Scientism: prospects and problems 9780190462758, 0190462752

Table of contents :
Introduction: Putting Scientism on the Philosophical Agenda Rene van Woudenberg, Rik Peels, and Jeroen de Ridder1. A Conceptual Map of Scientism Rik Peels2. Scientism and Its Rivals Mikael Stenmark3. Philosophical Challenges for Scientism (And How to Meet Them?) Alex Rosenberg4. Scientism with a Humane Face James Ladyman5. Philosophy, Science, and Common Sense Hillary Kornblith6. Is Scientism Epistemically Vicious? Ian James Kidd7. An Epistemological Critique of Scientism Rene van Woudenberg8. Kinds of Knowledge, Limits of Science Jeroen de Ridder9. Scientism: Who Needs It? Alvin Plantinga10. Cognitive Science and Moral Philosophy: Challenging Scientistic Overreach William FitzPatrick11. Physicalism, not Scientism Alyssa Ney12. Moderate Scientism in Philosophy Wesley Buckwalter and John Turri

Citation preview

i

SCIENTISM

ii

iii

SCIENTISM Prospects and Problems

Edited by Jeroen de Ridder, Rik Peels, and René van Woudenberg

1

iv

1 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries. Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America. © Oxford University Press 2018 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer. CIP data is on file at the Library of Congress ISBN 978–​0–​19–​046275–​8 9 8 7 6 5 4 3 2 1 Printed by Sheridan Books, Inc., United States of America

v

CONTENTS

Acknowledgments

vii

Contributors

ix

Introduction: Putting Scientism on the Philosophical Agenda—​R ené van Woudenberg , Rik Peels , and Jeroen de Ridder

1

1. A Conceptual Map of Scientism—​R ik Peels

28

2. Scientism and Its Rivals—​M ikael Stenmark

57

3. Philosophical Challenges for Scientism (and How to Meet Them?)—​A lex Rosenberg

83

4. Scientism with a Humane Face—​James Ladyman

106

5. Philosophy, Science, and Common Sense—​H ilary Kornblith

127

6. Is Scientism Epistemically Vicious?—​I an James Kidd

149

7. An Epistemological Critique of Scientism—​R ené van Woudenberg

167

8. Kinds of Knowledge, Limits of Science—​J eroen de Ridder

190

9. Scientism: Who Needs It?—​A lvin Plantinga

220

10. Cognitive Science and Moral Philosophy: Challenging Scientistic Overreach—​W illiam FitzPatrick

233

vi

v i   •  C o n t e n t s

11. Physicalism, Not Scientism—​A lyssa Ney

258

12. Moderate Scientism in Philosophy—​W esley Buckwalter and John Turri

280

Index

301

vi

ACKNOWLEDGMENTS

The editors would like to thank the Templeton World Charity Foundation for their generous support for the research project Science Beyond Scientism, of which this book is a major result. Without it, work on this publication would not have been possible. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of Templeton World Charity Foundation. We are grateful to the speakers and participants at the 2014 conference All You Need Is Science? at Vrije Universiteit Amsterdam, who provided invaluable input for several of the contributions to the present volume. We are also thankful to all our colleagues, collaborators, and students in the Science Beyond Scientism project—​Gijsbert van den Brink, Gerrit Glas, Leon de Bruin, Kelvin McQueen, Miriam Kyselo, Terence Cuneo, Russ Shafer-​ Landau, Lieke Asma, Hans van Eyghen, Naomi Kloosterboer, Josephine Lenssen, and Scott Robbins—​for many hours of excellent discussions, productive collaboration, and philosophical companionship. Our consecutive project and event managers, Marije Zeldenrijk and Irma Verlaan, deserve a big thank you. Without them, we would almost certainly have gotten lost in the practicalities of running a project of this size. Finally, we’re grateful to Peter Ohlin, Emily Sacharin, and Isla Ng at Oxford University Press for their unfailing support (and patience) in making this volume possible.

vi

ix

CONTRIBUTORS

Wesley Buckwalter, University of Pittsburgh Jeroen de Ridder, Vrije Universiteit Amsterdam William FitzPatrick, University of Rochester Ian James Kidd, University of Nottingham Hilary Kornblith, University of Massachusetts James Ladyman, University of Bristol Alyssa Ney, University of California, Davis Rik Peels, Vrije Universiteit Amsterdam Alvin Plantinga, University of Notre Dame/​Calvin College Alex Rosenberg, Duke University Mikael Stenmark, Uppsala University John Turri, University of Waterloo René van Woudenberg, Vrije Universiteit Amsterdam

x

xi

SCIENTISM

xi

1

INTRODUCTION

Putting Scientism on the Philosophical Agenda René van Woudenberg, Rik Peels, and Jeroen de Ridder

I.1 Riddles of Scientism This book discusses prospects and problems of scientism. Scientism is, roughly, the view that only science can provide us with knowledge or rational belief, that only science can tell us what exists, and that only science can effectively address our moral and existential questions.1 As Alex Rosenberg says, scientism “is the conviction that the methods of science are the only reliable ways to secure knowledge of anything,” the view that “science provides all the significant truths about reality” (Rosenberg 2011:  6–​7). Or, as Leslie Stevenson and Henry Byerly have it, it is “the view that knowledge obtainable by scientific method exhausts all knowledge . . . that whatever is not mentioned in the theories of science does not exist or has only a subordinate, secondary kind of reality” (Stevenson and Byerly 2000: 246–​247).2 Scientism can thus stand for a number of exclusivity claims about science. 1. Rik Peels’s chapter in the present volume identifies several versions of scientism in more systematic detail. 2. Similar definitions abound in the literature: “Scientism is the belief that all valid knowledge is science. Scientism says, or at least implicitly assumes, that rational knowledge is scientific, and everything else that claims the status of knowledge is just superstition, irrationality, emotion, or nonsense” (Hutchinson 2011:  1). And: “A totalizing attitude that regards science as the ultimate standard and arbiter of all interesting questions; or alternatively that seeks to expand the very definition and scope of science to encompass all aspects of human knowledge and understanding” (Pigliucci 2013: 144).

2

2   • 

René van Woudenberg, Rik Peels, and Jeroen de Ridder

More often than not, scientism is adopted implicitly and at best half-​wittingly. The view is usually not proposed or adopted on the basis of carefully crafted and explicitly formulated arguments in its favor. It is a view that appears to be more “in the air” than pinned down on paper as a philosophical position.3 Moreover, friends of the view (or rather family of views) only rarely embrace the label “scientism.” Thus, we have a view, scientism, which is widely but often implicitly adopted, that is rarely stated explicitly or defended rigorously, and that mostly does not go by the name “scientism.” These are the riddles of scientism. Fortunately, there is a ready solution to at least the last of these riddles. “Scientism” used to figure exclusively as a term of abuse. Labeling a view as “scientistic,” or a person an advocate of scientism, was to issue an accusation, or worse, to make an insult. Susan Haack, for example, describes “scientism” as “an exaggerated kind of deference towards science, an excessive readiness to accept as authoritative any claim made by the sciences, and to dismiss every kind of criticism of science or its practitioners as anti-​scientific prejudice” (Haack 2007: 17–​18). It is clear that no one will accept this notion of “scientism” as an adequate characterization of their own views, as no one will think that their deference to science is exaggerated, or their readiness to accept claims made by the sciences is excessive. Haack surely latches on to a standard, pejorative, use of “scientism.” Sometimes, however, active attempts are made to rid words of their pejorative overtones and to re-​appropriate them as badges of honor. An example is “impressionist,” a term once used to express abhorrence about a characteristic style of painting, but soon consciously self-​applied with pride. What we see nowadays—​and this is the solution for one of the riddles of scientism—​is that something similar is happening with “scientism.” Various philosophers self-​ consciously present themselves as advocates of scientism, for instance, Alex Rosenberg and James Ladyman—​both of whom contribute to this volume. It is in this use that we are interested: the problems and prospects of scientism as a serious philosophical position, rather than a derogatory label used to dismiss one’s opponents with a quick gesture.4,5

3. The subtitle of another recent collection of essays devoted to a critical discussion of scientism even goes so far as to label scientism “the new orthodoxy” (Williams and Robinson 2015). 4. We do not mean to accuse Haack of this. Her carefully argued criticisms of scientism are anything but a quick dismissal of it. 5. This recent use of scientism as a more neutral term or a badge of honor is one reason why we think scientism merits attention now. While older pioneering works on the topic (such as

3

Introduction  • 

3

This volume aims to contribute to understanding scientism as an explicitly formulated position by giving an overview of the claims and ideas that constitute the view, by signaling the varieties of scientism and their interrelations, by presenting and discussing arguments in favor of scientism, as well as by presenting criticisms of the view(s). In this introduction, we explain why scientism is an important and timely topic, both inside and outside of philosophy, and both inside and outside of academia. Next, in order to bring scientism into sharper focus, we sketch some of the (pre)history of scientism, explain what advocates of scientism take science to be, summarize the most important argument that is used to buttress scientism, and give an overview of how critics have responded to it. Finally, we present an overview of the coming chapters.

I.2 Why Scientism Should Be on the Philosophical Agenda There are at least five reasons why an in-​depth discussion of scientism is needed. First, scientism, if true, has devastating epistemic consequences. If science is our only source of knowledge or rational belief, we do not know and do not rationally believe many of the things that we take ourselves to know and rationally believe. Most people take themselves to know and rationally believe lots of things about their mental life through introspection; and they also assume they know and rationally believe a great number of things about their everyday experiences; and many also presume they have rational beliefs or even knowledge about political, moral as well as religious matters. To the extent that such presumed knowledge and rational belief is not supported by science—​nor capable of being so supported—​it would all have to go. Much of our ordinary epistemic self-​understanding is at stake here. Second and related, if only science provides rational belief and knowledge, then the humanities don’t have anything to offer that is of epistemic value. Thus, no rational belief or knowledge is to be found in the interpretations of texts; in accounts of historical developments; in philosophical theories about, for example, value, personhood, and rationality; in aesthetics or art theory, linguistics, and so on. At the very least, then, scientism should lead to a radically different self-​interpretation of the humanities. For what is it one

Sorrell 1991, Trigg 1993, and Stenmark 2001) contain much that is valuable and interesting, they did not anticipate this development.

4

4   • 

René van Woudenberg, Rik Peels, and Jeroen de Ridder

is doing in studying and analyzing something when one holds that there is no knowledge or rational belief to be gained from doing so? But maybe scientism implies that more drastic measures are in order. Since these academic disciplines don’t deliver knowledge or rational belief, they should perhaps be banned from universities altogether. Obviously, that would have radical implications for future research and higher education. Third, the truth of scientism would have severe repercussions for a number of well-​established social practices. We single out two: the legal and the psychiatric practice. Common sense tells us that psychologically healthy human beings have free will and bear responsibility for their freely performed actions. However, a vocal minority urges that science has given us strong reason to deny that we have free will. But if belief in free will lacks positive epistemic standing, then it is high time for us to revisit our legal practice of holding people responsible for at least some of their actions, and of assessing their behavior on the basis of common-​sense beliefs about freedom and responsibility. Perhaps we should treat humans as beings whose behavior is to be manipulated so that the odds of good behavior are maximized and those of bad behavior minimized and abandon our talk of crime and punishment. Second, in psychiatry scientism finds expression in how the DSM (Diagnostic and Statistical Manual of Mental Disorders) is sometimes applied. The DSM describes the criteria of a few hundred psychiatric disorders. Often, different criteria are given, not all of which have to be satisfied in order to diagnose someone with an illness. In addition, criteria for different types of illness frequently overlap, so that patients can easily be diagnosed with more than one illness. Psychiatrists thus have to tread carefully when working with the DSM and apply its criteria judiciously. While most psychiatrists are acutely aware of this, in other contexts the DSM is treated very differently. DSM diagnoses sometimes play a crucial role in forensic contexts or in decisions about medical insurance. In those contexts, DSM categories are treated in much the same way as the categories of the periodic system of elements, viz. as descriptions of fixed realities. Whoever satisfies the criteria of an illness is considered to “really” have it. This scientistic use of the DSM has real-​world effects: It may or may not exculpate someone before a jury or judge and it may or may not lead to compensation for psychiatric treatments or medication. Fourth, scientism used to be advocated by scientists or philosophers in academic venues. Nowadays, however, scientistic ideas are made available to a much larger audience in popular venues, such as popular science books, newspapers, blogs, and online discussion forums, which has led to what we could call a scientistic “pop-​epistemology.” By that, we mean a somewhat

