. When I first read Singer’s piece on Goodman Ace, I thought that the word ‘malaprop’, though the name of Sheridan’s character, was not a common noun that could be used in place of ‘malapropism’. It turned out to be my mistake. Not that it mattered: I knew what Singer meant, even though I was in error about the word; I would have taken his meaning in the same way if he had been in error instead of me. We could both have been wrong and things would have gone as smoothly. This talk of error or mistake is not mysterious nor open to philosophical suspicions. I was wrong about what a good dictionary would say, or what would be found by polling a pod of experts whose taste or training I trust. But error or mistake of this kind, with its associated notion of correct usage, is not philosophically interesting. We want a deeper notion of what words, when spoken in context, mean;

Ramberg and Davidson

163

and like the shallow notion of correct usage, we want the deep concept to distinguish between what a speaker, on a given occasion, means, and what his words mean. The widespread existence of malapropisms and their kin threatens the distinction, since here the intended meaning seems to take over from the standard meaning. I take for granted, however, that nothing should be allowed to obliterate or even blur the distinction between speaker’s meaning and literal meaning. In order to preserve the distinction we must, I shall argue, modify certain commonly accepted views about what it is to ‘know a language’, or about what a natural language is. In particular, we must pry apart what is literal in language from what is conventional or established. Here is a preliminary stab at characterizing what I have been calling literal meaning. The term is too incrusted with philosophical and other extras to do much work, so let me call what I am interested in first meaning. The concept applies to words and sentences as uttered by a particular speaker on a particular occasion. But if the occasion, the speaker, and the audience are ‘normal’ or ‘standard’ (in a sense not to be further explained here), then the first meaning of an utterance will be what should be found by consulting a dictionary based on actual usage (such as Webster’s Third). Roughly speaking, first meaning comes first in the order of interpretation. We have no chance of explaining the image in the following lines, for example, unless we know what ‘foison’ meant in Shakespeare’s day: Speak of the spring and foison of the year, The one doth shadow of your beauty show, The other as your bounty doth appear….3

Little here is to be taken literally, but unless we know the literal, or first, meaning of the words we do not grasp and cannot explain the image. But ‘the order of interpretation’ is not at all clear. For there are cases where we may first guess at the image and so puzzle out the first meaning. This might happen with the word ‘tires’ in the same sonnet: On Helen’s cheek all art of beauty set, And you in Grecian tires are painted new.

164

Siblings Under the Skin

And of course it often happens that we can descry the literal meaning of a word or phrase by first appreciating what the speaker was getting at. A better way to distinguish first meaning is through the intentions of the speaker. The intentions with which an act is performed are usually unambiguously ordered by the relation of means to ends (where this relation may or may not be causal). Thus the poet wants (let us say) to praise the beauty and generosity of his patron. He does this by using images that say the person addressed takes on every good aspect to be found in nature or in man or woman. This he does in turn by using the word ‘tire’ to mean ‘attire’ and the word ‘foison’ to mean ‘harvest’. The order established here by ‘by’ can be reversed by using the phrase ‘in order to’. In the ‘in order to’ sequence, first meaning is the first meaning referred to. (‘With the intention of ’ with ‘ing’ added to the verb does as well.) Suppose Diogenes utters the words ‘I would have you stand from between me and the sun’ (or their Greek equivalent) with the intention of uttering words that will be interpreted by Alexander as true if and only if Diogenes would have him stand from between Diogenes and the sun, and this with the intention of asking Alexander to move from between him and the sun, and this with the intention of getting Alexander to move from between him and the sun, and this with the intention of leaving a good anecdote to posterity. Of course these are not the only intentions involved; there will also be the Gricean intentions to achieve certain of these ends through Alexander’s recognition of some of the intentions involved. Diogenes’ intention to be interpreted in a certain way requires such a self-referring intention, as does his intention to ask Alexander to move. In general, the first intention in the sequence to require this feature specifies the first meaning. Because a speaker necessarily intends first meaning to be grasped by his audience, and it is grasped if communication succeeds, we lose nothing in the investigation of first meaning if we concentrate on the knowledge or ability a hearer must have if he is to interpret a speaker. What the speaker knows must correspond to something the interpreter knows if the speaker is to be understood, since if the speaker is understood he has been interpreted as he intended to be interpreted. The abilities of the speaker that go beyond what is required of an interpreter—invention and motor control—do not concern me here.

Ramberg and Davidson

165

Nothing said so far limits first meaning to language; what has been characterized is (roughly) Grice’s non-natural meaning, which applies to any sign or signal with an intended interpretation. What should be added if we want to restrict first meaning to linguistic meaning? The usual answer would, I think, be that in the case of language the hearer shares a complex system or theory with the speaker, a system which makes possible the articulation of logical relations between utterances, and explains the ability to interpret novel utterances in an organized way. This answer has been suggested, in one form or another, by many philosophers and linguists, and I assume it must in some sense be right. The difficulty lies in getting clear about what this sense is. The particular difficulty with which I am concerned in this paper (for there are plenty of others) can be brought out by stating three plausible principles concerning first meaning in language: we may label them by saying they require that first meaning be systematic, shared, and prepared. (1) First meaning is systematic. A competent speaker or interpreter is able to interpret utterances, his own or those of others, on the basis of the semantic properties of the parts, or words, in the utterance, and the structure of the utterance. For this to be possible, there must be systematic relations between the meanings of utterances. (2) First meanings are shared. For speaker and interpreter to communicate successfully and regularly, they must share a method of interpretation of the sort described in (1). (3) First meanings are governed by learned conventions or regularities. The systematic knowledge or competence of the speaker or interpreter is learned in advance of occasions of interpretation and is conventional in character. Probably no one doubts that there are difficulties with these conditions. Ambiguity is an example: often the ‘same’ word has more than one semantic role, and so the interpretation of utterances in which it occurs is not uniquely fixed by the features of the interpreter’s competence so far mentioned. Yet, though the verbal and other features of the context of utterance often determine a correct interpretation, it is not easy or perhaps even possible to specify clear rules for disambiguation. There are many more questions about what is required

166

Siblings Under the Skin

of the competent interpreter. It does not seem plausible that there is a strict rule fixing the occasions on which we should attach significance to the order in which conjoined sentences appear in a conjunction: the difference between ‘They got married and had a child’ and ‘They had a child and got married.’ Interpreters certainly can make these distinctions. But part of the burden of this paper is that much that they can do ought not to count as part of their basic linguistic competence. The contrast in what is meant or implied by the use of ‘but’ instead of ‘and’ seems to me another matter, since no amount of common sense unaccompanied by linguistic lore would enable an interpreter to figure it out. Paul Grice has done more than anyone else to bring these problems to our attention and to help sort them out. In particular, he has shown why it is essential to distinguish between the literal meaning (perhaps what I am calling first meaning) of words and what is often implied (or implicated) by someone who uses those words. He has explored the general principles behind our ability to figure out such implicatures, and these principles must, of course, be known to speakers who expect to be taken up on them. Whether knowledge of these principles ought to be included in the description of linguistic competence may not have to be settled: on the one hand they are things a clever person could often figure out without previous training or exposure and they are things we could get along without. On the other hand they represent a kind of skill we expect of an interpreter and without which communication would be greatly impoverished. I dip into these matters only to distinguish them from the problem raised by malapropisms and the like. The problems touched on in the last two paragraphs all concern the ability to interpret words and constructions of the kind covered by our conditions (l)—(3); the questions have been what is required for such interpretation, and to what extent various competencies should be considered linguistic. Malapropisms introduce expressions not covered by prior learning, or familiar expressions which cannot be interpreted by any of the abilities so far discussed. Malapropisms fall into a different category, one that may include such things as our ability to perceive a well-formed sentence when the actual utterance was incomplete or grammatically garbled, our ability to interpret words we have never heard before,

Ramberg and Davidson

167

to correct slips of the tongue, or to cope with new idiolects. These phenomena threaten standard descriptions of linguistic competence (including descriptions for which I am responsible). How should we understand or modify (1)—(3) to accommodate malapropisms? Principle (1) requires a competent interpreter to be prepared to interpret utterances of sentences he or she has never heard uttered before. This is possible because the interpreter can learn the semantic role of each of a finite number of words or phrases and can learn the semantic consequences of a finite number of modes of composition. This is enough to account for the ability to interpret utterances of novel sentences. And since the modes of composition can be iterated, there is no clear upper limit to the number of sentences utterances of which can be interpreted. The interpreter thus has a system for interpreting what he hears or says. You might think of this system as a machine which, when fed an arbitrary utterance (and certain parameters provided by the circumstances of the utterance), produces an interpretation. One model for such a machine is a theory of truth, more or less along the lines of a Tarski truth definition. It provides a recursive characterization of the truth conditions of all possible utterances of the speaker, and it does this through an analysis of utterances in terms of sentences made up from the finite vocabulary and the finite stock of modes of composition. I have frequently argued that command of such a theory would suffice for interpretation.4 Here however there is no reason to be concerned with the details of the theory that can adequately model the ability of an interpreter. All that matters in the present discussion is that the theory has a finite base and is recursive, and these are features on which most philosophers and linguists agree. To say that an explicit theory for interpreting a speaker is a model of the interpreter’s linguistic competence is not to suggest that the interpreter knows any such theory. It is possible, of course, that most interpreters could be brought to acknowledge that they know some of the axioms of a theory of truth; for example, that a conjunction is true if and only if each of the conjuncts is true. And perhaps they also know theorems of the form ‘An utterance of the sentence “There is life on Mars” is true if and only if there is life on Mars at the time of the utterance.’ On the other hand, no one now has explicit knowledge of a fully satisfactory theory for interpreting the speakers of any natural language.

168

Siblings Under the Skin

In any case, claims about what would constitute a satisfactory theory are not, as I said, claims about the propositional knowledge of an interpreter, nor are they claims about the details of the inner workings of some part of the brain. They are rather claims about what must be said to give a satisfactory description of the competence of the interpreter. We cannot describe what an interpreter can do except by appeal to a recursive theory of a certain sort. It does not add anything to this thesis to say that if the theory does correctly describe the competence of an interpreter, some mechanism in the interpreter must correspond to the theory. Principle (2) says that for communication to succeed, a systematic method of interpretation must be shared. (I shall henceforth assume there is no harm in calling such a method a theory, as if the interpreter were using the theory we use to describe his competence.) The sharing comes to this: the interpreter uses his theory to understand the speaker; the speaker uses the same (or an equivalent) theory to guide his speech. For the speaker, it is a theory about how the interpreter will interpret him. Obviously this principle does not demand that speaker and interpreter speak the same language. It is an enormous convenience that many people speak in similar ways, and therefore can be interpreted in more or less the same way. But in principle communication does not demand that any two people speak the same language. What must be shared is the interpreter’s and the speaker’s understanding of the speaker’s words. For reasons that will emerge, I do not think that principles (1) and (2) are incompatible with the existence of malapropisms; it is only when they are combined with principle (3) that there is trouble. Before discussing principle (3) directly, however, I want to introduce an apparent diversion. The perplexing issue that I want to discuss can be separated off from some related matters by considering a distinction made by Keith Donnellan, and something he said in its defense. Donnellan famously distinguished between two uses of definite descriptions. The referential use is illustrated as follows: Jones says ‘Smith’s murderer is insane’, meaning that a certain man, whom he (Jones) takes to have murdered Smith, is insane. Donnellan says that even if the man that Jones believes to have murdered Smith did not murder Smith, Jones has

Ramberg and Davidson

169

referred to the man he had in mind; and if that man is insane, Jones has said something true. The same sentence may be used attributively by someone who wants to assert that the murderer of Smith, whoever he may be, is insane. In this case, the speaker does not say something true if no one murdered Smith, nor has the speaker referred to anyone. In reply, Alfred MacKay objected that Donnellan shared Humpty Dumpty’s theory of meaning: ‘“When I use a word”, Humpty Dumpty said, ...” it means just what I choose it to mean.”’ In the conversation that went before, he had used the word ‘glory’ to mean ‘a nice knockdown argument’. Donnellan, in answer, explains that intentions are connected with expectations and that you cannot intend to accomplish something by a certain means unless you believe or expect that the means will, or at least could, lead to the desired outcome. A speaker cannot, therefore, intend to mean something by what he says unless he believes his audience will interpret his words as he intends (the Gricean circle). Donnellan says, If I were to end this reply to MacKay with the sentence ‘There’s glory for you’ I would be guilty of arrogance and, no doubt, of overestimating the strength of what I have said, but given the background I do not think I could be accused of saying something unintelligible. I would be understood, and would I not have meant by ‘glory’ ‘a nice knockdown argument’?5

I like this reply, and I accept Donnellan’s original distinction between two uses of descriptions (there are many more than two). But apparently I disagree with some view of Donnellan’s, because unlike him I see almost no connection between the answer to MacKay’s objection and the remarks on reference. The reason is this. MacKay says you cannot change what words mean (and so their reference if that is relevant) merely by intending to; the answer is that this is true, but you can change the meaning provided you believe (and perhaps are justified in believing) that the interpreter has adequate clues for the new interpretation. You may deliberately provide those clues, as Donnellan did for his final ‘There’s glory for you.’ The trouble is that Donnellan’s original distinction had nothing to do with words changing their meaning or reference. If, in the referential use, Jones refers to someone who did not murder Smith by using the

170

Siblings Under the Skin

description ‘Smith’s murderer’, the reference is none the less achieved by way of the normal meanings of the words. The words therefore must have their usual reference. All that is needed, if we are to accept this way of describing the situation, is a firm sense of the difference between what words mean or refer to and what speakers mean or refer to. Jones may have referred to someone else by using words that referred to Smith’s murderer; this is something he may have done in ignorance or deliberately. Similarly for Donnellan’s claim that Jones has said something true when he says ‘Smith’s murderer is insane’, provided the man he believes (erroneously) to have murdered Smith is insane. Jones has said something true by using a sentence that is false. This is done intentionally all the time, for example in irony or metaphor. A coherent theory could not allow that under the circumstances Jones’ sentence was true; nor would Jones think so if he knew the facts. Jones’ belief about who murdered Smith cannot change the truth of the sentence he uses (and for the same reason cannot change the reference of the words in the sentence). Humpty Dumpty is out of it. He cannot mean what he says he means because he knows that ‘There’s glory for you’ cannot be interpreted by Alice as meaning ‘There’s a nice knockdown argument for you.’ We know he knows this because Alice says ‘I don’t know what you mean by “glory”’, and Humpty Dumpty retorts, ‘Of course you don’t—till tell you.’ It is Mrs. Malaprop and Donnellan who interest me; Mrs. Malaprop because she gets away with it without even trying or knowing, and Donnellan because he gets away with it on purpose. Here is what I mean by ‘getting away with it’: the interpreter comes to the occasion of utterance armed with a theory that tells him (or so he believes) what an arbitrary utterance of the speaker means. The speaker then says something with the intention that it will be interpreted in a certain way, and the expectation that it will be so interpreted. In fact this way is not provided for by the interpreter’s theory. But the speaker is nevertheless understood; the interpreter adjusts his theory so that it yields the speaker’s intended interpretation. The speaker has ‘gotten away with it’. The speaker may or may not (Donnellan, Mrs Malaprop) know that he has got away with anything; the interpreter may or may not know that the speaker intended to get away with anything. What is common to the cases is that the speaker expects to be, and is,

Ramberg and Davidson

171

interpreted as the speaker intended although the interpreter did not have a correct theory in advance. We do not need bizarre anecdotes or wonderlands to make the point. We all get away with it all the time; understanding the speech of others depends on it. Take proper names. In small, isolated groups everyone may know the names everyone else knows, and so have ready in advance of a speech encounter a theory that will, without correction, cope with the names to be employed. But even this semantic paradise will be destroyed by each new nickname, visitor, or birth. If a taboo bans a name, a speaker’s theory is wrong until he learns of this fact; similarly if an outrigger canoe is christened. There is not, so far as I can see, any theory of names that gets around the problem. If some definite description gives the meaning of a name, an interpreter still must somehow add to his theory the fact that the name new to him is to be matched with the appropriate description. If understanding a name is to give some weight to an adequate number of descriptions true of the object named, it is even more evident that adding a name to one’s way of interpreting a speaker depends on no rule clearly stated in advance. The various theories that discover an essential demonstrative element in names do provide at least a partial rule for adding new names. But the addition is still an addition to the method of interpretation—what we may think of as the interpreter’s view of the current language of the speaker. Finding a demonstrative element in names, or for that matter in mass nouns or words for natural kinds, does not reduce these words to pure demonstratives; that is why a new word in any of these categories requires a change in the interpreter’s theory, and therefore a change in our description of his understanding of the speaker. Mrs. Malaprop and Donnellan make the case general. There is no word or construction that cannot be converted to a new use by an ingenious or ignorant speaker. And such conversion, while easier to explain because it involves mere substitution, is not the only kind. Sheer invention is equally possible, and we can be as good at interpreting it (say in Joyce or Lewis Carroll) as we are at interpreting the errors or twists of substitution. From the point of view of an ultimate explanation of how new concepts are acquired, learning to interpret a word that expresses a concept we do not already have is a far

172

Siblings Under the Skin

deeper and more interesting phenomenon than explaining the ability to use a word new to us for an old concept. But both require a change in one’s way of interpreting the speech of another, or in speaking to someone who has the use of the word. The contrast between acquiring a new concept or meaning along with a new word and merely acquiring a new word for an old concept would be salient if I were concerned with the infinitely difficult problem of how a first language is learned. By comparison, my problem is simple. I want to know how people who already have a language (whatever exactly that means) manage to apply their skill or knowledge to actual cases of interpretation. All the things I assume an interpreter knows or can do depend on his having a mature set of concepts, and being at home with the business of linguistic communication. My problem is to describe what is involved in the idea of ‘having a language’ or of being at home with the business of linguistic communication. Here is a highly simplified and idealized proposal about what goes on. An interpreter has, at any moment of a speech transaction, what I persist in calling a theory. (I call it a theory, as remarked before, only because a description of the interpreter’s competence requires a recursive account.) I assume that the interpreter’s theory has been adjusted to the evidence so far available to him: knowledge of the character, dress, role, sex, of the speaker, and whatever else has been gained by observing the speaker’s behavior, linguistic or otherwise. As the speaker speaks his piece the interpreter alters his theory, entering hypotheses about new names, altering the interpretation of familiar predicates, and revising past interpretations of particular utterances in the light of new evidence. Some of what goes on may be described as improving the method of interpretation as the evidential base enlarges. But much is not like that. When Donnellan ends his reply to MacKay by saying ‘There’s glory for you’ not only he, but his words, are correctly interpreted as meaning ‘There’s a nice knockdown argument for you.’ That’s how he intends us to interpret his words, and we know this, since we have, and he knows we have, and we know he knows we have (etc.), the background needed to provide the interpretation. But up to a certain point (before MacKay came on the scene) this interpretation of an earlier utterance by Donnellan of the same words would have been

Ramberg and Davidson

173

wrong. To put this differently: the theory we actually use to interpret an utterance is geared to the occasion. We may decide later we could have done better by the occasion, but this does not mean (necessarily) that we now have a better theory for the next occasion. The reason for this is, as we have seen, perfectly obvious: a speaker may provide us with information relevant to interpreting an utterance in the course of making the utterance. Let us look at the process from the speaker’s side. The speaker wants to be understood, so he utters words he believes can and will be interpreted in a certain way. In order to judge how he will be interpreted, he forms, or uses, a picture of the interpreter’s readiness to interpret along certain lines. Central to this picture is what the speaker believes is the starting theory of interpretation the interpreter has for him. The speaker does not necessarily speak in such a way as to prompt the interpreter to apply this prior theory; he may deliberately dispose the interpreter to modify his prior theory. But the speaker’s view of the interpreter’s prior theory is not irrelevant to what he says, nor to what he means by his words; it is an important part of what he has to go on if he wants to be understood. I have distinguished what I have been calling the prior theory from what I shall henceforth call the passing theory. For the hearer, the prior theory expresses how he is prepared in advance to interpret an utterance of the speaker, while the passing theory is how he does interpret the utterance. For the speaker, the prior theory is what he believes the interpreter’s prior theory to be, while his passing theory is the theory he intends the interpreter to use. I am now in a position to state a problem that arises if we accept the distinction between the prior and the passing theory and also accept the account of linguistic competence given by principles (l)—(2). According to that account, each interpreter (and this includes speakers, since speakers must be interpreters) comes to a successful linguistic exchange prepared with a ‘theory’ which constitutes his basic linguistic competence, and which he shares with those with whom he communicates. Because each party has such a shared theory and knows that others share his theory, and knows that others know he knows (etc.), some would say that the knowledge or abilities that constitute the theory may be called conventions.

174

Siblings Under the Skin

I think that the distinction between the prior and the passing theory, if taken seriously, undermines this commonly accepted account of linguistic competence and communication. Here is why. What must be shared for communication to succeed is the passing theory. For the passing theory is the one the interpreter actually uses to interpret an utterance, and it is the theory the speaker intends the interpreter to use. Only if these coincide is understanding complete. (Of course, there are degrees of success in communication; much may be right although something is wrong. This matter of degree is irrelevant to my argument.) The passing theory is where, accident aside, agreement is greatest. As speaker and interpreter talk, their prior theories become more alike; so do their passing theories. The asymptote of agreement and understanding is reached when passing theories coincide. But the passing theory cannot in general correspond to an interpreter’s linguistic competence. Not only does it have its changing list of proper names and gerrymandered vocabulary, but it includes every successful—i.e. correctly interpreted—use of any other word or phrase, no matter how far out of the ordinary. Every deviation from ordinary usage, as long as it is agreed on for the moment (knowingly deviant, or not, on one, or both, sides), is in the passing theory as a feature of what the words mean on that occasion. Such meanings, transient though they may be, are literal; they are what I have called first meanings. A passing theory is not a theory of what anyone (except perhaps a philosopher) would call an actual natural language. ‘Mastery’ of such a language would be useless, since knowing a passing theory is only knowing how to interpret a particular utterance on a particular occasion. Nor could such a language, if we want to call it that, be said to have been learned, or to be governed by conventions. Of course things previously learned were essential to arriving at the passing theory, but what was learned could not have been the passing theory. Why should a passing theory be called a theory at all? For the sort of theory we have in mind is, in its formal structure, suited to be the theory for an entire language, even though its expected field of application is vanishingly small. The answer is that when a word or phrase temporarily or locally takes over the role of some other word

Ramberg and Davidson

175

or phrase (as treated in a prior theory, perhaps), the entire burden of that role, with all its implications for logical relations to other words, phrases, and sentences, must be carried along by the passing theory. Someone who grasps the fact that Mrs. Malaprop means ‘epithet’ when she says ‘epitaph’ must give ‘epithet’ all the powers ‘epitaph’ has for many other people. Only a full recursive theory can do justice to these powers. These remarks do not depend on supposing Mrs. Malaprop will always make this ‘mistake’; once is enough to summon up a passing theory assigning a new role to ‘epitaph’. An interpreter’s prior theory has a better chance of describing what we might think of as a natural language, particularly a prior theory brought to a first conversation. The less we know about the speaker, assuming we know he belongs to our language community, the more nearly our prior theory will simply be the theory we expect someone who hears our unguarded speech to use. If we ask for a cup of coffee, direct a taxi driver, or order a crate of lemons, we may know so little about our intended interpreter that we can do no better than to assume that he will interpret our speech along what we take to be standard lines. But all this is relative. In fact we always have the interpreter in mind; there is no such thing as how we expect, in the abstract, to be interpreted. We inhibit our higher vocabulary, or encourage it, depending on the most general considerations, and we cannot fail to have premonitions as to which of the proper names we know are apt to be correctly understood. In any case, my point is this: most of the time prior theories will not be shared, and there is no reason why they should be. Certainly it is not a condition of successful communication that prior theories be shared: consider the malaprop from ignorance. Mrs. Malaprop’s theory, prior and passing, is that ‘A nice derangement of epitaphs’ means a nice arrangement of epithets. An interpreter who, as we say, knows English, but does not know the verbal habits of Mrs. Malaprop, has a prior theory according to which ‘A nice derangement of epitaphs’ means a nice derangement of epitaphs; but his passing theory agrees with that of Mrs. Malaprop if he understands her words. It is quite clear that in general the prior theory is neither shared by speaker and interpreter nor is it what we would normally call a language. For the prior theory has in it all the features special to the idiolect of the

176

Siblings Under the Skin

speaker that the interpreter is in a position to take into account before the utterance begins. One way to appreciate the difference between the prior theory and our ordinary idea of a person’s language is to reflect on the fact that an interpreter must be expected to have quite different prior theories for different speakers—not as different, usually, as his passing theories; but these are matters that depend on how well the interpreter knows his speaker. Neither the prior theory nor the passing theory describes what we would call the language a person knows, and neither theory characterizes a speaker’s or interpreter’s linguistic competence. Is there any theory that would do better? Perhaps it will be said that what is essential to the mastery of a language is not knowledge of any particular vocabulary, or even detailed grammar, much less knowledge of what any speaker is apt to succeed in making his words and sentences mean. What is essential is a basic framework of categories and rules, a sense of the way English (or any) grammars may be constructed, plus a skeleton list of interpreted words for fitting into the basic framework. If I put all this vaguely, it is only because I want to consider a large number of actual or possible proposals in one fell swoop; for I think they all fail to resolve our problem. They fail for the same reasons the more complete and specific prior theories fail: none of them satisfies the demand for a description of an ability that speaker and interpreter share and that is adequate to interpretation. First, any general framework, whether conceived as a grammar for English, or a rule for accepting grammars, or a basic grammar plus rules for modifying or extending it—any such general framework, by virtue of the features that make it general, will by itself be insufficient for interpreting particular utterances. The general framework or theory, whatever it is, may be a key ingredient in what is needed for interpretation, but it can’t be all that is needed since it fails to provide the interpretation of particular words and sentences as uttered by a particular speaker. In this respect it is like a prior theory, only worse because it is less complete. Second, the framework theory must be expected to be different for different speakers. The more general and abstract it is, the more difference there can be without it mattering to communication. The

Ramberg and Davidson

177

theoretical possibility of such divergence is obvious; but once one tries to imagine a framework rich enough to serve its purpose, it is clear that such differences must also be actual. It is impossible to give examples, of course, until it is decided what to count in the framework: a sufficiently explicit framework could be discredited by a single malapropism. There is some evidence of a more impressive sort that internal grammars do differ among speakers of ‘the same language’. James McCawley reports that recent work by Haber shows that there is appreciable variation as to what rules of plural formation different speakers have, the variation being manifested in such things as the handling of novel words that an investigator has presented his subjects with, in the context of a task that will force them to use the word in the plural…Haber suggests that her subjects, rather than having a uniformly applicable process of plural formation, each have a ‘core’ system, which covers a wide range of cases, but not necessarily everything, plus strategies…for handling cases that are not covered by the ‘core’ system…Haber’s data suggest that speakers of what are to the minutest details ‘the same dialect’ often have acquired grammars that differ in far more respects than their speech differs in.6

I have been trying to throw doubt on how clear the idea of ‘speaking the same dialect’ is, but here we may assume that it at least implies the frequent sharing of passing theories. Bringing in grammars, theories, or frameworks more general than, and prior to, prior theories just emphasizes the problem I originally presented in terms of the contrast between prior theories and passing theories. Stated more broadly now, the problem is this: what interpreter and speaker share, to the extent that communication succeeds, is not learned and so is not a language governed by rules or conventions known to speaker and interpreter in advance; but what the speaker and interpreter know in advance is not (necessarily) shared, and so is not a language governed by shared rules or conventions. What is shared is, as before, the passing theory; what is given in advance is the prior theory, or anything on which it may in turn be based. What I have been leaving out of account up to now is what Haber calls a ‘strategy’, which is a nice word for the mysterious process by

178

Siblings Under the Skin

which a speaker or hearer uses what he knows in advance plus present data to produce a passing theory. What two people need, if they are to understand one another through speech, is the ability to converge on passing theories from utterance to utterance. Their starting points, however far back we want to take them, will usually be very different— as different as the ways in which they acquired their linguistic skills. So also, then, will the strategies and strategems that bring about convergence differ. Perhaps we can give content to the idea of two people ‘having the same language’ by saying that they tend to converge on passing theories, degree or relative frequency of convergence would then be a measure of similarity of language. What use can we find, however, for the concept of a language? We could hold that any theory on which a speaker and interpreter converge is a language; but then there would be a new language for every unexpected turn in the conversation, and languages could not be learned and no one would want to master most of them. We just made a sort of sense of the idea of two people ‘having the same language’, though we could not explain what a language is. It is easy to see that the idea of ‘knowing’ a language will be in the same trouble, as will the project of characterizing the abilities or capacities a person must have if he commands a language. But we might try to say in what a person’s ability to interpret or speak to another person consists: it is the ability that permits him to construct a correct, that is, convergent, passing theory for speech transactions with that person. Again, the concept allows of degrees of application. This characterization of linguistic ability is so nearly circular that it cannot be wrong: it comes to saying that the ability to communicate by speech consists in the ability to make oneself understood, and to understand. It is only when we look at the structure of this ability that we realize how far we have drifted from standard ideas of language mastery. For we have discovered no learnable common core of consistent behaviour, no shared grammar or rules, no portable interpreting machine set to grind out the meaning of an arbitrary utterance. We may say that linguistic ability is the ability to converge on a passing theory from time to time—this is what I have suggested, and I have no better proposal. But if we do say this, then we should realize

Ramberg and Davidson

179

that we have abandoned not only the ordinary notion of a language, but we have erased the boundary between knowing a language and knowing our way around in the world generally. For there are no rules for arriving at passing theories, no rules in any strict sense, as opposed to rough maxims and methodological generalities. A passing theory really is like a theory at least in this, that it is derived by wit, luck, and wisdom from a private vocabulary and grammar, knowledge of the ways people get their point across, and rules of thumb for figuring out what deviations from the dictionary are most likely. There is no more chance of regularizing, or teaching, this process than there is of regularizing or teaching the process of creating new theories to cope with new data in any field—for that is what this process involves. The problem we have been grappling with depends on the assumption that communication by speech requires that speaker and interpreter have learned or somehow acquired a common method or theory of interpretation—as being able to operate on the basis of shared conventions, rules, or regularities. The problem arose when we realized that no method or theory fills this bill. The solution to the problem is clear. In linguistic communication nothing corresponds to a linguistic competence as often described: that is, as summarized by principles (1) - (3). The solution is to give up the principles. Principles (1) and (2) survive when understood in rather unusual ways, but principle (3) cannot stand, and it is unclear what can take its place. I conclude that there is no such thing as a language, not if a language is anything like what many philosophers and linguists have supposed. There is therefore no such thing to be learned, mastered, or born with. We must give up the idea of a clearly defined shared structure which languageusers acquire and then apply to cases. And we should try again to say how convention in any important sense is involved in language; or, as I think, we should give up the attempt to illuminate how we communicate by appeal to conventions.

180

Siblings Under the Skin

Notes 1. 2. 3. 4. 5.

6.

The New Yorker, 4 April, 1977, p. 56. Jonathan Bennett, Linguistic Behavior (Cambridge 1976). p. 186. Shakespeare, Sonnet 53. See the essays on radical interpretation in my Inquiries into Truth and Interpretation, (Clarendon Press, Oxford, 1984). Keith Donnellan, ‘Putting Humpty Dumpty Together Again’. The Philosophical Review, 77 (1968), p. 213. Alfred MacKay’s article, ‘Mr Donnellan and Humpty Dumpty on Referring’ appeared in the same issue of The Philosophical Review, pp. 197-202. James McCawley, ‘Some Ideas Not to Live By’, Die Neuen Sprachen, 75 (1976), p. 157. These results are disputed by those who believe the relevant underlying rules and structures are pre-wired. My point obviously does not depend on the example, or the level at which deviations are empirically possible.

Biographical notes for Donald Davidson follow the first of his essays in this collection, “A Coherence Theory of Truth and Knowledge.”

Sobstyl and Sellars

181

Lost in Logical Space? Wilfrid Sellars and Feminist Empiricism Edrie Sobstyl

In among the diverse, sometimes contradictory positions occupied by feminist epistemologists, one might wish to seek a unifying theme or set of themes to tie their projects together. Of course, the explicit political content of feminist epistemology is one such theme. But since part of the ongoing debate in this field concerns how (or whether) politics can be an epistemological issue, one might still inquire as to the nature of the terrain of feminist epistemology. If we take anti-foundationalism in the sense of opposition to the isolated, atomistic knower basic to the Cartesian tradition, then we can say that anti-foundationalism is an important commitment that binds feminist epistemologies together. However, this is not how the term anti-foundationalism is typically used in mainstream epistemology. Some feminists are anti-foundationalists in the sense that they are against treating epistemology as First Philosophy, basic to all other modes of inquiry, but this too is a non-standard use of the term. Most epistemologists contrast foundationalism with coherentism, and disagree about the nature of observation claims and the role of justification. However, feminist theorists often do not enter directly into specific conversations about foundationalism and coherentism, or about the status of observation claims, so the continuity of feminist epistemology with this tradition is sometimes missed. One writer whose work offers parallels with some concerns raised by feminists is Wilfrid Sellars, who, in “Empiricism and the Philosophy of Mind” and later in “More on Givenness and Explanatory Coherence”, tries to evade the charge of circularity that has traditionally been the bugbear of coherence theories. He does so by introducing the notions of logical space or dimension, explanatory coherence, and effective agency, three ideas that are also present, implicitly or not, in many feminist approaches to epistemology. Sellars claims that certain beliefs,

182

Siblings Under the Skin

that appear to be inferentially dependent on other beliefs, in fact bear an explanatory relation to those other beliefs in the logical space of reasons, and that our role as effective knowers legitimates this relation in the logical space of agency. In what follows, I outline Sellars’ position, and show that the challenge of linking effective agency to an epistemic framework, or logical space, in order to break out of the circle of coherentist justification, is not easily met. But I also suggest that feminist epistemology and Sellarsian explanatory coherence have much to offer one another. Indeed, attention to the work of Sellars, among others, demonstrates what many feminist empiricists are discovering: that the space in which to consider agents of knowledge as socially situated, and influenced by politics and morality, has long been present in the analytic tradition. The task for feminists is to make use of this logical space.

*** I. An attack on the Myth of the Given Sellars’ primary target in “Empiricism and the Philosophy of Mind” (hereafter EPM), and the aspect of his work which is most well known, is the so-called “Myth of the Given”, i.e., the myth that observation reports are the primary instance of non-inferential knowledge claims. Thus Sellars’ work is often read as an important contribution to the defeat of foundationalism, correspondence theories of truth, and the realism that typically accompanies such theories.1 In this myth, the authoritativeness of observation reports arises from the fact that in the circumstances in which such reports are made, they express the presence of the state of affairs being reported, and serve as the ineffable basis on which all other knowledge claims are based. The Myth of the Given is the view that: There is, indeed must be, a structure of particular matter of fact such that (a) each fact can not only be noninferentially known to be the case, but presupposes no other knowledge of either particular matter of fact, or of general truths; and (b) such that the noninferential knowledge of facts belonging to this structure constitutes the ultimate court of appeals for all factual claims—particular and general—about the world (EPM pp. 68-69).