5

Introduction  • 

5

simplified scientistic epistemology that is articulated and embraced by many non-​academics. This has led to public debates in favor of and against scientism. Some of the more visible contributions include the following: Sam Harris’s book The Moral Landscape; Alex Rosenberg’s book The Atheist’s Guide to Reality; Curtis White’s book The Science Delusion; Steven Pinker’s essay “Science Is Not Your Enemy” in The New Republic;6 Leon Wieseltier’s and Philip Kitcher’s replies in The New Republic: “Crimes against Humanities” and “The Trouble with Scientism”;7 • Adam Frank’s essay “The Power of Science and the Danger of Scientism” at NPR;8 • Gloria Origgi’s essay “The Humanities Are Not Your Enemy” in The Berlin Review of Books;9 • Oliver Burkeman’s column “ ‘Scientism’ Wars” in The Guardian;10 • • • • •

What is lacking in these discussions, however, is a careful and systematic evaluation by professional philosophers. Since scientism is essentially a philosophical position, philosophers should rise to the occasion.11 Fifth, scientism can affect the prestige of  and public trust in science. Adherents of scientism obviously have a great deal of faith in science and recommend this attitude for everyone. While this may be a good thing, the worry is that it may also backfire. To the extent that scientism raises the expectations of what science can deliver too high, it paves the way for disillusionment and frustration, which can easily contribute to undermining public trust in science. This might happen in a number of ways. By prematurely presenting provisional scientific results and theories as if they have

6. http://​newrepublic.com/​article/​114127/​science-​not-​enemy-​humanities 7.  Wieseltier: http://​newrepublic.com/​article/​114548/​leon-​wieseltier-​responds-​steven-​pinkers-​ scientism; Kitcher: http://​newrepublic.com/​article/​103086/​scientism-​humanities-knowledge-​theory​everything-​arts-​science 8.  http://​www.npr.org/​sections/​13.7/​2013/​08/​13/​211613954/​the-​power-​of-​science-​and-​the-​ danger-​of-​scientism 9. http://​berlinbooks.org/​brb/​2013/​09/​the-​humanities-​are-​not-​your-​enemy/​ 10. http://​web.archive.org/​web/​20150527003228/​http://​www.theguardian.com/​news/​ oliver-​burkeman-​s-​blog/​2013/​aug/​27/​scientism-​wars-​sam-​harris-​elephant 11. This increasing airtime of scientistic views in public discourse is another reason why scrutiny of scientism is called for now. Cf. note 5 above.

6

6   • 

René van Woudenberg, Rik Peels, and Jeroen de Ridder

been established beyond doubt, some adherents of scientism risk having to backpedal on earlier claims. This can give the false impression that scientific ideas are more in flux than they in fact are and that there is hardly any progress in science—​that it’s all “just theories.” Also, some adherents of scientism entangle scientific results with naturalistic or physicalistic worldviews in which humans are “nothing but” the atoms that make them up, their brains, their unconscious impulses, etc., and in which the universe is devoid of meaning and purpose. Even if such worldviews should ultimately turn out to be correct, it is too early now to proclaim that we know so for sure. Since many people find these worldviews unattractive, they become distrustful of the science from which it allegedly follows. So, there is more than enough reason to put scientism on the philosophical agenda.12

I.3 Scientism’s (Pre)History Science has left its marks all over the world. We see the stamp of science on technological artifacts like computers and bombs; on institutions like hospitals, armies, and journalism; on the physical face of the world: roads, islands on the coast of Dubai, skyscrapers, polluted rivers, destroyed natural sites; and on what we think about the world and ourselves:  We no longer believe in geocentrism and we think our bodies are composed of countless small and complex cells. Without science, the world would have looked very different from the way it in fact looks, its history over the last 500 years would have been very different from its actual history, and we wouldn’t have thought and believed a vast array of things that we now think and believe. All of this can be taken to indicate that from the scientific revolution onwards science has been spectacularly successful. The perceived success of science has inspired many philosophers, scientists, policy makers, and proverbial “men in the street” to put ever higher hopes on science and its deliverances. This isn’t scientism, but it can herald it. Glimmerings of scientism can be found in shattered remarks of many philosophers. We single out David Hume, August Comte, and the

12. Indeed, it appears that we were not the only ones sensing the urgency. The current volume has been a long time in the making and since its initial inception, three other collections of essays devoted to analyses and discussions of scientism have appeared: Williams and Robinson (2015), Beale and Kidd (2017), and Boudry and Pigliucci (2018), as well as a new e-​book by Susan Haack (2017).

7

Introduction  • 

7

logical positivists. In a famous exhortation in his Enquiry Concerning Human Understanding Hume says: When we run over libraries, persuaded of these principles, what havoc must we make? If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it to the flames: for it can contain nothing but sophistry and illusion.13 Although there is no explicit use of the term here, a number of chords are struck that are close to the heart of scientism, and that, we will see, recur over and over: opposition to theology and metaphysics, and allegiance to empirical modes of inquiry. To be sure, Hume was by no means an advocate of scientism; the skeptical ferment in his work is far too strong for it. Also, his remarks about abstract reasoning concerning quantity and number don’t sit well with scientism’s concentration on empirical modes of inquiry. Still, the Hume quotation does convey part of the spirit that animates scientism. More than a glimmering of scientism can be found in the works of the French philosopher Auguste Comte (1798–​ 1857), who formulated his famous Law of Three Stages:  “The law is this:  that each of our leading conceptions—​each branch of knowledge—​passes through three different theoretical conditions:  the Theological, or fictitious; the Metaphysical, or abstract; and the Scientific, or positive” (Comte 2009 [1853]: 1). The theological condition, Comte explains, is the stage in which humans, in search of first and final causes, explain the universe and its constituents by reference to supernatural agents. This stage, which is the necessary starting point of the human cognitive endeavor gradually develops into the metaphysical stage in which supernatural causes are replaced by “abstract forces” as the items that perform explanatory functions: final causes, form and matter, etc. The second stage, Comte tells us, is merely transitional, as the human mind is unable to jump directly from the theological to the positive mode of thinking. The third, positive stage, which Comte thought was breaking through in his own days, is the stage in which humans realize that laws govern both nature and human behavior, and that these laws can be discovered through observation 13. Quoted by Alfred J. Ayer in his introduction to a volume with key papers from the logical positivist movement (Ayer 1959: 10). The quotation is taken from the concluding paragraph of Hume’s Enquiry Concerning Human Understanding (1975 [1777]: 165).

8

8   • 

René van Woudenberg, Rik Peels, and Jeroen de Ridder

and reasoning. In this stage it is believed that science is key to all our cognitive endeavors. Comte did acknowledge that there is a hierarchy between the sciences, ranging from mathematics, being the most perfect science, through astronomy, to terrestrial physics, chemistry, and physiology. He conceived of physiology as the basis of sociology, which studies social structures, and which he held will enable us to terminate the crises in which many civilized countries found themselves. These Comtean ideas cry out for stage setting, precisification, and comment. But this is not the occasion. We want to draw attention to some themes close to the heart of scientism: averseness to theology and metaphysics (conceived of as the study of what lies “behind” what can be experienced), and a concentration on “positive facts,” that is, facts that are revealed in direct sense experience. Comte saw mathematics as the most perfect science, as it delivers theorems that, if true, are necessarily true. Comte thus acknowledged the validity of a priori reasoning, which fits poorly with scientism’s concentration on the empirical sciences. Friends of scientism, traditionally, have problems with mathematics and logic—​as these appear to have non-​empirical subject matters. Yet, the spirit of scientism animates Comte’s works.14 A next phase in the (pre)history of scientism is the logical positivist’s movement associated with the Vienna Circle that started in the 1920s. Among its many well-​known members were Moritz Schlick, Rudolf Carnap, Hans Hahn, and Felix Kaufmann. The logical positivists, famously, made a distinction between two classes of significant propositions, reminiscent of Hume’s exhortation quoted above. On the one hand, there are factual propositions, the hallmark of which is that they are empirically verifiable. On the other, there are formal propositions (of logic and mathematics), the hallmark of which is that they are tautological in the sense that they are true in virtue of the meaning of the words that compose them, and not in virtue of the way the world is. This bifurcation was held to be exhaustive. If a sentence expresses neither a formal truth or falsehood, nor something that is verifiable, it was held to express no proposition at all. Such sentences were deemed literally nonsensical, even though some of them give the impression of expressing factual propositions. The positivists held that the prime examples of factual propositions are the propositions of science, while prime examples of nonsensical propositions are the propositions of metaphysics, 14. Or it did so up to the time that he started devising a new world religion, of which he himself would be the high priest—​that project is much less congenial to scientistic tenets, to say the least.

9

Introduction  • 

9

ethics, and religion—​so propositions such as that the world is spiritual, that torturing others for the fun of it is morally wrong, or that the actual world is the best of all possible worlds. Whatever merits such nonsensical statements may have on an emotional level, as they don’t state facts, they have no truth value. The famous Verification Criterion was proposed as an instrument to reliably distinguish between factual and nonsensical statements. It said that no statement is meaningful unless it is either a tautology or in principle empirically verifiable.15 One of the most-​recounted parts of 20th-​century philosophy is how logical positivism developed and ran into problems (see, e.g., Godfrey-​Smith 2003: 19–​121; Losee 2001: 143–​196). It did so, first, because the positivists, in staying true to the spirit of empiricism, tended to be suspicious of theories that posit unobservable entities (such as electrons, quarks, or positrons), as such theories go beyond what can be observed. Such antirealist tendencies did not square well with the widespread realism among actual practitioners of the sciences. Scientists normally think of themselves as trying to find out what the structures and mechanisms behind the observable phenomena are. Positivist antirealism doesn’t seem true to scientific practice. Relatedly, positivism’s focus on the observable arguably leaves little room for what many think is the glory and singular significance of science: providing explanations. Science describes phenomena, it doesn’t explain them, the positivists said. Second, logical positivism ran into trouble because of at least three problems that beset the Verification Criterion. A first problem is that a number of principles that seem basic to science are not empirically testable (and neither are they tautologies). That nature is uniform, for example, cannot be established by experiment, as it must be presupposed by every experiment. This means that a principle that is essential to science is classified as meaningless by the Verification Criterion that was supposed to demarcate scientific from metaphysical statements. A  second problem is that the Criterion declares many statements cognitively empty that seem perfectly meaningful and possibly even true, such as moral statements. This raised a dilemma: Should the Criterion be discarded, or should moral and other seemingly meaningful statements be declared meaningless? Many have maintained that it is more rational to think that moral statements are cognitively meaningful than to think that the Criterion is correct. Third and most devastating, the Principle

15. Ayer (1946: 35–​4 0) offers a formulation of the criterion. An informative discussion of the vicissitudes of its various formulations is Hempel (1950).

10

1 0   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

was argued to be self-​referentially incoherent: If one accepts it, one thereby obtains a good reason to reject it. The Criterion itself is a statement. But it is not a statement that is empirically testable. And statements that are not empirically testable, says the Criterion, are meaningless unless they are tautologies. But the Criterion surely is no tautology either. Hence the statement that states the Criterion is meaningless (cf. Plantinga 1967: 162–​168). The spirit of logical positivism lives on the anti-​metaphysical and anti-​ theological stance of scientism, as is obvious in both Rosenberg’s and Ross and Ladyman’s scientism.16 And even though the Verification Criterion has been given up, new occupants for its role have been proposed, for example Ross, Ladyman, and Spurrett’s pragmatic verificationism (Ross, Ladyman, and Spurrett 2007: 29) and Herman Philipse’s Principle R (Philipse 2013: 97).17 Also, ethical and other normative statements continue to look problematic to adherents of scientism—​and for much the same reason as they were deemed problematic by the logical positivists. In many ways, then, scientism harks back to older philosophical movements and ideas. It has a history.

I.4 Scientism’s Science Scientism gives pride of place to science: Science is supposed to be our absolute authority in the cognitive domain and leaves no room for competitors. This thus puts a lot of pressure on the notion of science and, by implication, on the issue of how science is to be demarcated from non-​science. In what follows we will try to clarify the notion of science as it figures in scientism. We do this by, first, contrasting it with what others have held are epistemically upstanding but nonscientific modes of cognition, and next by contrasting it with a millennia-​old conception of science. As indicated above, scientism issues exclusivity claims about science. These claims primarily concern the natural sciences, with physics taking a unique place among them.18 The first thing to note about this claim is that the complement of the natural sciences—​that is, all cognitive endeavors that are not natural science—​is a very mixed bag. It contains rather obviously dubious 16. This point is discussed in more detail in René van Woudenberg’s chapter in this volume. 17. For discussion of this principle, see Van Woudenberg and Rothuizen–​van der Steen (2016). 18. This is explicitly the case in Ross, Ladyman, and Spurrett’s scientism, which involves the Principle of the Primacy of Physics (see their 2007:  38–​45). This is also true of Rosenberg (2011: 45–​70).

1

Introduction  • 

11

“fields” and “methods” such as alchemy, astrology, the reading of tarot cards or tea leaves, and crystal ball gazing. But it also contains fields and methods that are held in high regard by many and have a venerable tradition within academia, such as the various branches of the humanities, ethics, philosophy, and theology. It contains, moreover, perception, proprioception, introspection, memory, and other (alleged) everyday sources of knowledge. The contents of this mixed bag are all supposed to fall short of some (possibly complex) standard that the natural sciences don’t fall short of. It is in virtue of this that they, in contrast with the sciences, are supposed not to give us knowledge or rational belief. But what is this standard, and in virtue of what do the sciences satisfy it, while all these other things don’t? The natural sciences, or rather their products (so: empirical generalizations, statements of laws, predictions, theories, explanations, etc.) have been claimed to possess a number of characteristics that are unique to them. For instance, they have been supposed to be: (a) theoretical and explanatory as opposed to practical (Nagel 1961: 1–​14); (b) abstract as opposed to concrete (Dooyeweerd 1953–​1958); (c) general and lawlike, as opposed to individual and singular (Windelband 1894); (d) objective or value free, as opposed to value laden (Weber 1919); (e) obtained by the application of validated research methods, as opposed to rhapsodically obtained by common sense (Mill 1974 [1843]). We can think of this list as posing a challenge to scientism. For the features in virtue of which the products of the natural sciences are supposed to meet the standard, are possessed by the products of many of the items in the mixed bag (minus the dubious ones) as well. For instance, many explanations that we think are sound explanations, are not scientific explanations. You ask why the windows are wet; the explanation is that it has been raining. But that hardly qualifies as a scientific explanation—​unless we stretch the concept of “scientific explanation” in such a way that even small children can provide them. Also, non-​science sometimes concerns something abstract. Counting the natural numbers concerns abstract objects—​that is, objects not located in space-​time and causally inert. Yet counting the natural numbers is not science. Natural science is not limited to studying only what is general. The Big Bang is a singular event, as is the evolutionary history of life on earth. And again, science is by no means the value-​free enterprise that it is sometimes held to be. There are values that good scientific theories embody, such as simplicity, explanatory power, fertility, etc. Arguably, social and ethical values play

12

1 2   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

important indirect roles in science, for instance, in choosing which problems to study or in constraining acceptable methodological choices. And, finally, there is much that is brought to light by the application of validated methods of research but that isn’t science. Policing, calculating one’s share of a restaurant bill, figuring out how many rooms are rented out in a hotel by counting the keys that are “in” (while the policy of the hotel is to have guests leave their key at the counter when they go out), preparing a legal case to present for a jury or judge. None of this qualifies as doing science unless we stretch the meaning of science considerably. The point of saying all this is to re-​emphasize a point made forcefully by Larry Laudan (1983): It is well-​nigh impossible to demarcate science from non-​science. Still, in order for scientistic exclusivity claims to be more than innocuous rhetoric, the notion of “science” should be demarcated reasonably sharply from its complement. This is a significant challenge.19 This systematic challenge can be deepened by a historical challenge, deriving from the fact that scientism’s use of “science” departs from a historically extremely influential ideal of science, codified in what has been called the “classical model of science” (De Jong and Betti 2010). Let us explain. In the Western world, the classical model dominated philosophical thinking about science for at least two millennia. Three milestones in the model’s life are Aristotle’s Analytica Posteriora, especially book 1; the very influential so-​ called Logic of Port-​Royal (1662), especially part IV on method, written by Antoine Arnauld and relying in many respects on Pascal and Descartes; and, finally, Bernard Bolzano’s Wissenschafslehre (1837). According to the model, a science is a system S of propositions and concepts (or terms) that satisfies the following conditions:20 (1) All propositions and all concepts of S are about a certain domain of being(s). (2a) S contains a number of so-​called fundamental concepts. (2b) All other concepts occurring in S are composed of these fundamental concepts. (3a) S contains a number of fundamental propositions. (3b) All other propositions of S are provable or demonstrable from these fundamental propositions.

19. See Pigliucci and Boudry (2013) for various attempts to meet the demarcation challenge. 20. This is a slightly revised version of the model as presented in De Jong and Betti (2010: 186).

13

Introduction  • 

13

(4) All propositions in S are true. (5) All propositions in S are universal and necessary in some sense. (6) All propositions in S are known to be true. A non-​fundamental proposition is known to be true through its proof in S. (7) All concepts are adequately known. A non-​fundamental concept is adequately known through its composition. De Jong and Betti present the model as a reconstruction of an influential way of thinking about what a proper science is and what its methodology should be. It aims to summarize an ideal of proper science that has been defended by philosophers and scientists alike for over two millennia. Among its champions were such luminaries as Newton, Spinoza, Leibniz, Wolff, Kant, Husserl, Frege, and Lesniewski. Even a quick glance at the model indicates that it does not pick out current science, that is, it doesn’t pick out the referent of “science” as used by the advocates of scientism. According to the model, science is a deductive system of propositions that are derived from first principles, but current science, or the products thereof, don’t constitute such a system. Current science acknowledges precious few first principles, and involves numerous forms of non-​deductive reasoning such as induction, inference to the best explanation, probabilistic reasoning, statistical inference, and more. A second reason why the classical model doesn’t refer to what advocates of scientism have in mind when they refer to “science,” is that the model allows metaphysics, ethics, and theology to be sciences. Or rather, the history of metaphysics, ethics, and theology has seen many efforts to cast metaphysical, moral, and theological systems in the format of the classical model. Some examples are Christian Wolff ’s Philosophia Prima, sive Ontologica (1730); Baruch de Spinoza’s Ethica, Ordine Geometrico Demonstrata (1678); and Gisbertus Voetius’s Synopsis Purioris Theologiae (1625). When the friends of scientism speak about science, they typically focus on the best and most established parts of science as we now know it as the paragon of science.21 Our presentation of the classical model of science serves a twofold purpose: (i) In focusing on current best science, an implicit choice is made to understand science in a particular way; and (ii) while there may be good reasons for this choice, it is incumbent on the friends of scientism to present those reasons, for their preferred conception of science is historically

21. This is abundantly clear in Rosenberg (2011) and Ross, Ladyman, and Spurrett (2007).

14

1 4   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

contingent. There are other reasonable and historically prominent construals of science. The upshot of this section is that significant unclarity surrounds scientism’s understanding of science. The demarcation problem remains as daunting as ever, and it is historically far from obvious that we should accept what currently passes as science as the correct understanding. Both the proponents and the friends of scientism thus have their work cut out for them.

I.5 Arguments for Scientism We have said that scientism is often adopted implicitly and only half-​wittingly. There are exceptions, however. Some friends of scientism have offered explicit arguments in support of scientism.22 The most promising one, it seems to us, draws on the impressive success of modern science when compared to other attempts to find out truths about the world and ourselves.23 But we should try to get clearer about what this means. After all, both “science” and “success” have multiple senses. When the proponents of scientism state that science has been successful, they don’t mean to stress that certain scientific institutions have been around for a long time or that they operate effectively or efficiently. Nor do they mean to draw attention to the fact that scientists have conjured up so many hypotheses, theories, and explanations; or that they have published an unbelievable amount of articles and books. Rather, the point is to emphasize that science is epistemically successful; that it has succeeded in uncovering and establishing important truths about the world and ourselves. It has produced a great number of claims, theories, models, explanations, and predictions that have significant positive epistemic status, which is to say that they are justified, (likely to be) true, or empirically adequate. Typically, these are claims that have been severely tested, survived critical experiments, are supported by strong evidence, cohere with other results that have been tried and tested, and so on. Of course not everything that science produces has this elevated status, but the argument for scientism turns on a comparative claim: Science has a better track record in discovering and establishing truths than any other method humans have employed, such as relying on common sense, trusting religious or other authorities, a priori reasoning, etc. Note that spelling

22. Peels (2017) discusses 10 possible arguments for scientism. 23. Rosenberg (2011: 26) hints at such an argument.

15

Introduction  • 

15

out the precise sense in which science is epistemically more successful than common sense, a priori reasoning, etc., would actually take serious work, since a straightforward track record of science shows that most scientific claims published in respectable journals and books simply turn out to be false sooner or later. This is the basis for Laudan’s (1981) famous pessimistic meta-​induction and Samuel Arbesman’s (2012) somewhat misleadingly phrased claim that scientific facts have a “half-​life.” It may well be that the track record of mundane common-​sense claims looks much better in such purely quantitative terms. Someone seeking to defend claims about the comparative success of science will thus need to qualify (if not gerrymander) her claims. Perhaps science is particularly successful in the long run in unearthing certain kinds of “deep,” “fundamental,” or significant truths. We’ll assume for the sake of argument that this can be done in a non-​arbitrary or ad hoc manner. In addition to its epistemic success, proponents of scientism also like to stress what we can call science’s technological success. Many results of science have contributed directly or indirectly to effective technical applications:  applications that successfully do the job for which they have been designed and created.24 It is an open question to what extent all this technology has had an overall positive effect on the world and our quality of life, but this is something adherents of scientism can readily admit. To say that science has often been successfully applied in technology that is effective in the narrow sense of functioning properly, is not to be committed to naïve optimism about the overall effects of technology. Is the argument for scientism from the comparative success of science compelling? As we noted at the very beginning of the chapter, scientism is associated with a number of exclusivity claims about the role of science and its deliverances:  It maintains that only science can give us knowledge, rational belief, etc. Now it hardly needs saying that a claim about the comparative success of science nowhere near implies that science is our only route to knowl­edge, rational belief, etc. A  smartphone is extremely successful in doing quick calculations, but surely that doesn’t mean that only smartphones should be used to calculate one’s share of the restaurant check. This is such an obvious mistake that we should not dwell on it.

24. But note that the view that all technology just is applied science has fallen on hard times. Technology often develops independently of science and the direction of influence is sometimes the other way around: Successful technological applications can also lead to new or better scientific theories, explanations, etc. (Kline 1992; Kroes and Bakker 1992; Vincenti 1990).

16

1 6   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

What the proponent of scientism needs to plug the hole in this argument, are further premises with the following features:  (i) They should identify other alleged sources of knowledge, rational belief; and (ii) they should make the case that these sources either fail to deliver the required goods altogether or are so much less successful than science that we would do best to avoid them altogether. The challenge for the advocate of the argument from the success of science is, first, to state these premises in detail; and, second, to argue that they are plausible. Friends of scientism have taken up this challenge in a number of ways. First, they have marshaled psychological evidence indicating that our “pre-​ scientific” or common-​sense modes of thinking are riddled with biases and hence their deliverances highly unreliable. These biases pertain partly to the ways we think about ourselves (“folk psychology,” see, e.g., Wilson 2002). Also, they argue that there is compelling scientific evidence against many of our most cherished beliefs about ourselves, such as that we have free will, and that when we act we mostly do so for reasons (ibid.). Science, so it is suggested, shows these notions to be illusory. Moreover, they have offered broadly science-​based arguments against the reliability of our moral belief formation. One prominent recent branch of this endeavor are evolutionary debunking arguments ( Joyce 2006:  179–​219; Street 2006). The core idea behind these arguments is that evolutionary considerations provide a complete satisfactory explanation of human moral behavior and belief formation. Behaving and believing in certain ways—​or possessing the tendencies to do so—​so the thought goes in a nutshell, was evolutionarily advantageous and that is how morality evolved among humanoids. If this is correct, moral truth doesn’t factor into the process at all and hence we have no reason to think that our moral beliefs track the truth or constitute knowledge. Next, they argue that religious belief results from dubious sources (Bering 2011). Also, they argue that the humanities offer us nothing but entertaining illusions (Rosenberg 2011: 299–​308). Whether or not these arguments are convincing is very much a topic of ongoing debate.25 But for the argument for scientism from the success of science to gain traction, its supporters must find ways of buttressing these additional premises.