Sobstyl and Sellars

183

This Myth is familiar enough, and Sellars is quite correct to assert its dominance in the history of traditional empiricism. It is found among many rationalists as well (especially Descartes). Statements that express either analytic truths or observations have been taken as bearing an ultimate, that is, non-inferential, authority. Sellars insists that knowledge does have a non-inferential foundation, but may still presuppose knowledge of other particular or general facts about the world. This appears to be a tenuous claim. Isn’t knowledge that relies on other knowledge by definition inferential? Sellars argues that this question is just another episode in the Myth (EPM 69). He asks us to consider other kinds of connections in assessing the warrant of our knowledge claims, including both other kinds of justification, and what we might call human relations. Sellars elaborates on the further ramifications of the Myth of the Given, and his response to them is significant for feminist purposes. The Myth, he writes, contains three demi-myths. First, it treats the idea of an observation report in what he calls a technical sense. That is, where in ordinary usage we think of reports as being made by and to someone, traditional epistemology has abstracted away both the concrete verbal performativity of observation reports, and their function in a socio-linguistic context. This view is clearly a precursor to the feminist objection that epistemology which takes “S knows that p” statements as paradigmatic removes both knower and context from consideration (see Code, 1991). Sellars insists that both knower and context are important, although he focuses more attention on the latter than on the former. Nevertheless, it is worth stressing that this idea has been present in the analytic tradition from the outset: one’s observation reports will be judged and used by others, and one therefore has an epistemic responsibility to make such reports properly.2 A second and related point is that, despite traditional empiricism’s retreat from the ordinary active sense of report making, it continues to treat the making of observation reports as analogous to actions, and the making of correct reports as analogous to right action. Sellars warns that this analogy is confusing, and he will argue that other kinds of correctness may be more useful. There is an obvious connection to be drawn here between Sellars’ critique and contemporary vir-

184

Siblings Under the Skin

tue epistemology, which has its feminist proponents.3 I will, however, leave this connection to one side. The final ingredient in the Myth is its emphasis on rule following. Sellars mentions this only in order to emphasize that making authoritative observation reports counts as an instance of rule-following behavior only if one knows (or believes) that the state of affairs being reported obtains, and not if the state of affairs obtains but one does not know (or believe) it. Otherwise rule following would amount to mere uniformity of stimulus and response (EPM 73). Taken together, these ingredients comprise the Myth of the Given: our nonverbal awarenesses form an epistemic foundation, which may be expressed in verbal episodes, and these episodes are authoritative providing that they have been made correctly. All other knowledge claims are inferred, inductively or deductively, from this basis, and warranted when they are inferred correctly. Sellars notes that there is no meaningful difference between internal and external awareness with respect to this Myth. The chain of authority remains the same whether its basis is awareness of a particular (internal) sense datum, or awareness of a particular (external) physical object. In other words, it makes no difference to the mythiness of the Myth whether one is a phenomenalist or a representationalist. In “More on Givenness and Explanatory Coherence” (MGEC), he points out that this position, insisting as it does that directly apprehended states of affairs constitute knowledge, is paradoxical. Direct apprehension is a source of warrant for further claims, but is itself neither warranted nor unwarranted (MGEC para. 24). It is, in other words, “given”. But this, he argues, involves “denying that the direct apprehension of a fact is itself knowledge”, (para. 24) for to count as knowledge just is to have a warrant of some kind. The Given, therefore, can neither constitute knowledge, nor provide warrant for further knowledge claims. There is thus little point in worrying over the question that occupies many epistemologists, namely how one gets from intrinsically authoritative nonverbal awareness through correct expression to equally authoritative observation statements, and thence to warranted higher level inferences. The problem is found at step one, not at either of steps two, three, or four. Referring to nonverbal awareness as “given”, or “self-authenticating”, or “knowledge by acquaintance” re-describes this problem, but does not solve it.

Sobstyl and Sellars

185

Coherentism is a common response to this line of argument. Rather than building a justificatory pyramid on a base of self-authenticating awarenesses, the coherentist tries to spin a web of inter-related beliefs whose ultimate justification is a property of the way the whole set of beliefs hangs together, rather than of individual logical relations between base and edifice. However, coherentism also has its problems. Critics object that coherentism is circular, that it lacks a distinct connection with the “real” world, and that it offers no criteria for deciding either when to accept a particular belief, or how to choose between equally coherent sets of beliefs. Sellars’ project attempts to address these objections, as we shall see.

*** II. First pause for feminism Many feminist epistemologists espouse some version of empiricism, or coherentism, or both. Yet vestiges of the Myth of the Given and the foundationalism it supports remain hidden in many feminist positions. Many feminist (as well as many non-feminist) epistemologists often distinguish between “knowing that” and “knowing how”, in order to expand the workspace of traditional epistemology, and to accommodate the problem of the Given. But Sellars dismisses this distinction. He argues that, in order to express observational knowledge, one must already know that overt verbal episodes are reliable indicators of the presence of the things they report: It follows, as a matter of simple logic, that one couldn’t have observational knowledge of any fact unless one knew many other things as well. And let me emphasize that the point is not taken care of by distinguishing between knowing how and knowing that, and admitting that observational knowledge requires a lot of ‘know how.’ For the point is specifically that observational knowledge of any particular fact, e.g. that this is green, presupposes that one knows general facts of the form X is a reliable symptom of Y. And to admit this requires an abandonment of the traditional empiricist idea that observational knowledge ‘stands on its own feet’. (EPM 75-76)

186

Siblings Under the Skin

Now, no feminist empiricist is so traditional that s/he believes that observational knowledge stands entirely on its own feet. Yet it is not always obvious in the feminist literature what observational knowledge is in fact supposed to stand upon. Many feminists, especially standpoint theorists, simply do not address this question. Standpoint theory begins with the assumption that things look different “from below”. Sandra Harding is careful to insist that perceiving this difference in how things look is an achievement, not automatic, but this presupposes that there must be a gap between appearances themselves and the achievement of understanding them as a source of epistemic privilege.4 Among those feminists who do take on the question of observational evidence, two strategies are found that deal with perceptual knowledge, but only obliquely. First, there may be an appeal to the theory-ladenness of observation, which is sometimes used as the thin end of the wedge that permits politics to enter epistemology.5 But if we neglect Sellars’ argument against the Myth of the Given, this could in effect amount to claiming that, among the general facts presupposed by the reliable utterance of “this is green” are facts about gender relations. This is a stretch, and it will be useful to feminist epistemology to flesh such claims out in more detail. Even if we examine higher-level observation statements like “the men are hunting and the women gathering”, feminists still struggle to explain exactly how it is that observation statements are contaminated by gender politics (although we seem sure that they are). We also struggle to say exactly how a different perspective might presuppose a different set of general facts about observational knowledge, and hence give rise to a different set of observation statements. Much feminist work focuses on the layers of interpretation that necessarily exist between observation statements and full-blown theories. Also of interest are the ways in which some observations are simply not made by patriarchal science. Yet often, feminists, including those feminist empiricists who insist that all evidence for science is sensory evidence (Nelson 1990, 86), appeal indirectly to the Myth of the Given that Sellars rejects. That is, they hold that observation forms a foundation on which all other beliefs are built, and if there is a problem with sexist contamination, it emerges in the building, not in the observations themselves. Trying to show that observation

Sobstyl and Sellars

187

statements themselves are contaminated by androcentric bias, which some feminists do, is not only impossible; it presupposes the Myth of the Given. Of course Nelson would insist that, following Quine, she is not a foundationalist. She does not accept the view that physical objects provide an extra- or pre-theoretic foundation for science. (1990, 103) However, by endorsing Quine’s view that sense data are evidentially fundamental, she is still a foundationalist of the sort Sellars is concerned to defeat. The problem that Sellars identifies as the Myth of the Given is not a metaphysical claim about physical objects being foundational, but an epistemological claim about whether our experience of physical objects can be taken as foundational. The second strategy is a shift in epistemology away from the traditional Cartesian model of the knower. In “Subjects, Power and Knowledge”, Helen Longino points out that the usual response to the problem of the theory-ladenness of observation, and the related problem of underdetermination, is to repudiate empiricism (1993, 111). However, she argues that what ought to be rejected instead of empiricism is its “silent partner”, epistemic individualism. She holds that feminist empiricists should see science as “an outcome of the critical dialogue in which individuals and groups holding different points of view engage with each other” (1993, 112). Nelson, who insists that it is the community, not the individual, who knows, (1990) also defends a shift away from individualism. Clearly there is something to be gained in questioning the unconditioned subject, although there are risks as well. Leaving those difficulties aside for now, I believe that Sellars’ work shows that rejecting empiricism’s “silent partner”, epistemic individualism, is insufficient on its own to avoid the problems of theory-ladenness and underdetermination. We must also repudiate perceptual foundationalism as the “even-more-silent partner”. If Sellars is right about the Myth of the Given, the problem of elaborating the relationship between evidence and knowledge claims will not disappear just by switching to the community as our primary agent of knowledge. At best, the switch conceals implicit reliance on the Myth, in that a focus on the community as knower permits us to elide individual perception. Communities as such do not make observations in the traditional foundational sense, but the further ques-

188

Siblings Under the Skin

tion of what sense they make observations in is unaddressed. At worst, observational foundationalism is simply assumed without acknowledgement in the shift to a communally based epistemology. I return to this point, below. By now it should be clear that Sellars’ argument, if successful, could be of use to feminists caught in the foundationalist bind. He offers us a way to dispense with the notion of the Given, while still allowing us the conceptual apparatus to make well-warranted knowledge claims. I turn now to the notion of logical space, to explore further the tools and insights Sellars provides.

*** III. Logical Space The idea of logical space is, at first blush, a minor note in Sellars’ enormously dense and sweeping corpus of philosophical work. Logical space and logical dimension seem like metaphors, not concrete epistemological concepts. Yet if we succumb to the temptation to dismiss this metaphor too quickly, it is much more difficult to notice how explanatory coherence differs from traditional epistemological coherence.6 As I argue below, Sellars relies heavily on logical space to link explanatory coherence and effective agency. All three concepts are mutually reinforcing. In this section, I establish the ways in which logical space differs from a similar Quinean metaphor, and from the related idea of “mere” social practice. In “Epistemology Naturalized”, Quine asserts that the relationship between science and epistemology ought to be thought of as one of “reciprocal containment”. He argues that science should be seen as part of epistemology, while at the same time that epistemology should be seen as part of science.7 The ensuing mental image of a jumble of nested dolls is not especially helpful in understanding what Quine means, nor is his excuse that when we are working at the “philosophical fringes of science”, old idioms are bound to fail us and we must turn to metaphor.8 Let’s compare what Sellars has to say about logical space. In an important and widely quoted section of EPM, called “Does Empirical Knowledge Have a Foundation?” Sellars writes that

Sobstyl and Sellars

189

The essential point is that in characterizing an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says (76, emphasis in original).

It is easy to interpret “logical space” as a metaphorical description of Sellars’ commitment to the importance of social practices to justification, in just the way that Quine’s “reciprocal containment” is a metaphorical description of his commitment to the interdependency of science and epistemology. Unsurprisingly, some writers take Sellars in exactly this way. Richard Rorty, for example, says that Sellars’ attack on ‘givenness’ is one of the crucial steps in undermining the possibility of epistemology. He approvingly quotes the passage about the logical space of reasons, just cited, and several pages later credits Sellars with vital support for his argument that “justification is not a matter of a special relation between ideas (or words) and objects, but of conversation, of social practice”(1979, 170). Rorty is quite right that Sellars’ attack on the Given forbids us from taking knowledge of self-authenticating states as premises from which other knowledge may be inferred (Rorty, 177). But he errs in drawing the further conclusion that, for Sellars, questions of language and social practice are the only questions interestingly thought of as epistemological. Rorty says that if we’re behaviorists of the sort he thinks Sellars is, and If we understand the rules of a language-game, [then] we understand all that there is to understand about why moves in that language-game are made (all that is, save for the extra understanding obtained from inquiries nobody would call epistemological–into, for example, the history of the language, the structure of the brain, the evolution of the species, and the political or cultural ambience of the players). (Rorty 1979, 174, emphasis added)

It is far from certain that Sellars would call such inquiries nonepistemological, although it remains an open question, and one worth pursuing, as to how he views their epistemic impact (see notes 17 and 18, below).

190

Siblings Under the Skin

Robert Brandom and John McDowell are two more writers who embrace Sellars’ attack on the given, and who defend the role of logical space in Sellars’ work. However, both treat logical space only as a normative abstraction from the social practices of justification, and neglect explanatory coherence altogether.9 This is unfortunate. The error that McDowell makes in his work on Sellars, which Brandom amplifies, is a mistake that needn’t have been made if both had noted the importance of logical space to Sellars’ defense of explanatory coherence. McDowell raises worries about logical space, interpreted as “a certain sort of standing in the space of reasons.” He asserts that “at least for rational animals, a satisfactory standing in the space of reasons is a necessary condition for knowledge” (1995, 881). Sellars explicitly concurs in MGEC. McDowell becomes uncomfortable, however, with the way observational reports are warranted for Sellars. McDowell thinks that appeals to judgements about how the world is constitute an “extra condition” for Sellars. As will become evident, however, this is exactly the objection that the distinction between logical spaces is supposed to rule out. McDowell says that EPM contains An account of the authority of observational reports which expresses a good thought–that the capacity to make observational reports requires general knowledge of the world, even in cases as conceptually undemanding as saying what color something is–in what seems to me a suspect way, in terms of the subject’s ability to infer a judgement about the world from her own tokening (or propensity towards tokening) of an observational form of words. I am suspicious of this avoidance of the straightforward idea that the authority of the report consists in the fact that things are manifestly so (1995, 890 n.24). But Sellars doesn’t avoid the straightforward idea that the authority of the report consists in the fact that things are manifestly so–he expressly denies this. That’s what the attack on the Myth of the Given is all about. Sellars introduces explanatory coherence as a way to rescue the warrant of observational reports from the foundationalist scrapheap, and different logical spaces are what allow him to juggle several different forms of warrant at one time. That’s what it means to say that an observation statement can be justified by an empirical proposition in one logical dimension, and empirical propositions by observation statements in another (see EPM 78, below). Talk of logical space as single,

Sobstyl and Sellars

191

unitary, and internal—a certain standing in the space of reasons—does violence to the way Sellars writes, and especially to the way he uses the multiplicity of logical space to support explanatory coherence as a source—the ultimate source—of epistemic warrant. Similarly, Alessandra Tanesini explicitly embraces Sellars’ antifoundationalism as the first step on the path toward seeing justification as social.10 Her interpretation of logical space as a metaphor for social practice is so strong that she paraphrases Sellars’ discussion in EPM as a claim about the social: “since beliefs are the psychological states we attribute to individuals when they make assertions, the social practice of giving and asking for reasons plays an important role in their constitution” (1999, 52). Her own conception of justification, like Sellars’, is free from self-authenticating perceptual foundations, because she sees all knowledge as dependent on other knowledge. She sees some of those relations of dependency, also like Sellars, as non-inferential. She concludes that her position therefore “amounts to a rejection of both foundationalism and coherentism. The latter is the view that justification is always a matter of inference” (1999, 35 n8). Sellars, however, does not reject both foundationalism and coherentism. On the contrary, he uses the notion of logical space to allow him to preserve elements of both, albeit in a non-standard way: There is clearly some point to the picture of knowledge as resting on a level of propositions – observation reports – which do not rest on other propositions in the same way as other propositions rest on them. On the other hand, I do wish to insist that the metaphor of ‘foundation’ is misleading in that it keeps us from seeing that if there is a logical dimension in which other empirical propositions rest on observation reports, there is another logical dimension in which the latter rest on the former (EPM 78).

Like Quine’s metaphor of reciprocal containment, logical space may by now seem more puzzling than helpful. Sellars sees a relation of some kind between observation reports and knowledge, but it is not always, or not only, a relationship of “resting on”. But Tanesini is mistaken in her assertion that, for coherentism, justification is always a matter of inference. Sellars does not see coherentism this way—which is why he calls it explanatory coherence. Like Brandom and McDowell,

192

Siblings Under the Skin

Tanesini misses the significance of explanatory coherence because she reads logical space as a mere metaphor for social practice. This is even more unfortunate for Tanesini than it is for Brandom and McDowell, because her innovative positive project in epistemology resonates strongly with explanatory coherence.11 Even those commentators who attend to Sellars’ efforts to maintain a middle ground between foundationalism and coherentism sometimes slide past the importance of the explanatory part of explanatory coherence. In the best available exposition of Sellars’ epistemology, C.F. Delaney recognizes that Sellars made a conscious effort to integrate foundationalism and coherentism, rather than rejecting them both (Delaney 1977, 3). Delaney asserts that “Sellars develops a theory of knowledge that embodies a holistic theory of meaning, a correspondence theory of truth, and an account of justification that could be described either as a mitigated foundationalism—or a mitigated coherentist position depending on one’s categorial proclivities” (1977, 3-4). Thus, according to Delaney, Sellars sees “a point to the concerns of both the foundationalist and the coherentist, but each loses sight of what is legitimate in the other’s position” (1977, 31). Delaney adds two more important considerations. First, he describes Sellars’ epistemology as developing within the context of a “fully elaborated philosophy of mind” (1977, 3-4). This is important because it helps us to see the extent to which Sellars’ work exhibits a degree of systematic unity unusual in modern philosophy. His ideas of knowledge are inextricably interwoven with his views on mind, and are further tied to his metaphysics and theory of action. This is an important lesson for feminists, who, like their non-feminist colleagues, are more likely to take a piecemeal approach to philosophical problems. Such an approach can lead to debilitating disjunctions between categories of feminist discourse, including the troubled gap between perception and bias, described above.12 Sellars’ “synoptic vision” means that his views on cognition, realism, and action are not arbitrary addenda that we can ignore or distort while embracing his attack on the Given. He means them to be taken as interconnected.13 The second issue is illustrated through Delaney’s use of the passage on competing logical dimensions taken from EPM (78), above. Unlike Brandom, McDowell, and Tanesini, Delaney does not address

Sobstyl and Sellars

193

the idea of logical space.14 But his exposition is more thorough because he includes the bit that comes next in Sellars’ original: “Above all, the picture is misleading because of its static character” (EPM p. 78, quoted in Delaney 1977, 31; emphasis in Sellars). Sellars’ diplomatic attitude to foundationalism and coherentism is much less important than his insistence that “both factions share an impoverished picture of knowledge as an abstract structure existing in the instantaneous present rather than seeing knowledge as concretely embedded in an ongoing temporal process of inquiry.” (Delaney, 31). Thus Sellars is definitely a forerunner of those feminists who insist that we must break away from the tired Hempelian mode of thinking about science as a set of propositions, and consider it instead as a socially embedded process.15 When we place Sellars’ emphasis on process in the context of his views about explanatory coherence and effective agency, we will see that he offers feminists a useful stance on authority and responsibility, and that he has some success in evading the problems of conventional coherentism. We have seen what logical space is not. What, then, is it? The widely quoted passage from EPM is not the only reference to logical space. His text is peppered with them. In a discussion of appearances (“Explaining Looks”, EPM sec. IV), Sellars says that the philosophical, as opposed to the every day notion of appearance has an important contribution to make, “in its place”. “But this place is in the logical space of an ideal scientific picture of the world and not in the logical space of ordinary discourse” (52-3, emphasis in text). This sounds as though logical spaces embed different social practices, but it also looks as though he sees science and common sense as very different forms of social practice. This means that repudiating the Myth of the Given might not be such a desirable end for feminists who want to establish that science and common sense differ not in kind but in degree of authority and prestige.16 In a long and especially vexing passage, Sellars writes, When we picture a child…learning his first language, we, of course, locate the language learner in a structured logical space in which we are at home. Thus we conceive of him as a person (or, at least, a potential person) in a world of physical objects, colored, producing sounds, existing in Space and Time. But though it is

194

Siblings Under the Skin

we who are familiar with this logical space, we run the danger, if we are not careful, of picturing the language learner as having ab initio some degree of awareness—‘pre-analytic’, limited and fragmentary though it may be—of this same logical space. We picture his state as though it were rather like our own when placed in a strange forest on a dark night. In other words, unless we are careful, we can easily take for granted that the process of teaching a child to use a language is that of teaching it to discriminate elements within a logical space of particulars, universals, facts, etc., of which it is already undiscriminatingly aware, and to associate these discriminated elements with verbal symbols. And this mistake is in principle the same whether the logical space of which the child is supposed to have this undiscriminating awareness is conceived by us to be that of physical objects or of private sense contents (EPM pp. 64-5).

Here logical space is less like a collection of dimensions where differing social practices occur than it is an actual space, the container of reality. And while Sellars clearly wants us not to think of language learning as occurring in a pre-analytic space, it also appears that he wants us to think of it as the real space that we are in. It’s just that we are not in it prior to or independently of the acquisition of language (EPM 66). This puzzle will not be easily resolved, but a deeper examination of the notion of explanatory coherence will help.

*** IV. Explanatory coherence On first reading, MGEC might not appear to be congenial to a feminist reading. In terms of style, with its WPs, IPMs and MJs1-5, MGEC takes an approach to epistemology quite different from the one normally favored by feminist writers. If my argument so far has been correct, however, feminists would lose a valuable analytic ally if we allowed Sellars’ important insights to hide in the alphabet soup. The lesson here is that not everything that looks like “S knows that p” epistemology is “S knows that p” epistemology. In EPM, Sellars at most alludes to the idea that later emerges as explanatory coherence.

Sobstyl and Sellars

195

He calls MGEC his effort to cash out the promissory note he drew in EPM (MGEC, para. 87). He explains that the distinction between explanatory and justificatory coherence is “the linchpin” of his view (para. 84). He sets out to establish the “primacy of the concept of explanatory coherence in epistemic evaluation” (para. 43), and concludes by calling explanatory coherence the “ultimate criterion of truth” (para. 89). Thus he clearly sees explanatory coherence, not inferential coherence, as a focal epistemological concept. Sellars begins MGEC by re-launching his old attack on the Myth of the Given in new terms. Once again, he quickly dismisses the traditional empiricist idea that direct apprehension can both be knowledge and provide a foundation for knowledge. Now the stage is set for the entrance of explanatory coherence. Here, Sellars distinguishes between two kinds of warrant-increasing properties (WPs): the inferential and non-inferential. Inferential WPs are relatively simple and familiar: A statement S has an inferential WP either if it is validly inferable from certain other statements of a specified kind, or if it is validly inferable from a metastatement that is so warranted, asserting the likelihood of S’s truth given that it has property P.

Non-inferential warrant is a bit trickier: A statement S has a non-inferential WP if the metastatement “if S has property P, then S is likely to be true” is itself ultimately non-inferential, that is, it possesses neither of the two kinds of inferential WP described above.

So, for example, a statement S (“this thing is green”) has non-inferential WP, if the metastatement ‘if S (“this thing is green”) has P (the property of being a judgment exclusively about my present experience) then S is likely to be true,’ is ultimately non-inferential. Take the distinction between inferential WP and non-inferential WP as step one in his argument for explanatory coherence. Sellars, borrowing from Roderick Firth, asserts that, “some statements have some degree of warrant which is independent of (and in this sense prior to) the warrant (if any) which they derive from other statements” (MGEC

196

Siblings Under the Skin

para. 56). Thus Sellars establishes that there are two kinds of WP. Although this may sound like a relapse into the Myth, we shall soon see that it is not. Now, step two. What Sellars means by “ultimately non-inferential” is that the metastatements in question are explanatory. They explain certain regularities, but are not “inference-conducive”; that is, they cannot be used to provide inferential WP. Remember that he does not mean by this that some statements are self-justifying. He has already ruled out such statements. Next Sellars introduces a few metajudgments: MJ1 Judgements which have P (e), the property of being a judgement exclusively about my present experience, are likely to be true. MJ3 If a person ostensibly perceives (without ground for doubt) something to be Ø (for appropriate values of Ø) then it is likely to be true that he perceives something to be Ø. MJ4 If a person ostensibly remembers (without ground for doubt) having ostensibly perceived something to be Ø (for appropriate values of Ø) then it is likely to be true that he remembers ostensibly perceiving something to be Ø.

He contrasts these metajudgments, which he calls non-inductively warranted, with MJ2 Judgements which are believed by certain scholars are likely to be true.

He calls this an inductively warranted principle. The difference between these two classes of metajudgments appears to be that MJ2 supports an empirical statement (for example, the property of being a judgement about geography believed by the members of the American Geographical Society), while MJ1, MJ3, and MJ4 are seemingly not empirical. They are principles that provide criteria for adjudicating empirical knowledge claims. To assert otherwise would be to fall into vicious circularity. But perhaps this circle can be avoided. Does it help to note that MJ2 is warranted by induction, but the others are not? No. Sellars says it is “obvious” that MJ1, MJ3, and MJ4 cannot be inductively

Sobstyl and Sellars

197

warranted, and it would indeed be odd to think of one’s judgements of introspection, perception and memory (IPM) as having been established by instances. Sellars restates this circularity several times, to drive home to the reader that giving up foundationalism for coherentism is no epistemological panacea. (Feminist coherentists, take note.) What Sellars is after, he says, is a way for MJ1, MJ3, and MJ4 to be both criteria for adjudicating empirical knowledge claims, and empirical knowledge claims in their own right, without falling into the trap of the Given. To put it another way, he is groping for “a way in which it could be independently reasonable to accept MJ1, MJ3, and MJ4 in spite of the fact that a ground for accepting them is the fact that they belong to T, which we suppose to be an empirically-well confirmed theory” (MGEC para. 66). Or to put it yet another way, how “could it be reasonable … to accept T because it is supported by our introspective, perceptual and memory judgements (IPM judgements), if it is because they fall under MJ1, MJ3 and MJ4 that it is reasonable to accept these IPM judgements?” (para. 62) The answer for Sellars will be that it is reasonable to accept metajudgments like MJ1, MJ3, and MJ4 because “they are elements in a conceptual framework which defines what it is to be a finite knower in a world one never made. …To be one who makes epistemic appraisals is to be in this framework” (para. 73). As Sellars recognizes, this solution makes the circle look even more vicious. A statement S, call it an observation report, cannot be independently supported by any part of theory T when S is itself also a member of T. This is where logical space and explanatory coherence must meet. It would indeed be circular to explain or justify S with reference to T when S is a member of T. S is inferentially justified by some metajudgement(s) of T in the logical space or dimension in which it is a member of T. But in another logical space those metajudgments are non-inferentially explained by T. T hangs together with metajudgments and observation reports in the traditional coherentist sense of justified inference, but it also provides explanations for why it is reasonable to think of observation reports as likely to be true. Such explanations justify our acceptance of these metajudgments—they provide WPs—but not via inference. Explanation is a non-inferential WP, and when expla-

198

Siblings Under the Skin

nations mutually support one another, explanatory coherence is the result. (It might have helped if Sellars had used “explanatorily warranted” instead of “non-inferentially warranted”.)17 So far we can see that each component of Sellars’ argument (two kinds of WP, different logical spaces, and explanatory coherence) is necessary to give up the Given without falling into a coherentist circle. Sellars also insists that we separate questions about how we get into our epistemic framework from questions about justifying our acceptance of it. The first question, Sellars opines, will have a causal, probably evolutionary answer.18 The second question requires the notion of effective agency, the last piece of the Sellarsian picture.

*** V. Second pause—for feminist realism Realism occupies a tense position within feminist epistemology. Some feminists, including Richmond Campbell (1998, 49-53) and Naomi Scheman (1993), indicate that it is hard to make sense of feminist claims, or any scientific or other claims, in the absence of some kind of realism. Others, such as Anne Seller, object that feminist realism is a shallow political tactic, motivated by the desire to persuade but not by any epistemic commitments or their consequences. (1988, 170) Sellars is a scientific realist, and it is important for feminist epistemology to notice both why and in what way he is a scientific realist. As Gary Gutting puts it, Sellars’ realism “is not a dogma or a slogan but a carefully analyzed and justified position,” It follows inevitably from his epistemology (Delaney 1977, 74). Feminists who wish to be anti-foundationalists in the sense that they reject the Myth of the Given must, if Sellars is right, also embrace scientific realism. As a result, they will also inherit a Sellarsian connection between epistemology and politics for which we have struggled. Sellars is well known for his distinction between the “scientific” and the “manifest” images, or the theoretical and observational frameworks. Like the positivists, he attempts to correlate the language of the two frameworks via correspondence rules. Unlike the positivists, he has a distinctive understanding of what makes a theory good. Sellars

Sobstyl and Sellars

199

holds that a good theory must explain the observed phenomena, in just the sense described in the previous section. (Taken as a criterion for adjudicating empirical knowledge claims like S, MJ4 is a good criterion because it’s justified by theory T. Theory T is a good theory because it explains MJ4, taken as an empirical knowledge claim.) Merely providing empirical generalizations is not sufficient. Now, it is possible that a complete explanation of the world could have been given strictly in terms of the ontology of the manifest image. If this had been the case, as Gutting shows, then scientific realism would have been a contingent thesis for Sellars, as would the claim that we can replace the ontology of the manifest framework with that of the scientific one.19 But we don’t have such an explanation of the world. We have the scientific one, and it is the touchstone of reality because only scientific theory directly accounts for observed phenomena. Scientific realism is therefore necessary, not contingent. Sellars argues that the theoretical entities and explanations of science must be posited, because empirical generalizations by themselves are never adequate as descriptions of the observed facts. “Theory is the only source of genuine explanation and hence must be regarded as directly accounting for the observed phenomena. In so doing, it also explains the approximate correctness of the empirical generalizations” (Gutting 1977, 77). Note the similarity between this defense of scientific realism and the account of explanatory coherence. More important than this, however, is the fact that Sellars’ scientific realism is also connected to his rejection of givenness. In the debate over eliminative materialism, philosophers borrow from Sellars the idea that there is no need to hang onto the everyday observational idiom that we use to talk about our mental lives. The theoretical entities of a mature cognitive science have all the explanatory power of the observational terms, so why not just use the theoretical terms? (Gutting 1977, 80-81) Sellars, therefore, construes the correspondence rules between the domains of observation and theory as asserting that observational terms do not denote. As Gutting observes, this “view of the nature of correspondence rules is very unpositivistic.”20 But what does this have to do with the attack on the Given? Gutting explains:

200

Siblings Under the Skin

The major obstacle to such a view of correspondence rules is the belief that the observational framework has a privileged foundational role in knowing, precisely because it is the only possible locus of the experience on which all our knowledge about the world must be based. But, Sellars replies, this belief is just the myth of the given and, as such, simply false (Gutting, 81).

Thus, Gutting insists, “two of Sellars’ most fundamental doctrines, scientific realism and the rejection of the given, are seen to be inextricably interconnected.” (1977, 81) It might be tempting to wiggle out of these connections by returning to the distinction between the scientific and manifest images and dismissing it as unnecessary. But Sellars’ arguments against the Given and for explanatory coherence shows us that we cannot dismiss it so easily; explanatory coherence relies on it, and you cannot have noncircular coherentism without it. Can we then give up on coherentism? Again, no. As Sellars shows, we cannot have effective agency without all of these notions. Not only has Sellars shown that it is possible to be a coherentist with a correspondence theory of truth (even if it’s “very unpositivistic”), he establishes that it is impossible to be a good coherentist without such a theory of truth. Feminists who want to be anti-foundationalists about perception must therefore be scientific realists.

*** VI. Effective agency, at last Explanatory coherence and the support it receives from logical space are adequate to escape the charge of circularity within the epistemic framework, or theory T, in which we find ourselves. But they do not by themselves give us good reasons for accepting metajudgments such as those provided by Sellars. In order to provide good reasons for these, he has to move outside the framework. As Sellars puts it, “what does all of the above metaphysical chatter about frameworks have to do with the rationality of accepting MJ1, MJ3 and MJ4?” (MGEC, para. 81) It can’t be the case that just any explanation of why our judgements are likely to be true will do. The explanations in question must

Sobstyl and Sellars

201

be rational to accept. That is, they also must be likely to be true. “The answer, according to the proposed strategy, lies in the necessary connection between being in the framework of epistemic evaluation and being agents. It is this connection which constitutes the objective ground for the reasonableness of accepting something like theory T.” (MGEC, para. 80) Sellars calls this the “moment of truth”. He says that agency, to be effective, involves having reliable cognitive maps of our environment and our selves. This means, “the concept of effective agency involves that of our IPM judgements being likely to be true, i.e., to be correct mappings of ourselves and our circumstances.” Therefore, it is reasonable to accept metajudgments like MJ5 (“IPM judgments are likely to be true”) “simply on the ground that unless they are likely to be true, the concept of effective agency has no application” (MGEC, para. 84). Sellars claims that in order to be effective agents in the world, we must be capable of bringing about “changes in ourselves and our environment in order to realize specific purposes or intentions”(MGEC para.68). It isn’t just that a background epistemic framework like T is useful in helping us to realize our purposes, it’s that we can’t realize our purposes without such a framework. There is a necessary connection between being in the framework of epistemic evaluation and being agents. Since we are necessarily connected to the framework, one may well wonder why we still need to bother justifying our acceptance of it. If Sellars is correct, it is not as though we have any real choice in the matter. On his view it is unlikely that we could ever choose not to be effective agents. But if his coherentism is to have any appeal as a justificatory model, there should be the possibility of a meaningful choice between coherent systems of belief. By making effective agency a necessary connection, he seems to rule out the possibility of such a choice. But if we want an epistemology in which it is no accident that we get things right, it is not persuasive to say that this is just the kind of epistemology that we have, necessarily. Can Sellars respond to this serious objection? Logical spaces and explanatory coherence are again the connections through which this necessary connection to agency makes sense. Coherence is a property to be maximized primarily by our belief systems, but when it comes to choices about beliefs, it is not systems that

202

Siblings Under the Skin

we are faced with; it is individual beliefs. The logical space in which belief systems are explained is a different one than the logical space in which individual beliefs are accepted or rejected. Both spaces, and the different forms of justification they utilize, are necessary for us to meet the aims of being effective agents. But Sellars insists that the logical space of explanatory coherence is primary. Unless we take seriously the possibility of a broad Cartesian skepticism, we can conclude that at a certain level of generality, i.e., the level of metajudgment, there will be little variation in the framework provided by theory T. This is quite compatible with there being lots of variation at the level of specific beliefs, that can be adopted or rejected in light of both inference from and to other beliefs, and whether they are supported noninferentially through the metajudgments found in the logical space of explanations. The necessity of our connection to the framework, and our functioning effectively within it, is not a logical constraint for Sellars. It is more like a logical compulsion. The key is that he places the objective ground for the reasonableness of accepting T neither in a set of fundamental propositions nor in inferential relations between propositions, but in a connection between our epistemic framework and us. It is the quality of that connection that determines our success as knowers and Sellars’ success as an epistemologist. Epistemic responsibility is promoted by Sellars through this connection—the epistemic framework of theory T is shared. The agent is accountable for his or her beliefs because other agents in the framework will take them to be warranted. Part of what it means to be an effective agent is to recognize that one’s verbal reports will be taken as authoritative, and that others will infer the presence of the things that one’s verbal assertions report. This is where the shift away from individual knowers to epistemic communities by feminists becomes relevant once again. The model of the atomistic knower repudiated by feminists as solipsistic is a model that is contingent upon perceptual foundationalism. The self-authenticating reports that the attack on the Myth of the Given rejects are based on direct apprehensions, which are unavoidably individualistic—it is individuals who apprehend. The step from individual perception to authoritative verbal report in the Myth is a step from the atomistic individual knower to the community in which language is learned and used.

Sobstyl and Sellars

203

But if we reject the Myth, we set aside the isolated agent on whom it rests as well—to use Sellars’ terms, we dispose of the associated demi-myth of observation treated in the technical sense. But what remains after we reject the foundationalist individual is not just the community. It is a coherentist form of the knowing individual, freed of the constraints of the Myth of the Given. The authority of this agent’s verbal reports is not traced to self-authenticating observations. It is traceable to inferential warrant in one logical dimension (the person making the reports believes them), and to non-inferential warrant in another (the person making the report believes that the metajudgements of theory T support such claims). The role of the community is to support the reliable acquisition of the whole framework by each individual. Sellars’ account of language learning thus includes a vital role for the community, but not in the manner that social feminist epistemologists like Hankinson Nelson endorse. A problem for traditional empiricist accounts of language acquisition is that they rely on either ostension or innatism.21 Sellars rejects both. He holds that to know the meaning of a word is to know a language, which at first seems both counter-intuitive and innatist. But it only remains counterintuitive and innatist if we continue to rely on the foundationalist model of the individual that the Myth of the Given entails. Once we accept the coherentist knower instead, we can see that the language one acquires is already known, “not of course by the learner but by the community which is doing the training”(Delaney 1977, 25). The reliability of one’s training in a language supports the authority of one’s verbal reports, and is in turn linked to one’s effective agency. (A related objection is sometimes raised to coherence in general; that for the coherentist in order to know anything one has to know everything. Sellars’ response to the case of language applies, mutatis mutandis, to knowledge as well. In order to know anything, the community has to know everything, and reliably train the individual.) The general point is that the rejection of the foundationalist individual knower is unneccessary labour for feminist epistemologists—giving up perceptual foundationalism takes care of the featureless knower. But that does not mean that we can just switch to

204

Siblings Under the Skin

talking about the community as the primary agent of knowledge. It means that we now have a clear path to thinking of the individual, connected to his or her community and epistemic framework, very differently than we have before.

*** VII. Conclusions By vanquishing the Myth of the Given, and establishing that knowledge still has a non-inferential foundation, Sellars breaks coherence theory out of its vicious circle. Logical space and explanatory coherence are crucial to the success of this argument. They are also critical to his argument for realism, which connects coherentism to the world, and for effective agency, which responds to questions about choosing between belief systems and the role of the community. All of these arguments have feminist utility. The defeat of the Myth disposes of the featureless knower as well, and creates an opportunity for the development of a new model of a coherentist subject, one which specifies the nature of the connections between the individual and his or her community. Explanatory coherence in particular provides feminist epistemology with a decisive direction for research. If there is a logical space in which empirical propositions rest on observation reports, and another logical dimension in which observation reports rest on the empirical propositions Sellars calls metajudgments, it is to those metajudgments and their role in theory T that feminists should turn. These, rather than appearances and observations, are the most likely place where the influence of gender bias will be uncovered.

Sobstyl and Sellars

205

Notes 1.

2. 3. 4. 5. 6.

7. 8. 9.

Not everyone draws such sweeping conclusions from Sellars’ attack on the Given. Those, like Richard Rorty who ignore Alvin Goldman’s advice that knowledge, reality and truth must be kept distinct, attribute a much stronger position to Sellars than his work supports. As C. F. Delaney and Gary Gutting demonstrate at some length, Sellars may be an observational anti-foundationalist, but he is still a principled scientific realist and even endorses a version of a correspondence theory of truth. See Goldman (1999), Rorty (1979), and Delaney and Gutting in Delaney et. al. (1977). Arch-positivist A.J. Ayer, for example, says that other knowers adjudicate one’s knowledge claims. (“The Right to Be Sure”, reprinted in Alcoff (1998 pp. 20-25). Lorraine Code, for example, is a feminist virtue epistemologist, since she argues that knowing well is more important than being justified. See Code (1987 and 1991). See Harding 1996. Similarly, Donna Haraway argues that the kind of observation that underpins feminist standpoint is learned (1996). This, too, presupposes a vestigial perceptual foundationalism. See the discussion in Longino (1993). Some confusion arises from the fact that there are two extant versions of Sellars’ essay on explanatory coherence. The first, “Givenness and Explanatory Coherence”, appeared in 1973 and does not contain the crucial emphases on explanatory coherence that I focus on here. This is the version of the essay available to Delaney, e.t al. (1977) and Castaneda, e.t al. (1975). (None of the contributors to these two volumes refer to this essay; it appears in the bibliography only.) The second version, “More on Givenness and Explanatory Coherence”, contains a clearer elaboration and much more definitive commitment to the view that there are two distinct kinds of coherence. This version of the essay, excerpted here, is the one available to Brandom (1994 and 1995), Rorty in his introduction to the 1997 edition of EPM, and Tanesini and McDowell. Quine, “Epistemology Naturalized” (1969) p. 83. “Postscript on metaphor” 1981, quoted in Hankinson Nelson (1990) p. 85. There is one passage in Brandom’s Making it Explicit (1994, 90) that relates Sellars’ challenge to the Given very indirectly to something like explanatory coherence, but it requires that we very charitably interpret Brandom’s use of “understanding”.

206

Siblings Under the Skin

10. Tanesini (1999), p. 8; she relies on Brandom’s reading of Sellars’ logical space. 11. See her “The Practices of Justification” (1998), in which she argues for a normative but non-cognitivist function for epistemological language, analogous to non-cognitivism in ethics. This is similar to Sellars’ approach of treating justification as normative but non-inferential. 12. I am guilty of abstracting from Sellars’ work only the parts of relevance to the feminist project as outlined here. 13. Rorty is particularly guilty of this, claiming as he does that “Sellars’s psychological nominalism does not stem from behaviorism as a thesis about what the mind is or is not” (1977, 188n.19). He thinks that behaviorism in Sellars’ sense is just the view that our practices of justification are social and not in need of any empirical or ontological ground. This claim makes it sound as though Sellars doesn’t have a well thought out philosophy of mind, or that it’s unrelated to his epistemology and metaphysics, which isn’t true. It also overstates the extent of Sellars’ repudiation of a belief ’s being “grounded.” See Sellars EPM, p. 66, and 78 14. Remember that, also unlike Brandom, McDowell and Tanesini, he is writing prior to the 1979 publication of the revised essay MGEC. 15. All feminist philosophers of science think this. 16. Most feminist philosophers of science think this. Sellars says, “in the dimension of describing and explaining the world, science is the measure of all things, of what is that it is, and of what is not that it is not” (EPM 83). It is easy to see how Sellars’ ontological stance influences Paul Churchland’s eliminative materialism, but much harder to see how his views support Rorty’s rejection of epistemology. Note also, however, that he is using the notion of logical dimension to support this claim, which is compatible with another dimension in which science is not the measure of all things. 17. Sellars does not think that we currently have complete explanations of this sort; he calls this “unfinished business.” But he thinks that such explanations will involve “finding inductive support for hypotheses concerning the mechanisms involved and how they evolved in response to evolutionary pressures” (MGEC, para. 85). Therefore, contrary to Rorty, the causal evolutionary explanation of our beliefs is epistemological, because both inference and explanation are forms of justification for Sellars, and explanatory coherence is primary. 18. Presumably this separation is what Rorty has in mind when he claims that “nobody” would find questions about the evolution of the species to be epistemological. It is clear by now that Sellars would not agree, however. See n.17, above.

Sobstyl and Sellars

207

19. This is the claim that Paul Churchland (1979) borrows from Sellars, although he does not explicitly state that it is a contingent thesis. 20. In Delaney et al (1977), p. 79. Gutting’s discussion is based mostly on “Theoretical Explanation” (Sellars 1967) and “The Language of Theories” (Sellars 1963). 21. Sellars’ account of language acquisition is found in Delaney et al (1977).

References Ayer, A.J. 1998. The right to be sure. In Epistemology: The big questions, edited by L. M. Alcoff. Oxford: Blackwell. Brandom, Robert. 1994. Making it explicit. Cambridge: Harvard University Press. ——. 1995. Knowledge and the social articulation of the space of reasons. Philosophy and Phenomenological Research 55 (4) 895-908. Campbell, Richmond. 1998. Illusions of paradox: A feminist epistemology naturalized. New York: Rowman & Littlefield. Castaneda, Hector-Neri. 1975. Action, knowledge and reality: Critical studies in honor of Wilfrid Sellars. Indianapolis: Bobbs-Merrill. Churchland, Paul M. 1979. Scientific realism and the plasticity of mind. London: Cambridge University Press. Code, Lorraine. 1987. Epistemic responsibility. Hanover: University Press of New England. ——. 1991. What can she know? Feminist theory and the construction of knowledge. Ithaca: Cornell UP. Delaney, C.F. 1977. Theory of knowledge. In The synoptic vision: Essays on the philosophy of Wilfrid Sellars, edited by C. F. Delaney, Michael J. Loux, Gary Gutting, and W. David Solomon. Notre Dame: University of Notre Dame Press. Delaney, C.F., Michael J. Loux, Gary Gutting, and W. David Solomon. 1977. The synoptic vision: Essays on the philosophy of Wilfrid Sellars. Notre Dame: University of Notre Dame Press. Goldman, Alvin I. 1999. Knowledge in a social world. Oxford: Oxford University Press. Gutting, Gary. 1977. Philosophy of Science. In The synoptic vision: Essays on the philosophy of Wilfrid Sellars, edited by C. F. Delaney, Michael J.

208

Siblings Under the Skin

Loux, Gary Gutting, and W. David Solomon. Notre Dame: University of Notre Dame Press. Haraway, Donna. 1996. Situated knowledges: The Science Question in Feminism and the privilege of partial perspective. In Feminism and science, edited by Evelyn Fox Keller and Helen E. Longino. Oxford: Oxford University Press. Harding, Sandra. 1996. Rethinking standpoint epistemology: What is ‘Strong Objectivity’? In Feminism and science, edited by E. F. Keller and H.E. Longino. Oxford: Oxford University Press. Keller, Evelyn Fox and Helen E. Longino. 1996. Feminism and science, Oxford readings in feminism. Oxford: Oxford University Press. Longino, Helen. 1993. Subjects, power and knowledge: Description and prescription in feminist philosophies of science. In Feminist epistemologies, edited by L. M. Alcoff and E. Potter. New York: Routledge. McDowell, John. 1995. Knowledge and the internal. Philosophy and Phenomenological Research 55 (4) 877-893. Nelson, Lynn Hankinson. 1990. Who knows: From Quine to a feminist empiricism. Philadelphia: Temple University Press. Quine, W.V.O. 1969. Epistemology naturalized. In Ontological relativity and other essays. New York: Columbia University Press. Rorty, Richard. 1979. Philosophy and the mirror of nature. Princeton: Princeton University Press. Scheman, Naomi. 1993. Though this be method, yet there is madness in it. In A mind of one’s own: Feminist essays on reason and objectivity, edited by L. M. Armstrong and Charlotte Witt. Boulder: Westview Press. Sellars, Wilfrid. 1963. The language of theories. In Science, perception and reality. London: Routledge and Kegan Paul. ——. 1967. Theoretical explanation. In Philosophical perspectives. Springfield Illinois: Charles Thomas. ——. 1973. Givenness and explanatory coherence. Journal of Philosophy 70: 612-24. ——. 1979. More on givenness and explanatory coherence. In Justification and knowledge, edited by George S. pappas. Dordrecht: D. Reidel. 169-182. ——. 1997. Empiricism and the philosophy of mind. Cambridge: Harvard University Press. (Original edition, 1956, Minnesota studies in the Philosophy of Science, vol. 1. Herbert Feigl and Michael Scriven, eds. Minneapolis: University of Minnesota Press .)

Sobstyl and Sellars

209

Seller, Anne. 1988. Realism versus relativism: Towards a politically adequate epistemology. In Feminist perspectives in philosophy, edited by M. Griffith and M. Whitford. Bloomington: Indiana University Press. Tanesini, Alessandra. 1998. The practices of justification. In Epistemology: The big questions, edited by L. M. Alcoff. Oxford: Blackwell. ——. 1999. An introduction to feminist epistemologies. Oxford: Blackwell.

Biographical note Unlike Heather Douglas, about whom more following her essay “Hempelian Insights for Feminism,” Edrie Sobstyl (b. 1961) has been a cat person from earliest childhood. Cats, it will be noted, are inherently non-philosophical, despite their frequent occupation of mats in the literature. While this explains the idiosyncracy of Edrie’s career path (award-winning General Manager of an alternative FM radio station, in Calgary, Canada, 1986-89), her interest in epistemology can nonetheless be traced to her first cat, Mittens. Acquired when Edrie was four, Mittens lived for 17 happy years in Calgary where she raised a family of three. (Mittens, not Edrie). One day Mittens was smuggled to the vet and euthanized and then smuggled back home. Attempts were made to reassure Edrie with an elaborate tale of Mittens’ death from natural causes. You can see how this would incline a budding philosopher towards an interest in knowledge and truth. Edrie’s next cat, Dennis, wrote her doctoral dissertation. Well, he sat on the keyboard while it was written, which amounts to the same thing. The University of Alberta (Edmonton) overlooked his assistance and granted Edrie her Ph.D. in philosophy in 1995. Edrie’s essay on Wilfrid Sellars was written while she was a Rockefeller fellow in the Ecological Conversations program at the University of Oregon in Eugene, where her current cats, Otis and Lotus, wisely declined to contribute in light of concerns over intellectual property. His anti-foundationalism has always seemed to Edrie to be an obvious aid for feminist epistemologists, especially in an attempt to articulate a coherent role for empirical evidence, however Edrie is one of the first feminist theorists to make links with Sellars’ philosophy. When not enjoying the temperate rainforests of the NorthWest, Edrie, Otis and Lotus normally reside in Dallas, where Edrie teaches in the History of Ideas program at the University of Texas.

210

Siblings Under the Skin

Sobstyl and Sellars

211

More on Givenness and Explanatory Coherence Wilfred Sellars

1. Historically, there have been two competing strands in the concept of a ‘self-presenting’ state of affairs, i.e., the kind of state of affairs access to which is supposed to provide empirical knowledge with its ‘foundation’. 2. According to the first strand, a self-presenting state of affairs is a fact (an obtaining state of affairs) which is known to obtain, not by virtue of an act of warranted belief, but by virtue of a unique cognitive act which is more basic than that of any believing however warranted. 3. The main thrust of this position is directed against what it decries as ‘representationalism,’ the view that whatever other conditions they must satisfy to constitute knowledge, cognitive acts are, in the first instance, representations. 4. Representationalism can take different forms. That which is most prevalent today stresses the quasi-linguistic character of cognitive acts, the idea that they belong to a framework of signs and symbols which, whether innate or acquired, enables organisms which possess it to construct representations of themselves in their environment. 5. A more traditional form stresses a distinction between two ‘modes of being’ which objects and states of affairs may have: (a) actual being— roughly, the being which something has independently of being an object of thought; (b) intentional inexistence, the being something has when it is thought of, qua thought of—Descartes’ ‘representative’ or ‘objective’ reality. 6. According to the latter form, our cognitive access to the world consists exclusively in the occurrence of mental acts in which objects and events have intentional in-existence. The term ‘exclusively’ is of the essence, for it is representationalism in the (supposedly) pejorative sense by virtue of denying that we have any other cognitive access to the world than by such acts. Whatever other features these acts may

212

Siblings Under the Skin

have, in no case do they involve, nor are they accompanied by what the anti-representationalist would characterize as a ‘direct’ cognitive access to ‘the facts themselves’ in their character as factual. 7. The representationalist grants, of course, that certain cognitive acts have a special character by virtue of which they are capable of yielding knowledge of certain privileged matters of fact. The possession of this special character would either guarantee that a representational act is true, or at least give a high antecedent probability to the proposition that this was the case. 8. Just how this special character is to be conceived, classical representationalists found it difficult to say. As a matter of fact, careful reading of the texts reveals that most, if not all, representationalists covertly introduced a non-representational mode of cognitive access— interestingly enough, to representational acts themselves. Because of this fact, Immanuel Kant might well have been the first thoroughgoing representationalist. 9. They took it for granted that one could ‘notice,’ ‘compare’ and ‘consider’ one’s mental acts, and to do these things was, in effect, to have nonrepresentational knowledge of these acts as being what they are, e.g., a representation of a tree. Such awareness would simply be a special case of what anti-representationalists refer to as unmediated or direct apprehension of matters of fact. 10. That Descartes, like Locke, covertly introduced a nonrepresentational element into his theory of knowledge along the above lines is, I think, clear. He did, however, also insist that certain representational acts have, in addition to their character as representations, a special property—which he referred to as ‘clarity and distinctness’—which plays the role described in paragraph 7 above. 11. Such a property might be called a knowledge-making property, by analogy with what in moral philosophy are called right-making characteristics.1 12. Of course, a coherent representationalist who argues that all cognitive access to the world is a matter of the occurrence of representational acts need not take clarity-cum-distinctness to be the knowledge-making property, or even a knowledge-making property.

Sobstyl and Sellars

213

In any case it seems more appropriate to a priori knowledge than to knowledge of particular matters of fact. 13. With respect to the latter, the representationalist might propose some mode of causal confrontation of the knowing by the known. The connection would be a knowledge-making property of the representational act, and would give it a cognitive virtue which other kinds of representations do not have. 14. The contemporary representationalist who stresses the linguistic analogy might well argue that the causal relationship between the knowing and the known in introspective, perceptual and memory judgments is reflected by the presence of demonstrative components in the representational act. 15. To develop this theme, however, would require the formulation of a theory of intentionality which does justice to both the logical and the causal dimensions of discourse about mental acts, and this topic is much too large to be more than adumbrated on the present occasion.2 16. Representationalists typically become touchy when asked whether our only access to the fact, when it is a fact, that an act has a knowledgemaking property (whatever it might be) is by the occurrence of a further representational act. And, indeed, the question whether representationalism can be so formulated as to alleviate this touchiness is a central theme in disputes pertaining to ‘foundationalism’. 17. From this perspective, the alternative to representationalism is the view that we have a direct access to the factuality of certain privileged facts unmediated by representational acts, whether quasi-linguistic episodes (e.g., tokens of Mentalese) or conceptual acts in which states of affairs have ‘representative being’ or intentional in-existence. 18. According to this alternative, our direct or non-representational access to these privileged facts (call it ‘direct apprehension’) provides a cognitive stratum which ‘underlies,’ ‘supports’ or ‘provides a foundation for’ cognitive acts of the representational category. If we call the latter ‘beliefs,’ then while some justified true beliefs may merit the term ‘knowledge,’ they rest on a foundation of direct apprehensions which are not special cases of beliefs, but belong to a radically different category.

214

Siblings Under the Skin

19. According to this line of thought, then, a self-presenting state of affairs is one which is either directly apprehended, or of such a kind as to be capable of being directly apprehended should the corresponding question arise. Direct apprehension or direct apprehendibiity would be a source of epistemic authority. 20. Notice that for a state of affairs to be self-presenting, as thus construed, it must obtain, i.e., be a fact rather than a mere possibility. 21. Now one who takes this line might deny that direct apprehensions themselves have intrinsic epistemic value and restrict terms of epistemic appraisal, thus ‘evident’ or ‘warranted,’ to beliefs, and, hence, to propositions whose epistemic status does not require that they be true.3 22. One might accordingly argue that the self-presentingness of self presenting states of affairs is a ‘prime mover unmoved’ (to borrow Chisholm’s useful metaphor) of epistemic authority (Chisholm 1976), i.e., that the direct apprehension or apprehendibility of a state-of-affairs is a source of evidentness or warrant, but itself neither warranted nor unwarranted. 23. Or, to put it differently, one might claim that their special relation to self-presenting facts is a knowledge-making property of certain beliefs. 24. To make this move, however, would be paradoxical, for it involves denying that the direct apprehension of a fact is itself knowledge. For direct apprehension, by those who have evolved the concept, is almost invariably taken to be the very paradigm of knowledge ‘properly socalled’. 25. On the other hand, to hold that an act of direct apprehension is subject to epistemic appraisal as, for example, evident seems to point in the direction of representationalism. How could items so sharply contrasted, as apprehensions and representations are by the anti-representationalist, both be ascribed epistemic value without equivocation. The anti-representationalist program is not without its problems. 26. The classical alternative to this conception of a self-presenting state of affairs as a fact which is either directly apprehended or capable of

Sobstyl and Sellars

215

being directly apprehended should the appropriate question arise is, of course, that of the representationalist. According to the latter account a self-presenting state of affairs is one which is such that if the relevant person at the relevant time occurrently believes (judges) it to obtain, the believing would have high epistemic worth and, indeed, would be non-inferentially warranted or self-evident. 27. Notice, as pointed out above, that this alternative is compatible with the idea that self-presenting states of affairs need not obtain (be facts). It is also compatible with the idea that when a self-presenting state of affairs does obtain, it is a factor which contributes (causally) to bringing about the occurrent belief (should such arise) that it does obtain. (See paragraphs 13 to 15 above). 28. The distinctive feature of this account is that the self-presentingness of a self-presenting state of affairs is to be understood in terms of the idea that the factual category to which the state of affairs belongs, e.g., that of being one’s occurrent mental state of the present moment, is an (epistemic) value-making property of the state of affairs. 29. Whereas ‘direct apprehension’ can masquerade as a non-value-laden term (see paragraph 24 above). II 30. Now it seems eminently clear to me that in his excellent paper on ‘Coherence, Certainty and Epistemic Priority’ (1964), Roderick Firth rejects the radical distinction between beliefs and direct apprehensions which is central to the first account of self-presenting states. 31. In his reconstruction of the concept of givenness the central theme is the idea that certain judgments have, in his terminology, an ultimately non-inferential warrant increasing property. 32. He alternates between speaking of ‘judgments’ (which I have been calling ‘occurrent beliefs’) and of ‘statement;’ but it seems clear that he is using the term ‘statement’ not to refer to verbal performances, but as a surrogate for ‘proposition’ in that sense in which propositions are contents of beliefs (i.e., are that which is believed). I think that he is implicitly making an analogical use of Strawson’s distinction

216

Siblings Under the Skin

between sentences and the statements they can be used to make on particular occasions, to emphasize that judgments have the conceptual counterparts of the indexical features of statement-utterances. 33. Firth distinguishes (pp. 549—50) two modes of inferential warrant increasing properties which statements may have. I shall paraphrase his distinction, referring to the two kinds simply as ‘the first kind’ and ‘the second kind’. (1) A statement, S, has an inferential warrant increasing property, P, of the first kind, if P consists in the fact that S is “validly inferable from certain other statements of a specified kind.” (2) A statement, S, has an inferential warrant increasing property, P, of the second kind, if the meta-statement, ‘if S has the property P, then S is likely to be true’4 has an inferential warrant increasing property, P’, of the first kind. 34. Firth refers to both ‘deductive’ and ‘inductive’ inference. The former is reasonably unproblematic. On the other hand he does not clearly specify the scope of ‘inductive.’ The example he gives is a case of what is usually called ‘simple’ or ‘instantial’ induction. Nevertheless the fact that he parenthetically ties ‘inferable from’ to ‘coheres with’ (1964, 550, 1.3) leaves the door open to the possibility of expanding the scope of ‘inductive inference’ to include other modes of non-deductive explanatory reasoning. 35. Whether or not he is prepared to avail himself of this option—and if so, to what extent—is a matter of crucial importance, for on it hinges the ultimate significance of his distinction between inferential and noninferential warrant increasing properties. 36. In any event, Firth illustrates inferential warrant increasing properties of the second kind by “the property of being believed by certain scholars,” e.g., I take it, by members of the relevant academy: It belongs to the second kind because the meta-statement, If a statement, S, has the property of being believed by certain scholars, then S is likely to be true, has the property of being inductively inferable from a statement to the effect that “these scholars have usually held correct beliefs about

Sobstyl and Sellars

217

statements similar to S’ in certain respects;” (Firth, p. 550) and this property is an inferential warrant increasing property of the first kind. 37. What, then, is a non-inferential warrant increasing property? Clearly, as far as the extension of the concept is concerned, it is a warrant increasing property (WP) which belongs to neither of the above two kinds. As for the intension of the concept, the contrast between non-inferential WP’s and inferential WP’s of the first kind is unproblematic. 38. Again, if the ‘inferential’ WP, P' of the meta-statement, Statements which have P are likely to be true, is construed as the property of having inductive support in the instantial sense, the contrast between inferential WPs of the second kind and noninferential WPs is equally unproblematic. 39. As a matter of fact, even if one counts the acquisition of a theory by a substantial degree of confirmation as a variety of acquiring inductive support, the distinctions remain reasonably straightforward. 40. But suppose that P' is the property of belonging to a theory of persons as representers of themselves-in-the-world, which, although it has good explanatory power and is capable of refinement by inductive procedures, was not (and, indeed, could not have been) arrived at by inferences guided by inductive canons however broadly construed. Would P' be an inferential WP or an explanatory but not inferential WP? 41. It might be thought that the question as to how the theory was ‘arrived at is one which belongs to the ‘order of discovery’ rather than ‘the order of justification.’ But reflection on the fact that to answer a question of the form ‘Is x justified in Ø-ing?’ requires taking x’s historical situation into account should give one pause. 42. I shall return to this theme in a moment. For the time being I note only that in the essay in question, Firth does not touch on these topics, although the issues they raise suggest possibilities which are in keeping with one aspect of his enterprise, which is, as I would put it, to reconcile as far as possible the claims of those who stress warrantedness grounded in explanatory coherence (among whom I count myself )

218

Siblings Under the Skin

with the claims of those who stress the non-inferential warrantedness5 of certain empirical statements (among whom I also count myself ). I shall attempt to push Firth in the direction of Firth-Bosanquet. III 43. But before arguing the case for the primacy of the concept of explanatory coherence in epistemic evaluation, it will be helpful to take a careful look at Firth’s account of the non-inferential warrantedness of certain empirical statements. It will also help in defining our problem to examine the connection he finds between the concept of non-inferential warrant and that of what he calls ‘epistemic priority’ (Firth, p. 553 ff.) 44. He begins by suggesting that …the statement… ‘it looks to me as if I am seeing something red’...has a certain degree of warrant for me because it is a statement (whether true or false) that purports to characterize (and only to characterize) the content of my present experience. To which he adds, finding the above to be too permissive, …a statement about my present experience can have some degree of ultimate non-inferential warrant for me only if I believe it to be true (p. 553). 45. Separating out the chaff introduced by the ambiguities of the word ‘statement’, this amounts to the idea that the property (PE ) of being a judgment about, and exclusively about, my present experience is a noninferential WP. 46. This requires that the meta-judgment, MJ1 Judgments which have PE are likely to be true, if itself warranted, have a warrant which, as Firth is using the term, is non-inferential. 47. Assuming, for the moment, that MJ1 is warranted, and having in mind the question raised in paragraph 40 above, let me provisionally characterize MJ1 as a non-inductively warranted warrant principle and

Sobstyl and Sellars

219

contrast it with MJ2 Judgments which are believed by certain scholars are likely to be true. Which can be characterized as an inductively warranted warrant principle. 48. Notice that in the second passage quoted in paragraph 44, Firth refers to “degree[s] of ultimate non-inferential warrant” (ital. mine). What he has in mind, of course, is that while the property of being believed by certain scholars (e.g., members of the American Geographical Society) is not an inferential property of geographical statements, so that it would be incorrect to classify its character as a WP as its being an inferential WP, yet it would be equally incorrect to classify it as a non-inferential WP, for it owes its character as a WP to its own inferential property of being an inductively established sign of probable truth. Although itself a non-inferential property of geographical statements, it has a higher order inferential property which is the ground of its ability to contribute warrant to geographical statements. It is by virtue of this fact that it is an inferential WP of the second kind. And it is in this sense that it is not an ultimately noninferential WP. 49. Thus, when Firth tells us that PE, the property of being exclusively about one’s own experience of the present moment, is an ultimately non-inferential WP, he means that PE is not only an inferential property of the judgments in question, but that, unlike being believed by members of the A.G.S., it is not a WP by virtue of itself having an inferential property. 50. In other words, Firth is committed to the idea that MJ1, which characterizes PE as a WP, either itself has no WP or, if it does, has a WP which is non-inferential. 51. Since, however, Firth does not press the question as to whether MJ1 does or does not have a WP, the possibility arises that he is emphasizing that it does not belong in the same box as MJ2, i.e., that its WP (if it has one) is not inductive. This would leave open the possibility that its WP, if any, is either non-inferential, or inferential but non-inductively so (cf. paragraph 47).

220

Siblings Under the Skin

52. These considerations make it clear that the question ‘Does MJ1 have a WP?’ or, to put the matter bluntly, ‘Is there any reason to accept MJ1?’ can no longer be deferred. 53. A negative answer would, on the face of it, bring to shipwreck the enterprise of making sense of the epistemic evaluation of empirical propositions. On the other hand, an affirmative answer would immediately raise the question ‘What WP does MJ1 have?’ or ‘What reason is there to accept MJ1?’ 54. And to these questions the only available answers, given the Firthian context in which they arise, are: (1) it is self-evident or exiomatic that it is reasonable to accept MJ1; (2) it is reasonable to accept MJ1 because if it is false, no empirical statements are warranted. 55. Both of these answers would turn us aside with a stone instead of the bread which, in spite of all the dialectical niceties, we intuitively feel must be there. Self-evidence is too atomistic an interpretation of the authority of epistemic principles; while the second answer—which amounts to the old slogan ‘This or nothing’ is too weak, in that we do seem to have some insight into why something like the epistemic principles so lovingly polished by Firth and Chisholm are true. What has gone wrong? IV 56. Firth formulates what he calls “the central thesis of epistemic priority” as “the thesis that some statements have some degree of warrant which is independent of (and in this sense ‘prior to’) the warrant (if any) which they derive from other statements” (p. 553). He thinks that this thesis is correct, and concludes that “Lewis has always been right in maintaining that the major task of a theory of empirical knowledge is to show how it is possible ... for statements that have independent non-inferential warrant to serve as the ground of all of the rest of empirical knowledge” (p. 556). 57. Now it is clear that some principles which assert that a certain property of empirical statements is a WP are themselves statements which belong to the content of empirical knowledge thus

Sobstyl and Sellars

221

MJ2 Statements which are accepted by the A.G.S. are likely to be true is not only a criterion for assessing geographical knowledge claims; it is itself an empirical statement and, indeed, an inductively confirmed knowledge claim in its own right. 58. But what of MJ1? And what of such principles6 as MJ3 If a person ostensibly perceives (without ground for doubt) something to be Ø (for appropriate values of Ø) then it is likely to be true that he perceives something to be Ø. MJ4 If a person ostensibly remembers (without ground for doubt) having ostensibly perceived something to be Ø (for appropriate values of Ø) then it is likely to be true that he remembers ostensibly perceiving something to be Ø. Might not these also be both principles which provide criteria for adjudicating certain empirical knowledge claims and empirical knowledge claims in their own right?7 59. Now if an affirmative response took the form of a claim that MJ1, MJ3 and MJ4 are empirically confirmed knowledge claims—thus putting them in a box with MJ2—a sensitive nerve would be touched. Would not such a claim involve a vicious circularity? 60. Since it is obvious that they cannot be empirical generalizations which owe their epistemic authority to confirmation by instances, one might look for a less direct mode of confirmation by experience. 61. Even if indirectly, however, an appeal must ultimately be made to the fruits of introspection, perception and memory. Sooner or later we would be confronted by such pairs of statements as It is reasonable to accept MJ1, MJ3 and MJ4 because they are elements in a theory T which coheres with our introspections, perceptions and memories. Our ostensible introspections, perceptions and memories are likely to be true because they fall under MJ1, MJ3 and MJ4. 62. How, we are inclined to expostulate, could it be reasonable (at t) to accept T because it is supported by our introspective, perceptual and memory judgments (IPM judgments), if it is because they fall under

222

Siblings Under the Skin

MJ1, MJ3 and MJ4 that it is reasonable to accept these IPM judgments? 63. Consider Jones accepts T for the reason that IPM judgments support T; Jones accepts IPM judgments for the reason that they fall under T. This would seem to be no more rational on Jones’s part than the situation described by Jones saves money in order to maintain his bank account; Jones maintains his bank account in order to facilitate his saving money. 64. We might say that in these two cases Jones is being unreasonable because his motivation is circular. They can be compared to cases in which a person is unreasonable because acting from a selfcontradictory intention. Of course, the circle must not be too small (or the contradiction too blatant) if the concept of such a case is to be coherent. 65. To be distinguished from such ‘subjective’ unreasonableness is the objective ungroundedness8 which would obtain if the following per impossible were true: What makes it reasonable for Jones to accept T is simply the fact that T is supported by the IPM judgments which it is reasonable for Jones to accept. What makes it reasonable for Jones to accept these IPM judgments is simply the fact that they are likely to be true by virtue of falling under T, which it is reasonable for Jones to accept. 66. It should therefore be clear, if it was not already, that what we were groping for in paragraph 58 was a way in which it could be independently reasonable to accept MJ1 , MJ3 and MJ4 in spite of the fact that a ground for accepting them is the fact that they belong to T, which we suppose to be an empirically well-confirmed theory. 67. I think that such a way can be found by following a strategy developed in two essays on the reasonableness of accepting inductive

Sobstyl and Sellars

223

hypotheses.9 As a matter of fact, the following considerations are necessary to round out the argument of those essays. 68. Such an expanded account might well be called ‘Epistemic Evaluation as Vindication’. Its central theme would be that achieving a certain end or goal can be (deductively) shown to require a certain integrated system of means. For the purposes of this necessarily schematic essay, the end can be characterized as that of being in a general position, so far as in us lies, to act, i.e., to bring about changes in ourselves and our environment in order to realize specific purposes or intentions. 69. In the above-mentioned essays I argued that among the necessary means to this end is the espousal of certain patterns of reasoning, specifically those involved in the establishing of statistical hypotheses, laws and theories. 70. Although I did have something to say about how these various patterns of reasoning are interrelated, I treated each ‘mode of probabilification’ as though it stood on its own feet and said relatively little about how these dimensions of ‘prima facie’ reasonableness combine and interact to generate probabilities ‘all things considered’.10 71. And, in particular, I had nothing to say about the probability of observation statements—though it is obvious that the probability of an inductive hypothesis is a function of the probability of the observational premises which are mustered to support it. 72. In the language of the present essay, I had nothing to say about the probability which attaches to ostensible introspections, perceptions and memories (IPM judgments). 73. If challenged, I would have appealed to something like PJ1, PJ3 and PJ4, and argued that they are true. If asked why it is reasonable to accept them, I would have argued that they are elements in a conceptual framework which defines what it is to be a finite knower in a world one never made. 74. In short I would have appealed to a more encompassing version of what I have been calling theory T. 75. To be one who makes epistemic appraisals is to be in this framework. And to be in this framework is to appreciate the interplay of the reasonablenesses of inductive hypotheses and of IPM judgments.