25. The literature on each of the topics mentioned above is vast, but some good starting points include Wegner (2002), Schloss and Murray (2009), Bergmann and Kain (2014), and Mele (2014). Hilary Kornblith’s and William FitzPatrick’s contributions to this volume also touch on these issues.

17

Introduction  • 

17

I.6 Arguments against Scientism Let us now turn to arguments against scientism. Again, we will not attempt to evaluate these arguments here. We merely sketch them. The reader will find that several of these arguments return in one guise or another in various essays in this volume. We look at four arguments. First, scientism, at least on its stronger versions, has counterintuitive consequences. Is it really true that I cannot know what my birthday is, merely because I have never done any scientific research on it? Can I not know that killing for fun is morally wrong just because that belief of mine is not based on scientific research—​and maybe cannot even be based on scientific research? As one of us has pointed out, there are also many general things that we seem to know even though they are not based on any kind of scientific research: Next there are many very general truths that I happen to know, such as:  that there are very many people, that they live on the surface of the earth, that they need food and liquids to keep themselves alive, that they need love and respect, that there are very many countries in which these people live, and that these countries have governments, some of which are very bad, but others of which are tolerably good. (Van Woudenberg 2011: 183) To deny that we know such things seems to fly in the face of reason. This counterintuitiveness might be taken to provide an argument against scientism.26 Whether such an argument can be successful depends in large part on the extent to which the proponents of scientism are able to establish the unacceptability of extra-​scientific sources of knowledge. Second, one might worry that scientism is self-​referentially incoherent.27 The thesis of scientism itself, it seems, is not based on scientific results. But then it follows, on scientism, that it cannot be rationally believed or known. But if it cannot be rationally believed or known, then how could a rational person ever embrace it? We find short versions of this problem in the literature. One of us, for instance, writes: Scientism suffers from self-​referential problems. Not being a scientific claim itself, it would seem scientism cannot be known by anyone. This 26. Another way of pursuing this point would be to argue that scientific knowledge has inherent limits and there are other modes of knowledge next to it. This is Rescher’s (1999) line; cf. also Jeroen de Ridder’s contribution to this volume. 27. For more on this notion, see Mavrodes (1985).

18

1 8   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

raises the question of why anyone should assert or believe it in the first place. (De Ridder 2014: 27) And Mikael Stenmark reasons as follows: How do you set up a scientific experiment to demonstrate that science or a particular scientific method gives an exhaustive account of reality? I cannot see how this could be done in a non-​question begging way. What we want to know is whether science sets the limits for reality. The problem is that since we can only obtain knowledge about reality by means of scientific methods . . ., we must use those methods whose scope is in question to determine the scope of these very same methods. If we used non-​scientific methods we could never come to know the answer to our question, because there is according to scientistic faith no knowledge outside science. We are therefore forced to admit either that we cannot avoid arguing in a circle or that the acceptance of [scientism] is a matter of superstition or blind faith. (Stenmark 2001: 22–​23) The adherent of scientism may have replies available. She could argue that there is scientific evidence for scientism; that we should make an exception for scientism; that scientism is pragmatically rather than epistemically justified; or that scientism escapes self-​referential incoherence because it is not itself a belief, but rather a stance in Van Fraassen’s (2002) sense.28 One of us has explored these options elsewhere.29 Third, there is what one of us has called the fundamental problem. Here, the idea is that the basis or foundation of science is itself nonscientific. It consists of all sorts of beliefs based on perception, memory, introspection, and so on. If these beliefs were not rational, then the science that is based on them could not be rational either. Hence, either they are rational and scientism has to go, or both scientific and nonscientific beliefs are equally irrational (and scientism has to go as well). One of us makes a similar point elsewhere: Another response . . . might be to bite the bullet and deny that extra-​ scientific beliefs ever amount to knowledge. This, however, would be

28. This is James Ladyman’s considered view; see Ladyman (2011) and his contribution to this volume. 29. See Rik Peels (2019).

19

Introduction  • 

19

deeply problematic. For scientific knowledge depends in many ways on extra-​scientific knowledge, for instance, on what we know through perception, such as that the thermometer now reads 118 degrees Fahrenheit. Without such extra-​scientific knowledge it is hard to see how science could even get started. (Van Woudenberg 2013: 26) Again, though, the adherent of scientism could push back. Following Otto Neurath (1973: 199), she could argue that the results of science don’t need to be based on nonscientific beliefs, for science systematically checks and replaces any nonscientific beliefs on which it initially might depend, in the process transforming them to scientific beliefs that are sufficiently reliable to qualify as rational belief or knowledge. The idea is that we can use scientific beliefs from one area to scientifically test nonscientific beliefs in another area. In doing so systematically, we eventually make all of our beliefs more reliable, in the same way as we can replace the planks of a ship one by one, each time relying on the other planks to keep us afloat. One of us has explored this suggestion and similar ones elsewhere (Peels 2018). Fourth, there is the argument from nonscientific assumptions and principles in science. The basic idea is this: There are certain principles that are indispensable for doing science, but that are not themselves the result of the science. These principles are held to be indispensable for science in the sense that if they cannot be rationally believed or known, then science cannot deliver rational belief or knowledge. According to Mary Midgley, for instance: Science cannot stand alone. We cannot believe its propositions without first believing in a great many other  .  .  .  things, such as the existence of the external world, the reliability of our senses, memory and informants, and the validity of logic. If we do believe in these things, we already have a world far wider than that of science. (Midgley 1992: 108) Among the basic assumptions and principles that scientists would seem to have to embrace in order to do science are epistemic theses—​for example, that our cognitive faculties, such as perception or logical reasoning, are broadly reliable or that we should prefer those theories that exhibit virtues such as simplicity, explanatory power, or broad scope to a greater extent—​metaphysical theses—​for example, that the world behaves in a regular way, has done so for times immemorial, and will continue to do so—​and semantic principles—​for example, that if a name refers to a thing and another name refers to a thing,

20

2 0   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

and the things referred to have all the same properties, then the names refer to numerically the same thing.30

I.7 Preview of the Chapters Having explained why scientism should be on the philosopher’s agenda, sketched its prehistory, discussed an argument in its favor, and mentioned four challenges that it faces, in this final section we present a preview of the coming chapters. As we said above, scientism is an elusive participant in the intellectual life. It is not only natural but also necessary, then, to try to pinpoint it more before it is developed, defended, and assessed further. Hence, the first two chapters lay the conceptual groundwork for the rest of the volume. Both Rik Peels and Mikael Stenmark develop detailed overviews of what scientism is and what it could be. Rik Peels’s chapter focuses on distinguishing a great variety of scientistic theses and clarifying their logical and conceptual connections. In doing so, it offers a conceptual roadmap to scientism. According to Peels, a key scientistic thesis is that the boundaries of the natural sciences should be expanded in order to include academic disciplines or realms of life that are widely considered not to belong to the realm of the natural sciences. He furthermore introduces a number of helpful distinctions that can serve to specify different versions of scientism:  (a) academic or universal scientism; (b) eliminative, methodological, epistemological, ontological, moral, or existential scientism; (c) full or partial scientism; and (d) in the case of moral and existential scientism: replacement or illusion scientism. Both would-​be proponents and critics of scientism can help themselves to these distinctions to clarify their positions and targets. In addition to surveying different versions of scientism and richly illustrating them through quotations from the literature, Mikael Stenmark’s chapter also widens the perspective by contrasting scientism to rival and equally encompassing views. In particular, Stenmark suggests that that liberal naturalism, humanism, social constructionism, religious naturalism, and theism are all best understood as rivals to scientism. That does not mean that they are necessarily incompatible with scientism on all accounts, but they do contain elements that are in serious tension with the epistemological and ontological commitments of many versions of scientism or with scientism’s

30. For these and similar examples, see Van Woudenberg (2011: 177).

21

Introduction  • 

21

tendency to be suspicious about everything in reality that cannot be described, understood, or explained by the natural sciences. The next three chapters develop scientism and defend it against objections. Alex Rosenberg, perhaps the most vocal champion of a strong version of scientism, expounds his preferred version of scientism and then takes on two major challenges to the view:  the epistemology of mathematics and eliminativism about cognition. While Rosenberg doesn’t claim to provide definitive answers to how we can come to have knowledge of the necessary truths of mathematics, he does sketch the contours of a possible fictionalist solution that is consistent with scientistic commitments. According to Rosenberg, the second challenge is just as important, but often goes unnoticed, even among those who endorse scientism, since they don’t recognize their own commitment to eliminativism. To confront it, Rosenberg describes the outlines for a thoroughly scientistic account of intentionality and semantic evaluability. He admits that more work needs to be done to develop his sketches, but that is as it should be for the adherent of scientism. James Ladyman is on record as a defender of a scientistic stance in metaphysics, according to which “serious metaphysics” shouldn’t rely on intuitions and simplified toy examples and models, but should always have its starting point in established physics. In his contribution to this volume, he proposes a form of scientism as a belief-​forming stance with universal scope. This version of scientism, however, has a humane face. According to him, it is not dogmatic or uncritical, nor does it ignore the actual limitations to current scientific knowledge. There are other modes of inquiry that deserve epistemic respect, and scientists should not be deferred to about matters beyond their expertise. Nonetheless—​and this is where scientism’s real bite is—​no a priori limits should be placed on what science can study and we cannot say in advance what the limits of future science will be. Where science conflicts with common sense, religion, and tradition, science should be regarded as authoritative for the purposes of education and public policy as well as objective inquiry; and scientific knowledge is even relevant to moral and political deliberation. This is the core of Ladyman’s scientism. Hilary Kornblith cashes out his preferred version of scientism in terms of Wilfrid Sellars’s well-​known distinction between “the scientific image”: of our place in the world, and “the manifest image.” While Sellars wanted to join these views together in spite of their apparent conflict, Kornblith thinks we ought to endorse features of the manifest image only to the extent that they are part of the scientific image. He then explores what this version of scientism entails for how we should think of doxastic deliberation, that is, deliberation

2

2 2   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

about what to believe. He argues that the manifest image of such deliberation is flatly in conflict with the best current scientific theorizing about the nature of deliberative processes. Hence, he argues, we should embrace the scientific account and reject our first-​personal view of deliberation as illusory. The next three chapters turn a critical eye to scientism. In popular discussions of scientism, it is often suggested that adherents of scientism are unduly dogmatic, closed-​minded, or intellectually arrogant. Ian James Kidd employs the framework of virtue epistemology to examine this charge in more detail, arguing that it sticks. Like several of scientism’s proponents, Kidd characterizes scientism as a stance and then argues that the epistemically vicious dispositions mentioned above are demonstrably among the components of a scientistic stance, so that those who adopt it can be led to manifest these vices. He concludes that determining whether or not any given stance is indeed vicious requires sensitivity to the ontology of that stance and the psychology of the agents who adopt them. René van Woudenberg takes a critical look at two prominent philosophical defenses of epistemological scientism in the literature, one by Alex Rosenberg in his book The Atheist’s Guide to Reality; and the other by Don Ross, James Ladyman, and David Spurrett in Every Thing Must Go. Rosenberg’s scientism is the view that the methods of science are the only reliable ways to secure knowledge of anything. Van Woudenberg argues that the view faces many counterexamples, that Rosenberg’s arguments in its favor are weak, and that it is self-​referentially incoherent. Scientism as propounded in Every Thing Must Go is the view that science is our only guide to the objective features of the world. The worked-​out view includes, among other things, an institutional demarcation criterion to distinguish bona fide science from non-​ science, a non-​positivistic form of verificationism, and the idea that scientism should not be understood as a thesis but as a stance. Van Woudenberg argues that this view, too, faces counterexamples. He also raises problems for the institutional demarcation criterion, the proposed neo-​verificationism, and the idea of scientism as a stance. Jeroen de Ridder’s chapter embodies a more indirect approach to criticizing scientism. He doesn’t take on a specific version of scientism, but defends an idea that goes against the spirit of many forms of scientism, to wit the idea that science has limits. He does so first by investigating the distinct nature of scientific knowledge, as contrasted with other kinds of knowledge. De Ridder develops two plausible and mutually compatible proposals for understanding what’s special about scientific knowledge: the idea that scientific knowledge is high-​grade knowledge and Bas van Fraassen’s proposal that