224

Siblings Under the Skin

76. Now in the case of particular theories, e.g., the corpuscular theory of light, one can imagine that one gets into the conceptual framework of the theory by a process of inductive reasoning. One entertains the framework and finds it inductively reasonable to espouse it. One espouses it for the reason that it is inductively reasonable to do so, and one is being reasonable in so doing. 77. But can one espouse theory T for the reason that it is inductively reasonable to do so, and be reasonable in so doing? We have already seen that the answer is ‘No!’. 78. Clearly we must distinguish the question ‘How did we get into the framework?’ from the question ‘Granted that we are in the framework, how can we justify accepting it?’ In neither case, however, is the answer ‘by inductive reasoning’ appropriate. 79. Presumably the question ‘How did we get into the framework?’ has a causal answer, a special application of evolutionary theory to the emergence of beings capable of conceptually representing the world of which they have come to be a part. 80. As to the second question, the answer, according to the proposed strategy, lies in the necessary connection between being in the framework of epistemic evaluation and being agents. It is this connection which constitutes the objective ground for the reasonableness of accepting something like theory T. 81. I say ‘something like’ theory T, for we are now at the moment of truth and must get down to specifics. Thus, what does all of the above metaphysical chatter about frameworks have to do with the rationality of accepting MJ1, MJ3 and MJ4? 82. The answer is that since agency, to be effective, involves having reliable cognitive maps of ourselves and our environment, the concept of effective agency involves that of our IPM judgments being likely to be true, i.e., to be correct mappings11 of ourselves and our circumstances. 83. Notice, then, that if the above argument is sound, it is reasonable to accept MJ5 IPM judgments are likely to be true, simply on the ground that unless they are likely to be true, the concept of effective agency has no application.

Sobstyl and Sellars

225

84. Now for the linchpin. We must carefully distinguish between having good reason to accept MJ5 and having good reason to accept a proposed explanation of why IPM judgments are likely to be true. 85. To explain why IPM judgments are likely to be true does involve finding inductive support for hypotheses concerning the mechanisms involved and how they evolved in response to evolutionary pressures. And this obviously presupposes the reasonableness of accepting IPM judgments. 86. To borrow a Firthian locution, MJ5 is epistemically prior to the reasonableness of particular IPM judgments, whereas particular IPM judgments are epistemically prior to explanations of the likely truth of IPM judgments. 87. Some twenty-two years ago I wrote: If I reject the framework of traditional empiricism, it is not because I want to say that empirical knowledge has no foundation. For to put it this way is to accept that it is really ‘empirical knowledge so-called,’ and to be put in a box with rumors and hoaxes. There is clearly some point to the picture of knowledge as resting on a level of propositions observation reports—which do not rest on other propositions in the same way as other propositions rest on them. On the other hand, I do wish to insist that the metaphor of ‘foundation’ is misleading in that it keeps us from seeing that if there is a logical dimension in which other empirical propositions rest on observation reports, there is another logical dimension in which the latter rest on the former.12

To the extent that this passage was one of my notorious promissory notes, I hope that the present essay provides some of the cash. 88. I will conclude by rounding out the considerations advanced in paragraphs 85—6. They explain why I wrote in paragraph 80 that there is an objective ground for accepting ‘something like’ theory T; for, as it exists at any one time, theory T is a complex which includes MJ5 and attempts to explain why IPM judgments are likely to be true. The latter enterprise is still unfinished business. 89. It is in the former respect that it constitutes the conceptual framework which spells out the ‘explanatory coherence’ which is the ultimate criterion of truth.13

226

Siblings Under the Skin

Notes This paper is a revised version of the central argument of ‘Givenness and Explanatory Coherence’, The Journal of Philosophy, 70 (1973). 1. Compare Chisholm, Perceiving, pp. 30-32. 2. I have, however, discussed it at length on a number of occasions, thus Science and Metaphysics, chapters 3-5; most recently in Naturalism and Ontology, chapters 4-5. 3. Otherwise put, ‘to states of affairs whose epistemic status does not require that they obtain’. Compare the preceding paragraph. 4. Firth writes ‘true’ where I have put ‘likely to be true’—but that is simply a symptom of his lack of concern for inferential warrant increasing properties of the second kind. 5. A warrantedness, however, which may be inferential in something like Firth’s account of inferential WP’s of the second kind. 6. Adapted from Chisholm’s principles B and D (1976, pp. 80-1). 7. After all, can it not be argued that law-like statements are both empirical hypotheses and material rules of inference? 8. It is important to distinguish between two possible constructions of ‘It is reasonable for Jones to do A because A-ing would bring about X. Thus R1 The fact that A-ing would bring about X makes it reasonable for Jones to do A. R2 It is reasonable for Jones to do A for the reason that by A-ing he (himself ) would bring about X. In R1 what is reasonable is simply the doing of A, and the fact that A would bring about X is the ground of this reasonableness. In R2 what is reasonable is the doing of A with a certain intention or purpose, i.e., that of bringing about X by doing A. 9. ‘Induction as Vindication’, in ([1964]1974); and ‘Are There Non-deductive Logics?’ (1970). 10. After all, my philosophical purpose was to exhibit inductive reasoning as a form of practical reasoning, and by so doing to throw light on the very concept of probability. 11. May I call them pictures? 12. ‘Empiricism and the Philosophy of Mind,’ in Minnesota Studies in the Philosophy of Science, Vol.1 (1956); reprinted with minor alterations in Science, Perception and Reality (1963). The passage occurs on page 170 of SPR. 13. I have, of course, neglected in this essay the equally important and no less difficult topic of deductive reasonableness.

Sobstyl and Sellars

227

References Chisholm, Roderick. 1957. Perceiving. Ithaca: Cornell University Press. ——. 1976. Theory of knowledge (2nd edition). Englewood Cliffs: Prentice Hall. Firth, Roderick. 1964. “Coherence, certainty, and epistemic priority.” The Journal of Philosophy, 61. Sellars, Wilfred. [1956] 1963. “Empiricism and the philosophy of mind.” Reprinted with minor alterations in Science, perception and reality, London. ——. [1964] 1974. “Induction as vindication.” Reprinted in Essays in Philosophy and its History. Dordrecht. ——. 1970. “Are there non-deductive logics?” In Essays in Honor of C. G. Hempel Ed. by N. Rescher. Dordrecht. ——. 1979. Naturalism and Ontology. Ridgeview Press.

Biographical note Wilfred Sellars (1912-1989), was the son of Canadian-born philosopher Roy Wood Sellars and was raised and educated in the American midwest. In the years prior to US involvement in WWII he went to Oxford as a Rhodes scholar and earned a second bachelor’s and master’s degree. He saw active duty in the US Naval Reserves, and then taught at a number of institutions, including Yale. In 1963 he moved to the University of Pittsburg, where he remained until his death from chronic alcohol poisoning. At a memorial service held at the University Chapel, his second wife Susanna Downie (his first wife had died in 1970), spoke of his strong support for her feminist politics, though he often chided her for her more New Age moments, which marked their only disagreements. He is best known for his criticism of what he called the “myth of the given”, a foundationalist view of the role of empirical evidence that was widely-held in analytic philosophy until the mid-century. He had a naturalist, pragmatic view of epistemology, which he used, some think, paradoxically, to supported a realist view of science. His most famous work is Empiricism and the Philosophy of Mind (1956, republished by Harvard University Press, 1997).

228

Siblings Under the Skin

Lacey and Popper

229

Where values interact with science Hugh Lacey

I. Popper and the ideal of scientific objectivity In countless ways the practices and theories of science bear the marks of moral and social values. This is a fact but one that does not, in Popper’s philosophy of science, call into question the ideal of science as “objective” or “beyond human authority”: …we should…admit that all knowledge is human; that it is mixed with our errors, our prejudices, our dreams, and our hopes; that all we can do is to grope for truth even though it be beyond our reach. We may admit that our groping is often inspired, but we must be on our guard against the belief…that our inspiration carries any authority, divine or otherwise. If we thus admit that there is no authority beyond the reach of criticism to be found within the whole province of our knowledge, …then we can retain, without danger, the idea that truth is beyond human authority. …without this idea there can be no objective standards of inquiry; no criticism of our conjectures; no groping for the unknown; no quest for knowledge (Popper, “On the Sources of Knowledge,” 1963a: 29-30).

Popper’s confidence in the ideal of science as objective is rooted in several theses about scientific methodology that I put as follows: Thesis 1. We may distinguish two stages in the conduct of scientific practices: that in which a theory is put forward and that in which it is tested (Popper, 1959: 30-31).1 Values may play a role at the first but not the second stage. Thesis 2. Testing a theory depends solely on its logical relations with its empirical basis. Theories are falsifiable: they may be inconsistent with statements accepted in their empirical bases; and they should

230

Siblings Under the Skin

be rejected when falsified and accepted (considered corroborated) when attempts to falsify them have been unsuccessful (Popper, 1959: chs. 4, 5). Thesis 3. The conduct of research is responsive to antecedently made decisions about what to observe and about the categories to be employed in observational reports (or basic statements)—i.e., about the selection of data or the items of the empirical basis—and reciprocally about the kinds of theories to put forward as candidates for testing. Consider: Observation is always selective. It needs a chosen object, a definite task, an interest, a point of view, a problem. And its description presupposes a descriptive language, with property words; it presupposes similarity and classification, which in turn presupposes interests, points of view, and problems.… For an animal a point of view is provided by its needs, the task of the moment, and its expectations; for the scientist by his theoretical interests, the special problems under investigation, his conjectures and anticipations, and the theories which he accepts as a kind of background: his frame of reference, his ‘horizon of expectations’ (Popper, 1963b: 46-47).

In my own terminology (SVF),2 (logically) prior to engaging in research a decision is made to adopt a strategy. The principal roles of a strategy are to constrain the kinds of theories that may be entertained— and thus to specify the kinds of possibilities that may be explored in the course of the inquiry—and to select the appropriate empirical basis for testing theories constrained in this way. Without deciding which strategy to adopt we cannot address coherently and systematically: What questions to pose? What puzzles to resolve? What classes of possibilities to attempt to identify? What kinds of explanations to explore? What phenomena to observe, measure and experiment upon? What language to use in observational reports? What procedures to use? Thesis 4. Basic statements deploy categories of a “mechanistic” or “materialistic” kind, categories applied in the light of use of measuring instruments, typically in experimental contexts:

Lacey and Popper

231

Basic statements are…statements asserting that an observable event is occurring in a certain individual region of space and time. …it is possible to interpret the concept of an observable event in a psychologistic sense. But I am using it in such a sense that it might just as well be replaced by ‘an event involving position and momentum of macroscopic physical bodies’. Or we might lay it down, more precisely, that every basic statement must either be itself a statement about relative positions of physical bodies, or it must be equivalent to some basic statement of this ‘mechanistic’ or ‘materialistic’ kind (Popper, 1959: 103).

Basic statements, so decided upon, can only be brought into logical contact with theories articulated with certain categories which I summarize as those that can be deployed in conjectures or representations of law (mathematically expressed), of the underlying structures of things and their components, and of their processes and interactions—in abstraction from the place of things in human experience, social relations and value outlooks. In my terminology, scientific methodology involves the adoption of materialist strategies (SVF). Theses 2 and 4 express decisions, and like any decisions they are rationalized by appeal to values. …I freely admit that in arriving at my proposals I have been guided, in the last analysis, by value judgments and predilections. But I hope that my proposals may be acceptable to those who value not only logical rigor but also freedom from dogmatism; who seek practical applicability, but are even more attracted by the adventure of science, and by discoveries which again and again confront us with new and unexpected questions, challenging us to try out new and hitherto undreamed-of answers (Popper, 1959: 38).

The values appealed to Popper are, we might say, values held to be constitutive of the practices of science, reflecting (on the face of it) no particular moral and social values. Practical applicability is one of the values, but not applicability in the service of any particular valueoutlook. The decisions, for him, express commitment to the value of objectivity, of “truth beyond human authority.”

232

Siblings Under the Skin

Note that the adoption of materialist strategies, since they involves abstraction from human life and experience, ensures that there are no consequences that can be drawn from scientific theories in the realm of values.3 Whether or not theories put forward under materialist strategies can express “truth beyond human authority,” they cannot (except derivatively) have authority in the realm of values. These theories are cognitively neutral among value-outlooks. This is ensured by the decision to adopt ‘materialistic’ categories only, categories that have been deployed precisely because they carry no value import. Cognitive neutrality often seems to imply another kind of neutrality: neutrality-in-the-context-of-application—theories may be applied to serve the interests of any value-outlook, and/or the theoretical products of science are, considered as a totality, evenhanded in application.4 That is why it appears sufficient for Popper, in the quotation above, simply to cite the value of practical applicability. Science that is responsive to his methodological decisions plays no moral favorites; otherwise it could not aspire to a truth beyond human authority. A challenge to neutrality-in-application would thus seem to call the objectivity that Popper values into question. Does the decision to engage in scientific practices that are geared to ensure cognitive neutrality also lead to neutrality-in-application? If not, is the decision well grounded? Science for the progress of Brazilian society A few years ago a major scientific convention held in Brazil had as its theme: “Science for the progress of Brazilian society”.5 The organizers, I am sure, wished to evoke the idea of science as neutral-in-application: any value-outlook can be served by scientific knowledge, and so—presumably—science can serve the progress of Brazilian society. At the same time, the theme hints at a certain urgency, that science may not be serving Brazilian progress very well. It suggests that currently science is falling short of the ideal represented by this kind of neutrality, but that it could be re-structured and re-directed so as to fit it better. Might it not also raise the question that—when we consider the needs of Brazilian development—we are able to see that it is not an appropriate ideal to guide the practices of science?

Lacey and Popper

233

This hint of urgency evokes further considerations. First, “progress” is a value-laden term. What would constitute progress for Brazilian society? What values are constitutive of “progress”? Those involved with progressive incorporation into the global neoliberal order? Those linked with the liberation of vast numbers of poor Brazilians from sufferings that have systemic causes? (Something else?) Secondly, can science in principle serve progress, regardless of the interpretation given to “progress”? Is science really neutral-in-application? Or, is science, or its currently predominant modes (Thesis 4), especially well adapted to serve the interests of some value-outlooks rather than others, perhaps at the present time those of the neoliberal project of “globalization”?6 Clearly the interest and importance of these questions generalizes to issues about the place of science in development projects throughout the “third world,” and to whether the answers vary with different conceptions of development (SVF: ch. 8; Lacey, 1994). In this chapter I will focus upon the second set of questions concerning the neutrality-in-application of science and implications that then arise for the objectivity of science. Objectivity and its critics Theses 2 and 4 capture the core of Popperian “objectivity”. They complement each other: adopting the constraints of 4 serves to make transparent any violations of 2, and making the decision expressed in 4 would make little sense if factors other than empirical data and “appropriate” (logical) relations between theory and data were properly brought to bear in judgments of theory choice. Thesis 2 is a severe version of a thesis I call “impartiality.” Impartiality: Accept a theory of a domain of phenomena if and only if it manifests the cognitive values to a suitably high degree (according to the highest standards) in the light of available empirical data; and reject one if and only if a theory inconsistent with it has been soundly accepted. In the statement of this thesis it is presupposed that the cognitive values (empirical adequacy, informativeness, explanatory power, capability to identify the possibilities of things, minimization of recourse to ad hoc hypotheses, consistency with other accepted theories,

234

Siblings Under the Skin

etc.—whatever characteristics of theories, relations between theory and data, and relations between theory and other theories are held to be the criteria of soundly accepted theories) are distinct from moral, social and other kinds of values. For present purpose we may take it that neutrality-in-application presupposes impartiality and cognitive neutrality.7 I will use objectivity to refer to the conjunction of impartiality and neutrality (both cognitive and in-application), and the more severe Popperian objectivity to refer to the conjunction of Theses 2 and 4. In recent years the idea that the sciences are objective (and a fortiori Popperian objective) has been strongly criticized. The critics of objectivity form an oddly assorted group, including religious fundamentalists, new age enthusiasts and deep ecologists, feminists and third world social movements. Some say that scientific theories are “social constructions”, open to sociological explanation but not to decisive rational appraisal. Many also suggest that particular value-outlooks are in play in the choice of theories, especially socially dominant values, often attributing to modern science (or parts of it) influences drawn from the menu: western, patriarchal, capitalist, racist and imperialist, as well as from directly practical interests. For the adherents of science as objective, the criticisms represent a return to irrationality. From the critics’ perspective, however, the view that science is objective is inherently ideological, false consciousness that serves the interests of dominant values. Irrationality or ideology: Is that really the choice? I will explore the dialectical structure of this conflict, and propose a way to transcend it, not by means of a defense of Popperian objectivity, but in a way that draws upon the distinctness of impartiality and neutrality. In part, the conflict derives from conflicting views about the nature of scientific understanding. Does modern scientific understanding constitute the only form of rational understanding of the world, or is it just one among several alternative forms?8 II. Scientific understanding and the metaphysics of scientific materialism Understanding in general can take different forms and, in so doing, be responsive to the different interests of different practices. Un-

Lacey and Popper

235

derstanding phenomena involves being able to answer: What are they? Why are they the way that they are? What possibilities do they permit (including hitherto unrealized ones), both in virtue of their own powers to develop and through their interactions with other things? (And sometimes how are these possibilities to be realized?)9 The questions are deeply interrelated. To get an image of how they are, consider crop seeds. Understanding crop seeds We cannot answer what crop seeds are without pointing to the possibilities that growing them allow. What are they? They are at least at one and the same time: (a) Biological entities—under appropriate conditions they will grow into mature plants from which (e.g.) grain will be harvested; so they are subject to genetic, physiological, biochemical, cellular, developmental, etc analyses. (b) Parts of various ecological systems. (c) Human-produced entities. (d) Implicated in social relations. Items (a) - (d) provide parameters for specific answers to the questions: What? Why? What possible? For researchers in current biotechnology crop seeds are primarily what they are shown to be in genetic, cellular, molecular biological, etc studies that are informed by a particular interpretation of (c), which lends interest to the investigation of possibilities of genetically modifying seeds for the sake of such further possibilities as increased crop yields or growing pesticide-resistant plants. Then seeds become entities developed recently by scientists10 by means of genetically modifying seeds that (for the most part) have previously been selected in the course of the practices by numerous farmers over the course of centuries, and produced now (largely) by capital-intensive corporations. Then crop seeds cease to be what they are for many traditional farmers and for contemporary agroecologists: the “natural” product of plants, simply a part of the grain harvested, entities that regenerate themselves annually as a normal part of the crop, and integral to sustainable agroecological systems in which biodiversity is nurtured and to social arrangements which emphasize local well being (Kloppenburg, 1991; Shiva, 1991; Lacey, 2000a; Lacey & de Oliveira, 2001). When crop seeds become (in social practice) what they are considered to be in biotechnology,

236

Siblings Under the Skin

the possibilities of maintaining sustainable ecological systems of which they are a part is threatened. They also become more entrenched as commodities (Kloppenburg, 1991; Shiva, 1991): entities bought and sold on the market, “property” whose users may not be their owners, and (in some cases) available be patented and otherwise regulated in accord with intellectual property rights. These seeds—unlike the seeds of sustainable agroecological systems—need not contribute to the possibilities of meeting the food and nourishment needs of communities in the locale of the production. What crop seeds are, then, draws upon all four of the deeply interrelated dimensions. Dimension (a), e.g., cannot be investigated so that the outcomes of the investigation can coexist with any interpretations of the other dimensions. It is part of the biological/socio-economic causal nexus today that crop seeds, investigated as they are in current biotechnology, will become entrenched as commodities. The possibilities of genetic modification (e.g., pesticide-resistant plants) cannot be realized today without realizing more completely the possibility for seeds to become ever more exclusively commodities, with effects on the environment that are not yet well understood and evaluated (Rissler & Mellon, 1996; Nodari & Guerra, 2001). Similarly, the possibilities of expanded crop production in sustainable agroecosystems (Altieri, 1987) cannot be realized under social arrangements in which the global market has primacy and in which agribusiness has a monopoly of agricultural research. Seeds as objects for genetic modification and seeds as components of sustainable agroecosystems cannot readily coexist in the same agro-ecological systems; their respective possibilities cannot be realized together (SVF: chs. 8, 10; Lacey, 2000a). Understanding gained under materialist strategies It is characteristic of modern natural scientific theories to express a particular form of understanding, that gained from conducting research under what I called above (elaborating Thesis 4) materialist strategies, strategies that constrain theories to those able to represent phenomena as lawful, typically as generable from underlying structure, process, interaction and law. Laws express relations among quantities, and theories identify the possibilities of things in terms of the generative

Lacey and Popper

237

power of the underlying order. Since their fundamental categories are those of law and quantity, theories represent phenomena in abstraction from any place they may have in human experience and practical activity, and from any links with social values. How a phenomenon relates to value-outlooks is irrelevant to its theoretical representation. Thus, for theoretical purposes, an arrow is considered in abstraction from its role in war or sport and from the social processes of its fabrication and its movements are represented only as a function of such variables as velocity, mass and aerodynamic properties. Similarly, under materialist strategies the seed is represented in abstraction from its having become a commodity and from its consequences in agroecosystems. Reciprocally, theories that are constrained in this way, are tested in virtue of their relations with other accepted theories and, most importantly, with the appropriately selected empirical data: those represented in basic statements, which in addition to passing tests of intersubjectivity and replicability, tend to be quantitative (obtained from measurement operations) and to describe experimental phenomena, and their descriptive categories ensure abstraction from contexts of intentionality and value. To accept a theory is to judge that it is sufficiently well supported empirically (or, following Popper, sufficiently corroborated) so that, skeptical considerations aside, no further appraisal of it is needed at present, including with respect to the context of application. If the theory has been soundly accepted, i.e., accepted in accordance with impartiality, it has been judged that it manifests the cognitive values to a suitably high degree, that it adequately maintains the appropriate relations with the available empirical data and with other accepted theories (SVF: 13-17; ch. 3). Are theories, developed and accepted under materialist strategies, neutral-in-application; can any value-outlook be informed by applications of some of these theories so that their interests are significantly served? Or, do they (when considered as a totality) favor particular value orientations, perhaps those in which technological innovations, or specifically modern ways of conceiving the control of natural objects, are deemed highly valuable? Is the abstraction from the context of values an indication of applicability across value-outlooks, or a mask to favor a particular outlook? Note: that theories are neutral-in-applica-

238

Siblings Under the Skin

tion cannot be inferred from their being cognitively neutral; that they are the former is an empirical not a conceptual thesis. Why would one think that theories developed under materialist strategies are neutral-in-application? One response comes from commitment to the metaphysics of scientific materialism: The aim of science is to represent the world “as it really is”; and the way the world really is is independent of human perception, desires, values, practices and interests; the world, the spatio-temporal totality and/or the totality of possibilities, simply is constituted of objects, characterized by a set of quantities and relations, and interacting with other objects in accordance with laws. From this interaction among objects and the components of their underlying structures and processes, phenomena are generated—and the totality of possibilities is encapsulated in the generative power of the underlying order. This materialist image of the world is well known. Since Galileo and Descartes it has had widespread appeal, and I suspect that it is this appeal that enabled Popper to propose making the decision to adopt materialist strategies without explicitly considering other possible strategies, that might constrain theories and select data in different ways.11 From the viewpoint of scientific materialism, only the categories deployed under materialist strategies can be adequate to represent the world as it is. Then, neutrality-in-application follows from the claim that acceptable theories represent the world as it is—for, ceteris paribus, a theory that did not so represent the world could not be relied upon to inform the practices of any value-outlook. Representing the world as it is While the appeal of scientific materialism has been strong, it also confronts difficulties. How can we know that the world is the way that scientific materialism says that it is?12 Moreover, even if it is, how can we know that our best theories represent it adequately? These questions arise because we cannot compare our theories directly with the world. We produce representations of the world, especially in the course of our measuring and experimental interactions with it. The world is not like a book, as Galileo thought, lying open to be read, waiting for its categories to be transposed into our theories. Rather we fit our categories to the world.

Lacey and Popper

239

A paradox threatens: the aim is to represent the world as it is, and materialist metaphysics holds that the way it is is independent of human perceptions and interests; but, for us, representation is linguistic or symbolic and its categories are ours. Representations are human products, historical constructions of scientific practice, produced with the aid of strategies that themselves emerged historically; and they deploy theoretical categories that have been created, structured, developed, refined, transformed and applied in the course of our interactions with the world. In order to represent, we must establish contact with the world and engage with it. Our experience is never simply of the world, but of the world in interaction with us. Thus, we can know directly only that theories represent the world as it is encountered in and appears from the perspective of scientific practices. How, then, can one affirm that what theories represent is independent of the practices in which theories are produced—that theories represent the way the world is? How can one affirm that an object [X], as it is grasped in scientific practices, is identical to the object as it is independently of human values and practices? I.

X as grasped in scientific practice = X as it really is. [?]

I said that we cannot compare a theory directly with the world. In scientific practices, the empirical data provide our point of contact. We accept and reject theories—according to impartiality—principally on the ground of their relations, as expressed in key cognitive values, with the empirical data. Then, an object, as grasped in scientific practice, is identical to the object as represented in a soundly accepted theory [T] one which, with respect to the appropriate domain of phenomena, manifests the cognitive values to a suitably high degree in the light of the available empirical data. II. X as grasped in scientific practices = X as represented in soundly accepted T. Is there a further argument that an object as represented in a soundly accepted theory is identical to the object as it is?

240

Siblings Under the Skin

III. X as represented in soundly accepted T = X as it really is. [?] Without (III) we cannot infer (I) from (II), and we lack an argument for (III). We now arrive at a new picture. We cannot ground neutrality-inapplication in materialist metaphysics, for—apart from the question of the grounds for materialist metaphysics—we have no assurance that scientific practices enable us to grasp the world as it is, and thus that the aim of science stated above, “to represent the world as it is,” can be realized. Let us replace this aim with one we can more confidently expect to be realizable, e.g., “to construct and consolidate theories that accord with impartiality.”13 Then, we remain with three key questions: 1) Can neutrality-in-application (or even cognitive neutrality) be derived from impartiality? 2) Can impartiality be adequately defended? 3) Why adopt materialist strategies, and does doing so accord with neutralityin-application (SVF: ch. 6)? III. “Postmodern” criticisms of objectivity Certain criticisms of science as objective, that I will call “postmodern,”14 maintain that not only is neutrality-in-application not manifested in fact of the theories accepted in modern science, but also moral and social values are in fact playing roles in the choice of scientific theories. The latter claim is supported by socio-historical studies; recent feminist philosophy of science, e.g., has documented numerous cases, principally in biology and psychology, in which racist and sexist biases have played such roles (Longino, 1990; Nelson & Nelson, 1994). More importantly these criticisms also involve epistemological arguments that, in principle, impartiality is an unrealizable and therefore not a normatively binding ideal; the roles that are played in fact by moral and social values (or factors other than the cognitive values, e.g., metaphysics) cannot be avoided. We are thus left only with a choice of which values should play the role, not with an option to eliminate the role. We may consider (leaving aside a lot of details) such arguments to proceed in two steps. The first step aims to refute the Popperian Thesis 2, and also Theses 1 and 4, drawing upon the theses of underdetermina-

Lacey and Popper

241

tion and invulnerability to refutation of theory by data. It readily generalizes to apply to those versions of logical empiricism that affirm the epistemic primacy of empirical data, the hypothetico-deductive analysis of scientific theories (a theory is a deductively organized structure of generalizations and hypotheses where the available data of the relevant domain are among its deductive consequences),15 and that the empirical support of a theory increases with the quantity of data deductively linked with it. The second step maintains that Kuhnian alternatives to critical rationalism and logical empiricism do not permit impartiality to be regained. Underdetermination Thesis 2 affirms that theories are falsifiable. The same account of the logical structure of scientific theories that supports falsifiability, however, also supports (logically) that, regardless of the number of tests to which a theory has been submitted successfully and thus of the magnitude of the set of empirical data included among its deductive consequences, there always remains the possibility that there is an incompatible theory (perhaps not yet formulated) with the same set of data among its deductive consequences. Then, if such a theory were to be developed, a choice between the two theories could not be made on the basis of these data. No amount of data can eliminate the possibility of discovering a rival theory to match even a highly empirically adequate theory. Theory choice is always underdetermined by the data, so that—it is inferred—empirical adequacy can only be a necessary condition for sound theory choice. Then, that values may play a role alongside empirical adequacy supplementing the data, as diagnosed in the socio-historical studies, may come as no surprise and perhaps appear to be unavoidable. After all, if the data do not suffice to ground the choice, something else must also be playing a role. (When only one relevant theory, that is highly empirically adequate, has been formulated it may be difficult to discern that role.) Underdetermination, while not challenging that acceptable theories must be empirically adequate, apparently underlies that there may be “ideological” critique of scientific theories, and also “legitimation” for investigation that is explicitly driven by values. Underdetermination follows from the logi-

242

Siblings Under the Skin

cal structure of theories posited in Thesis 2; then the inference drawn from it above denies, as not realizable, the conditions for acceptance of theories affirmed there. The inference ignores, however, that underdetermination is only one side of the coin. It is a thesis about the continued possibility of theoretical conflict. The other side of the coin is that, when there is an actual (explicit) theoretical conflict, it can be resolved (in principle) decisively by appeal to the empirical data. We may expect that, when developed adequately, contradictory predictions may be generated at the observational level from the conflicting theories. Then it would be sufficient, in principle if not always in practice, to make the relevant observations (in the relevant experimental or natural space, using the appropriate instruments) in order to resolve the conflict by falsifying one of the theories—here there is no room for the intervention of values. When both sides of the coin are considered, rather than being drawn inevitably to the inference that values must play a role in matters of theory choice, we see that any influence of values in theory choice may be attenuated by a methodology that insists on the multiplication of theories (Feyerabend, 1981) that are actively put into confrontation with one another. Then theoretical conflict actively generated by investigators with different value-outlooks can provide the means to diagnose and attenuate any improper role that values may be playing (cf. Longino, 1990: 76-81; Lacey, 1999b). Invulnerability Given the logical structure of theories posited in Thesis 2, considerations of empirical adequacy cannot pick out a unique acceptable theory; however, they suffice (in principle) for choice between actual theoretical competitors. But in the absence of efforts to develop alternative theories, in the light of underdetermination it is plausible that moral and social values may actually play a role alongside the cognitive values—their acceptance might even explain why there is no serious effort to entertain alternatives. Be that as it may, the argument continues, actual scientific theories do not have the logical structure posited in Thesis 2. Although many theories may be considered as deductively organized structures

Lacey and Popper

243

of hypotheses, generally they do not contain empirical data among their deductive consequences; but data can be among the deductive consequences of a theory in combination with various auxiliary hypotheses, which include hypotheses about the mode of operation of instruments and experimental apparatus, about the boundary conditions of experimental spaces, about the acceptable limits of accuracy, and about ceteris paribus conditions. Theoretical and auxiliary hypotheses function jointly in the production of predictions. Thus, the experimental refutation of a prediction derived from a theory and a set of auxiliary hypotheses shows only that either the theory or one of the auxiliary hypotheses is false. Then, there is no logical obstacle to replacing the auxiliary hypotheses with alternatives in order to preserve the empirical adequacy of the theory. In general, any theory can be protected from falsification in this way. Theories are invulnerable to refutation by the data. Then, no logical obstacle remains to us accepting a theory in view of its links with our values and, since introducing auxiliary hypotheses in this way leaves underdetermination intact, to others accepting a rival theory in view of its links with their values.16 Then, separation of the stage in which a theory is put forward, in which values may have a role, and that in which it is tested, where they do not, cannot be sustained, so that Thesis 1 is also rejected. Incommensurability The second step of the argument is to challenge Thesis 4 as well as Theses 1 and 2. Kuhn (1962/1970) is the focal text. Kuhn accepts both Thesis 3, that research is conducted under strategies17 that constrain the kind of theory entertained and select the data of interest, and the resulting mutual dynamic behind theory-development and data-gaining activities in which both draw upon a common “lexicon”. The strategies that are adopted bear the mark of history (Lacey, forthcoming)—so that holding Thesis 4 reflects the mistaken view that materialist strategies must frame all scientific research. Furthermore, e.g., there have been certain passages from theory to theory in which the lexicon as well as theories change (e.g., at the time of the “scientific revolution” when materialist strategies came to ascendancy)—in such a way that the key terms of succeeding lexicons are not intertranslatable, and this

244

Siblings Under the Skin

failure of translation is reflected all the way down to the empirical data, so that there is not a common relevant stock of empirical data to bring to bear on theories across the passage. Theories formulated in different lexicons are incommensurable. Nevertheless, they are incompatible, not because they have inconsistent implications regarding empirical data, but because the strategies, under which the categories of the respective lexicons are applied, are incompatible. One cannot follow incompatible strategies simultaneously in the same context; one cannot simultaneously apply to the same things only the categories of quantity and law, and the categories of the Aristotelian cosmos, or (referring back to my earlier example) those of biotechnology and of agroecology (SVF: ch. 7). It is a practical incompatibility, rather like that involved in trying to play soccer and rugby simultaneously on the same field (Taylor 1982). Incommensurability of theories derives from incomparability of strategies (Lacey 2001a). Theory choice, therefore, is intimately linked with choice of strategy. Why adopt a particular strategy? Noting that Thesis 4 has been widely accepted, why are materialist strategies widely (almost exclusively) adopted in modern science? Kuhn himself emphasizes that a strategy should be fruitful, i.e., under it ever more theories are entertained that come to manifest the cognitive values highly. An old strategy is replaced by a new one when research under the new becomes fruitful and that under the old ceases to be so. Since underdetermination and invulnerability remain in place, there are readily available mechanisms for maintaining the empirical adequacy of theories of the old paradigm, although at the price of a lower degree of manifestation of some of the cognitive values (e.g., minimization of use of ad hoc hypotheses) and perhaps lending them support from their links with favored values. For Kuhn fruitfulness always takes precedence over such values, and that is why it is rational to adopt materialist rather than Aristotelian strategies. Even so fruitfulness is only a necessary condition rationally for the (long term ) adoption of a strategy.18 Kuhn does not discuss sufficient conditions because he did not think that two conflicting fruitful strategies could be in play at the same time even in different institutions. I have argued (SVF: ch. 7; Lacey, 2001a; forthcoming) that, while there are not many instances of conflicting fruitful strategies co-existing, there is no general (logical or methodological) reason why this should

Lacey and Popper

245

be so; and there are instances of it: agroecological and materialist (biotechnological) strategies, although incompatible in important ways, are both fruitful while—not without creating considerable tensions—supported in different institutional contexts which have interests in different kinds of practical applications in agriculture (Lacey 2000a).19 In such cases choice of strategy is grounded in values (or other factors, e.g., metaphysics, distinct from cognitive values and their strategic counterpart, fruitfulness). I have argued, e.g., that the almost exclusive adoption of materialist strategies (and certainly their predominance in contemporary agricultural science) is properly explained in terms of mutually reinforcing relations with specifically modern values concerning the control of nature;20 and the adoption of agroecological strategies21 in terms of relations with values, commonly found in “third world” popular organizations, such as local empowerment (well being, agency, participation and community) and ecological sustainability.22 Theories accepted under one kind of strategy, e.g., serve especially well—on application—the interests of the associated value-outlook; and the strategy is adopted in significant part for this reason. Incommensurability, then, is linked with lack of neutrality-in-application. Remember that the argument for incommensurability already assumed the rejection of Thesis 2. It might seem that impartiality has thus been refuted too. Certainly it is not uncommon for the inference to be drawn from lack of neutrality-in-application to the unrealizability of impartiality; the role of values in the choice of strategies is seen to penetrate right down to the choice of theories. Certainly, if one kind of strategies predominates, then theories not fitting its constraints will not be investigated seriously. It is a small move from this to the rejection of such theories;23 but, then, the values linked with the strategies are part of the ground for the rejection, in violation of impartiality. If impartiality goes, nothing is left of objectivity. IV. The success of modern science Together the theses of underdetermination, invulnerability and incommensurability provide the epistemological basis for the postmodern critique that I am considering, that values must play an essential role in choices of theories as well as of research strategies, so that impartiality is

246

Siblings Under the Skin

(in principle) unrealizable. Then, the critical task becomes that of identifying the values that are, in fact, playing that role, and—as indicated above—the values actually identified by the postmodern sociological critique are quite problematic ones. In the light of this critique, the ideal of science as objective, and the aim of science proposed by scientific materialism and substitute proposed for it, appear to be illusions. In science, we do not obtain representations of the world as it is or consolidate theories that accord with impartiality. Instead, we gain images of the world through which our presence and our value penetrate pervasively, thus casting doubt on whether science can be distinguished from opinion, ideology, dogma, or value judgments. It may appear to be this way! But the postmodern critique does not adequately deal with one striking phenomenon: the success of modern science in application. Practices based in scientific knowledge have been remarkably successful; they have changed the face of the earth. Scientific knowledge has made possible modern technology by providing the appropriate understanding of the material design and workings of technological objects. Technology works. What characterizes this knowledge that permits us to understand the material workings of technological objects? One response is: It is knowledge of the world as it is, consisting of representations of the components, structures, processes and laws underlying phenomena of the world. The success of technology and applied science appear to vindicate that, at least in some domains, we obtain knowledge of the world as it is (Taylor 1982, criticized in SVF: ch. 6), genuinely impartial knowledge, not something that has been shaped to fit the interests of the powerful. According to this response, successful application in social practice comes to the fore as the key cognitive criterion. Scientific materialism has not yet left the scene. We now find ourselves in the following predicament: On the one hand, scientific materialism is able to explain the success of science but it cannot explain how scientific methodology could produce knowledge of the world as it is. On the other hand, the postmodern critique argues to the impossibility, in principle, of scientific methodology producing knowledge of the world as it is, but it cannot explain the success of science. Neither position is satisfactory. Can we get out of this predicament?