23

Introduction  • 

23

scientific knowledge is objectifying knowledge. He then investigates what these two proposals entail for whether scientific knowledge is limited. It turns out that, on both proposals, there are various in-​principle limits to what can be known scientifically. This of course spells trouble for any form of scientism denying this. The final four chapters explore what scientism might mean for morality, religious belief, and philosophy, and what the prospects and problems are for scientistic views in these areas. Alvin Plantinga considers the relation between scientism and religious, particularly Christian, belief. After surveying several possible versions of scientism and arguing that most of them are false, he homes in on an apparently more moderate version of scientism: the idea that when any nonscientific belief comes into conflict with a scientific belief, it is always the scientific belief that should prevail. Plantinga argues that this form of scientism, too, fails. Building on his well-​known work in religious epistemology, he explains how it can be reasonable to hold on to Christian belief even when it fits ill with scientific orthodoxy. It turns out that the moderate scientism in question isn’t so moderate on closer inspection, for it implies that Christian belief is false. According to Plantinga, the right way to think about things is that science is a wonderful institution and a wonderful source of knowledge or warranted belief; but it is not our only source of knowledge about ourselves and the world. William FitzPatrick explores how the scientific study of morality, as carried out in cognitive science and moral psychology, is relevant to matters in normative ethics and meta-​ethics. For the proponent of scientism, there can be no doubt that science is highly relevant here. After all, it is pretty much all we have to go on, according to her view. But Fitzpatrick takes a dimmer view. First, he looks at two prominent recent attempts to show that experimental work in moral psychology debunks widely held ethical or meta-​ethical intuitions, one by Joshua Green, the other by Shaun Nichols. Fitzpatrick argues that the cases they make for this sort of strong impact of experimental work on moral philosophy suffer from a problematic form of scientism and ultimately fail; indeed, that they fail for reasons that likely apply to other projects with similar ambitions as well. Next, he sets out to clarify the nature of the dialectical situation with respect to empirically driven attempts to debunk traditional philosophical views, which leads to a general challenge to such debunking projects going forward. He ends by suggesting a more modest and plausible role for experimental work in connection with moral philosophy—​one that is still significant, particularly for some audiences, but which gives up the overreaching debunking ambitions.

24

2 4   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

While scientism has not received much sustained philosophical attention, a number of views in its neighborhood have, among them physicalism, materialism, and empiricism. In fact, these views have sometimes been accused of being scientistic in the pejorative sense of putting too much trust in science. In her chapter, Alyssa Ney investigates whether this charge is correct in the case of physicalism. She shows how standard formulations of physicalism, which take it to be something like the view that the world is in totality the way physics says it is, can indeed make physicalism look like a reductionistic form of scientism. But, she argues, more subtle formulations of physicalism, which we have independent reason to prefer, reveal the difference between physicalism and scientism. Physicalism neither entails imperialistic claims about the scope of physics, nor exclusivist claims about science being our only mode of discovery. The final chapter considers what scientism might mean for philosophy. In particular, Wesley Buckwalter and John Turri recommend a moderate form of scientism in philosophy. Their moderate scientism is the view that empirical science can help answer questions in nonscientific disciplines. The chapter showcases several ways that science has contributed to research in epistemology, action theory, ethics, philosophy of language, and philosophy of mind; and it reviews several ways that science has contributed to our understanding of how philosophers make judgments and decisions. On the basis of these rich inductive grounds, Buckwalter and Turri conclude that the case for moderate philosophical scientism is strong: Scientific practice has promoted significant progress in philosophy and its further development should be welcomed and encouraged. Needless to say, such moderate scientism has no tendency to support stronger imperialistic or exclusivist forms of scientism.31

References Arbesman, Samuel. 2012. The Half-​Life of Facts:  Why Everything We Know Has an Expiration Date. New York: Current. Ayer, Alfred J. 1946. Language, Truth, and Logic. New York: Dover. Ayer, Alfred J. ed. 1959. Logical Positivism. New York: Free Press. Beale, Jonathan and Ian James Kidd. eds. 2017. Wittgenstein and Scientism. London:  Routledge.

31. This publication was made possible through the support of a grant from the Templeton World Charity Foundation. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Templeton World Charity Foundation.

25

Introduction  • 

25

Bergmann, Michael and Patrick Kain. eds. 2014. Challenges to Moral and Religious Belief: Disagreement and Evolution. New York: Oxford University Press. Bering, Jesse. 2011. The God Instinct: The Psychology of Souls, Destiny, and the Meaning of Life. London: Nicholas Brealey. Boudry, Maarten and Massimo Pigliucci. eds. 2018. Science Unlimited? The Challenges of Scientism. Chicago: University of Chicago Press. Comte, August. 2009 (1853). The Positive Philosophy of Auguste Comte, Vol. 1. Translated by H. Martineau. Cambridge: Cambridge University Press. De Jong, Willem R. and Arianna Betti. 2010. “The Classical Model of Science I:  A Millennia-​Old Model of Scientific Rationality.” Synthese 174(2): 185–​203. De Ridder, Jeroen. 2014. “Science and Scientism in Popular Science Writing.” Social Epistemology Review and Reply Collective 3(12): 23–​39. Dooyeweerd, Herman. 1997 (1953–​1958). A New Critique of Theoretical Thought, 4 Vols. Lewiston, NY: Edwin Mellen Press. Godfrey-​Smith, Peter. 2003. Theory and Reality. Chicago: University of Chicago Press. Haack, Susan. 2007. Defending Science within Reason: Between Scientism and Cynicism. Amherst, NY: Prometheus Books. Haack, Susan. 2017. Scientism and Its Discontents. Open Access e-​book, available at https://​roundedglobe.com/​. Harris, Sam. 2010. The Moral Landscape: How Science Can Determine Human Values. New York: Free Press. Hempel, Carl G. 1950. “The Empiricist Criterion of Meaning.” In Ayer 1959: 108–​129. Hume, David. 1975 [1777]. An Enquiry Concerning Human Understanding. Edited by L.A. Selby-​Bigge, 3rd ed. edited by P. H. Nidditch. Oxford: Clarendon Press. Hutchinson, Ian. 2011. Monopolizing Knowledge. Belmont, MA: Fias. Joyce, Richard. 2006. The Evolution of Morality. Cambridge MA: MIT Press. Kline, Ronald. 1992. Steinmetz:  Engineer and Socialist. Baltimore:  Johns Hopkins University Press. Kroes, Peter and Martijn Bakker. eds. 1992. Technological Development and Science in the Industrial Age. Dordrecht: Springer. Ladyman, James. 2011. “The Scientistic Stance: The Empirical and Materialist Stances Reconciled.” Synthese 178(1): 87–​98. Laudan, Larry. 1981. “A Confutation of Convergent Realism.” Philosophy of Science 48(1): 19–​49. Laudan, Larry. 1983. “The Demise of the Demarcation Problem.” In Physics, Philosophy and Psychoanalysis, edited by Robert S. Cohen and Larry Laudan, 111–​ 127. Dordrecht: Reidel. Losee, John. 2001. An Historical Introduction to the Philosophy of Science. 4th ed. Oxford: Oxford University Press. Mavrodes, George I. 1985. “Self-​Referential Incoherence.” American Philosophical Quarterly 22(1): 65–​72. Mele, Alfred. 2014. Free: Why Science Hasn’t Disproved Free Will. New York: Oxford University Press.

26

2 6   • 

René van Woudenberg, R ik Peels, and Jeroen de R idder

Midgley, Mary. 1992. Science as Salvation:  A Modern Myth and Its Meaning. London: Routledge. Mill, John Stuart. 1974 (1843). A System of Logic Ratiocinative and Inductive. In The Collected Works of John Stuart Mill, edited by John M. Robson. Vol. VIII. Toronto: University of Toronto Press; London: Routledge & Kegan Paul. Nagel, Ernest. 1961. The Structure of Science. Problems in the Logic of Scientific Explanation. London: Routledge & Kegan Paul. Neurath, Otto. 1973. “Anti-​Spengler.” In Empiricism and Sociology, edited by Marie Neurath and Robert S. Cohen, 158–​213. Dordrecht: Reidel. Peels, Rik. 2017. “Ten Reasons to Embrace Scientism.” Studies in the History and Philosophy of Science A 63: 11–​21. Peels, Rik. 2018. “The Fundamental Argument Against Scientism.” In Science Unlimited? The Challenges of Scientism, edited by Maarten Boudry and Massimo Pigliucci, 165–​184. Chicago: University of Chicago Press. Peels, Rik. 2019. “Should We Accept Scientism? The Argument from Self-Referential Incoherence.” In Knowing in Science: An Introduction, edited by Kevin Ray McCain and Kostas Kampourakis. London: Routledge. Pigliucci, Massimo and Maarten Boudry. eds. 2013. The Philosophy of Pseudoscience. Chicago: University of Chicago Press. Philipse, Herman. 2013. “The Real Conflict Between Science and Religion:  Alvin Plantinga’s Ignoratio Elenchi.” European Journal for Philosophy of Religion 5(2): 87–​110. Pigliucci, Massimo. 2013. “New Atheism and the Scientistic Turn in the Atheism Movement.” Midwest Studies in Philosophy 37(1): 142–​153. Plantinga, Alvin. 1967. God and Other Minds. Ithaca, NY: Cornell University Press. Rescher, Nicholas. 1999. The Limits of Science. 2nd ed. Pittsburgh, PA: University of Pittsburgh Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality: Enjoying Life without Illusions. New York: W. W. Norton. Ross, Don, James Ladyman, and David Spurrett. 2007. “In Defence of Scientism.” In Every Thing Must Go, edited by James Ladyman and Don Ross, with David Spurrett and John Collier, 1–​65. Oxford: Oxford University Press. Schloss, Jeffrey and Michael Murray. eds. 2009. The Believing Primate. New York: Oxford University Press. Sorell, Tom. 1991. Scientism:  Philosophy and the Infatuation with Science. London:  Routledge. Stenmark, Mikael. 2001. Scientism: Science, Ethics and Religion. Aldershot: Ashgate. Stevenson, Leslie and Henry Byerly. 2000. The Many Faces of Science. 2nd ed. Boulder, CO: Westview. Street, Sharon. 2006. “A Darwinian Dilemma for Realist Theories of Value.” Philosophical Studies 127(1): 109–​166. Trigg, Roger. 1993. Rationality and Science: Can Science Explain Everything? Oxford:  Blackwell.

27

Introduction  • 

27

Van Fraassen, Bas C. 2002. The Empirical Stance. New Haven, CT: Yale University Press. Van Woudenberg, René. 2011. “Truths that Science Cannot Touch.” Philosophia Reformata 76(2): 169–​186. Van Woudenberg, René. 2013. “Limits of Science and the Christian Faith.” Perspectives on Science and Christian Faith 65(1): 24–​36. Van Woudenberg, René and Joëlle Rothuizen-​van der Steen. 2016. “Science and the Ethics of Belief:  An Examination of Philipse’s Rule R.” Journal for General Philosophy of Science 47(2): 349–​362. Vincenti, Walter. 1990. What Engineers Know and How They Know It. Baltimore: Johns Hopkins University Press. Weber, Max. 1919. Wissenschaft als Beruf. In Gesamtausgabe, Vol. 17. Tübingen: Mohr Siebeck. Wegner, Daniel. 2002. The Illusion of Conscious Will. Cambridge, MA: MIT Press. Williams, Richard and Daniel Robinson. eds. 2015. Scientism:  The New Orthodoxy. London: Bloomsbury. Wilson, Timothy D. 2002. Strangers to Ourselves. Cambridge, MA:  Harvard University Press. Windelband, Wilhelm. 1894. Geschichte und Naturwisschenschaft. Strassburg: Heitz.

28

1

A CONCEPTUAL MAP OF SCIENTISM

Rik Peels

1.1 Introduction Few people living in Western societies today would deny that science has great value. It is also widely believed, though, that the scope and value of science can be exaggerated; science has its boundaries and those boundaries should not be crossed—​below, I  say more about how this is to be understood. Philosophers or scientists who do cross what are thought to be the boundaries are often referred to as subscribing to or practicing scientism. Here is a quote from the American historian of science William Provine that many would consider as an instance—​or even several instances—​of scientism: Modern science directly implies that the world is organized strictly in accordance with mechanistic principles. There are no purposive principles whatsoever in nature. There are no gods and no designing forces that are rationally detectable. . . . modern science directly implies that there are no inherent moral or ethical laws, no absolute guiding principles for human society.  .  .  .  human beings are marvelously complex machines. . . . when we die, we die and that is the end of us. . . . Free will as it is traditionally conceived—​the freedom to make uncoerced and unpredictable choices among alternative possible courses of action—​simply does not exist.  .  .  .  There is no ultimate meaning for humans. (Provine 1988: 27–​29) I said that many would describe this as a clear case of scientism. But precisely what is scientism? If we consider the passage just quoted,

29

A Conceptual Map of Scientism  • 

29

we discover a wide variety of different claims: Science implies that the world is purely mechanistic; that free will, as traditionally conceived, is an illusion; that there is no ultimate meaning for us; and so on. This gives rise to all sorts of questions. Is each of these theses an instance of scientism? Are there other kinds of scientism that are not found in this quote? Is there an underlying basic idea in virtue of which these claims are widely considered to be instances of scientism? How do different kinds of scientism relate to each other? The aim of this chapter is to provide a framework for answering questions like these by construing a conceptual map of scientism. By “scientism” I mean, roughly, the view that the boundaries of science should be expanded in order to encompass other academic disciplines and/​or other realms of reality, such as human cognition in general or morality. What such expansion amounts to depends on the variety of scientism in question. It can mean, for instance, that only science can tell us what exists or that science should replace common sense in a domain like morality. In what follows, I confine myself to the natural sciences, such as biology, physics, and cosmology, because paradigm instances of scientism are cashed out in terms of these disciplines, even though one could make similar claims for disciplines such as sociology and economics. To get sharper into focus what I  mean by “scientism,” let me formulate three constraints on something to count as an instance of scientism. These constraints are based on how words like “scientism” and “scientistic” are widely used. First, I treat scientism as a particular claim or thesis. This is not the only way one could think of scientism. One might also think of scientism as some kind of attitude, affection, stance, or still something else.1 For two reasons, I nonetheless prefer to treat scientism as a thesis. First, as evidenced by the quotations given in this chapter, scientism as a thesis is frequently found in the writings of scientists and philosophers. Second, it seems that every attitude, affection, or stance, at least if it is to be up to discussion, can be translated into a thesis, namely, the thesis that it is good to have that affection, attitude, or stance. No matter how one understands “scientism,” then, it will always imply some scientistic thesis or other. Second, every instance of scientism puts the natural sciences or even a specific natural science, such as physics, center stage. Each case of scientism, then, is a claim about the relation that should obtain between the natural sciences

1.  See, for instance, Haack (2007:  17–​18); Rescher (1999:  1); Ross, Ladyman, and Spurrett (2007: 57–​59).

30

3 0   •  R i k   P e e l s

on the one hand and something else—​another academic discipline or another realm of reality—​on the other. This means that the claim that scientists themselves are somehow superior to other people falls outside the scope of this chapter, even though this claim might in some way be related to scientism (see Snow 1972: 11, 48). Third, even though the word “scientism” is often used pejoratively, it need not be. For instance, Don Ross, James Ladyman, and David Spurrett, in Every Thing Must Go say expressis verbis that they adhere to scientism and go on to defend it (Ross et al. 2007: 1–​65; see also Rosenberg 2011: 6). Thus, to claim that something is an instance of scientism is not thereby to take a positive or negative stance toward the relevant assertion.2 In fact, I think that most philosophers and scientists will embrace at least some of the weaker versions of scientism described in this chapter. Thus, a thesis is an instance of scientism only if it is formulated in such a way that it is up to discussion. When I say that I provide a conceptual map of scientism, I mean that I analyze the varieties in which scientism comes and how these varieties relate to each other, in order consequently to display the results of these analyses in a diagram. In doing so, I contrast my view with that of others who have written on scientism, especially Mikael Stenmark, since his valuable work on scientism is both detailed and influential.3 It is not my aim to draw a map of all possible instances of scientism. Rather, I aim to draw a map of the most important varieties of scientism that we find in the literature. In construing the map I use the words “variety,” “version,” and “instance” of scientism. By a “variety of scientism” I  mean a species of the genus scientism:  The variety entails scientism, but not vice versa. By a “version of scientism” I mean a particular way of understanding a variety of scientism. And by an “instance of scientism” I mean a particular person’s written or spoken verbal expression of her scientism. The project of providing a conceptual map of scientism is important for at least two reasons. First, the word “scientism” is often used in science, philosophy, and in the wider culture and frequently in a pejorative sense. However, it is often unclear what is meant when someone is labeled as an adherent of scientism. We can judge whether such labeling is correct only if we have some grip on the term “scientism” and the varieties in which it comes. Second, it

2. Stevenson and Byerly (1995: 212). 3. For a recent version of his views on the varieties of scientism and how scientism relates to various alternatives, see ­chapter 2 of this volume.

31

A Conceptual Map of Scientism  • 

31

is important to see what an adherent of a particular variety of scientism is committed to. If a particular variety of scientism commits one to another variety with unpalatable implications, the position might be less plausible than initially thought. Or an adherent of scientism might falsely assume that a particular kind of scientism commits her to another kind of scientism. Whether this is indeed the case is something that the conceptual map will tell us. This chapter is structured as follows. First, I sketch the main varieties of scientism. I distinguish between academic and universal scientism. Academic scientism comes in two varieties: methodological and eliminative scientism; whereas universal scientism comes in four varieties: epistemological, ontological, moral, and existential scientism. Subsequently, I show how these varieties relate to each other. A defense of each of the entailments can be found in the Appendix to this chapter. Next, I argue that there is a nontrivial set of necessary and sufficient conditions that a claim should meet in order to count as an instance of scientism. Finally, I draw the threads of this chapter together by giving four conclusions.

1.2 Varieties of Scientism 1.2.1 Academic Scientism 1 and 2: Methodological and Eliminative Scientism The main distinction that we need to make is that between what I call academic scientism and universal scientism. Academic scientism is restricted to the academic disciplines,4 whereas universal scientism is meant to apply both inside and outside of the academy. We will see in a moment what these claims amount to. The first distinction we need to make with respect to academic scientism is between methodological and eliminative scientism. Whereas the methodological variety grants that, say, philosophy and psychology are proper academic disciplines that ask sensible questions, it asserts that they are so only if they adopt the methods of the natural sciences, such as observation and experimentation. Thus, the traditional questions of, say, theology or philosophy, can be answered only by using the methods of the natural sciences.5 The eliminative version is stronger in that it claims that academic disciplines

4. Hayek (1979), for instance, treats scientism as a claim about natural science and other academic disciplines. 5. Thus also Stenmark (2001: 3).

32

3 2   •  R i k   P e e l s

other than the natural sciences, such as the humanities, have nothing to add to the natural sciences if properly carried out. The questions asked in, say, psychology and philosophy, are nonsensical or obscure. We should abandon the subject matters of these disciplines altogether. For example, Otto Neurath gives a rather rhetorical statement of his view that metaphysics should be given up in favor of physics when he says: how does the elimination of metaphysics proceed in practice? Men are induced to give up senseless sentences and freed from metaphysics. But must this always remain so? Must everyone in turn go through metaphysics as through a childhood disease—​perhaps the earlier he gets it, the less dangerous it is—​to be led back to unified science? No. Every child can in principle learn to apply the language of physicalism correctly from the outset, first in a crude form, then in a more refined and precise way. (Neurath 1987: 9) Neurath is rather explicit about his eliminative academic scientism. Others, such as Patricia Churchland and Stephen Stich, are less explicit. They argue that, since no consensus is forthcoming after two thousand years of discussion, we should abandon traditional philosophical problems, such as how knowledge is to be analyzed.6 These problems should be left aside altogether, since they cannot be solved by means of natural science. The assumption here is, clearly, that only natural science delivers what we are looking for (consensus) and that we should, therefore, give up any academic disciplines that do not employ the methods of natural science.7 A second distinction that is relevant here is that between partial and full academic scientism. Whereas partial academic scientism makes a scientistic claim about only some of the academic disciplines that are distinct from the 6. See, for instance, Churchland (1987) and Stich (1983). Haack (1995: 158–​181) characterizes these two views as revolutionary scientism. 7. In the same spirit, Stephen Hawking and Leonard Mlodinow open their book The Grand Design by asking: What is the nature of reality? Where did all this come from? Did the universe need a creator?  .  .  .  Traditionally these are questions for philosophy, but philosophy is dead. Philosophy has not kept up with modern developments in science, particularly physics. Scientists have become the bearers of the torch of discovery in our quest for knowledge. (Hawking and Mlodinow 2010: 5) Hughes (2012: 33) takes methodological scientism to be crucial to scientism generally.

3

A Conceptual Map of Scientism  • 

33

natural sciences, full academic scientism is a claim about all other academic disciplines than the natural sciences. Thus, Churchland’s and Stich’s claim that traditional epistemology should be replaced with neuroscience and cognitive psychology is an instance of partial scientism, whereas E. O. Wilson adopts a version of full scientism when he says: “It may not be too much to say that sociology and the other social sciences, as well as the humanities, are the last branches of biology waiting to be included in the Modern Synthesis [that is, Neo-​Darwinism; RP]” (Wilson 1975: 4). His idea seems to be that all academic disciplines should be reduced to the natural sciences, especially to biology. An exhaustive map of scientism would encompass different degrees of scientism. For instance, Francis Crick’s claim that everything can be explained by physics and chemistry (Crick 1966: 14, 98) seems weaker than Rosenberg’s claim that physics is the whole truth (Rosenberg 2011:  25), but stronger than those versions of eliminative scientism that claim that all academic disciplines are reducible to some natural science. I  leave it to the reader to make these further distinctions, since they will fit into the conceptual map that I provide below. 1.2.2 Universal Scientism 1: Epistemological Scientism I use the expression “universal scientism” as a term of art in that it refers to a variety of scientism that is supposed to apply both within and outside of the academy. Some philosophers call this variety of scientism “academic-​external scientism.” That seems misguiding, though. For, as we shall see, universal scientism applies not only to matters external to the university, but also within the academic realm. However, since universal scientism is as such not a claim about academic methodology or the reduction of one academic discipline to another, academic scientism is not a variety of universal scientism. In what follows, I distinguish four varieties of universal scientism.8 The first kind of universal scientism that we ought to distinguish is epistemological scientism. Epistemological scientism is scientism about the

8.  The reader who is familiar with Stenmark’s work will notice that I  follow him in distinguishing epistemological, ontological, moral, and existential scientism. As we shall see, though, the conceptual map I provide also differs on crucial points from Stenmark’s account of scientism.