Lacey and Popper

247

The way in which objects are grasped in scientific practices Earlier in the essay I raised the question: Does scientific practice grasp objects the way that they are (I) and, if so, how could we know this? Then, I pointed out that, according to impartiality, objects as grasped in modern scientific practice are identical to objects as represented in theories that, of the relevant domains, manifest the cognitive values to a high degree (II). The argument based on underdetermination, invulnerability and incommensurabilty is intended to cast doubt that objects as represented in theories are identical to the way they are independently of human interests, perceptions and values (III), and this has been taken to cast doubt on the realizability of impartiality (and thus even of (II)). So we remain without a positive answer to the question: Does scientific practice grasp objects the way they are? (Without III we cannot move from II to I.) What might we draw from reflecting on the success of science? Could it be that objects as grasped in scientific practice are identical to objects as interacted with in technological applications? Or, objects as successfully brought under control? Or objects as potential objects of control? In general, of course, they are not, for in modern science we gain knowledge of some objects, like distant galaxies, that are not potential objects of control.24 Nevertheless, the understanding we gain under materialist strategies is the kind of understanding needed to bring objects under technological control (SVF: ch. 6). Perhaps we may say: In scientific practices objects are grasped as they must be grasped if they are to be (become) objects of technological control. IV. X as grasped in scientific practice = X as it must be grasped in order to be an object of technological control. Scientific materialism proposes that it is because we understand the way an object is, as an object of the world independently of human perceptions and practices (including practices of control), that we are able to control it. It proposes, for those objects that might come within the reach of our power to control, that the way they are is identical to the way they must be described when considered as potential objects of control.

248

Siblings Under the Skin

V. X as it really is = X as it must be grasped in order to be an object of technological control. From (V) and (IV), (I) follows. Is (V) true? Long ago, Bacon proposed a similar idea: we are able to put objects “under constraint” for purposes of social utility because we have grasped principles of objects “free and at large”. He also offered the complementary idea: we come to understand the principles of “nature free and at large” by “putting nature under constraint” in experimental spaces. What supports the proposal that the way these objects are is identical to the way they are described when considered as potential objects of control? It would have to be that it explains technological success. Do we really need to appeal to (V) in order to explain technological success? Bacon’s complementary idea suggests an alternative. To bring objects under control, why do we need to know how they are in nature free and at large? Why is it not enough to know their behaviors, tendencies and possibilities that come to the fore in the spaces we construct, both experimental and technological, in order to exercise control?25 Or, why is it not enough to hold that the theories we accept can be expected to identify the possibilities that can be realized under conditions of “constraint”? Is there anything about modern scientific practices that ensures this? Remember first that characteristically they are conducted under materialist strategies that select with high salience data that describe experimentally controlled phenomena, and secondly that theories are accepted as providing understanding of specific domains of objects or phenomena. Because of the salience of experimental data, the domains, of which theories manifest the cognitive values highly, include experimental spaces and spaces that resemble them, such as technological (and some natural) ones. It is an empirical (historical) fact of the modern scientific tradition that some wide-ranging theories of such spaces have become well established. Then, technological application of a theory may be considered as one more concrete replication of the experiments that provide the theory’s empirical support. No “deeper”, metaphysical explanation (as in (V)) of the success of technology is needed, only that the world has proven itself to be amenable to grasp under materialist strategies, a grasp that progressively enables more of its “abstracted” possibilities to be identified.26

Lacey and Popper

249

VI. X as grasped in scientific practices = X as it must be grasped in order to be an object of technological control conducted under material strategies. Of course these are genuine possibilities of the world. But no ground has been provided to identify the world with the totality of “abstracted” possibilities; their realization may prevent the realization of other possibilities (not reducible to “abstracted” ones)—you cannot play soccer and rugby on the same field at the same time, and the seed cannot be both a commodity and play an integral role in sustainable agroecosystem in the same location at the same time; and the realization and interest of “abstracted” possibilities to a large extent depends on us (Lacey, 2000a; 2001a). The extent to which these possibilities are actually realized outside of contexts created by human beings, while it is extensive, always remains a matter for further empirical inquiry and speculation. Problems with scientific materialism and the postmodern criticism The postmodern criticism denied that scientific practices grasp objects the way that they are, independently of human values and practices—based in an argument that also denied impartiality (as well as Thesis 2). But this argument could not explain the success of science. My argument preserves impartiality, both as a worthy ideal and as concretely realized at least in the cases of our most entrenched theories. It does not claim that scientific practices grasp objects the way that they are, but it affirms that they grasp the “abstracted” possibilities of things. In general, there is no reason to hold that the possibilities of an object are exhausted by its “abstracted” possibilities (Lacey & de Oliveira, 2001). Why are we interested in the “abstracted” possibilities of things? More importantly, why has modern science developed under strategies that focus on “abstracted” possibilities almost exclusively? I think it is because that is how we need to grasp things in order to augment our capability to control things.27 Research under the materialist strategies grasps things as they must be grasped for the sake of furthering the social value of augmenting our capability to control natural objects. Like the postmodern critique, then, I hold that a social value is playing an important role. But the social values do not play their role

250

Siblings Under the Skin

alongside the cognitive values on a par with them, or as supplements to them when their play is insufficient to ground a theoretical choice. Rather they function at the level of strategies (constraining theories/ selecting data), at a logical moment prior to the play of the cognitive criteria. We adopt the strategies that we do because of their links with social values: materialist strategies, e.g., because of their links with the modern valuation of control towards natural objects. We adopt whatever strategy will best serve to identify the possibilities the realization of which would further the interests of those holding the values. Then, within the adopted strategies, the empirical data and the cognitive values are often sufficient to ground the acceptance of theories of certain domains of phenomena. I emphasize that the social and cognitive values do not interact at the same level. The social values do not determine which specific theories we accept; they only constrain the type of theory that we will investigate or, what is effectively equivalent, frame the choice of the domain of possibilities that we find it interesting to investigate. If we wish to investigate the possibilities of augmenting human control over natural objects, then it is rational to adopt materialist strategies. But exactly what the possibilities of control are will be represented in theories that manifest the cognitive values highly. Our desires and our values play no legitimate role in the acceptance of theories.28 The postmodern critique, that I have discussed, has three weaknesses. In the first place it tends to reduce the cognitive values to one only, empirical adequacy (and thus impartiality to Thesis 2) and does not consider how the deployment of the other cognitive values resolves, to a considerable degree, the problems raised above by underdetermination, invulnerability and incommensurability.29 Secondly, it does not recognize the distinction of logical moments of choice: the adoption of strategy and the concrete choice of theories. Since the strategies adopted reflect what are the possibilities of interest, it is proper to attribute a role in scientific practice to social values, but not at the moment of choice between theories that equally meet the restrictions of the adopted strategy. (Thus, its rejection of Thesis 1 must also be reconsidered.) Thirdly, and derivative from the second, it does not recognize that a theory may be accepted in accordance with impartiality (and even with cognitive neutrality) but not accord with

Lacey and Popper

251

the thesis of neutrality-in-application (SVF: ch. 10; Lacey, 1997a; Lacey, 2002). Scientific materialism also continues to have weaknesses. Theories developed under materialist strategies do not represent the world as it is, but the “abstracted” possibilities of things, and thus the possibilities of control of natural objects; and, for some other things beyond the realm of control (e.g., galactic phenomena), they provide an understanding that is projected from understanding gained in contexts of control (SVF: ch. 6; see also Note 24).30 V. Research and the social values it serves The conclusion that I have just drawn makes possible a questioning of modern science that is deeper than that of the postmodern critique which, when the clutter is cleared away, remains a skeptical questioning. My conclusion is compatible with impartiality but not with neutrality-in-application. The application of modern science, for the most part, serves especially well those value-outlooks and moral projects that embody the modern valuation of control. What form of scientific understanding serves the progress of Brazilian society? Is understanding gained under materialist strategies adequate for contributing to “the progress of Brazilian society”? Or does it need to be supplemented (or even supplanted) by other forms of understanding, which deploy different strategies (constraining theories/selecting data)? At the present time, the practices of the control of nature are largely in the hands of the forces of neoliberalism and so they serve certain values well and others not so well. At the risk of being provocative, they serve: individualism rather than solidarity; private property and profits rather than social goods; the market rather than the well-being of all persons; commodification rather than strengthening a plurality of values; individual liberty and economic efficiency rather than human liberation; the interests of the rich rather than the rights of the poor; formal rather than participatory democracy; and civil and political rights dissociated from any dialectical relation with social, economic and social rights.

252

Siblings Under the Skin

They serve, in other words, those values broadly associated with neoliberalism, but run counter to the values aspired to in many popular organizations throughout Latin America (Lacey 1997; de Oliveira 2001; see Note 22). Today, modern science, and the augmenting of our capability to control nature, serve for the most part the interests of neoliberalism. Could they also (or instead) serve the interests of alternative values? Are the possibilities of interest to alternative value-outlooks of a kind that generally might be discovered under materialist strategies? Once again, let us consider the seed. In the context of contemporary neoliberalism, the seed is becoming ever more completely a commodity. Its use has become increasingly part of the logic of profit and capital investment, and broken from the logic of the well being of the people in the regions of production and of ecological sustainability (Lacey 2000a). The values of the popular organizations underline an interest in research that addresses questions such as (Notes 21, 23): What are the possibilities of producing crops so that all the people in the region of production gain access to a nutritious diet, in a social context that reinforces local empowerment and sustains the environment? That question does not abstract from the prevailing conditions of daily experience and practical activities, and does not presuppose that issues of the social order must be subordinated to the implementation of innovative controls pertaining to production and distribution. It does not consider the chemistry, biology, ecology and sociology separately. It locates questions about crop yields in the midst of other questions like these: What are the socio-economic conditions and social effects of agricultural production? Who controls the production and the product? How is it utilized and distributed? How are the conditions of production and distribution related to each other? What are their effects on health and the ecology? How were these things done in the past, what were their conditions and effects, and why did they change? Could the methods of production and distribution of the past be developed to meet the needs of today? In the context of such questions, crop yields come to be investigated not only as a function of quantities, abstracted from their relations with human beings—as occurs under materialist strategies—but also as a function of human and social variables. When questions are posed in this way, attention is turned to the local and its particularity; and sound empirically-based answers may differ with

Lacey and Popper

253

locale. While such answers will lack neutrality-in-application, there is no reason why they, like answers generated under materialist strategies, should not accord with impartiality (SVF: ch. 8; Lacey, 2001a; forthcoming). The question, “What are the possibilities that serve the empowerment of all in a region?” may not receive a satisfactory response from research conducted exclusively under materialist strategies. In order to carry out the relevant investigations we will need to have recourse to some forms of understanding that do not deploy materialist strategies, and ask questions like those I have just put (Lacey, 2000a). In any case, in order to identify what type of science, or what type of systematic, empirically-based understanding, serves the “progress of Brazilian society”, we must first identify what is progress for Brazilian society, what values best express what it is. That, of course, is a question to be discussed not only in scientific circles, but also in democratic debates. Without that debate we cannot discern how values should properly interact with scientific practices; but, however they may interact, the proper place of doing so is at the moment of adoption of strategy. Popper’s legacy What remains intact of the Popperian legacy summarized at the outset? Thesis 3 remains fundamental. The conduct of scientific inquiry depends upon antecedently made decisions about the adoption of a strategy. But the decision—to adopt materialist strategies exclusively—expressed in Thesis 4 does not represent one that should generally be made to frame research unless one appeals to a particular (albeit widely endorsed and deeply socially embodied) social value, the modern valuation of control. Thesis 4 supports cognitive neutrality but not neutrality-in-application. Alternative strategies may be adopted (subject to their long-term fruitfulness) in view of mutually reinforcing relations that they may have with value-outlooks that conflict with the modern valuation of control, in particular their capability to generate theories that, on application, are likely to serve practical interests connected with the value-outlooks (e.g., the one outlined in the previous subsection). Considerations from the “stage of application” are relevant to decisions about adoption of strategy, and they support a strategic

254

Siblings Under the Skin

pluralism rather than the exclusivity of research under materialist strategies. Although Thesis 1 needs qualification, it expresses an important insight. I emphasize a third stage, that of application, which strongly influences the stage in which a theory is put forward, and I think that the relevant distinction is better put as that between the stage of adoption of strategy and choice of theory (with respect to a domain of phenomena and class of possibilities). Moreover, when a theory is considered in isolation from the strategy under which it has been developed, it is difficult to define the kinds of possibilities that it may be expected to encapsulate; the interpretation of a soundly accepted theory is not readily dissociated from its consolidation under a specified strategy. Finally, I have argued that Thesis 2 should be replaced by that of impartiality. The strategic pluralism that I urged to replace Thesis 4 is compatible with impartiality. They go hand in hand, thus avoiding the dire consequences apparently diagnosed by the postmodern criticism of Thesis 2, as Theses 2 and 4 do in the Popperian analysis.31 That is what remains of objectivity: impartiality without neutrality-in-application. The role of values in strategy adoption, that derives from the stage of application, need not undermine impartiality. That is enough to ensure that theories can be soundly accepted, thus bypassing the threat that all claims to knowledge include some play of values among their grounds, and the threat that rational criticism articulated by the cognitive values remains viable and, I believe, of crucial importance for all aspirations to liberation (SVF: ch. 9). That legacy, now deepened, is worth preserving.32

Lacey and Popper

255

Notes 1.

2. 3. 4. 5. 6.

The same distinction is often indicated using different words: the context of discovery (invention, conception, provisional entertainment, advancement of theories) vs. that of justification (validation). I take it that Popper considers the stage of testing to be distinct also from the stage of “application” of theories in social projects, a stage about which he is largely silent, and one at which values clearly have a role. “SVF” refers to Is Science Value Free? (Lacey, 1999a). See also Lacey (2002). I will use “values” as shorthand for “moral, social and other non-cognitive values”. I presuppose a distinction between cognitive and other kinds of values (SVF: ch. 3; Lacey, 1997a, in press). See SVF: ch. 10 for details and critical discussion; also Lacey (1997a) 46th annual meeting of SBPC (Sociedade Brasileira para o Progresso da Ciência), 1996. An earlier version of this article was published in Portuguese (Lacey 1997b, 1998: ch. 1) in response to SBPC’s theme. The term, “neoliberalism”, is not widely used in the US, though the reality it signifies pervades much of the economic practice and thinking that engulfs our lives here and that lies behind policies such as those embodied in “welfare reform.” By way of a summary statement, neoliberalism is identified first by commitment to the international free market (“globalization”), expanded commodification, individual initiative, private control of the economy, deregulation, removing restrictions on capital flow across borders, production for profit in the global market, reduced role and responsibilities for government (especially concerning social issues, though not concerning fiscal and monetary policies), formal electoral democracy and the separation of powers, and (where convenient) promoting civil and political rights; and secondly by programs and policies to further implement these objectives, shaped by organizations responsible for planning and overseeing them (e.g., International Monetary Fund, World Trade Organization, North American Free Trade Association and the proposed Free Trade Association of the Americas). The legitimating claim of neoliberalism is that a rising standard of living will gradually spread throughout a country’s population. The reality has tended to be a growing of the gap between the handful of the rich and the vast majority of the poor, increased impoverishment and social disruption, the selling off of national patrimonies to foreign owners, decreased social services and educational opportunities for the poor, the devastation of hope and the accompanying rise of criminal violence, and the exposure of the poor periodically to especially intense sufferings induced by fluctuations of the

256

Siblings Under the Skin

global economy. (See SVF: ch. 8; Lacey 1997.) Note that “globalization” is a contested term: on the one hand, referring to the fact of increased interaction among peoples from all parts of the world, and to the possibilities for cooperation that this occasions; on the other hand, referring to the structuring of this interaction within the neoliberal order. 7. For arguments about why impartiality is preferable to the narrower Thesis 2 (which may be considered, roughly, as what impartiality becomes when the cognitive values are limited to one: empirical adequacy) and more generally why the criteria used in the cognitive (epistemic) appraisal of theories should be considered “values”—”cognitive values”—see SVF: ch. 3; for details and qualifications about impartiality and neutrality (SVF: chs. 4 and 10; also Lacey, 2002). There I have argued that the idea of science as value free is best understood as compounded of three ideas: impartiality, neutrality (cognitive and in-application), and autonomy, which may be stated: “The characteristics of scientific methodology, the adoption of strategies in research, and the priorities and direction of basic research are set, without ‘outside’ interference, by cognitive interests alone.” I will not discuss autonomy in this chapter; nor will I address the contested question of what is the proper list of cognitive values. 8. I will use the term “science” to designate any systematic form of empirical inquiry that aims to provide understanding of phenomena (SVF: ch. 5). Thus, accord with Thesis 4 is not part of the meaning of science. 9. For elaboration see SVF: ch. 5. 10. Mainly in corporate but also in university, Non Governmental Organization and other laboratories. 11. 11 Its current version is adapted to the best available theories: quantum mechanics, relativity and evolution. It also provides the motivation for those recent developments of cognitive psychology which deploy computer models of the mind. Remember that, for Popper, falsifiability is considered a criterion for the demarcation of scientific conjectures and metaphysics, and that metaphysics plays an important role in his account of scientific practices—it is just that, like values, metaphysics has no legitimate role at the stage of testing theories. Popper’s methodology is designed to eliminate theories that falsely represent the way the world is. 12. Much of the argument in this section draws upon well known discussions of difficulties that confront theses of “scientific realism.” Note also that there are value-outlooks with presuppositions (e.g., religious ones) that deny that the world is this way, and whose proponents claim that there is soundly based knowledge that is not reducible to that gained under materialist strategies. Theories, soundly accepted under materialist strategies, may not apply evenhandedly within these and other value-outlooks.

Lacey and Popper

13.

14.

15.

16.

17.

257

(That is a matter for empirical investigation.) Then, the argument for neutrality-in-application in general would dissolve into an argument for neutrality-in-application among those value-outlooks whose presuppositions do not clash with materialist metaphysics. Without an independent compelling argument for materialist metaphysics (that there is not such an argument, see SVF: 126-130), this would hardly suffice to support to adopt Thesis 4, to support the exclusive adoption of materialist strategies. For similar reasons, Popper’s aim for science may perhaps be put: “to develop and consolidate theories that resist persistent and rigorous attempts to falsify them” or “… theories that maintain and increase the degree to which empirical adequacy is manifested in them after persistent …” (I will not consider how this relates to Popper’s own account in terms of “verisimilitude.”) This is simply a label that I give to the criticisms I discuss. I have not surveyed all criticisms of science that have been called “postmodern” or “poststructuralist,” so I do not claim that my objections to the criticisms stated in the text apply to all these criticisms. Strictly speaking, as Popper (1959: 101) makes clear, certain truth functions of basic statements [of the form: (B1&B2&…&Bn) ——> Bn+1], not atomic basic statements, are supposedly among the deductive consequences of theories. We may regard a datum as a statement of this form. I will say that a theory manifests empirical adequacy to the extent that (under Thesis 2) available relevant data are among its logical consequences or, more generally, (later when Thesis 2 is questioned) they are among the logical consequences of the theory and associated auxiliary hypotheses—where the degree of manifestation is “measured” in the light of the standards proposed in SVF: 62-64. The details of the first step of the argument presented here are well known (see, e.g., Longino, 1990; Okruhlik, 1994), and I have not introduced any novel moves. What I have called the thesis of invulnerability is a version of proposals that are often grouped under the “Duhem/Quine thesis”. What I have called strategies derive from and captures important elements of Kuhn’s paradigms, disciplinary matrices and structured lexicons. (See SVF: 261.) My notion of strategy has affinities with Hacking’s notion of “form of knowledge.” In each case questions can be asked about its adoption as a frame for research (logically) prior to the appraisal of any theories developed within the frame. Hacking says: “By a form of a branch of scientific knowledge I mean a structured set of declarative sentences that stand for possibilities, that is, sentences that can be true or false, together with techniques for finding out which ones are true and which ones false.

258

18. 19.

20.

21.

Siblings Under the Skin … A form of knowledge represents what is held to be thinkable, to be possible, at some point in time” (Hacking, 1999: 170-171). In my language, a strategy identifies the general features of the range of possibilities that may be explored within a research program; and, if its claims become hegemonic in a society, that will set limits to the “thinkable” in that society, so that questions that make sense in one frame may be unintelligible in another. Hacking’s and my emphases are different: he emphasizes how a form of knowledge may be altered by radically new inventions, I that commitment to conflicting value-outlooks may lead to the adoption of different strategies, though Hacking is sensitive to the contingency of the adoption of a form of knowledge and the role of social factors in explaining their adoption. Corresponding to underdetermination of theories by data, I suggest, is a thesis of “underdetermination of strategy by fruitfulness”. Perhaps the reason why there are few of these cases is that, in view of the material and social conditions required for conducting research, only strategies that promise products that, on application, may serve the practical interests of dominant or ascendent social values receive the support necessary to develop. Elsewhere I have discussed analogous cases in experimental psychology in Lacey (1980; 2001b). I am referring to the distinctive role played by control of natural objects and the way in which it is valued in modernity (“the modern valuation of control”): its scope and the value of its augmented scope, centrality in daily life, relative insubordination to other social values, and the deep sense that control is the characteristic human stance towards natural objects, so that the expansion of technologies into more and more spheres of life and their becoming the means for the solution of more and more problems are highly valued. (For detailed elaboration, see SVF: ch. 6; Lacey, in press.) Agroecological strategies may be considered as particular instances of general ecological strategies that enable us to identify the possibilities that things (e.g., seeds) have in virtue of their place in agroecological systems. Elaborating: Under agroecological strategies research aims to confirm generalizations concerning the tendencies, capacities and functioning of agroecosystems and their components, and relations among them. These include generalizations in which (e.g.) “mineral cycles, energy transformations, biological processes and socioeconomic relationships” are considered in relationship to the whole system; generalizations concerned not with “maximizing production of a particular system, but rather with optimizing the agroecosystem as a whole” and so with “complex interactions among and between people, crops, soil and livestock” (Altieri, 1987: xiv-xv). Of particular salience are generalizations that help to identify pos-

Lacey and Popper

22. 23.

24.

25.

259

sibilities connected with: “Maintenance of the productive capacity of the ecosystem; preservation of the natural resource base and functional biodiversity; social organization and reduction of poverty; empowerment of local communities, maintenance of tradition, and popular participation in the development process” (Altieri et al. 1996: 367-68). See also Lacey, 2000b; forthcoming—where it is discussed how agroecological strategies do not exclude, and indeed require, input from theories confirmed under materialist strategies of certain domains of phenomena; and how they may exhibit continuity with traditional forms of knowledge in “third world” communities. Elaborated below in the section titled, “Research and the social values it serves.” (See also SVF: ch. 8.) Agroecological theories are under-investigated and, typically, this is tantamount to their rejection. A recent study (Zhu, et al., 2000)—conducted on rice crops in Yunnan Province, China—demonstrated that “a simple, ecological approach to disease control can be used effectively at large spatial scale to attain environmentally sound disease control” without loss of productivity (compared to chemically-intensive farming based on monocultures). Commenting on this study Wolfe (2000), after observing that Darwin was aware that mixed cropping (of wheat) is more productive than monocultures, asked: “Why is the mixed approach not widely used?” He answered with the rhetorical question: “Is it just too simple, not making enough use of high technology?” He continued: “Variety mixtures may not provide all the answers to the problems of controlling diseases and producing stable yields in modern agriculture. But their performance so far in experimental situations merits their wider uptake. More research is needed to find the best packages for different purposes and to breed varieties specifically for use in mixtures. … Mixtures of species provide another layer of crop diversity, with half-forgotten advantages waiting to be exploited in contemporary approaches.” This study exemplifies the potential fruitfulness of research under agroecological strategies. In order to gain understanding of astronomical phenomena, materialist strategies are adopted. This makes clear that the adoption of these strategies is not (only, always) for the sake of exercising control over the objects of which we gain understanding. This is consistent with my thesis that adopting materialist strategies almost exclusively is best explained by reference to mutually reinforcing relationships between them and the modern valuation of control (Note 20). For the details of what is involved in these mutually reinforcing relationships, see SVF: ch. 6. The same theories that express this knowledge might also enable us to gain understanding of objects outside of these spaces (see previous Note).

260

Siblings Under the Skin

26. I call the possibilities that may be identified under materialist strategies, those that are encapsulated in the generative power of underlying structure, process, interaction and law in abstraction from connections with human life and values and social and ecological relations, “abstracted possibilities.” Many things also have possibilities derived from their places in human life, social relations, etc (“non-abstracted” possibilities). The possibilities of seeds in agroecological systems, e.g., are not reducible to those that may be identified in biotechnological research. Note that (VI) does not serve as a premise from which to derive (I); and since scientific strategies and not limited to materialist ones—SVF: chs 8, 9; Lacey, 2000b; 2001a; forthcoming—it follows that (IV) is false in general. 27. Notes 20 and 24 gives references to my more extensive and nuanced analysis. 28. For elaboration of the previous two paragraphs, see Lacey (2002). 33. I have made this point only implicitly in the text. (a) If T is empirically adequate of D, then there may be T' (inconsistent with T) where T' matches the extent of empirical adequacy of T of D, and where predictions derived from it may lead to the conclusion that T can no longer be judged empirically adequate of D. On the other hand, if T manifests all the cognitive values highly of D, the discovery of a T' that also manifests highly them of D, and even of some larger domain D' (which contains D), does not lead to withdrawing the judgment about T. Newton’s theory manifests the cognitive values very highly of some D (which can be specified sharply from the perspective of relativity and quantum theories), even though relativity manifests them highly of a more encompassing domain. That is why Newton’s theory still plays a practical role. T (of D) may come to manifest the cognitive values highly, and come to be properly considered part of the repository of settled knowledge, even though a more encompassing T' (of D'), inconsistent (or incommensurable) with it, may later also come to be part of settled knowledge. Underdetermination does not rule this out. (b) Invulnerability says we can maintain the empirical adequacy of T pretty much come what may by adjusting auxiliary hypotheses. Often the price of doing so is, e.g., that the cognitive value “minimize use of ad hoc hypotheses” ceases to be manifested to a high degree. Adjusting auxiliary hypotheses may suffice to preserve empirical adequacy, but not to maintain the high manifestation of all the cognitive values. Popper (1959: 82-83) makes a similar point in a somewhat different context. (c) The argument in (a) holds even when T and T' are incommensurable, articulated under competing strategies (Lacey 2001a).

Lacey and Popper

261

30. Adopting strategies that are incompatible with materialist strategies (e.g., agroecological ones) does not commit one to deny that “abstracted” possibilities are reliably encapsulated under materialist strategies; it only involves a general questioning of the value of realizing these possibilities. 31. Popper maintains that the decisions that he endorses (Theses 2 and 4) enable us to judge of a theory, “even before it has been tested,” that it is potentially progressive, because “it tells us more” (Popper, 1963: 217). This ignores that Thesis 4 imposes constraints that preclude the investigation of “non-abstracted” possibilities. No one strategy can encapsulate all the possibilities of things. 32. I wish to thank NSF (SES-9905945) for partial support in preparing this chapter.

References Altieri, M.A. 1987. Agroecology: The scientific basis of alternative agricultures. Boulder, CO: Westview. Altieri, M.A., Yurjevic, A., Von der Weid, J.M., Sanchez, J. 1996. Applying agroecology to improve peasant farming systems in Latin America, in R. Costanza, O. Segura & J. Martinez-Alier (eds), Getting down to earth: practical applications of ecological economics. Washington, DC: Island Press. Hacking, I. 1999. The social construction of what? Cambridge: Harvard University Press. Feyerabend, P.K. 1981. Problems of empiricism. Cambridge: Cambridge University Press. Kloppenburg, J. Jr. 1991. The plant germplasm controversy. Bioscience, 37: 190-198. Kuhn, T. [1962] 1970. The structure of scientific revolutions, 2nd ed. Chicago: University of Chicago Press. Lacey, H. 1980. Psychological conflict and human nature: The case of behaviorism and cognition, Journal for the Theory of Social Behavior, 10: 131-155. ——. 1994. ¿Que tipo de ciencia le sirve al desarrollo auténtico? Realidad, 39: 369-382.

262

Siblings Under the Skin

——. 1997a. Neutrality in the social sciences: On Bhaskar’s argument for an essential emancipatory impulse in the social sciences. Journal for the Theory of Social Behavior, 27: 213-241. ——. 1997b. Ciência e valores. Manuscrito, 20: 9-36. ——. 1998. Valores e atividade científica. São Paulo: Discurso Editorial. ——. 1999a. Is science value free? Values and scientific understanding. London: Routledge. ——. 1999b. Values and the conduct of science: Principles. Principia, 3: 5785. ——. 2000a. Seeds and the knowledge they embody. Peace Review, 12: 563569. ——. 2000b. Author’s response [to John Forge, Science and the ‘modern values of control’, a review of Is Science Value Free]. Metascience, 9: 334-338. ——. 2001a. Incommensurability and ‘multicultural science’. In P. Hoyningen-Huene & H. Sankey (eds), Incommensurability and Related Matters. Dordrecht: Kluwer. ——. 2001b. Psicologia Experimental e Natureza Humana: Ensaios de filosofia da psicologia. Florianópolis: NEL/Universidade Federal de Santa Caterina. ——. 2002. The ways in which the sciences are and are not value free. In P. Gardenfors, K. Kijania-Placek & J. Wolenski eds., Proceedings of the 11th International Congress of Logic, Methodology and Philosophy of Science, pp. 513-526. Dordrecht: Kluwer. ——. (forthcoming). The social location of scientific practices. In T. Rockmore & J. Margolis (eds.), The Historicity of Science. Lacey, H. & de Oliveira, M.B. 2001. Prefácio a Vandana Shiva, Biopirataria, Petrópolis: Editora Vozes. [Preface to the Portuguese translation of Biopiracy]. Longino, H.E. 1990. Science as social knowledge: Values and objectivity in scientific inquiry. Princeton: Princeton University Press. Nelson, L.H. & Nelson, J. 1995. Feminist values and cognitive virtues. In D. Hull, M. Forbes & R. M. Burian (eds) PSA 1994, volume 2. East Lansing MI: Philosophy of Science Association. Nodari, R.O. & Guerra, M.P. 2001. Avaliação de riscos ambientais de plantas transgênicas. Cadernos de Ciência e Tecnologia, 18. Okruhlik, K. 1994. Gender and the biological sciences. Canadian Journal of Philosophy, 20: 21-42. Popper, K. R. 1959. The logic of scientific discovery. New York: Harper. ——. 1963a. On the sources of knowledge and of ignorance. Conjectures and refutations: The growth of scientific knowledge. New York: Harper.

Lacey and Popper

263

——. 1963b. Science: Conjectures and refutations. In Conjectures and refutations: The growth of scientific knowledge. New York: Harper. ——. 1963c. Truth, rationality and the growth of scientific knowledge. Conjectures and refutations: The growth of scientific knowledge. New York: Harper. Rissler, J. & Mellon, M. 1996. The ecological risks of engineered crops. Cambridge: MIT Press. V. Shiva. 1991. The violence of the green revolution. London: Zed Books. Taylor, C. 1982. Rationality. In M. Hollis & S. Lukes (eds.), Rationality and relativism. Cambridge: MIT Press. Wolfe, M.S. 2000. Crop strength through diversity. Nature, 406: 681-682. Zhu, Y., et al. 2000. Genetic diversity and disease control in rice. Nature, 406: 718-722.

Biographical note Hugh Lacey (b. 1939) is descended from an Irish peasant who was transported by the British crown to New South Wales in 1835 for the crime of perjury. Hugh’s studies in History and Philosophy of Science were carried out at University of Melbourne and Indiana University (Ph.D., 1966). He has held positions at the University of Sydney and the University of São Paulo, and remains a regular visitor to the latter, from which his wife (Maria Ines Lacey) holds a Ph.D. Since 1972 he has taught at Swarthmore College—which, with good Quaker tolerance, does not discriminate against those who have genes for criminality—where he now holds the position of Scheuer Family Professor of Humanities. In recent years his writings have mainly been concerned with the ways in which values play roles (legitimate and illegitimate) in scientific practices—see Valores e Atividade Científica (São Paulo, 1998) and Is Science Value Free? Values and Scientific Understanding (London, 1999). He has often (with Brazilian collaborators) applied his conclusions to matters that arise in current controversies about the use of genetically modified organisms in agriculture, and about the role of science for grassroots organizations in South America that are struggling for emancipation from the structures of the “global” market. Hugh’s interest in Popper began with his first course in Philosophy of Science—the same year that Popper’s “The Logic of Scientific Discovery” was published (yes, Hugh is that old). Popper presented positions that Hugh

264

Siblings Under the Skin

opposed from the outset, but Hugh always found that a closer reading of Popper’s texts showed that he was aware of the deeper problems. For Hugh, Popper has been a constant point of reference: on the one hand, Hugh finds it important to test his views against Popper’s and, on the other hand, Hugh has found Popper’s texts regularly worth mining for fresh insights into old problems. It is worth mentioning that, unlike some of the other analytic philosophers represented in this collection, Popper was not known for being particularly supportive of younger colleagues with fresh perspectives on his work, or supportive of, well, anyone really. But Hugh is happy to report that while Popper is known to have hated a great many things, including cats (!) and smokers, he especially liked kids and they even liked him. So, he wasn’t all bad. And, of course, as Popper’s book The Open Society and its Enemies, makes clear, he worked throughout his life to support liberal democracy. With grumpy “enemies” like him, who needs friends?