34

3 4   •  R i k   P e e l s

cognitive realm, both within and outside of the academy. There are numerous versions of epistemological scientism. Here are some of them: (a) All genuine knowledge is to be found only through (methods of ) natural science.9 (b) Natural science provides the only reliable path to knowledge.10 (c) All questions can in principle be answered by natural science.11 (d) Everything that can be known can be known through natural science.12 There is a lot to be said about each of these theses. Here, I will only note that they are distinct theses. For instance, to claim that all genuine knowledge is to be found through the natural sciences is different from claiming that the natural sciences are the only reliable path to knowledge, for one might think that other methods than the natural sciences incidentally (unreliably) lead to knowledge.13 For practical reasons, in what follows, when I mention “epistemological scientism” I confine myself to (a), the claim that all genuine knowl­ edge is to be found only through the (methods of ) natural sciences. As with academic scientism, we can distinguish between full and partial epistemological universal scientism. Since this might sound somewhat paradoxical—​does “universal” not exclude “partial”?—​let me explain this. I  have used the word “universal” in the rather restricted sense of “applying to both the academic and non-​academic realms.” Now, one might think, for instance, that all knowledge about anything whatsoever is to be acquired by the natural sciences. That would count as full epistemological universal scientism. But one might also make the more restricted claim that all knowledge about, say, consciousness is to be acquired by the natural sciences. That would 9.  For a similar characterization of this version of scientism, see Stenmark (2001:  vii–​viii). Nielsen (1997: 441), even defines “scientism” entirely along these lines. 10. Barbour (1990: 3–​5) uses (b) as equivalent with the thesis that the scientific method is the only reliable form of understanding. However, Barbour continues to argue that this view is mistaken. For (b), see Rosenberg (2011: 6). 11.  This view is sometimes ascribed to Rudolf Carnap on the basis of Carnap (1961:  254; 1967: 290). Carnap, however, makes clear that his claim is limited to questions that are formed from scientific concepts (Carnap 1967: 292). A better example is to be found in Atkins (1995). 12. See Russell (1946: 863): “Whatever can be known, can be known by means of science.” Russell admits that certain issues are beyond the scope of science, but those issues concern such things as feelings, and in the realm of feelings and values, Russell seems to think, and there is no knowledge to be had. 13. The claim that only natural science provides true explanations seems a weaker version of (a). For, this claim can plausibly be interpreted as saying that only natural science provides us with knowledge about why some state of affairs obtains.

35

A Conceptual Map of Scientism  • 

35

be a version of partial epistemological universal scientism. An example of the latter is what Bertrand Russell says about God and immortality: God and immortality, the central dogmas of the Christian religion, find no support in science. . . . No doubt people will continue to entertain these beliefs, because they are pleasant, just as it is pleasant to think ourselves virtuous and our enemies wicked. But for my part I cannot see any ground for either. . . . no one of these hypotheses is more probable than any other; they lie outside the region of even probable knowledge, and therefore there is no reason to consider any of them. (Russell 1967: 44, my italics) Russell’s point here seems to be that since natural science cannot deliver any knowledge about God and immortality, surely anything other than natural science will not deliver such knowledge either. This is a variety of partial epistemological universal scientism, because it is restricted to the supernatural realm, namely, God and immortality. It does not say that the only knowledge we could possibly have about anything whatsoever is to be produced by the natural sciences. It is nonetheless universal because it applies both within and outside of the academy: If Russell is right, then theology and philosophy will not be able to deliver any knowledge about God or immortality either. 1.2.3 Universal Scientism 2: Ontological Scientism The second variety of universal scientism is not a claim about our knowledge, but about what does and does not exist. Now, some philosophers, such as Roger Trigg, have taken this variety of scientism to amount to the claim that only those things exist that are at some point discovered by science (Trigg 1993: 70). As I said in the Introduction, though, the aim of my conceptual map is to provide an overview of options that could plausibly be defended, not an overview of any possible position. And the claim that only those things exist that are discovered at some point by science is clearly implausible, given that there are vast stretches of the universe that we might never be able to explore. Adherents of scientism sometimes conflate epistemological and ontological scientism.14 Can we nonetheless distinguish ontological scientism as

14.  For such a conflation, see, for instance, Peterson, Hasker, Reichenbach, and Basinger (1991:  36); Stevenson and Byerly (1995:  212); Quine (1992:  9). And sometimes, we find a blend of epistemological and ontological scientism. According to Wilfrid Sellars, for instance, “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.” (See Sellars 1963: 173.)

36

3 6   •  R i k   P e e l s

a separate variety of scientism? I think we can. Take the opening statement of Carl Sagan’s classic book Cosmos:  “The cosmos is all that is or ever was or ever will be” (Sagan 2002:  4). Although he does not say this explicitly, the idea seems to be that every aspect of the cosmos can in principle be investigated by science. Other things, such as free will, universals, and immaterial souls do not exist, for all that exists is matter and can be investigated by science. Full varieties of ontological scientism will be rare, but partial ones will not. On partial ontological scientism, a specific kind of thing is nothing but (a collection of ) those things acknowledged by the natural sciences. Undoubtedly, the most popular variety of partial ontological scientism is scientism about human beings. Carl Sagan, for instance, describes himself as a collection of water, calcium, and organic molecules (ibid.: 127). Francis Crick has called the idea, which he himself advocates, that we are nothing but a pack of neurons “The Astonishing Hypothesis.” It is astonishing, because it means that our memories, ambitions, beliefs, desires, choices, and sorrows are nothing but the collection or behavior of a large sum of nerve cells and their associate molecules (Crick 1994: 3, 258–​259). Another version of partial ontological scientism is the idea that free will is an illusion because science can explain every decision without appeal to free will, as advocated, for instance, by William Provine in that quote that I gave in the Introduction. 1.2.4 Universal Scientism 3: Moral Scientism A third variety of universal scientism is moral scientism. There are two varieties of moral scientism. On the first variety, the natural sciences lead or will lead us to the good life. Here are some versions of it: (a) The natural sciences guide us toward the morally good life. (b) Common-​sense morality should be replaced with scientific morality (Harris 2010). (c) Our moral personal and social problems can be solved by the  natural sciences.15 The second variety says something rather different. Here the basic idea is that science shows us or makes it sufficiently probable that morality is an illusion. Here are some versions of it:

15. This is how Peacocke partly characterizes scientism (see Peacocke 1993: 7–​8).

37

A Conceptual Map of Scientism  • 

37

(a) Science shows us that morality is an illusion. (b) Science shows us that good and evil are merely social conventions. (c) Science shows us that moral intuitions and beliefs are nothing but evolutionarily adaptive features of humans. The second variety of moral scientism is clearly a case of partial ontological scientism. I will nonetheless also treat it as a variety of moral scientism, since it is explicitly about morality. I will call these two varieties respectively the R-​ variety (from “Replacement”) and I-​variety (from “Illusion”). According to some philosophers, such as Stenmark, there is a further variety of moral scientism, namely, the claim that evolutionary theory can fully explain our moral sense. I agree that adherents of moral scientism usually also make this claim. But it seems to me misguided to take this claim itself to be an instance or part of moral scientism. Our moral beliefs are a natural phenomenon and it is, thus, not at all controversial that there might be some kind of scientific explanation for many or maybe even all of our moral beliefs (not for the truth of those beliefs, but for our holding those beliefs), in terms of evolutionary theory or cultural history. In order for a view to count as moral scientism, it should make a stronger claim. According to E O. Wilson, for instance, scientists and humanists should seriously consider removing ethics from the hands of philosophers, in order to biologize it (Wilson 1975: 562). The idea seems to be that biological principles can be applied in the social realm and that they can be used to justify and not merely explain certain moral norms and values.16 Now, one might think that moral scientism is partial by its very nature, given that it is restricted to the moral realm. But this is mistaken. In the same way as there can be full and partial academic scientism, even though academic scientism is restricted to the academic realm, there can be full and partial moral scientism, even though moral scientism is restricted to the moral realm. One might think, for instance, that science can replace some of our morality, but not all of it, since we need to start from some moral intuitions. It does seem true, though, that on its I-​variety, moral scientism does not come in full and partial varieties. It seems hard to defend that, say, moral obligations are illusions, but that moral intuitions can nonetheless be true. Hence, R-​moral

16. Stenmark (2001: 34) falsely claims that such moral scientism is implied by the project of sociobiology. True, many sociobiologists adhere to moral scientism. However, other sociobiologists endorse the project of sociobiology without claiming or even while denying that ethics should or even can be biologized (see, for instance, Ruse 1979: 199–​204).

38

3 8   •  R i k   P e e l s

scientism can be full or partial, whereas I-​moral scientism can only be full scientism. 1.2.5 Universal Scientism 4: Existential Scientism The final variety of universal scientism that I would like to distinguish is existential scientism. As with moral scientism, this kind of scientism comes in two rather different varieties. The first variety, which I call the R-​variety, says that science should replace religion; mythology; secular ideologies, such as fascism and Marxism; and other nonscientific ways of answering our existential questions. Here are some versions of existential scientism: (a) Science should replace traditional religions and secular ideologies.17 (b) Salvation can be achieved by (the methods of ) science alone (cf. Midgley 1992: 37). Richard Dawkins articulates existential scientism when he says that the answers we give to the big questions of life are meaningless unless they are informed by natural science, especially evolutionary biology. The entire intellectual traditions of, say, ancient Greek philosophy and Medieval scholastics are without worth, because they are not based on scientific research (Dawkins 1989: 1). The second variety, which I will call the I-​variety says that the idea that there is ultimate meaning or purpose in life is illusory. Provine’s claim, quoted at the outset of this chapter, that modern science implies that there is no ultimate meaning for humans, that there are no gods, and that there are no absolute guiding principles for human society, is an example of this. Some versions of the R-​variety of existential scientism are full, others partial. One might think that science can answer all our existential questions or that science can replace all aspects of traditional religions. But one might also think that science can only replace certain aspects of traditional religions and secular ideologies, such as their answers to questions about the ultimate origin of human beings, but not answers to questions about meaning and purpose in life. Even its I-​variety seems to admit of full and partial varieties. One might think that science shows that God is an illusion, but that there is

17.  This is how Stenmark (2001:  viii) characterizes existential scientism. This kind of scientism is defended by Wilson (1978: 201–​207), who argues that traditional religion and secular ideologies should be replaced with what he calls “scientific materialism.”

39

A Conceptual Map of Scientism  • 

39

nonetheless objective meaning and purpose. One might think that properties concerning value and meaning supervene on natural properties, even though there is no God. Some have included other theses under the umbrella of “existential scientism.” According to Stenmark, for instance, the idea that evolutionary theory can explain religious beliefs and the view that it can undermine religious belief also count as part of existential scientism. It is true that adherents of existential scientism are likely to adopt these theses as well, but we should not conclude from that that these theses are instances of existential scientism. Several religious scientists and philosophers embrace the thought that our religious beliefs are produced by something like a Hyperactive Agency Detection Device,18 and many of them would agree that certain ideas that were inspired by religion, such as that the earth is at the center of the universe, have been undermined by science. Still, they do not seem thereby to count as adherents of scientism. 1.2.6 Further Varieties of Universal Scientism? Are these four versions of universal scientism exhaustive? Let me discuss two proposals for further varieties of universal scientism. First, Stenmark distinguishes the thesis that a belief is rational only if it is the deliverance of natural science as a separate variety of scientism and calls it rational scientism. He interprets “epistemological scientism” as the claim that only the natural sciences can deliver knowledge and rightly points out that that thesis is conceptually distinct from rational scientism. He fails to acknowledge, though, that the two are closely related to each other. Imagine that you believe that only the natural sciences can deliver knowledge. Imagine also that you consider some proposition p that you believe you have no scientific evidence for. Then, if you are rational, you will believe that you cannot know that p. You will realize that if you nonetheless believe that p, you might turn out to be lucky and hold a true belief, but you will also realize that in such a case you will not know that p. But if you are aware of the fact that you cannot know that p, then how could you possibly rationally believe that p? On most accounts of rationality, the belief that one cannot know that p provides a defeater for rationally believing that p. Someone who is sufficiently rational will realize that if (she believes that) she cannot know that p, then she

18. For more on this, see Hans van Eyghen, Rik Peels, and Gijsbert van den Brink (2019).

40

4 0   •  R i k   P e e l s

cannot rationally believe that p. This means that epistemological scientism, in conjunction with some highly plausible principles about knowledge and rationality, entails rational scientism. And that means that we have good reason not to treat rational scientism as a separate variety of universal scientism, but as a version of epistemological scientism. Second, Stenmark also distinguishes what he calls axiological scientism. The idea here is that natural science is more valuable than other ways of learning. Those other ways usually include at least the humanities, but some adherents of scientism add politics, sports, art, literature, and philosophy to the list.19 Some philosophers seem to understand scientism almost exclusively along the lines of axiological scientism. According to Tom Sorell, for instance, scientism is the view that natural science is much the most valuable part of human learning.20 Is axiological scientism truly a separate variety of scientism? I doubt it is. For, one might ask in what sense the natural sciences are supposed to be more valuable than other ways of learning. In other words, what kind of value is referred to in axiological scientism? Three possibilities come to mind: Such value could be epistemic, moral, or existential. In other words, the natural sciences can be more valuable in that they are more likely to lead to knowl­ edge, in that they guide us in leading the good life, or in that they help us to meet our existential needs. But axiological scientism will then be reducible to respectively epistemological, moral, and existential scientism. The only other kind of value that I  can think of is aesthetic value:  the value that beautiful music, painting, and architecture have. But it seems that (virtually) no one has made the rather bold assertion that science is aesthetically more valuable than other ways of learning. That means that we should not treat axiological scientism as an additional variety of scientism. 1.2.7 First Conceptual Map: The Varieties of Scientism I will now provide the first conceptual map of scientism. It displays scientism’s varieties. It should be read from left to right. In order to end up with a thesis

19. Radnitzky (1978: 1011) also considers the claim that science is more valuable than other realms of life as a variety of scientism. 20.  Sorell (1991:  9):  “What is crucial to scientism is not the identification of something as scientific or unscientific but the thought that the scientific is much more valuable than the non-​scientific, or the thought that the non-​scientific is of negligible value.” (See also Sorrell 1991: x, 1.)