Lacey and Popper

265

On the Sources of Knowledge and Ignorance Karl Popper

…it is impossible for us to think of any thing, which we have not antecedently felt, either by our external or internal senses. —David Hume

The title of my lecture is likely, I fear, to offend some critical ears. For although ‘Sources of Knowledge’ is in order, and ‘Sources of Error’ would have been in order too, the phrase ‘Sources of Ignorance ‘is another matter. ‘Ignorance is something negative: it is the absence of knowledge. But how on earth can the absence of anything have sources?’ This question was put to me by a friend when I confided to him the title I had chosen for the lecture. I was a little shaken by this for I had been, I confess, quite pleased with my title. Hard pressed for a reply I found myself improvising a rationalization, and explaining to my friend that the curious linguistic effect of my title was actually intended. I told him that I hoped to direct attention, through the phrasing of my title, to a number of historically important although unrecorded philosophical doctrines and among them, especially, to a conspiracy theory of ignorance which interprets ignorance not as a mere lack of knowledge but as the work of some mischievous power, the source of impure and evil influences which pervert and poison our minds and instill in us the habit of resistance to knowledge. I am not quite sure whether this explanation allayed my friend’s misgivings, but it did silence him. Your case is different since you are silenced by the rules of the present transactions. So I can only hope that I have allayed your misgivings sufficiently, for the time being, to allow me to begin my story at the other end—with the sources of knowledge rather than with the sources of ignorance. However, I shall presently come back to the sources of ignorance, and also to the conspiracy theory of these sources.

266

Siblings Under the Skin

I. The problem which I wish to examine afresh in this lecture and which I hope not only to examine but to solve, may perhaps be described as an aspect of the old quarrel between the British and the Continental schools of philosophy—the quarrel between the classical empiricism of Bacon, Locke, Berkeley, Hume, and Mill, and the classical rationalism or intellectualism of Descartes, Spinoza, and Leibniz. In this quarrel, the British school insisted that the ultimate source of all knowledge was observation, while the Continental school insisted that it was the intellectual intuition of clear and distinct ideas. Most of these issues are still very much alive. Not only has empiricism, still the ruling doctrine in England, conquered the United States, but it is now widely accepted even on the European Continent as the true theory of scientific knowledge. Cartesian intellectualism, alas, has been only too often distorted into one or another of the various forms of modern irrationalism I shall try to show in this lecture that the differences between classical empiricism and rationalism are much smaller than their similarities, and that both are mistaken. I hold that they are mistaken although I am myself both an empiricist and a rationalist of sorts. But I believe that, though observation and reason have each an important role to play, these roles hardly resemble those which their classical defenders attributed to them. More especially, I shall try to show that neither observation nor reason can be described as a source of knowledge, in the sense in which they have been claimed to be sources of knowledge, down to the present day. II. Our problem belongs to the theory of knowledge, or to epistemology, reputed to be the most abstract and remote and altogether irrelevant region of pure philosophy. Hume, for example, one of the greatest thinkers in the field, predicted that, because of the remoteness and abstractness and practical irrelevance of his results, none of his readers would believe in them for more than an hour.

Lacey and Popper

267

Kant’s attitude was different. He thought that the problem ‘What can I know?’ was one of the three most important questions a man could ask. Bertrand Russell, in spite of being closer to Hume in philosophic temperament, seems to side in this matter with Kant. And I believe that Russell is right when he attributes to epistemology practical consequences for science, for ethics, and even for politics. He points out, for example, that epistemological relativism, or the idea that there is no such thing as objective truth, and epistemological pragmatism, or the idea that truth is the same as usefulness, are closely linked with authoritarian and totalitarian ideas. Russell’s views are of course disputed. Some recent philosophers have developed a doctrine of the essential impotence and practical irrelevance of all genuine philosophy, and thus, one can assume, of epistemology. Philosophy, they say, cannot by its very nature have any significant consequences, and so it can influence neither science nor politics. But I think that ideas are dangerous and powerful things, and that even philosophers have sometimes produced ideas. Indeed, I have no doubt that this new doctrine of the impotence of all philosophy is amply refuted by the facts. The situation is really very simple. The belief of a liberal—the belief in the possibility of a rule of law, of equal justice, of fundamental rights, and a free society—can easily survive the recognition that judges are not omniscient and may make mistakes about facts, and that, in practice, absolute justice is hardly ever realized in any particular legal case. But his belief in the possibility of a rule of law, of justice, and of freedom, can hardly survive the acceptance of an epistemology which teaches that there are no objective facts; not merely in this particular case, but in any other case; and that the judge cannot have made a factual mistake because he can no more be wrong about the facts than he can be right. III. The great movement of liberation which started in the Renaissance and led through the many vicissitudes of the reformation and the religious and revolutionary wars to the free societies in which the English-speak-

268

Siblings Under the Skin

ing peoples are privileged to live, this movement was inspired throughout by an unparalleled epistemological optimism: by a most optimistic view of man’s power to discern truth and to acquire knowledge. At the heart of this new optimistic view of the possibility of knowledge lies the doctrine that truth is manifest. Truth may perhaps be veiled, and removing the veil may not be easy. But once the naked truth stands revealed before our eyes, we have the power to see it, to distinguish it from falsehood, and to know that it is truth. The birth of modern science and modern technology was inspired by this optimistic epistemology whose main spokesmen were Bacon and Descartes. They taught that there was no need for any man to appeal to authority in matters of truth because each man carried the sources of knowledge in himself; either in his power of sense-perception which he may use for the careful observation of nature, or in his power of intellectual intuition which he may use to distinguish truth from falsehood by refusing to accept any idea which is not clearly and distinctly perceived by the intellect. Man can know: thus he can be free. This is the formula which explains the link between epistemological optimism and the ideas of liberalism. This link is paralleled by the opposite link. Disbelief in the power of human reason, in man’s power to discern the truth, is almost invariably linked with distrust of man. Thus epistemological pessimism is linked historically, with a doctrine of human depravity, and it tends to lead to the demand for the establishment of powerful traditions and the entrenchment of a powerful authority which would save man from his folly and wickedness. (There is a striking sketch of this theory of authoritarianism, and a picture of the burden carried by those in authority, in the story of The Grand Inquisitor in Dostoevsky’s Brother’s Karamazov.) The contrast between epistemological pessimism and optimism may be said to be fundamentally the same as that between epistemological traditionalism and rationalism, using this latter term in the wider sense in which it is opposed to irrationalism, and in which it thus covers not only Cartesian intellectualism but empiricism also. For we can interpret traditionalism as the belief that, in the absence of an objective and discernible truth, we are faced with the choice between accepting

Lacey and Popper

269

the authority of tradition, and chaos; while rationalism has, of course, always claimed the right of reason and of empirical science to criticize, and to reject, any tradition, and any authority, as being based on sheer unreason or prejudice or accident. IV. It is a disturbing fact that even an abstract study like pure epistemology is not as pure as one might think (and as Aristotle believed) but that its ideas may, to a large extent, be motivated and unconsciously inspired by political hopes and Utopian dreams. This should be a warning to the epistemologist. What can he do about it? As an epistemologist I have only one interest—to find out the truth about the problems of epistemology, whether or not this truth fits in with my political ideas. But am I not liable to be influenced, unconsciously, by my political hopes and beliefs? It so happens that I am not only an empiricist and a rationalist of sorts but also a liberal; but just because I am a liberal, I feel that few things are more important for a liberal that to submit the various theories of liberalism to the most searching and critical examination. While I was engaged in a critical examination of this kind I discovered the part played by certain epistemological theories in the development of liberal ideas; and especially by the various forms of epistemological optimism. And I found that, as an epistemologist, I had to reject these epistemological theories as untenable. This experience of mine may illustrate the point that our dreams and our hopes need not necessarily control our results, and that, in searching for the truth, it may be our best plan to start by criticizing our most cherished beliefs. This may seem to be a perverse plan. But it will not seem so to those who want to find the truth and are not afraid of it. Editor’s note: Sections V through XII of Popper’s essay contain an historical survey that, in the interest of space, has been omitted.

270

Siblings Under the Skin

XIII. I will now leave all these largely historical reflections aside, and turn to the problems themselves, and to their solution. This part of my lecture might be described as a criticism of empiricism, as formulated for example in the following classical statement of Hume’s: ‘If I ask you why you believe any particular matter of fact, … you must tell me some reason; and this reason will be some other fact, connected with it. But as you cannot proceed after this manner, in infinitum, you must at last terminate in some fact, which is present to your memory or senses; or must allow that your belief is entirely without foundation.’ (Enquiry Concerning Human Understanding, Section v, Pt. I, edited by Selby-Bigge, p. 46; see also my motto, taken from Section vii, Pt. I, p.62.) The problem of the validity of empiricism may be roughly put as follows: is observation the ultimate source of our knowledge of nature? And if not, what are the sources of our knowledge? These questions remain, whatever I may have said about Bacon, and even if I should have managed to make those parts of his philosophy on which I have commented somewhat unattractive for Baconians and for other empiricists. The problem of the source of our knowledge has recently been restated as follows. If we make an assertion, we must justify it; but this means that we must be able to answer the following questions. ‘How do you know? What are the sources of your assertion?’ This, the empiricist holds, amounts in its turn to the question, ‘What observations (or memories of observations) underlie your assertion?’ I find this string of questions quite unsatisfactory. First of all, most of our assertions are not based upon observations, but upon all kinds of other sources. ‘I read it in The Times’ or perhaps, ‘I read it in the Encyclopaedia Britannica’ is a more likely and a more definite answer to the question ‘How do you know?’ than ‘I have observed it’ or ‘I know from an observation I made last year’. ‘But’, the empiricist will reply, ‘how do you think that The Times or the Encyclopaedia Britannica got their information? Surely, if you only carry on your inquiry long enough, you will end up with reports of the observations of eyewitnesses (sometimes called “protocol sentences”

Lacey and Popper

271

or—by yourself—”basic statements”). Admittedly’, the empiricist will continue, ‘books are largely made from other books. Admittedly, a historian, for example, will work from documents. But ultimately, in the last instance, these other books, or these documents, must have been based upon observations. Otherwise they would have to be described as poetry, or invention, or lies, but not as knowledge. It is in this sense that we empiricists assert that observation must be the ultimate source of our knowledge.’ Here we have the empiricist’s case, as it is still put by some of my positivist friends. I shall try to show that this case is as little valid as Bacon’s; that the answer to the question of the sources of knowledge goes against the empiricist; and, finally, that this whole question of ultimate sources—sources to which one may appeal, as one might to a higher court, or a higher authority—must be rejected as based upon a mistake. First I want to show that if you actually went on questioning The Times and its correspondents about the sources of their knowledge, you would in fact never arrive at all those observations of eyewitnesses in the existence of which the empiricist believes. You would find, rather, that with every single step you take, the need for further steps increases in snowball-like fashion. Take as an example the sort of assertion for which reasonable people might simply accept as sufficient the answer ‘I read it in The Times’; let us say the assertion ‘The Prime Minister has decided to return to London several days ahead of schedule’. Now assume for a moment that somebody doubts this assertion, or feels the need to investigate into its truth. What shall he do? If he has a friend in the Prime Minister’s office, the simplest and most direct way would be to ring him up; and if this friend corroborated the message, then that is that. In other words, the investigator will, if possible, try to check, or to examine, the asserted fact itself, rather than trace the source of the information. But according to the empiricist theory, the assertion ‘I have read it in The Times’ is merely a first step in a justification procedure consisting in tracing the ultimate source. What is the next step? There are at least two next steps. One would be to reflect that ‘I have read it in The Times’ is also an assertion, and that we might ask ‘What is the source of your knowledge that you read it in The Times and

272

Siblings Under the Skin

not, say, in a paper looking very similar to The Times?’ The other is to ask The Times for the sources of its knowledge. The answer to the first question may be ‘But we have only The Times on order and we always get it in the morning’ which gives rise to a host of further questions about sources which we shall not pursue. The second question may elicit from the editor of The Times the answer: ‘We had a telephone call from the Prime Minister’s office.’ Now according to the empiricist procedure, we should at this stage ask next: ‘Who is the gentleman who received the telephone call?’ and then get his observation report; but we should also have to ask that gentleman: ‘What is the source of your knowledge that the voice you heard came from an official in the Prime Minister’s office’, and so on. There is a simple reason why this tedious sequence of questions never comes to a satisfactory conclusion. It is this. Every witness must always make ample use, in his report, of his knowledge of persons, places, things, linguistic usages, social conventions, and so on. He cannot rely merely upon his eyes or ears, especially if his report is to be of use in justifying any assertion worth justifying. But this fact must of course always raise new questions as to the sources of those elements of his knowledge which are not immediately observational. This is why the program of tracing back all knowledge to its ultimate source in observation is logically impossible to carry through: it leads to an infinite regress. (The doctrine that truth is manifest cuts off the regress. This is interesting because it may help to explain the attractiveness of that doctrine.) I wish to mention, in parenthesis, that this argument is closely related to another—that all observation involves interpretation in the light of our theoretical knowledge, or that pure observational knowledge, unadulterated by theory, would, if at all possible, be utterly barren and futile. The most striking thing about the observationalist program of asking for sources—apart from its tediousness—is its stark violation of common sense. For if we are doubtful about an assertion, then the normal procedure is to test it, rather than to ask for its sources; and if we find independent corroboration, then we shall often accept the assertion without bothering at all about sources.

Lacey and Popper

273

Of course there are cases in which the situation is different. Testing an historical assertion always means going back to sources; but not, as a rule, to the reports of eyewitnesses. Clearly, no historian will accept the evidence of documents uncritically. There are problems of genuineness, there are problems of bias, and there are also such problems as the reconstruction of earlier sources. There are, of course, also problems such as: was the writer present when these events happened? But this is not one of the characteristic problems of the historian. He may worry about the reliability of a report, but he will rarely worry about whether or not the writer of a document was an eyewitness of the event in question, even assuming that this event was of the nature of an observable event. A letter saying ‘I changed my mind yesterday on this question’ may be most valuable historical evidence, even though changes of mind are unobservable (and even though we may conjecture, in view of other evidence, that the writer was lying). As to eyewitnesses, they are important almost exclusively in a court of law where they can be cross-examined. As most lawyers know, eyewitnesses often err. This has been experimentally investigated, with the most striking results. Witnesses most anxious to describe an event as it happened are liable to make scores of mistakes, especially if some exciting things happen in a hurry; and if an event suggests some tempting interpretation, then this interpretation, more often than not, is allowed to distort what has actually been seen. Hume’s view of historical knowledge was different: ‘... we believe’, he writes in the Treatise (Book I, Pt. III, sect. iv; SelbyBigge, p. 83), ‘that Caesar was kill’d in the senate-house on the ides of March... because this fact is establish’d on the unanimous testimony of historians, who agree to assign this precise time and place to that event. Here are certain characters and letters present either to our memory or senses; which characters we likewise remember to have been us’d as the signs of certain ideas; and these ideas were either in the minds of such as were immediately present at that action, and receiv’d the ideas directly from its existence; or they were deriv’d from the testimony of others, and that again from another testimony... ‘till we arrive at those who were eye-witnesses and spectators of the event.’ (See also Enquiry, Section x; Selby-Bigge, pp. iii ff.) It seems to me clear that this view must lead to the infinite regress described above. For the problem is, of course,

274

Siblings Under the Skin

whether ‘the unanimous testimony of historians’ is to be accepted, or whether it is, perhaps, to be rejected as the result of their reliance on a common yet spurious source. The appeal to ‘letters present to our memory or our senses’ cannot have any bearing on this or on any other relevant problem of historiography. XIV. But what, then, are the sources of our knowledge? The answer, I think, is this: there are all kinds of sources of our knowledge; but none has authority. We may say that The Times can be a source of knowledge, or the Encyclopaedia Britannica. We may say that certain papers in the Physical Review about a problem in physics have more authority, and are more of the character of a source, than an article about the same problem in The Times or the Encyclopaedia. But it would be quite wrong to say that the source of the article in the Physical Review must have been wholly, or even partly, observation. The source may well be the discovery of an inconsistency in another paper, or say, the discovery of the fact that a hypothesis proposed in another paper could be tested by such and such an experiment; all these non-observational discoveries are ‘sources’ in the sense that they all add to our knowledge. I do not, of course, deny that an experiment may also add to our knowledge, and in a most important manner. But it is not a source in any ultimate sense. It has always to be checked: as in the example of the news in The Times we do not, as a rule, question the eyewitness of an experiment, but, if we doubt the result, we may repeat the experiment, or ask somebody else to repeat it. The fundamental mistake made by the philosophical theory of the ultimate sources of our knowledge is that it does not distinguish clearly enough between questions of origin and questions of validity. Admittedly, in the case of historiography, these two questions may sometimes coincide. The question of the validity of an historical assertion may be testable only, or mainly, in the light of the origin of certain sources. But in general the two questions are different; and in general we do not test the validity of an assertion or information by tracing its sources or its

Lacey and Popper

275

origin, but we test them, much more directly, by a critical examination of what has been asserted—of the asserted facts themselves. Thus the empiricist’s questions ‘How do you know? What is the source of your assertion?’ are wrongly put. They are not formulated in an inexact or slovenly manner, but they are entirely misconceived: they are questions that beg for an authoritarian answer. XV. The traditional systems of epistemology may be said to result from yesanswers or no-answers to the questions about the sources of knowledge. They never challenge these questions, or dispute their legitimacy: the questions are taken as perfectly natural, and nobody seems to see any harm in them. This is quite interesting, for these questions are clearly authoritarian in spirit. They can be compared with that traditional question of political theory, ‘Who should rule?’, which begs for an authoritarian answer such as ‘the best’, or ‘the wisest’, or ‘the people’, or ‘the majority’. (It suggests, incidentally, such silly alternatives as ‘Whom do you want as rulers: the capitalists or the workers?’, analogous to ‘What is the ultimate source of knowledge: the intellect or the senses?’) This political question is wrongly put and the answers which it elicits are paradoxical (as I have tried to show in Chapter7 of my Open Society). It should be replaced by a completely different question such as ‘How can we organize our political institutions so that bad or incompetent rulers (whom we should try not to get, but whom we so easily might get all the same) cannot do too much damage?’ I believe that only by changing our question in this way can we hope to proceed towards a reasonable theory of political institutions. The question about the sources of our knowledge can be replaced in a similar way. It has always been asked in the spirit of: ‘What are the best sources of our knowledge—the most reliable ones, those which will not lead us into error, and those to which we can and must turn, in case of doubt, as the last court of appeal?’ I propose to assume, instead, that no such ideal sources exist—no more than ideal rulers—and that all ‘sources’ are liable to lead us into error at times. And I propose to

276

Siblings Under the Skin

replace, therefore, the question of the sources of our knowledge by the entirely different question: ‘How can we hope to detect and eliminate error?’ The question of the sources of our knowledge, like so many authoritarian questions, is a genetic one. It asks for the origin of our knowledge, in the belief that knowledge may legitimize itself by its pedigree. The nobility of the racially pure knowledge, the untainted knowledge, the knowledge which derives from the highest authority, if possible from God: these are the (often unconscious) metaphysical ideas behind the question. My modified question, ‘How can we hope to detect error?’ may be said to derive from the view that such pure, untainted and certain sources do not exist, and that questions of origin or of purity should not be confounded with questions of validity, or of truth. This view may be said to be as old as Xenophanes. Xenophanes knew that our knowledge is guesswork, opinion—doxa rather than epistēmē—as shown by his verses (DK, B, 18 and 34): The gods did not reveal, from the beginning, All things to us; but in the course of time, Through seeking, men find that which is the better. But as for certain truth, no man has known it, Nor will he know it; neither of the gods, Nor yet of all the things of which I speak. And even if by chance he were to utter Finality, he would himself not know it; For all is but a woven web of guesses. Yet the traditional question of the authoritative sources of knowledge is repeated even today—and very often by positivists, and by other philosophers who believe themselves to be in revolt against authority. The proper answer to my question ‘How can we hope to detect and eliminate error?’ is, I believe, ‘By criticizing the theories or guesses of others and—if we can train ourselves to do so—by criticizing our own theories or guesses.’ (The latter point is highly desirable, but not indispensable; for if we fail to criticize our own theories, there may be others to do it for us.) This answer sums up a position which I propose to call

Lacey and Popper

277

‘critical rationalism’. It is a view, an attitude, and a tradition, which we owe to the Greeks. It is very different from the ‘rationalism’ or ‘intellectualism’ of Descartes and his school, and very different even from the epistemology of Kant. Yet in the field of ethics, of moral knowledge, it was approached by Kant with his principle of autonomy. This principle expresses his realization that we must not accept the command of an authority, however exalted, as the basis of ethics. For whenever we are faced with a command by an authority, it is for us to judge, critically, whether it is moral or immoral to obey. The authority may have power to enforce its commands, and we may be powerless to resist. But if we have the physical power of choice, then the ultimate responsibility remains with us. It is our own critical decision whether to obey a command; whether to submit to an authority. Kant boldly carried this idea into the field of religion: ‘...in whatever way,’ he writes, ‘the Deity should be made known to you—even ... if He should reveal Himself to you: it is you … who must judge whether you are permitted to believe in Him, and to worship Him.’ (Religion Within the Limits of Pure Reason, 2nd edition, Chapter iv, Pt. 2, § i, the first footnote.) In view of this bold statement, it seems strange that Kant did not adopt the same attitude—that of critical examination, of the critical search for error—in the field of science. I feel certain that it was only his acceptance of the authority of Newton’s cosmology—a result of its almost unbelievable success in passing the most severe tests—which prevented Kant from doing so. If this interpretation of Kant is correct, then the critical rationalism (and also the critical empiricism) which I advocate merely puts the finishing touch to Kant’s own critical philosophy. And this was made possible by Einstein, who taught us that Newton’s theory may well be mistaken in spite of its incredible success. So my answer to the question ‘How do you know? What is the source or the basis of your assertion? What observations have led you to it?’ would be: ‘I do not know: my assertion was merely a guess. Never mind the source, or the sources, from which it may spring— there are many possible sources, and I may not be aware of half of them; and origins or pedigrees have in any case little bearing upon truth. But if you are interested in the problem which I tried to solve

278

Siblings Under the Skin

by my tentative assertion, you may help me by criticising it as severely as you can; and if you can design some experimental test which you think might refute my assertion, I shall gladly, and to the best of my powers, help you to refute it.’ This answer1 applies, strictly speaking, only if the question is asked about some scientific assertion as distinct from an historical one. If my conjecture was an historical one, sources (in the non-ultimate sense) will of course come into the critical discussion of its validity. Yet fundamentally, my answer will be the same, as we have seen. XVI. It is high time now, I think, to formulate the epistemological results of this discussion. I will put them in the form of nine theses. 1. There are no ultimate sources of knowledge. Every source, every suggestion, is welcome; and every source, every suggestion, is open to critical examination. Except in history, we usually examine the facts themselves rather than the sources of our information. 2. The proper epistemological question is not one about sources; rather, we ask whether the assertion made is true—that is to say, whether it agrees with the facts. (That we can work, without getting involved in antinomies, with a notion of objective truth in the sense of correspondence to the facts, has been shown by the work of Alfred Tarski.) And we try to find this out, as well as we can, by examining or testing the assertion itself; either in a direct way, or by examining or testing its consequences. 3. In connection with this examination, all kinds of arguments may be relevant. A typical procedure is to examine whether our theories are consistent with our observations. But we may also examine, for example, whether our historical sources are internally consistent. 4. Quantitatively and qualitatively by far the most important source of our knowledge—apart from inborn knowledge—is tradition. Most things we know we have learned by example, by being told, by reading books, by learning how to criticize, how to take and to accept criticism, how to respect truth.

Lacey and Popper

279

5. The fact that most of the sources of our knowledge are traditional condemns anti-traditionalism as futile. But this fact must not be held to support a traditionalist attitude: every bit of our traditional knowledge (and even our inborn knowledge) is open to critical examination and may be overthrown. Nevertheless, without tradition, knowledge would be impossible. 6. Knowledge cannot start from nothing—from a tabula rasa—nor yet from observation. The advance of knowledge consists, mainly, in the modification of earlier knowledge. Although we may sometimes, for example in archaeology, advance through a chance observation, the significance of the discovery will usually depend upon its power to modify our earlier theories. 7. Pessimistic and optimistic epistemologies are about equally mistaken. The pessimistic cave story of Plato is the true one, and not his optimistic story of anamnesis (even though we should admit that all men, like all other animals, and even all plants, possess inborn knowledge). But although the world of appearances is indeed a world of mere shadows on the walls of our cave, we all constantly reach out beyond it; and although, as Democritus said, the truth is hidden in the deep, we can probe into the deep. There is no criterion of truth at our disposal, and this fact supports pessimism. But we do possess criteria which, if we are lucky, often allow us to recognize error and falsity. Clarity and distinctness are not criteria of truth, but such things as obscurity and confusion indicate error. Similarly, coherence does not establish truth, but incoherence and inconsistency establish falsehood. And, when they are recognized, our own errors provide the dim red lights which help us in feeling our way out of the dark of our cave. 8. Neither observation nor reason are authorities. Intellectual intuition and imagination are most important, but they are not reliable: they may show us things very clearly, and yet they may mislead us. They are indispensable as the main sources of our theories; but most of our theories are false anyway. The most important function of observation and reasoning, and even of intuition and imagination, is to help us in the critical examination of those bold conjectures which are the means by which we probe into the unknown.

280

Siblings Under the Skin

9. Every solution of a problem raises new unsolved problems; the more so the deeper the original problem and the bolder its solution. The more we learn about the world, and the deeper our learning, the more conscious, specific, and articulate will be our knowledge of what we do not know, our knowledge of our ignorance. For this, indeed, is the true source of our ignorance—the fact that our knowledge can only be finite, while our ignorance must necessarily be infinite. We may get a glimpse of the vastness of our ignorance when we contemplate the vastness of the heavens: though the mere size of the universe is not the deepest cause of our ignorance, it is one of its causes. ‘Where I seem to differ from some of my friends’, F. P. Ramsey wrote in a charming passage of his Foundations of Mathematics (p. 291), ‘is in attaching little importance to physical size. I don’t feel in the least humble before the vastness of the heavens.’ The stars may be large but they cannot think or love; and these are qualities which impress me far more than size does. I take no credit for weighing nearly seventeen stones. I suspect that Ramsey’s friends would have agreed with him about the insignificance of sheer physical size; and I suspect that if they felt humble before the vastness of the heavens, this was because they saw in it a symbol of their ignorance. I believe that it would be worth trying to learn something about the world even if we merely learnt that we do not know much. This state of learned ignorance might be a help in many of our troubles. It might be well to remember that, while differing widely in the various little bits we know, in our infinite ignorance we are all equal. XVII. There is a last question I wish to raise. If only we look for it we can often find a true idea, worthy of being preserved, in a philosophical theory which we must reject as false. Can we find an idea like this in one of the theories of the ultimate sources of our knowledge? I believe we can; and I suggest that it is one of the two main ideas which underlie the doctrine that the source of all our knowledge is super-natural. The first of these ideas is false, I believe, while the second is true.

Lacey and Popper

281

The first, the false idea, is that we must justify our knowledge, or our theories, by positive reasons, that is, by reasons capable of establishing them, or at least of making them highly probable; at any rate, by better reasons than that they have withstood criticism. This idea implies, as I suggested, that we must appeal to some ultimate or authoritative source of true knowledge; which still leaves open the character of that authority—whether it is human, like observation or reason, or super-human (and therefore super-natural). The second idea—whose vital importance has been stressed by Russell—is that no man’s authority can establish truth by decree; that we should submit to truth; that truth is above human authority. Taken together these two ideas almost immediately yield the conclusion that the sources from which our knowledge derives must be super-human; a conclusion which tends to encourage self-righteousness and the use of force against those who refuse to see the divine truth. Some who rightly reject this conclusion do not, unhappily, reject the first idea—the belief in the existence of ultimate sources of knowledge. Instead they reject the second idea—the thesis that truth is above human authority. They thereby endanger the idea of the objectivity of knowledge, and of common standards of criticism or rationality. What we should do, I suggest, is to give up the idea of ultimate sources of knowledge, and admit that all knowledge is human; that it is mixed with our errors, our prejudices, our dreams, and our hopes; that all we can do is to grope for truth even though it be beyond our reach. We may admit that our groping is often inspired, but we must be on our guard against the belief however deeply felt, that our inspiration carries any authority, divine or otherwise. If we thus admit that there is no authority beyond the reach of criticism to be found within the whole province of our knowledge, however far it may have penetrated into the unknown, then we can retain, without danger, the idea that truth is beyond human authority. And we must retain it; for without this idea there can be no objective standards of inquiry; no criticism of our conjectures; no groping for the unknown; no quest for knowledge.

282

Siblings Under the Skin

Notes 1. This answer, and almost the whole of the contents of the present section XV, are taken with only minor changes from a paper of mine which was first published in The Indian Journal of Philosophy, i, No. I, 1959.

Biographical note Of the many 20th century philosophers of science, Karl R. Popper (19021994) has received the most acceptance among professional scientists. Born in Vienna, later a refugee from nazism, he then worked at the London School of Economics from 1946, the year in which he received British citizenship, until his retirement in 1969. His most important work is The Logic of Scientific Discovery (1959; shorter version in German in 1934), in which he developed his famous thesis that what demarcates science from other disciplines is that scientific theories are falsifiable, a thesis which he attempted to show is compatible with scientific realism and the growth of scientific knowledge. The ideas are developed in numerous books, of which Conjectures and Refutations (1963) and Objective Knowledge (1972) are the most well known. His book of political philosophy, The Open Society and Its Enemies (1945), which defends liberal democracy and, according to some interpretations, also economic liberalism, against all forms of totalitarianism (any society that does not nurture criticism), has been widely influential among intellectuals in the countries of the former Soviet block. He was knighted by the British crown in 1965 for services to philosophy

Douglas and Hempel

283

Hempelian Insights for Feminism1 Heather Douglas

I. Introduction With Carl G. Hempel’s death in 1997, one of the most clearheaded and reflective analytic philosophers of the twentieth century passed from our midst. Hempel, a leading figure in the logical empiricist school, is best known for the paradox of the ravens, among other paradoxes of confirmation, and for his groundbreaking work on explanation. All of this work appears in the classic collection, Aspects of Scientific Explanation and other Essays in the Philosophy of Science, published in 1965. Often referred to simply as “Aspects,” this collection also contains a little discussed gem, “Science and Human Values.” So ignored is “Science and Human Values” that a recent collection of essays in honor of Hempel and his work, Science, Explanation and Rationality: Aspects of the Philosophy of Carl G. Hempel (Fetzer 2000), doesn’t mention the piece once. In this essay, I argue that “Science and Human Values” holds great insight for feminist philosophers of science, and indeed for anyone who is interested in the question of values in science. First written in 19602, “Science and Human Values” was Hempel’s response to a debate over the role of values in science that had been ongoing in the 1950s. Below, I summarize that debate, review Hempel’s synthesis, and then turn to its implications for current critiques of science. II. Values in science: 1950s Even as debates continued over the nature of confirmation in science in the late 1940s, several philosophers began to ask under what conditions a scientist ought to accept a hypothesis, not simply to confirm it. While the degree of confirmation was still important, philosophers like C. West Churchman argued that it was insufficient for deciding whether to accept or reject a hypothesis: “But there would

284

Siblings Under the Skin

be cases where we would not want to accept an hypothesis even though the evidence gives a high d.c. [degree of confirmation] score, because we are fearful of the consequences of a wrong decisio.” (Churchman 1948, p. 256). Churchman argued that one must consider the ends to which one will use a hypothesis in order to make a full evaluation of the hypothesis. Because there are a range of ends to which one can use scientific hypotheses, including uses outside of science proper, Churchman claimed that “the complete analysis of the methods of scientific inference shows that the theory of inference in science demands the use of ethical judgments” (Churchman, p. 265) We must have “ethical criteria of adequacy” in order to find some theory or hypothesis is adequately supported. An essay published five years later by Richard Rudner developed Churchman’s argument further. In the essay “The Scientist Qua Scientist Makes Value Judgments,” Rudner argued that accepting or rejecting hypotheses was an essential component of what scientists did, and determining what level of evidence is sufficient for acceptance (or, conversely, rejection), should be made in part by “how serious a mistake might be” (Rudner 1953, p. 2). According to Rudner, determining the seriousness of a mistake is (in part) a value judgment, and thus value judgments play an essential role in science. Anticipating a possible avenue of criticism, Rudner also considered the position that the scientist does not reject or accept hypotheses, that the scientist merely assigns probabilities to hypotheses and leaves the acceptance or rejection to their audience. Rudner argued that even if this were the case, the scientist would have to reject or accept the probability attached to the hypothesis (p. 4). The scientist must say something, whether outright acceptance/rejection or a probability judgment. Either way, the scientist must decide that there is sufficient warrant for their stand, a decision that requires consideration of values to weigh the potential consequences of error. Several philosophers responded to this challenge to the traditional idea of science as “value-free.” Richard Jeffrey, for example, argued that scientists are removed sufficiently (or should be so removed) from the arena of public decision-making so that they do not accept or reject hypotheses to be used as the basis for action. Instead scientists assign probabilities to hypotheses, and turn the hypotheses and their assigned

Douglas and Hempel

285

likelihoods over to the public. As noted above, Rudner had already addressed this possible argument, and Jeffrey acknowledged this, stating “Rudner’s objection must be included as one of the weightiest [for the]...probabilistic view of science” (Jeffrey 1956, p. 246). Despite his acknowledgment of the problem, Jeffrey provided no direct response to Rudner’s objection. In order to exclude values from science, he insisted that scientists do no accepting or rejecting, holding out hope that a probabilistic view of scientific reasoning will show that accepting or rejecting hypotheses are unnecessary to science, even while in his own examples, the probabilities assigned are assumed to be reliable and “definite”(e.g., Jeffrey, p. 241). In addition to his push for a probabilistic view of science, Jeffrey attempted to show that considering values in the acceptance or rejection of scientific hypotheses is absurd. To do this, he argued that in general the acceptance or rejection of scientific hypotheses involves such a complex mixture of potential consequences from mistakes that it is unreasonable to expect scientists to consider them all: “it is certainly meaningless to speak of the cost of mistaken acceptance or rejection, for by its nature a putative scientific law will be relevant in a great diversity of choice situations among which the cost of a mistake will vary greatly” (p. 242, italics his). Thus it is useless to demand that scientists consider those costs. At the end of his essay, he argued that the acceptance or rejection of hypotheses (and the consideration of values involved) can even be harmful: “If the scientists is to maximize good he should refrain from accepting or rejecting hypotheses, since he cannot possibly do so in such a way as to optimize every decision which may be made on the basis of those hypotheses” (p. 245). Scientific pronouncements can be used in so many contexts, Jeffrey suggested, that the values used to accept a hypothesis in one context could be quite damaging when applied to another. Rather than involve the scientist in particular decision-making contexts where their work is used, and adjust the values accordingly, Jeffrey’s response to this complexity was to sever scientists from the context in which their hypotheses are used, and to remove them from considering the consequences of potential error altogether. The removal of scientists from the context in which their work has practical relevance was a powerful move to keep science value-free and was used in another line of argumentation against the Rudner/

286

Siblings Under the Skin

Churchman challenge. Instead of relying on not yet developed probabilistic reasoning methods that could eliminate acceptance or rejection of hypotheses in science, one could accept the idea that values are needed in judgments on the acceptance or rejection of hypotheses but limit the scope of those values. Churchman, in his 1956 defense of Rudner’s essay, opened this avenue of debate when he suggested that many of the decisions made by scientists on whether to accept or reject hypotheses could be justified in terms of values internal to the scientific process and its aims. Scientists whose work had no clear practical implications would want to make their decisions considering such things as “the relative worth of (1) more observations, (2) greater scope of his conceptual model, (3) simplicity, (4) precision of language, (5) accuracy of the probability assignment” (Churchman 1956, p. 248). But Churchman did not think such “epistemic virtues” (as philosophers came to call them) were necessarily the only considerations for all scientists. This was precisely the line argued by Isaac Levi, however. Levi argued that scientists should utilize only what we now term “epistemic” values in their judgments of whether there is sufficient evidence for accepting a hypothesis. Levi believed epistemic values (simplicity, explanatory power, etc.) are sufficient for setting decision criteria for scientists and that scientists should not go beyond those values (Levi 1962, p. 49.) In many respects, this became the standard position for the next thirty years. III. Hempel’s synthesis In “Science and Human Values,” Hempel attempts to resolve the tensions between the Rudner/Churchman position and the Jeffrey/Levi position.3 Much of the essay is devoted to the influence science can have on the values that we hold and the decisions we make.4 In section 6 of the essay, Hempel turns to the issue of what influence values should have on science, or “whether scientific knowledge and method presuppose valuation” (Hempel 1965, p. 90). Hempel first rapidly narrows the scope of the question. He accepts the role of values in both the selection of questions to pursue and the choice of a career by a scientist. He denies that values themselves could confirm scientific knowledge. However, the question of whether values play a role in the scientific method, or in the doing of science, is a more complex issue.

Douglas and Hempel

287

Following Churchman, Hempel separates the issue of confirmation from acceptance, recognizing that values have no direct role to play in the weighing of evidence for a hypothesis. (Hempel, p. 91). In line with Rudner’s argument, Hempel then recognizes the need for values in the acceptance or rejection of hypotheses. (Unlike Jeffrey, Hempel sees the acceptance or rejection of hypotheses as an important part of science.) The need for values in science arises from the “inductive risk” inherent in science: “Such acceptance [of a scientific law] carries with it the ‘inductive risk’ that the presumptive law may not hold in full generality, and that future evidence may lead scientists to modify or abandon it” (Hempel, p. 92). Because the chance of error is always present, decision rules for the acceptance of scientific hypotheses are needed. And, “the formulation of ‘adequate’ decision rules requires... specification of valuations that can then serve as standards of adequacy” (p. 92). Adequate decision rules would consider the possible outcomes of accepting or rejecting a hypotheses, and these outcomes include the following four main possibilities: 1) the hypothesis is accepted as true and is, in fact, true; 2) the hypothesis is rejected as false and is, in fact, false; 3) the hypothesis is accepted as true but really is false; and 4) the hypothesis is rejected as false but really is true (ibid.). The problem of inductive risk lies in the latter two possibilities, where values must be used to decide the seriousness of error in those cases. The issue then becomes what kind of values would be needed to set up adequate rules of acceptance. For Hempel, the values depend on the case. For some situations, non-epistemic values would be required, particularly for the use of science in industrial quality control: “In the cases where the hypothesis under test, if accepted, is to be made the basis of a specific course of action, the possible outcomes may lead to success or failure of the intended practical application; in these cases, the values and disvalues at stake may well be expressible in terms of monetary gains and losses.” (p. 93). Thus, under Hempel’s model, non-epistemic values can and sometimes do play an important role in scientific judgments. However, aside from such industrial applications of science, Hempel does not perceive the full importance of non-epistemic values in science or the full implications of inductive risk. Like Levi and Jeffrey, he still sees most of science as being removed from practical implica-

288

Siblings Under the Skin

tions, i.e., “pure scientific research, where no practical applications are contemplated.” (ibid.)5 Because of this perception, he believes that generally epistemic values will suffice when making judgments about inductive risk. In later essays on this topic,6 he is primarily concerned with making more precise the nature of epistemic values, trying to nail down the epistemic guidance they give. Yet, even as Hempel wrote these essays in the 1980s, feminist philosophers were turning a critical eye toward the biological and psychological sciences that were having very practical influence on women and men. IV. Locating values in science While modern scientific claims about the nature of women’s and men’s capabilities have proliferated and influenced public policy, feminist scholars have heavily critiqued many of these claims. Whether the critiques center on the claim that sexist science is “corrupted” science or that sexist science is business-as-usual, i.e., “conventional science” for a sexist society, central to the arguments has been that science is value-laden, and that sexist values have influenced the content of science (Fausto-Sterling, pp. 8-10). More difficult has been tracking down the precise location of those values in the scientific process, i.e., how and where those values enter into science. Sometimes the influence is obvious, in the creation of blind spots and bad reasoning. Examples abound of simple errors in biological research, of scientists who conduct research on men and then assume the results hold for women, or scientists who blatantly ignore confounding factors, thus concluding that characteristics of men or women are biologically essential, or of scientists who take absurd liberties with the interpretation of evidence. In these cases, values operate in the worst possible way in science, with the desire to achieve a certain result running roughshod over good evidence and sound reasoning. But these are the easy cases. More difficult are the cases where no obvious mistakes have been made, but there still seems to be something sexist about the results. In her groundbreaking work Science as Social Knowledge, Helen Longino attempts to lay out a framework for the role of values in these more subtle cases. Some of the roles that she cites for values in science have been readily acknowledged by other philosophers, such as the influence

Douglas and Hempel

289

of societal values on the choice of questions to pursue, limitations on methodological practices (to avoid unethical use of human subjects, for example), or in the uses towards which scientific technologies are put. (Longino 1990, pp. 83-85). Others are more controversial, particularly the role of values in data selection and characterization7 and the influence of values through both specific and global background assumptions. (p. 86) It is on the influence of values through background assumptions that I will focus here.8 The difficulty with considering background assumptions the vehicle for values in science is that it is not entirely clear how some background assumptions embody values. In Longino’s terms, values “can be expressed in or motivate” background assumptions, but for some background assumptions, their value-ladenness is unclear (p. 86). Take, for example, Longino’s discussion of the background assumption of gender dimorphism, i.e. the assumption that there are two norms for gender behavior around which individuals cluster. As she states: “The assumption of gender dimorphism makes the clustering of individuals around certain behavioral poles more significant than the amount of individual variation that is as much a feature of the data as the clustering” (p. 121). Longino sees the assumption of gender dimorphism as “directly encoding contextual values” (p. 128). However, it is difficult to see precisely how the assumption encodes values. It is ostensibly an empirical or factual assumption about the nature of the world: that there are two sexes and that gender behavior characteristics clusters around those two sexes. That this assumption is not necessarily imbued with patriarchal values is acknowledged by Longino a few pages later, when she notes that some feminists heartily embrace sexual and gender dimorphism (p. 132). We may not like the results of science performed with the assumption of gender dimorphism as background belief that assists structuring research, but that does not necessarily mean that this assumption is itself the problem, the source of sexist values. So how do ostensibly empirical or factual statements encode values and allow their entry into science? It is precisely to this problem that Hempel’s concept of inductive risk can provide insight, thus expanding and clarifying Longino’s framework.

290

Siblings Under the Skin

V. Hempelian insights for feminist critiques Instead of the somewhat vague understanding that apparently factual statements acting as background assumptions encode values, we can take a more direct approach by examining the values one must consider in deciding whether to accept certain background assumptions for use in scientific research. We can ask what the inductive risks are of accepting or rejecting the assumptions and what the inductive risks are for accepting or rejecting the theories or findings based on those assumptions. Following this line of thought, it is easy to identify clear and unambiguous sexism in the use of some background assumptions. I will first discuss some findings based on the assumption of gender dimorphism and then turn to the assumption itself. Consider the following example of scientific findings based on the assumption of gender dimorphism. In 1980, Drs. Camilla Benbow and Julian Stanley published an article entitled “Sex Differences in Mathematical Ability: Fact or Artifact?” in the journal Science, with much accompanying fanfare and press coverage, claiming to have found good evidence that men are inherently better at math than women (Benbow and Stanley 1980, Kolata 1980, Fausto-Sterling 1985 pp. 53-59, Longino 1990, pp. 124-125). The study tested girls and boys from seventh and eighth grade who were thought to be unusually talented at math and found that the boys did better than the girls on average at advanced math tests (the math SATs). The scientists presented their findings as generally reliable and strongly supportive of the hypothesis of “that sex differences in achievement in and attitude toward mathematics result from superior male mathematical ability” (Benbow and Stanley 1980, p. 1264). While Benbow and Stanley recognize that other factors influence mathematical achievement, they think they have managed to isolate mathematical aptitude or ability, a quality they see as essentially innate and strongly linked to gender/sex (no differentiation is made between gender and sex in their work). Thus their work is based on the assumption of gender dimorphism, i.e., that there are two different gender based norms that we should expect for human behavior. What are the inductive risks of accepting this science as reliable? This work, particularly with the heavy press coverage accompany-

Douglas and Hempel

291

ing it, did have a powerful impact on public beliefs about math and gender. For example, Science published an accompanying article covering the Benbow and Stanley piece entitled “Math and Sex: Are Girls Born with Less Ability?” In this piece, Stanley states that their work is being deliberately made prominently public: “We want our data out in the public domain so that they can’t be ignored” (Kolata 1980, p. 1234). Let us take the position of the scientists and consider what the potential impacts could be of presenting the work so strongly and of fostering press coverage. Most obviously, with the publication of this work (and other similar studies) there is science to back up the idea that boys and girls should be separated at some point in math education, or at least that different resources and challenges should be given to the different genders. Kolata reports that Benbow thinks such research suggests that women should stop trying so hard: But if there are genetically based differences in mathematical abilities between males and females, it may be more difficult for women to study math and it may be one reason why so few women receive Ph.D.’s in math. Benbow believes that socialization is a factor in women’s math achievement but that it is not the only factor. Women, she says, would be better off accepting their differences and working to encourage girls to achieve as much as they can than to constantly blame their lesser achievements in mathematics solely on social factors (p. 1235).

In other words, women should accept their inherent limitations and should not push girls to push past the limits of their abilities. Clamoring for better math education and less social sexism is pointless. And, if the scientists are right, such an approach will be better for everybody. Assuming they are right, boys will be able to progress at the pace they need, getting the resources available to do so, while the girls will not have to struggle miserably through material that is beyond them. Hempel’s concept of inductive risk requires that we consider more than this, however. We must also consider what would happen if we accepted those scientific findings as reliable and they are not. What if the idea that men are better than women at math is dead wrong (as many have argued)? What will we have done by accepting it? The boys will be doing okay, with the resources, attention, and intellectual challenge at

292

Siblings Under the Skin

their disposal. The girls, on the other hand, are trapped in a system that believes them incapable of achievement, exacerbating any self-doubts they might have, and with correspondingly fewer resources placed at their disposal. So the girls and women pay a high price if we accept the theory and we are wrong. This alone is not enough for a clear charge of sexism. We must first examine the converse situation: What if we reject the findings and they are right? What if we make no assumptions about mathematical ability based on gender and do not segregate resources along gender lines? If the Benbow and Stanley findings are right, this may hurt both the boys and the girls a bit. The boys would have to share the time and attention of teachers in the most challenging classes with struggling girls and the girls would have to struggle mightily with their inherent lack of ability. However, if boys really do have greater ability, they will manage, although not quite as well as they could otherwise. And girls will at least have the opportunity to attempt the challenge. So, if the findings are wrong and we accept them, women and girls pay a terrible price of a completely inadequate mathematical education. If the findings are right and we reject them, both genders pay a penalty, but not nearly as serious, that of a less efficient educational system. Without the belief in gender dimorphism in mathematical ability, we have to treat individuals as individuals, making no presuppositions on the basis of gender. We would not use gender as a presorting technique for getting at mathematical ability. Which outcome of error is worse; women being shut out of mathematics unnecessarily or a losing a presorting device? Only if one values the potential harm to boys more than the potential harm to girls can one find the first situation more acceptable than the second. Because the potential harm to boys is far less egregious than the harm to girls caused by excluding them from adequate educational opportunity, only if one values men/boys much more than women/girls would one ask for equivalent levels of confirmation before accepting the theories. Valuing what happens to men/boys more than what happens to women/girls simply because of gender is the clearest form of sexism. There is no ambiguity here about the nature of the values involved or their role in the scientific process.

Douglas and Hempel

293

The argument above does not mean we should never accept gender-based differences in ability regardless of evidence. Following the reasoning of inductive risk, there would have to be much better evidence (enough to overcome the terrible costs of being wrong) or higher confirmation in order to accept a gender-based difference in mathematical ability than in order to reject it. To not place a higher burden of proof on such studies is sexist. In general, those arguing for gender-based (or race-based) intellectual abilities should have a very high burden of proof indeed, given the incredible risks of harm presented by the acceptance of such theories if proven wrong. In accepting theories or findings based on the assumption of gender dimorphism, scientists should be very sure that they have strong evidence to support their findings—more sure than in other, less potentially damaging cases. One might respond to this argument with the claim that the publication and public notice of such studies does not carry societal inductive risks at all, that these discussions take place solely among scientists. Such an argument is wholly naive. Scientific discourse is an authoritative voice in modern society. Thus, the inductive risks are quite real. As recounted in Fausto-Sterling, a ten-year old girl heard the radio report on the Benbow-Stanley study. What she heard was “that boys are genetically better at math than are girls. Girls…would be less frustrated if they recognized their limits and stopped their fruitless struggle to exceed them.” She went concerned to her father and asked: “Daddy, I always wanted to be a math professor like you. Does this mean I can’t?” (Fausto-Sterling 1985, pp. 53-54). From Barbies who complain that math is hard to young women who come to believe math is beyond them, the potential consequences of error in accepting such work as reliable are serious indeed. This does not mean we should never consider evidence of such gender-based differences, but given the risks involved, we should consider the evidence very carefully, demanding a high level of support, before accepting hypotheses about essentialist gender-based differences. And scientists, when dealing with such risks, should be cautious about the implications of the data they have gathered. Benbow and Stanley were clearly not adequately cautious in their interpretation, putting forth the strongest thesis possible to drum up publicity, with no thought to the risks involved.

294

Siblings Under the Skin

Analysis using inductive risks reveals the role of values in the acceptance of scientific work based on the assumption of gender dimorphism. What of the inductive risks of accepting the assumption of gender dimorphism itself? Not surprisingly, a more complex analysis emerges. What are the risks of assuming that gender behaviors cluster around poles? What are the risks of not assuming this? I will first discuss the societal risks of making (or rejecting) this assumption in scientific work, and then turn to the epistemic risks of the assumption. Although the analysis provides less of a clear cut direction for decision-making in this case, it is revealing that the assumption will likely be more harmful than helpful, both societally and epistemically. The societal risks of accepting dimorphism as a background assumption depend in part on what other social values and assumptions abound in the context for which the assumption will operate. Adopting the assumption of dimorphism will tend to dichotomize how one perceives and categorizes behaviors, and if there is a prevailing hierarchy that accompanies the dichotomy (e.g., if, as has been true historically in Western societies, women are valued less than men), assuming a gender dimorphism will tend to reinforce the dichotomy and its associated hierarchy. For example, if one assumes gender dimorphism, and performs research on the basis of that assumption, the results will reinforce the view that men and women should do fundamentally different things, even if this means relegating women to less valued work. The history of the study of gender roles in anthropology exemplifies the potential societal risks of assuming dimorphism (see, e.g., Longino pp. 104112). With men’s work more valued, and dimorphism assumed, men became the originators of tool use, language, societal structure, etc. in anthropological theory. These theories then provided reinforcement for beliefs that the reason women didn’t earn as much as men or couldn’t break through glass ceilings was somehow biological. If the assumption is wrong, and the research flawed, then the research helps reinforce a system of ongoing damage to half of our population. The social risks of assuming gender dimorphism can be problematic even without the valuational hierarchy. In a society where deviations from norms create social pressures on individuals, scientific research in support of behavioral gender-based norms can harm “deviant” individuals. Suppose there really is no gender dimorphism in, for example,

Douglas and Hempel

295

play behavior. Assuming that there are two norms (even two equally valued norms) around which behaviors cluster would be harmful to individuals who do not fit those norms. They would be labeled “abnormal,” which despite its ambiguous meaning, carries a stigma with it, and anxious parents and teachers would push children back towards the behaviors they “should” exhibit. If one assumes gender dimorphism and one is wrong, one will likely cause a good deal of damage to individuals who appear aberrant under one’s theory. But what of the converse situation? What if we don’t rely on the background assumption of gender dimorphism and it really is there? Instead of coercion to individuals, we risk inefficiencies in helping individuals. Without beliefs about how individuals should behave, about what their capabilities are based on gender, we cannot assume that we know how to effectively distribute resources to assist them. For example, if little Suzy is miserable because she thinks she likes math and is having a hard time in math class, or because she thinks she likes an aggressive form of play, and if gender dimorphism is accurate and Suzy is attempting to exhibit abilities beyond the extent of her capabilities, the theory would assist us in quickly diagnosing Suzy and helping her come to terms with her inherent limitations. We might eventually come to the same conclusions without assuming gender dimorphism (that Suzy is just pushing herself too hard in some way), but it might take longer. A similar area of difficulty would be in our assumptions concerning what people want, from appropriate toys for kids to appropriate institutional resources in our education system. If we reject gender dimorphism and we are wrong to do so, we lose the ability to simplify our decision-making about individuals and resource distribution. Which is worse, coercing people into behaving along norms that don’t fit them or having an overly complicated view of people that makes us less efficient in dealing with them? I would much prefer the latter problem than the former, so I am much more willing to risk being wrong in rejecting the assumption of gender dimorphism than being wrong in accepting the assumption. This judgment is even more supportable when one considers that no authors who support gender dimorphism currently believe that everyone completely accords with one of the two norms for behavior. Everyone acknowledges that there are statistical variations among the two genders, and that the variations

296

Siblings Under the Skin

overlap. Thus, we shouldn’t assume anything about individuals we encounter even if one assumes behaviors are dimorphic based on gender. Those are only statistical averages, at best. In terms of societal risks, it is not clear we risk anything at all if we reject gender dimorphism and we are wrong. Scientists, however, may feel distressed about giving up the assumptions of gender dimorphism. What are the epistemic inductive risks of this assumption? What epistemic advantages does the assumption provide us? What are the potential errors of relying on the assumption? What would we risk by rejecting it? At its most basic, the assumption of gender dimorphism provides us with a conceptual tool to untangle the complexity of the world. Gender provides a basic category to be considered by scientists when collecting data and analyzing that data for correlations. If we expect to see behavior clustered around two gendered norms, we can use that assumption to look for those clusterings and to expound on what those norms are. Relying on the assumption, however, is not without epistemic risks. The assumption of gender dimorphism is a simplifying one, and as such, it may blind us to important complexities or alternative patterns in the world. For example, if there are, in fact, more than two norms around which behaviors cluster, those other norms may become invisible if we are only expecting two. Or, if there are more important influences on behavior than gender that produce a greater complexity of behavioral patterns, such complexity may be masked by the oversimplifying assumption of gender dimorphism. These are serious epistemic risks, ones that will hamper the development of the human sciences. What of the alternative set of risks? What do we risk if we reject the assumption and it is really accurate? If we reject the assumption and it is true, we lose a valuable conceptual lever with which to do science. The world will be harder to explore, data analysis more difficult, interpretations trickier. So which is worse: to be initially blinded to complexity or to be initially swamped by it? Having relied on the assumption for so long, and having been blind to so much, perhaps it would be a good idea to try doing science without the assumption for a while (if we can manage to get rid of the assumption). We should find out if the world really is too complex for us to deal with in the absence of the assumption. If we are, in fact, overwhelmed with the complexity of human

Douglas and Hempel

297

behavior without gender dimorphism, we may decide it is necessary. However, it seems a general rule that unless it proves problematically difficult to make something out of data, one should try not to rely on assumptions that could mask so much of the information contained in data. The epistemic risks appear more serious if we accept the assumption and are wrong than the converse. Thus, in the end, consideration of both the epistemic and societal inductive risks should make us lean the same way. Feminist critiques of research based on gender dimorphism should raise enough concern over the uncertainties surrounding that assumption to warrant serious consideration of the inductive risks. The inductive risks, both epistemic and societal, of accepting the assumption and being wrong, seem far more serious than the converse. Thus scientists should be cautious about relying on assumptions of gender dimorphisms. Such inductive risk considerations should weigh heavily on the minds of scientists when deciding both how to structure their research and how to present it to the public. At first glance, gender dimorphism appears to be a purely epistemic assumption. However, when considering the inductive risks of the assumption, it becomes apparent how sexist values, used to weigh the consequences of errors, underlie the acceptance of some risks rather than others. Hempel’s notion of inductive risk cleanly delineates where the sexist values enter science in some otherwise tricky cases (and cases heavily critiqued by feminist philosophers). It is a useful line of thought in other arenas as well. VI. Hempelian insights for EPA policy Scientific research on sex and gender differences has the potential to have a profound impact on individual practices and social policies. Other areas of research also have this capacity, and these areas are also illuminated by the concept of inductive risk. Wherever science will have an impact on decision-making, and there is uncertainty in the science, we must recognize the inductive risks we take, consider the consequences of error, both epistemic and non-epistemic, and reflect on which consequences are more acceptable. Some inductive risks are apparent only to scientists in the process of doing science, whereas other inductive risks are more public and thus more easily debated openly.

298

Siblings Under the Skin

(See Douglas 2000 for some examples of deeply internal inductive risks.) Values thus become an important part of the practice of science when real world decisions are likely to be influenced by the science. These values do not determine what the science will say; instead they determine which uncertainties are more worrisome, and which errors more tolerable. In the past fifty years, science has become an ever more important aspect of policy-making. This importance is particularly clear in the arena of environmental policy-making, where science plays a central role in both focusing our attention on particular problems and in the determination of policy details for addressing those problems. A recent example of the role in science for policy-making concerns the deliberation by the Environmental Protection Agency’s Scientific Advisory Board (SAB) over whether dioxin (in particular 2,3,7,8-tetrachlorodibenzo-p-dioxin or TCDD, the best studied and most potent form of dioxin) is a known human carcinogen. There are both significant uncertainties in and clear non-epistemic consequences for this decision, and understanding the role of values in the deliberations through inductive risk helps clarify how the SAB could ultimately fail to reach an agreement on the matter. The scientific uncertainties for this case are significant and hotly debated, but do not arise from a lack of study. TCDD has been extremely well-studied in laboratory animals and there is no debate over whether TCDD is an animal carcinogen. It is a somewhat unusual carcinogen in animals, however, because it is not a mutagen (i.e. it does not cause the genetic mutations that lead to cancer), but seems instead to act as a promoter, causing increased growth of already present mutated cells. Even more unusual, it causes cancer at multiple organ sites that vary depending on the species and strain tested. The precise mechanism of cancer promotion is not clear, but most scientists believe that the process begins at the cellular level when TCDD binds with an intercellular receptor (known as the “Ah” receptor) present in humans and animals. Unfortunately, the TCDD-receptor complex appears to interact with over 300 genetic sites, so untangling the complete causal process involved when the receptor-dioxin complex interacts with genetic material to cause cell proliferation will be difficult (Puga et. al. 2000).

Douglas and Hempel

299

The human epidemiological data is also complex. Because we cannot test a potent substance like TCDD on humans directly, we must rely on accidental and occupational exposures to TCDD. Such cases always come with confounding causal factors. Dioxins are not deliberately manufactured (except for scientific study), but instead are industrial by-products of a wide array of chemical processes, from paper bleaching to pesticide production to waste incineration. Thus, epidemiologists have studied human groups known to have had an elevated exposure to dioxin, usually from working in a pesticide manufacturing plant or from an accident at such a plant. Statistically significant elevated cancer rates have been found in such human populations. Critics, however, point out that 1) there are confounding chemical exposures for such human populations, e.g., the pesticides themselves and 2) there is no consistency among the types of cancer elevated among the populations studied.9 These arguments are not devastating, however. To the second critique, it can be pointed out that such a lack of consistency is not surprising given the results of the animal studies. To the first critique, it can be argued that to demand a human population study without confounders is to demand an ethically unacceptable study and thus to set the standard of proof too high. Precisely where the standard of proof should be, and which uncertainties are more acceptable, is what the debate over TCDD is about. What does a designation as a known human carcinogen mean? In 1984, the EPA designated TCDD as a probable human carcinogen, because it was thought at the time that not enough was known about the mechanism of TCDD carcinogenicity and that there was insufficient epidemiological data to designate TCDD as a known human carcinogen. Thus, the SAB’s debate in 2001 is over whether there is now adequate evidence to upgrade TCDD’s status from probable human carcinogen to known carcinogen. The epidemiological studies I mentioned above have been completed since 1984, and the role of the receptor in the carcinogenic process has become widely accepted. Is this enough to change the regulatory status of TCDD? Clearly, there are scientific uncertainties here. What are the inductive risks? Designating TCDD as a known human carcinogen will lead to stricter regulations for those industries which inadvertently produce

300

Siblings Under the Skin

dioxins. Leaving TCDD as merely a probable carcinogen will make it difficult to support stricter environmental standards. There are significant consequences of error here either way. On the one hand, if we designate TCDD as a known human carcinogen and it is not, we risk economic harm to those industries. Conversely, if we fail to designate TCDD as a known human carcinogen and it is, we risk failing to decrease cancer rates for millions of Americans who are regularly exposed to dioxins in the food chain. The significance of the issue has not gone unnoticed. As reported in the Washington Post: The chemical, beef, and poultry industries are waging an intense campaign to delay further an Environmental Protection Agency study showing that consumption of animal fat and dairy products containing traces of dioxin can cause cancer in humans. … The economic stakes in the dioxin controversy are high: The EPA’s issuance of a final report could result in federal and state regulations costly to chemical manufacturers. … Industry groups including the American Chemistry Council, the Chlorine Chemistry Council, the Chemical Manufacturers Association, the American Meat Institute and the National Cattelman’s Beef Association contend that the EPA’s study is seriously flawed and exaggerates the health risk dioxin poses. ‘We are alarmed at any study that reaches conclusions not based on science,’ said Gary Weber, executive director of regulatory affairs for the cattleman’s association (Pianin 2001).

The designation of dioxin as a known human carcinogen will allow the EPA to say that dioxin can, in fact, cause cancer in humans, instead of there being substantial doubt about the issue. What the SAB debated, and failed to agree on, was whether there is enough evidence to alter the designation. Note how the unavoidable scientific uncertainty is interpreted by the industry spokesperson as a lack of science in the process, rather than as the normal result of doing empirical research. What did the SAB ultimately say about the issue? The SAB reports states that “there is a lack of consensus in the Panel” over the issue of dioxin’s human carcinogen designation. Specifically, “there is disagreement about the strength of the epidemiological data indicating that dioxin is carcinogenic in humans (i.e., whether statistically significant

Douglas and Hempel

301

associations between exposure and cancer could be concluded to be causal), as well as the scientific data demonstrating similar modes of action in humans and laboratory animals.” Apparently, the disagreement was quite dramatic, with over one third of the twenty-member panel supporting the classification of dioxin as a human carcinogen, almost half disagreeing with the designation, and the remaining not taking a stand (SAB 2001, pp. 4-5). These points of disagreement are not new to dioxin debates, and yet they are persistent. Inductive risk can assist us in understanding the disagreement. Much of the disagreement concerns the level of proof needed for designation of a substance as a human carcinogen and which assumptions to use in interpreting the available evidence. Emphasize the strengths of the epidemiological studies and their consistency with animal studies, and the designation appears well supported. Emphasize the uncertainties and the designation appears weakly supported at best. If one considers the inductive risks of these choices, the disagreements make more sense. If one believes the risks to large segments of American industry the more serious threat here, then one will want the higher standard of proof and will tend to emphasize the weaknesses of the epidemiological studies, risking the error of underregulating dioxins. If one believes the potential health risks to human populations a more serious threat, then one will tolerate a lower standard of proof and will tend to emphasize the strengths of the epidemiological studies, risking the error of overregulation of dioxins. It is important to recognize that the role of values here is not irrational or unwarranted. The scientists sitting on the SAB panel should be thinking about which inductive risks are more acceptable and which are more problematic. What the uncertainties and inductive risks tell us is that “science” must incorporate these considerations into its judgments. However, this mean incorporating social values into science. Which values should the scientists emphasize? Currently, there is deep disagreement over which should be more valued, widespread but relatively low level human health risks or economic dislocation and industrial disruption. Studies linking poverty to increased rates of disease and death further complicate the picture. However, inductive risk confirms the importance of pursuing this debate further and of reaching acceptable tradeoffs between the two values (or, even better, ways of avoiding

302

Siblings Under the Skin

the conflict completely). Thus, we should foster the debates over the values at issue, e.g., human health risks vs. economic stability and benefits, rather than expecting scientists to simply tell us the answer to the question: is dioxin a human carcinogen? It is reasonable for the SAB panel members to have a range of positions concerning the values at stake. Because of the role of inductive risk in this dispute, it is reasonable (and even expected) that SAB members would disagree over the designation of dioxin as a human carcinogen. The uncertainties that arise in epidemiological studies are likely to persist, and thus we should expect the disagreements over the issue to persist until the value dispute is adequately addressed. Inductive risk helps to reveal the sources of the dispute and the values at stake. Hempel’s concept thus serves as a useful analytical tool both within feminist critiques of science and in socially relevant science generally.

Douglas and Hempel

303

Notes 1.

2.

3.

4.

5.

6. 7.

I could not have completed this work without the support of the University of Puget Sound’s Martin Nelson Junior Sabbatical Fellowship and the National Science Foundation (SDET grant #0115258). I would like to thank Douglas Cannon and Denis Arnold for their thoughtful comments on earlier drafts of this paper. The essay first appeared in a collection entitled Social Control in a Free Society (1960), edited by Robert Spiller. Although Hempel modified the essay for Aspects, the key sections discussed here are almost identical in the two versions. In a more recent piece from 1981, Hempel discussed the Rudner paper and Jeffrey’s response more directly. In this piece, he appears to accept Jeffrey’s view that scientists do not play a role in practical decision-making: “But, Jeffrey notes, the scientist qua scientist is not concerned with giving advice, or making decisions, on contemplated courses of action.” (p. 348) Hempel never disagrees with this statement, despite the large role scientists qua scientists played in federal decision-making by 1980. See “Turns in the Evolution of the Problem of Induction,” reprinted in The Philosophy of Carl G. Hempel (2001), Fetzer ed., pp. 347-352. Hempel here lays out the distinction between instrumental and categorical value judgments. Science is useful for the instrumental ones, but not the categorical. Science can inform our choices by telling us how to achieve our goals. It can even influence our ideas about which goals are achievable and thus which are worth pursuing intently. Science cannot tell us ultimately what we should value, however. Unlike the positivists, Hempel did not consider categorical value statements meaningless even though they could not be empirically confirmed. The irony of the view that science was generally removed from decisionmaking contexts, which took hold even as scientists took on an increasing public role as advisors and decision-makers in government, should not go unnoticed. See Fetzer 2001 collection, “Turns in the Evolution of the Problem of Induction” and “Valuation and Objectivity in Science.” For example, Longino discusses how descriptions that carry societal valuations of particular behaviors bring those societal values into the data. The use of terms like “tomboyism” to describe a girl who exhibits behavior supposedly acceptable only from boys “implies the inappropriateness of the behavior” (p. 131). Societal valuations of gender roles thus enters into the data, i.e., the descriptions of behaviors which will then be analyzed. Once the evaluative nature of some apparently descriptive term is recog-

304

8. 9.

Siblings Under the Skin nized, this mode of entry for values is fairly easy to head off, however, and Longino acknowledges scientists’ efforts to remove such value-laden terms from their accounts of behaviors (p. 131..) Coming to recognize descriptive categories as carrying evaluative import is more difficult. Inductive risk may assist with such recognition, as I suggest below. The analysis I present can also be applied to some aspects of values in data selection and interpretation. See Douglas (2000) for an example outside of feminist critiques. There are also debates over the strength of the statistical associations found in the studies. This issue is too complex to discuss here. The associations I discuss are statistically significant but also due to a relatively small number of actual cancers in the populations studied.

References Benbow, C. & Stanley, J. 1980. Sex differences in mathematical ability: Fact or artifact? Science, 210: 1262-1264. Churchman, C. W. 1948. Statistics, pragmatics, induction. Philosophy of Science, 15: 249-268. ——. 1956. Science and decision making. Philosophy of Science, 22: 247-249. Douglas, H. 2000. Inductive risk and values in science. Philosophy of Science, 67: 559-579. Fausto-Sterling, A. 1985. Myths of gender. New York: Basic Books. Fetzer, J. H. (ed.) 2000. Science, explanation, and rationality: Aspects of the philosophy of Carl G. Hempel. New York: Oxford University Press. Hempel, C.G. [1981] 2001. Turns in the evolution of the problem of induction. Reprinted in Fetzer, J. H. (ed.) The philosophy of Carl G. Hempel: Studies in science, explanation, and rationality. New York: Oxford University Press. ——. 1965. Science and human values. In Hempel, C. G. Aspects of scientific explanation. New York: The Free Press. Jeffrey, R. 1956. Valuation and acceptance of scientific hypotheses. Philosophy of Science, 22: 237-246. Kolata, G. 1980. Math and sex: Are girls born with less ability? Science, 210: 1234-1235. Levi, I. 1962. On the seriousness of mistakes. Philosophy of Science, 29: 4765.

Douglas and Hempel

305

Longino, H. 1990. Science as Social Knowledge. Princeton: Princeton University Press. Pianin, E. 2001. Dioxin report by EPA on hold. Washington Post, April 12: A01. Puga, A., et al. 2000. The transcriptional signature of dioxin in human hepatoma HepG2 cells. Biochemical Pharmacology, 60: 1129-1142. Rudner, R. 1953. The scientist qua scientist makes value judgments. Philosophy of Science, 20: 1-6. Scientific Advisory Board. 2001. Dioxin reassessment: An SAB review of the Office of Research and Development’s reassessment of dioxin. EPA-SABEC-01-006. Spiller, R. E. (ed.) 1960. Social control in a free society. Philadelphia: University of Pennsylvania Press.

Biographical note Heather Douglas was born in 1969 to an American animal physiologist and a Swedish flight attendant, brought together through an improbable case of frozen penguins. While living on the plains of Northern Illinois, Heather got her first dog at the age of seven. Yorick, a high spirited keeshond, was a wonderful companion throughout her later childhood years in New England. Due to Yorick’s peaceful passing in 1989, Heather was unable to share her thoughts with him on Time, which led to her B. A. in Philosophy and Physics from the University of Delaware (1991). Nor was she able to expound her ideas to Yorick on the role of values in the science policy concerning dioxin that led to her Ph. D. in History and Philosophy of Science from the University of Pittsburgh (1998). We are confident though that if Yorick was with us, he would have been the first to celebrate her being hired as the Phillip M. Phibbs Assistant Professor of Science and Ethics at the University of Puget Sound in 1998. Happily, Heather has recently been blessed with a new puppy, Corin, a high spirited black lab/ boxer mix. By the time that Heather was enrolled at the University of Pittsburgh, Hempel had of course retired, but she became well-versed in his writings on explanation and confirmation. However, it was not until she was working through the issue of values in science, that she came across Hempel’s essay on the topic, recommended to her by fellow grad student Erik Anger. Heather

306

Siblings Under the Skin

was impressed that Hempel had brought his usual clarity to an area that was so often muddled, and thus inspired, produced her dissertation and a number of essays including “Hempelian Insights for Feminism” written especially for this volume. Heather spends most of her free time with Corin and her husband Ted, digging in the dirt around their Tacoma, Washington home, discussing the finer points of science, values, and objectivity. The digging (and discussion) have inspired her first book, due out soon.

Douglas and Hempel

307

Science and Human Values Carl Hempel

I. The problem Our age is often called an age of science and of scientific technology, and with good reason: the advances made during the past few centuries by the natural sciences, and more recently by the psychological and sociological disciplines, have enormously broadened our knowledge and deepened our understanding of the world we live in and of our fellow men; and the practical application of scientific insights is giving us an ever increasing measure of control over the forces of nature and the minds of men. As a result, we have grown quite accustomed, not only to the idea of a physico-chemical and biological technology based on the results of the natural sciences, but also to the concept, and indeed the practice, of a psychological and sociological technology that utilizes the theories and methods developed by behavioral research. This growth of scientific knowledge and its applications has vastly reduced the threat of some of man’s oldest and most formidable scourges, among them famine and pestilence; it has raised man’s material level of living, and it has put within his reach the realization of visions which even a few decades ago would have appeared utterly fantastic, such as the active exploration of interplanetary space. But in achieving these results, scientific technology has given rise to a host of new and profoundly disturbing problems: The control of nuclear fission has brought us not only the comforting prospect of a vast new reservoir of energy, but also the constant threat of the atom bomb and of grave damage, to the present and to future generations, from the radioactive by-products of the fission process even in its peaceful uses. And the very progress in biological and medical knowledge and technology which has so strikingly reduced infant mortality and increased man’s life expectancy in large areas of our globe has significantly contributed to the threat of the “population explosion,”

308

Siblings Under the Skin

the rapid growth of tile earth’s population which we are facing today, and which, again, is a matter of grave concern to all those who have the welfare of future generations at heart. Clearly, the advances of scientific technology on which we pride ourselves, and which have left their characteristic imprint on every aspect of this “age of science,” have brought in their train many new and grave problems which urgently demand a solution. It is only natural that, in his desire to cope with these new issues, man should turn to science and scientific technology for further help. But a moment’s reflection shows that the problems that need to be dealt with are not straightforward technological questions but intricate complexes of technological and moral issues. Take the case of the population explosion, for example. To be sure, it does pose specific technological problems. One of these is the task of satisfying at least the basic material needs of a rapidly growing population by means of limited resources; another is the question of means by which population growth itself may be kept under control. Yet these technical questions do not exhaust the problem. For after all, even now we have at our disposal various ways of counteracting population growth; but some of these, notably contraceptive methods, have been and continue to be the subject of intense controversy on moral and religious grounds, which shows that an adequate solution of the problem at hand requires, not only knowledge of technical means of control, but also standards for evaluating the alternative means at our disposal; and this second requirement clearly raises moral issues. There is no need to extend the list of illustrations: any means of technical control that science makes available to us may be employed in many different ways, and a decision as to what use to make of it involves us in questions of moral valuation. And here arises a fundamental problem to which I would now like to turn: Can such valuational questions be answered by means of the objective methods of empirical science, which have been so successful in giving us reliable, and often practically applicable, knowledge of our world? Can those methods serve to establish objective criteria of right and wrong and thus to provide valid moral norms for the proper conduct of our individual and social affairs?

Douglas and Hempel

309

II. Scientific testing Let us approach this question by considering first, if only in brief and sketchy outline, the way in which objective scientific knowledge is arrived at. We may leave aside here the question of ways of discovery; i.e., the problem of how a new scientific idea arises, how a novel hypothesis or theory is first conceived; for our purposes it will suffice to consider the scientific ways of validation; i.e., the manner in which empirical science goes about examining a proposed new hypothesis and determines whether it is to be accepted or rejected. I will use the word ‘hypothesis’ here to refer quite broadly to any statements or set of statements in empirical science, no matter whether it deals with some particular event or purports to set forth a general law or perhaps a more or less complex theory. As is well known, empirical science decides upon the acceptability of a proposed hypothesis by means of suitable tests. Sometimes such a test may involve nothing snore than what might be called direct observation of pertinent facts. This procedure may be used, for example, in resting such statements as “It is raining outside,” “All the marbles in this urn are blue,” “The needle of this ammeter will stop at the scale point marked 6,” and so forth. Here a few direct observations will usually suffice to decide whether the hypothesis at hand is to be accepted as true or to be rejected as false. But most of the important hypotheses in empirical science cannot be tested in this simple manner. Direct observation does not suffice to decide, for example, whether to accept or to reject the hypotheses that the earth is a sphere, that hereditary characteristics are transmitted by genes, that all Indo-European languages developed from one common ancestral language, that light is an electromagnetic wave process, and so forth. With hypotheses such as these, science resorts to indirect methods of test and validation. While these methods vary greatly in procedural detail, they all have the same basic structure and rationale. First, from the hypothesis under test, suitable other statements are inferred which describe certain directly observable phenomena that should be found to occur under specifiable circumstances if the hypothesis is true; then those inferred statements are tested directly; i.e., by checking whether the specified phenomena do in fact occur;

310

Siblings Under the Skin

finally, the proposed hypothesis is accepted or rejected in the light of the outcome of these tests. For example, the hypothesis that the earth is spherical in shape is not directly testable by observation, but it permits us to infer that a ship moving away from the observer should appear to be gradually dropping below the horizon; that circumnavigation of the earth should be possible by following a straight course; that high-altitude photographs should show the curving of the earth’s surface; that certain geodetic and astronomical measurements should yield such and such results; and so forth. Inferred statements such as these can be tested more or less directly; and as an increasing number and variety of them are actually borne out, the hypothesis becomes increasingly confirmed. Eventually, a hypothesis may be so well confirmed by the available evidence that it is accepted as having been established beyond reasonable doubt. Yet no scientific hypothesis is ever proved completely and definitively; there is always at least the theoretical possibility that new evidence will be discovered which conflicts with some of the observational statements inferred from the hypothesis, and which thus leads to its rejection. The history of science records many instances in which a once accepted hypothesis was subsequently abandoned in the light of adverse evidence. III. Instrumental judgments of value We now turn to the question whether this method of test and validation may be used to establish moral judgements of value, and particularly judgments to the effect that a specified course of action is good or right or proper, or that it is better than certain alternative courses of action, or that we ought—or ought not—to act in certain specified ways. By way of illustration, consider the view that it is good to raise children permissively and bad to bring them up in a restrictive manner. It might seem that, at least in principle, this view could be scientifically confirmed by appropriate empirical investigations. Suppose, for example, that careful research had established (1) that restrictive upbringing tends to generate resentment and aggression against parents and other persons exercising educational authority, and that this leads to guilt and anxiety and an eventual stunting of the child’s ini-

Douglas and Hempel

311

tiative and creative potentialities; whereas (2) permissive upbringing avoids these consequences, makes for happier interpersonal relations, encourages resourcefulness and self-reliance, and enables the child to develop and enjoy his potentialities. These statements, especially when suitably amplified, come within the purview of scientific investigation; and though our knowledge in the matter is in fact quite limited, let us assume, for the sake of the argument, that they had actually been strongly confirmed by careful tests. Would not scientific research then have objectively shown that it is indeed better to raise children in a permissive rather than in a restrictive manner? A moment’s reflection shows that this is not so. What would have been established is rather a conditional statement; namely, that if our children are to become happy, emotionally secure, creative individuals rather than guilt-ridden and troubled souls then it is better to raise them in a permissive than in a restrictive fashion. A statement like this represents a relative, or instrumental, judgment of value. Generally, a relative judgment of value states that a certain kind of action, M, is good (or that it is better than a given alternative M1) if a specified goal G is to be attained; or more accurately, that M is good, or appropriate, for the attainment of goal G. But to say this is tantamount to asserting either that, in the circumstances at hand, course of action M will definitely (or probably) lead to the attainment of G, or that failure to embark on course of action M will definitely (or probably) lead to the nonattainment of G. In other words, the instrumental, value judgment asserts either that M is a (definitely or probably) sufficient means for attaining the end or goal G, or that it is a (definitely or probably) necessary means for attaining it. Thus, a relative, or instrumental, judgment of value can be reformulated as a statement which expresses a universal or a probabilistic kind of means-ends relationship, and which contains no terms of moral discourse—such as ‘good,’ ‘better,’ ‘ought to’—at all. And a statement of this kind surely is an empirical assertion capable of scientific test. IV. Categorical judgments of value Unfortunately, this does not completely solve our problem; for after a relative judgment of value referring to a certain goal G has been

312

Siblings Under the Skin

tested and, let us assume, well confirmed, we are still left with the question of whether the goal G ought to be pursued, or whether it would be better to aim at some alternative goal instead. Empirical science can establish the conditional statement, for example, that if we wish to deliver an incurably ill person from intolerable suffering, then a large dose of morphine affords a means of doing so; but it may also indicate ways of prolonging the patient’s life, if also his suffering. This leaves us with the question whether it is right to give the goal of avoiding hopeless human suffering precedence over that of preserving human life. And this question calls, not for a relative but for an absolute, or categorical, judgment of value to the effect that a certain state of affairs (which may have been proposed as a goal or end) is good, or that it is better than some specified alternative. Are such categorical value judgments capable of empirical test and confirmation? Consider, for example, the sentence “Killing is evil.” It expresses a categorical judgment of value which, by implication, would also categorically qualify euthanasia as evil. Evidently, the sentence does not express an assertion that can be directly tested by observation; it does not purport to describe a directly observable fact. Can it be indirectly tested, then, by inferring from it statements to the effect that under specified test conditions such and such observable phenomena will occur? Again, the answer is clearly in the negative. Indeed, the sentence ‘Killing is evil’ does not have the function of expressing an assertion that can be qualified as true or false; rather, it serves to express a standard for moral appraisal or a norm for conduct. A categorical judgment of value may have other functions as well; for example, it may serve to convey the utterer’s approval or disapproval of a certain kind of action, or his commitment to the standards of conduct expressed by the value judgment. Descriptive empirical import, however, is absent; in this respect a sentence such as ‘Killing is evil’ differs strongly from, say, ‘Killing is condemned as evil by many religions, which expresses a factual assertion capable of empirical test. Categorical judgments of value, then, are not amenable to scientific test and confirmation or disconfirmation; for they do not express assertions but rather standards or norms for conduct. It was Max Weber, I believe, who expressed essentially the same idea by remarking that science is like a map: it can tell us how to get to a given place, but it

Douglas and Hempel

313

cannot tell us where to go. Gunnar Myrdal in his book An American Dilemma ([1944] 1962, p. 1052), stresses in a similar vein that “factual or theoretical studies alone cannot logically lead to a practical recommendation. A practical or valuational conclusion can be derived only when there is at least one valuation among the premises.” Nevertheless, there have been many attempts to base systems of moral standards on the findings of empirical science; and it would be of interest to examine in some detail the reasoning which underlies those procedures. In the present context, however, there is room for only a few brief remarks on this subject. It might seem promising, for example, to derive judgments of value from the results of an objective study of human needs. But no cogent derivation of this sort is possible. For this procedure would presuppose that it is right, or good, to satisfy human needs—and this presupposition is itself a categorical judgment of value: it would play the role of a valuational premise in the sense of Myrdal’s statement. Furthermore, since there are a great many different, and partly conflicting, needs of individuals and of groups, we would require not just the general maxim that human needs ought to be satisfied, but a detailed set of rules as to the preferential order and degree in which different needs are to be met, and how conflicting claims are to be settled; thus, the valuational premise required for this undertaking would actually have to be a complex system of norms; hence, a derivation of valuational standards simply from a factual study of needs is out of the question. Several systems of ethics have claimed the theory of evolution as their basis; but they are in serious conflict with each other even in regard to their most fundamental tenets. Some of the major variants are illuminatingly surveyed in a chapter of G. G. Simpson’s book, The Meaning of Evolution (1949). One type, which Simpson calls a “toothand-claw ethics,” glorifies a struggle for existence that should lead to a survival of the fittest. A second urges the harmonious adjustment of groups or individuals to one another so as to enhance the probability of their survival, while still other systems hold up as an ultimate standard the increased aggregation of organic units into higher levels of organization, sometimes with the implication that the welfare of the state is to be placed above that of the individuals belonging to it. It is obvious that these conflicting principles could not have been validly inferred

314

Siblings Under the Skin

from the theory of evolution—unless indeed that theory were self-contradictory, which does not seem very likely. But if science cannot provide us with categorical judgments of value, what then can serve as a source of unconditional valuations? This question may either be understood in a pragmatic sense, as concerned with the sources from which human beings do in fact obtain their basic values. Or it may be understood as concerned with a systematic aspect of valuation; namely, with the question where a proper system of basic values is to be found on which all other valuations may then be grounded. The pragmatic question comes within the purview of empirical science. Without entering into details, we may say here that a person’s values—both those he professes to espouse and those he actually conforms to—are largely absorbed from the society in which he lives, and especially from certain influential subgroups to which he belongs, such as his family, his schoolmates, his associates on the job, his church, clubs, unions, and other groups. Indeed his values may vary from case to case depending on which of these groups dominates the situation in which he happens to find himself. In general, then, a person’s basic valuations are no more the result of careful scrutiny and critical appraisal of possible alternatives than is his religious affiliation. Conformity to the standards of certain groups plays a very important role here, and only rarely are basic values seriously questioned. Indeed, in many situations, we decide and act unreflectively in an even stronger sense; namely, without any attempt to base our decisions on some set of explicit, consciously adopted, moral standards. Now, it might be held that this answer to the pragmatic version of our question reflects a regrettable human inclination to intellectual and moral inertia; but that the really important side of our question is the systematic one: If we do want to justify our decisions, we need moral standards of conduct of the unconditional type—but how can such standards be established? If science cannot provide categorical value judgments, are there any other sources from which they might be obtained? Could we not, for example, validate a system of categorical judgments of value by pointing out that it represents the moral standards held up by the Bible, or by the Koran, or by some inspiring thinker or social leader? Clearly, this procedure must fail, for the factual informa-

Douglas and Hempel

315

tion here adduced could serve to validate the value judgments in question only if we were to use, in addition, a valuational presupposition to the effect that the moral directives stemming from the source invoked ought to be complied with. Thus, if the process of justifying a given decision or a moral judgment is ever to be completed, certain judgments of value have to be accepted without any further justification, just as the proof of a theorem in geometry requires that some propositions be accepted as postulates, without proof. The quest for a justification of all our valuations overlooks this basic characteristics of the logic of validation and of justification. The value judgments accepted without further justification in a given context need not, however, be accepted once and for all, with a commitment never to question them again. This point will be elaborated further in the final section of this essay. As will hardly be necessary to stress, in concluding the present phase of our discussion, the ideas set forth in the preceding pages do not imply or advocate moral anarchy; in particular, they do not imply that any system of values is just as good, or just as valid, as any other, or that everyone should adopt the moral principles that best suit his convenience. For all such maxims have the character of categorical value judgments and cannot, therefore, be implied by the preceding considerations, which are purely descriptive of certain logical, psychological, and social aspects of moral valuation. V. Rational choice: Empirical and valuational components To gain further insight into the relevance of scientific inquiry for categorical valuation let us ask what help we might receive, in dealing with a moral problem, from science in an ideal state such as that represented by Laplace’s conception of a superior scientific intelligence, sometimes referred to as Laplace’s demon. This fiction was used by Laplace, early in the nineteenth century, to give a vivid characterization of the idea of universal causal determinism. The demon is conceived as a perfect observer, capable of ascertaining with infinite speed and accuracy all that goes on in the universe at a given moment; he is also an ideal theoretician who knows all the laws of nature and has combined them into one universal formula; and finally, he is a perfect mathematician who, by means of that universal formula, is able to infer, from the

316

Siblings Under the Skin

observed state of the universe at the given moment, the total state of the universe at any other moment; thus past and future are present before his eyes. Surely, it is difficult to imagine that science could ever achieve a higher degree of perfection! Let us assume, then, that, faced with a moral decision, we are able to call upon the Laplacean demon as a consultant. What help might we get from him? Suppose that we have to choose one of several alternative courses of action open to us, and that we want to know which of these we ought to follow. The demon would then be able to tell us, for any contemplated choice, what its consequences would be for the future course of the universe, down to the most minute detail, however remote in space and time. But, having done this for each of the alternative courses of action under consideration, the demon would have completed his task: he would have given us all the information that an ideal science might provide under the circumstances. And yet he would not have resolved our moral problem, for this requires a decision as to which of the several alternative sets of consequences mapped out by the demon as attainable to us is the best; which of them we ought to bring about. And the burden of this decision would still fall upon our shoulders: it is we who would have to commit ourselves to an unconditional judgment of value by singling out one of the sets of consequences as superior to its alternatives. Even Laplace’s demon, or the ideal science he stands for, cannot relieve us of this responsibility. In drawing this picture of the Laplacean demon as a consultant in decision-making, I have cheated a little; for if the world were as strictly deterministic as Laplace’s fiction assumes, then the demon would know in advance what choice we were going to make, and he might disabuse us of the idea that there were several courses of action open to us. However that may be, contemporary physical theory has cast considerable doubt on the classical conception of the universe as a strictly deterministic system: the fundamental laws of nature are now assumed to have a statistical or probabilistic rather than a strictly universal, deterministic, character. But whatever may be the form and the scope of the laws that hold in our universe, we will obviously never attain a perfect state of knowledge concerning them; confronted with a choice, we never have more than a very incomplete knowledge of the laws of nature and of the state

Douglas and Hempel

317

of the world at the time when we must act. Our decisions must therefore always be made on the basis of incomplete information, a state which enables us to anticipate the consequences of alternative choices at best with probability. Science can render an indispensable service by providing us with increasingly extensive and reliable information relevant to our purpose; but again it remains for us to evaluate the various probable sets of consequences of the alternative choices under consideration. And this requires the adoption of pertinent valuational standards which are not objectively determined by the empirical facts. This basic point is reflected also in the contemporary mathematical theories of decision making. One of the objectives of these theories is the formulation of decision rules which will determine an optimal choice in situations where several courses of action are available. For the formulation of decision rules, these theories require that at least two conditions be met: (1) Factual information must be provided specifying the available courses of action and indicating for each of these its different possible outcomes—plus, if feasible, the probabilities of their occurrence; (2) there must be a specification of the values—often prosaically referred to as utilities—that are attached to the different possible outcomes. Only when these factual and valuational specifications have been provided does it make sense to ask which of the available choices is the best, considering the values attaching to their possible results. In mathematical decision theory, several criteria of optimal choice have been proposed. In case the probabilities for the different outcomes of each action are given, one standard criterion qualifies a choice as optimal if the probabilistically expectable utility of its outcome is at least as great as that of any alternative choice. Other rules, such as the maximin and the maximax principles, provide criteria that are applicable even when the probabilities of the outcomes are not available. But interestingly, the various criteria conflict with each other in the sense that, for one and the same situation, they will often select different choices as optimal. The policies expressed by the conflicting criteria may be regarded as reflecting different attitudes towards the world, different degrees of optimism or pessimism, of venturesomeness or caution. It may be said therefore that the analysis offered by current mathematical models indicates two points at which decision-making calls not solely for factual

318

Siblings Under the Skin

information, but for categorical valuation, namely, in the assignment of utilities to the different possible outcomes and in the adoption of one among many competing decision rules or criteria of optimal choice. (This topic is developed in more detail in section 10.2 of my essay “Aspects of Scientific Explanation,” 1965) VI. Valuational “presuppositions” of science The preceding three sections have been concerned mainly with the question whether, or to what extent, valuation and decision presuppose scientific investigation and scientific knowledge. This problem has a counterpart which deserves some attention in a discussion of science and valuation; namely, the question whether scientific knowledge and method presuppose valuation. The word “presuppose” may be understood in a number of different senses which require separate consideration here. First of all, when a person decides to devote himself to scientific work rather than to some other career, and again, when a scientist chooses some particular topic of investigation, these choices will presumably be determined to a large extent by his preferences, i.e., by how highly he values scientific research in comparison with the alternatives open to him, and by the importance he attaches to the problems he proposes to investigate. In this explanatory, quasi-causal sense the scientific activities of human beings may certainly be said to presuppose valuations. Much more intriguing problems arise, however, when we ask whether judgments of value are presupposed by the body of scientific knowledge, which might be represented by a system of statements accepted in accordance with the rules of scientific inquiry. Here presupposing has to be understood in a systematic-logical sense. One such sense is invoked when we say, for example, that the statement ‘Henry’s brother-in-law is an engineer’ presupposes that Henry has a wife or a sister: in this sense, a statement presupposes whatever can be logically inferred from it. But, as was noted earlier, no set of scientific statements logically implies an unconditional judgment of value; hence, scientific knowledge does not, in this sense, presuppose valuation. There is another logical sense of presupposing, however. We might say, for example, that in Euclidean geometry the angle-sum theorem for

Douglas and Hempel

319

triangles presupposes the postulate of the parallels in the sense that that postulate is an essential part of the basic assumptions from which the theorem is deduced. Now, the hypotheses and theories of empirical science are not normally validated by deduction from supporting evidence (though it may happen that a scientific statement, such as a prediction, is established by deduction from a previously ascertained, more inclusive set of statements); rather, as was mentioned in section II, they are usually accepted on the basis of evidence that lends them only partial, or “inductive,” support. But in any event it might be asked whether the statements representing scientific knowledge presuppose valuation in the sense that the grounds on which they are accepted include, sometimes or always, certain unconditional judgments of value. Again the answer is in the negative. The grounds on which scientific hypotheses are accepted or rejected are provided by empirical evidence, which may include observational findings as well as previously established laws and theories, but surely no value judgments. Suppose for example that, in support of the hypothesis that a radiation belt of a specified kind surrounds the earth, a scientist were to adduce, first, certain observational data, obtained perhaps by rocket-borne instruments; second, certain previously accepted theories invoked in the interpretation of those data; and finally, certain judgments of value, such as ‘it is good to ascertain the truth’. Clearly, the judgments of value would then be dismissed as lacking all logical relevance to tile proposed hypothesis since they can contribute neither to its support nor to its disconfirmation. But the question whether science presupposes valuation in a logical sense can be raised, and recently has been raised, in yet another way, referring more specifically to valuational presuppositions of scientific method. In the preceding considerations scientific knowledge was represented by a system of statements which are sufficiently supported by available evidence to be accepted in accordance with the principles of scientific test and validation. We noted that as a rule the observational evidence on which a scientific hypothesis is accepted is far from sufficient to establish that hypothesis conclusively. For example, Galileo’s law refers not only to past instances of free fall near the earth, but also to all future ones; and the latter surely are not covered by our present evidence. Hence, Galileo’s law, and similarly any other law in empirical science, is accepted on the basis of incomplete evidence. Such accep-

320

Siblings Under the Skin

tance carries with it the “inductive risk” that the presumptive law may not hold in full generality, and that future evidence may lead scientists to modify or abandon it. A precise statement of this conception of scientific knowledge would require, among other things, the formulation of rules of two kinds: First, rules of confirmation, which would specify what kind of evidence is confirmatory, what kind disconfirmatory for a given hypothesis. Perhaps they would also determine numerical degree of evidential support (or confirmation, or inductive probability) which a given body of evidence could be said to confer upon a proposed hypothesis. Secondly, there would have to be rules of acceptance: these would specify how strong the evidential support for a given hypothesis has to be if the hype. thesis is to be accepted into the system of scientific knowledge; or, more generally, under what conditions a proposed hypothesis is to be accepted, under what conditions it is to be rejected by science on the basis of a given body of evidence. Recent studies of inductive inference and statistical testing have devoted a great deal of effort to the formulation of adequate rules of either kind. In particular, rules of acceptance have been treated in many of these investigations as special instances of decision rules of the sort mentioned in the preceding section. The decisions in question are here either to accept or to reject a proposed hypothesis on the basis of given evidence. As was noted earlier, the formulation of “adequate” decision rules requires, in any case, the antecedent specification of valuations that can then serve as standards of adequacy. The requisite valuations, as will be recalled, concern the different possible outcomes of tile choices which the decision rules are to govern. Now, when a scientific rule of acceptance is applied to a specified hypothesis on the basis of a given body of evidence, the possible “outcomes” of the resulting decision may be divided into four major types: (1) the hypothesis is accepted (as presumably true) in accordance with the rule and is in fact true; (2) the hypothesis is rejected (as presumably false) in accordance with the rule and is in fact false; (3) the hypothesis is accepted in accordance with the rule, but is in fact false; (4) the hypothesis is rejected in accordance with the rule, but is in fact true. The former two cases are what science aims to achieve; the possibility of the latter two represents the inductive risk that any acceptance rule must involve. And the problem of formu-

Douglas and Hempel

321

lating adequate rules of acceptance and rejection has no clear meaning unless standards of adequacy have been provided by assigning definite values or disvalues to those different possible “outcomes” of acceptance or rejection. It is in this sense that the method of establishing scientific hypotheses “presupposes” valuation: the justification of the rules of acceptance and rejection requires reference to value judgments. In the cases where the hypothesis under test, if accepted, is to be made the basis of a specific course of action, the possible outcomes may lead to success or failure of the intended practical application; in these cases, the values and disvalues at stake may well be expressible in terms of monetary gains or losses; and for situations of this sort, the theory of decision functions has developed various decision rules for use in practical contexts such as industrial quality control. But when it comes to decision rules for the acceptance of hypotheses in pure scientific research, where no practical applications are contemplated, the question of how to assign values to the four types of outcome mentioned earlier becomes considerably more problematic. But in a general way, it seems clear that the standards governing the inductive procedures of pure science reflect the objective of obtaining a certain goal, which might be described somewhat vaguely as the attainment of an increasingly reliable, extensive, and theoretically systematized body of information about the world. Note that if we were concerned, instead, to form a system of beliefs or a world view that is emotionally reassuring or esthetically satisfying to us, then it would not be reasonable at all to insist, as science does, on a close accord between the beliefs we accept and our empirical evidence; and the standards of objective testability and confirmation by publicly ascertainable evidence would have to be replaced by acceptance standards of an entirely different kind. The standards of procedure must in each case be formed in consideration of the goals to be attained; their justification must be relative to those goals and must, in this sense, presuppose them. VII. Concluding comparisons If, as has been argued in section IV, science cannot provide a validation of categorical value judgments, can scientific method and knowledge play any role at all in clarifying and resolving problems of moral

322

Siblings Under the Skin

valuation and decision? The answer is emphatically in the affirmative. I will try to show this in a brief survey of the principal contributions science has to offer in this context. First of all, science can provide factual information required for the resolution of moral issues. Such information will always be needed, for no matter what system of moral values we may espouse—whether it be egoistic or altruistic, hedonistic or utilitarian, or of any other kind— surely the specific course of action it enjoins us to follow in a given situation will depend upon the facts about that situation; and it is scientific knowledge and investigation that must provide the factual information which is needed for the application of our moral standards. More specifically, factual information is needed, for example, to ascertain (a) Whether a contemplated objective can be attained in a given situation; (b) if it can be attained, by what alternative means and with what probabilities; (c) what side effects and ulterior consequences the choice of a given means may have apart from probably yielding the desired end; (d) whether several proposed ends are jointly realizable, or whether they are incompatible in the sense that the realization of some of them will definitely or probably prevent the realization of others. By thus giving us information which is indispensable as a factual basis for rational and responsible decision, scientific research may well motivate us to change some of our valuations. If we were to discover, for example, that a certain kind of goal which we had so far valued very highly could be attained only at the price of seriously undesirable side effects and ulterior consequences, we might well come to place a less high value upon that goal. Thus, more extensive scientific information may lead to a change in our basic valuations—not by “disconfirming” them, of course, but rather by motivating a change in our total appraisal of the issues in question. Secondly, and in a quite different manner, science can illuminate certain problems of valuation by an objective psychological and sociological study of the factors that affect the values espoused by an individual or a group; of the ways in which such valuational commitments change; and perhaps of the manner in which the espousal of a given value system may contribute to the emotional security of an individual or to the functional stability of a group.

Douglas and Hempel

323

Psychological, anthropological, and sociological studies of valuational behavior cannot, of course, “validate” any system of moral standards. But their results can psychologically effect changes in our outlook on moral issues by broadening our horizons, by making us aware of alternatives not envisaged, or not embraced, by our own group, and by thus providing some safeguard against moral dogmatism or parochialism. Finally, a comparison with certain fundamental aspects of scientific knowledge may help to illuminate some further questions concerning valuation. If we grant that scientific hypotheses and theories are always open to revision in the light of new empirical evidence, are we not obliged to assume that there is another class of scientific statements which cannot be open to doubt and reconsideration, namely, the observational statements describing experiential findings that serve to test scientific theories? Those simple, straightforward reports of what has been directly observed in the laboratory or in scientific field work, for example—must they not be regarded as immune from any conceivable revision, as irrevocable once they have been established by direct observation? Reports on directly observed phenomena have indeed often been considered as an unshakable bedrock foundation for all scientific hypotheses and theories. Yet this conception is untenable; even here, we find no definitive, unquestionable certainty. For, first of all, accounts of what has been directly observed are subject to error that may spring from various physiological and psychological sources. Indeed, it is often possible to check on the accuracy of a given observation report by comparing it with the reports made by other observers, or with relevant data obtained by some indirect procedure, such as a motion picture taken of the finish of a horse race; and such comparison may lead to the rejection of what had previously been considered as a correct description of a directly observed phenomenon. We even have theories that enable us to explain and anticipate some types of observational error, and in such cases, there is no hesitation to question and to reject certain statements that purport simply to record what has been directly observed. Sometimes relatively isolated experimental findings may conflict with a theory that is strongly supported by a large number and variety

324

Siblings Under the Skin

of other data; in this case, it may well happen that part of the conflicting data, rather than the theory, is refused admission into the system of accepted scientific statements—even if no satisfactory explanation of the presumptive error of observation is available. In such cases it is not the isolated observational finding which decides whether the theory is to remain in good standing, but it is the previously well-substantiated theory which determines whether a purported observation report is to be regarded as describing an actual empirical occurrence. For example, a report that during a spiritualistic séance, a piece of furniture freely floated above the floor would normally be rejected because of its conflict with extremely well confirmed physical principles, even in the absence of some specific explanation of the report, say, in terms of deliberate fraud by the medium, or of high suggestibility on the part of the observer. Similarly, the experimental findings reported by the physicist Ehrenhaft, which were claimed to refute the principle that all electric charges are integral multiples of the charge of the electron, did not lead to the overthrow, nor even to a slight modification, of that principle, which is an integral part of a theory with extremely strong and diversified experimental support. Needless to say, such rejection of alleged observation reports by reason of their conflict with well-established theories requires considerable caution; otherwise, a theory, once accepted, could be used to reject all adverse evidence that might subsequently be found—a dogmatic procedure entirely irreconcilable with the objectives and the spirit of scientific inquiry. Even reports on directly observed phenomena, then, are not irrevocable; they provide no bedrock foundation for the entire system of scientific knowledge. But this by no means precludes the possibility of testing scientific theories by reference to data obtained through direct observation. As we noted, the results obtained by such direct checking cannot be considered as absolutely unquestionable and irrevocable; they are themselves amenable to further tests which may be carried out if there is reason for doubt. But obviously if we are ever to form any beliefs about the world, if we are ever to accept or to reject, even provisionally, some hypothesis or theory, then we must stop the testing process somewhere; we must accept some evidential statements as sufficiently trustworthy not to require further investigation for the time being. And on the basis of such evidence, we can then decide what

Douglas and Hempel

325

credence to give to the hypothesis under test, and whether to accept or to reject it. This aspect of scientific investigation seems to me to have a parallel in the case of sound valuation and rational decision. In order to make a rational choice between several courses of action, we have to consider, first of all, what consequences each of the different alternative choices is likely to have. This affords a basis for certain relative judgments of value that are relevant to our problem. If this set of results is to be attained, this course of action ought to be chosen; if that other set of results is to be realized, we should choose such and such another course; and so forth. But in order to arrive at a decision, we still have to decide upon the relative values of the alternative sets of consequences attainable to us; and this, as was noted earlier, calls for the acceptance of an unconditional judgment of value, which will then determine our choice. But such acceptance need not be regarded as definitive and irrevocable, as forever binding for all our future decisions: an unconditional judgment of value, once accepted, still remains open to reconsideration and to change. Suppose, for example, that we have to choose, as voters or as members of a city administration, between several alternative social policies, some of which are designed to improve certain material conditions of living, whereas others aim at satisfying cultural needs of various kinds. If we are to arrive at a decision at all, we will have to commit ourselves to assigning a higher value to one or the other of those objectives. But while the judgment thus accepted serves as an unconditional and basic judgment of value for the decision at hand, we are not for that reason committed to it forever—we may well reconsider our standards and reverse our judgment later on; and though this cannot undo the earlier decision, it will lead to different decisions in the future. Thus, if we are to arrive at a decision concerning a moral issue, we have to accept some unconditional judgments of value; but these need not be regarded as ultimate in the absolute sense of being forever binding for all our decisions, anymore than the evidence statements relied on in the test of a scientific hypothesis need to be regarded as forever irrevocable. All that is needed in either context are relative ultimates, as it were: a set of judgments—moral or descriptive—which are accepted at the time as not in need of further scrutiny. These relative ultimates permit us to keep an open mind in regard to the possibility of making changes in

326

Siblings Under the Skin

our heretofore unquestioned commitments and beliefs; and surely the experience of the past suggests that if we are to meet the challenge of the present and the future, we will more than ever need undogmatic, critical, and open minds.

References Hempel, C. G. 1965. Aspects of scientific explanation. In Aspects of scientific explanation and other essays in the philosophy of science. New York: The Free Press. Myrdal, Gunnar. [1944] 1962. An American dilemma: The Negro problem and modern democracy. New York: Harper and Row. Simpson, G.G. 1949. The meaning of evolution. New Haven: Yale University Press.

Biographical note Carl G. Hempel was born near Berlin on January 8, 1905. Beginning in 1923, he studied at the Universities of Gottingen, Heidelberg, and Berlin. While at Berlin, he studied with Hans Reichenbach, who encouraged him to go to Vienna for the 1929-1930 academic year, where he studied with Carnap and met with the Vienna Circle. He then returned to Berlin, earning his doctorate in philosophy in 1934. He then left Germany for Brussels, where he collaborated with Paul Oppenheim. In 1937-1938, Carnap brought Hempel to the University of Chicago to act as his research associate. After a brief return to Brussels in 1938, Hempel emigrated permanently to the U.S. in 1939. For the first year, he taught at City University in New York. Then from 1940 to 1948, he taught at Queen’s University in New York and published

Douglas and Hempel

327

some his most influential work on confirmation and explanation. In 1948, Hempel moved to Yale University, and then in 1955, to Princeton as the Stuart Professor of Philosophy. Until mandatory retirement in 1973, he remained at Princeton, producing his Aspects of Scientific Explanation (1965) and his textbook, Philosophy of Natural Science (1966). He continued to teach after retirement at Berkeley, Irvine, and Jerusalem, and in 1977, accepted an appointment as University Professor at Pittsburgh. He retired from Pittsburgh in 1985, and moved back to Princeton, New Jersey, where he remained philosophically active, until his death on November 9, 1997.