41

A Conceptual Map of Scientism  • 

41

that is sufficiently precise to be assessed, one should make three or, in the case of moral and existential scientism, even four choices: (a) Is it academic or universal scientism? (b) In the case of academic scientism: Is it methodological or eliminative scientism? In the case of universal scientism: Is it epistemological, ontological, moral, or existential scientism? (c) Is it full or partial scientism? (d) In the case of moral and existential scientism:  Is it Replacement (R)-​ scientism or Illusion (I)-​scientism? Given that, as I argued, moral I-​scientism will always be full rather than partial, there are 15 varieties of scientism. The difference between ellipses and squares indicates that one should choose between certain descriptive adjectives (say, academic or universal, partial or full) to end up with a specific thesis (say, eliminative partial weak scientism). The arrow between methodological and eliminative scientism with a cross over it indicates mutual exclusivity. The other arrows denote implication. As I said, this figure displays which choices an adherent of scientism as well as those who critique scientism will have to make in specifying what they are talking about. I suspect that many philosophers and scientists will adhere to at least some variety of scientism and quite a few to several of them. The figure makes clear that there are six main varieties of scientism: methodological, eliminative, epistemological, ontological, moral, and existential. Given that there is only one instance of mutual exclusion, namely, that between methodological and eliminative scientism, the adherent of scientism could in principle combine five varieties of scientism: epistemological, ontological, moral, existential, and either eliminative or methodological scientism.

1.3 The Interrelations of Scientism’s Varieties How do these varieties of scientism relate to each other, apart from the relation of entailment that holds by definition between full and partial scientism? Rather than bothering the reader with a defense of each of the relations that I take to obtain, I merely present the results of my analysis and refer those who want to know more to the Appendix, in which I spell out my reasons for thinking that each of these relations obtains. The implicatory relations among the 15 varieties of scientism described in Section 1.2 and referred to on the right-​hand side of Figure 1.1 are also depicted in Figure 1.2. The second

42

full

1. Full methodological scientism

methodological partial

2. Partial methodological scientism

academic full

3. Full reductive scientism

reductive partial

full Scientism

4. Partial reductive scientism

5. Full epistemological scientism

epistemological partial

full

6. Partial epistemological scientism

7. Full ontological scientism

ontological partial

universal

full

8. Partial ontological scientism

9. Full R-moral scientism

R partial

10. Partial R-moral scientism

moral 11. I-moral scientism

I

full

12. Full R-existential scientism

R partial

13. Partial R-existential scientism

existential full

14. Full I-existential scientism

I partial

Figure 1.1  Varieties of scientism

15. Partial I-existential scientism

43

A Conceptual Map of Scientism  •  Full R-existential scientism

Full R-moral scientism

Full ontological scientism

43

Full epistemological scientism

I-Moral scientism Partial ontological scientism

Full I-existential scientism

Partial I-existential scientism

Partial epistemological scientism

Partial R-existential scientism Partial R-moral scientism

Partial reductive scientism

Full reductive scientism

Full methodological scientism

Partial methodological scientism

Figure 1.2  Logical relations between the varieties of scientism

figure hierarchically displays the relations of implication that hold between different versions of scientism. For reasons of simplicity, I have not drawn arrows where a full version of scientism implies a partial version of scientism (that saves us seven arrows). The dotted lines indicate a disjunctive implication. For example, full epistemological scientism entails either full eliminative or full methodological scientism. Let me also point out that partial R-​moral scientism and partial R-​existential scientism could also have been placed one level higher. A careful look at Figure  1.2 reveals some important facts about the relations between the varieties of scientism, as I defined them. There are four kinds of scientism that are at the top of the hierarchy in the sense that they imply other varieties of scientism without being implied by some other kind of scientism. At five different levels, there are nine kinds of scientism that are in the middle in that they are implied by certain kinds of scientism but also imply other kinds of scientism. And there are two varieties of scientism that are at the bottom in that they are implied by other kinds of scientism without

4

4 4   •  R i k   P e e l s

implying some other kind of scientism. Also, eliminative scientism and methodological scientism are clearly comparatively weak theses in that all of their varieties are found at the bottom of the hierarchy. Epistemological and ontological scientism, as well as the Replacement-​varieties of moral and existential scientism are, depending on its specific version, to be found at the top or in the middle and are, therefore comparatively strong or average. Also, the R-​versions of moral and existential scientism are not entailed by any other kind of scientism. They do entail certain varieties of epistemological, ontological, eliminative, and methodological scientism, but adopting one of the latter varieties of scientism does not commit one to being an adherent of scientism about morality or existential issues. Next, full ontological scientism and full epistemological scientism are not entailed by any other variety of scientism, but they do entail other kinds of scientism, especially certain varieties of academic scientism. Generally, then, these are fairly strong versions of scientism. The variety of scientism that is implied by the largest number of other varieties of scientism is partial eliminative scientism, the thesis that at least some academic disciplines can be reduced to the natural sciences. It is implied by five other varieties of scientism. Finally, given that eliminative scientism seems stronger than methodological scientism in that it leaves no place at all for academic disciplines other than the natural sciences, the strongest position one could adopt is a combination of full epistemological, full ontological, full moral, full existential, and full eliminative scientism. Such a strong version seems adopted by William Provine and Alex Rosenberg (see the quotes earlier in this chapter). Finally and most importantly, Figure 1.2 gives us good reason, in evaluating scientism, to focus on the variety of partial epistemological scientism. After all, it is entailed by most other varieties of scientism. On partial epistemological scientism, in a smaller or larger realm of life that goes significantly beyond what are widely considered to be the borders of the natural sciences, only the natural sciences provide rational belief and knowledge. This is implied by full epistemological scientism, full and partial ontological scientism, I-​moral scientism, full and partial R-​moral scientism, full and partial R-​existential scientism, and by full and partial I existential scientism.

1.4 A Unifying Definition of Scientism An important question that arises from this overview is whether all these varieties of scientism have something in common such that in virtue of meeting that condition they count as varieties of scientism. Slightly more

45

A Conceptual Map of Scientism  • 

45

precisely, is there a nontrivial condition—​not a disjunctive condition like being one of these varieties of scientism—​that they all satisfy such that they are varieties of scientism because of that? This is an important question, for an answer to it might give us insight into how a conceptual map of scientism is to be drawn. For example, if scientism should be understood as a family resemblance concept, on which something counts as an instance of scientism if it is sufficiently similar to other instances of scientism, the resulting conceptual map will be rather different from the one we get when there is a set of necessary and sufficient conditions for being an instance of scientism. 1.4.1 Peacocke’s Definition Many definitions of scientism suffer from the fact that they exclude one or several of the varieties of scientism that we distinguished above, theses that are widely thought to be instances of scientism. Take the following definition, provided by Arthur Peacocke: Scientism1: The view that the only kind of reliable knowledge is that provided by science, coupled with a conviction that all our personal and social problems are “soluble” by enough science. (Peacocke 1993: 8) With some flexibility, one could interpret the phrase that “our personal and social problems are “soluble” by enough science” as a statement of moral and existential scientism. Even then, though, methodological and eliminative academic scientism are only implicitly present in the definition and ontological scientism is completely absent from the picture. We need a broader definition, in order to do justice to the wide variety of scientistic theses that can be found in the writings of philosophers and scientists. 1.4.2 Radnitzky’s Definition More encompassing is Gerard Radnitzsky’s definition: Scientism2: The view that science has no boundaries, i.e. that eventually it will answer all theoretical questions and provide solutions for all our practical problems. (Radnitzky 1978: 1008; cf. also Churchland 2011: 3) This definition is broad enough to encompass at least eliminative scientism and each of the four varieties of universal scientism. For instance, ontological

46

4 6   •  R i k   P e e l s

and epistemological scientism are each plausibly interpreted as varieties of the view that science can answer all our theoretical questions. Radnitzky’s definition faces another problem, though. For it seems that we have now ruled out some versions of partial scientism. One might think, for instance, that only science provides us knowledge about the natural world, including human beings, but that we need our moral faculties rather than science to acquire knowledge about nonnatural properties, such as goodness and badness. Radnitzky’s definition would have the implausible implication that such a view would not count as an instance of scientism because it does not say that all our questions can be answered by science. Peacocke’s definition turned out to be too weak, whereas Radnitzky’s analysis of scientism turns out to be too strong. 1.4.3 Stenmark’s Definition Stenmark’s definition is broader than Peacocke’s but weaker than Radnitzky’s: Scientism3:  The view that the boundaries of science should be expanded to include disciplines (or answers to questions) that have not previously been considered a part of the domain of science. (Stenmark 2001: 20, 133) The idea that the conceptual core of scientism is the expansion of the natural sciences seems promising to me. After all, both academic and universal scientism claim that the boundaries of the natural sciences should be expanded, either to academic disciplines different from the natural sciences or to other realms of reality. How such an expansion is to be cashed out differs from one variety of scientism to the other. On eliminative academic scientism, for instance, the methods of one or several academic disciplines are to be transposed to that of another or several other academic disciplines. And on epistemological scientism, the natural sciences are to be expanded in the sense that they should also tell us what to believe about realms of life that seem to be radically different from nature, such as human culture. Unfortunately, the definition is too strong. Take religious belief. About two centuries ago, the explanation of why people believe in God was not considered to be a proper part of the natural sciences, but rather a proper part of theology. During the last few decades, scientists have offered several empirical explanations of belief in God. Justin Barrett, for instance, has argued

47

A Conceptual Map of Scientism  • 

47

that belief in God is the result of a Hyperactive Agency Detection Device, and Deborah Kelemen has argued that people are intuitive theists in that they are born with a strong, although resistible, tendency to give teleological explanations of natural events (Barrett 2012; Kelemen 2004). It seems false, though, to dub these scientific explanations scientistic. Many theists, for instance, have gladly embraced such empirical explanations as giving insight into the mechanisms that God has apparently used to produce belief in him. It seems that such explanations become versions of scientism only if they are debunking in the sense that they are taken to imply that belief in God is an illusion. 1.4.4 A Unifying Definition Stenmark’s definition should be revised in two regards in order to be tenable. First, it should not be spelled out in terms of which realms of reality were previously considered the domain of science. The history of science displays increasing insight into the proper domain of the sciences. It turned out that that domain was larger than initially thought. What matters, therefore, is not whether something was considered to be an autonomous domain, but whether something is widely regarded as an autonomous domain. This means that “scientism” is an indexical term; what was once an instance of scientism may no longer be so. Second, it is better to talk about entire “realms of life” and academic disciplines rather than merely “answers to questions.” For, as we saw, scientism makes claims not merely about specific questions and answers, but about entire domains. This leads to the following definition: Scientism4: The view that the boundaries of the natural sciences should be expanded to include academic disciplines or realms of life that are widely considered not to be the domain of science. Belief in God, for instance, does not fall under “disciplines or realms of life that are widely considered not to be the domain of science,” for many believers acknowledge that there may well be a good natural explanation of religious belief. Belief in God, after all, should be distinguished from God’s existence or the question whether or not God exists. How precisely we should understand something’s “not being the domain of science” differs from the one variety of scientism to the other. When it

48

4 8   •  R i k   P e e l s

comes to epistemological scientism, for instance, X’s being not being the domain of science means that there are nonscientific or maybe even only nonscientific ways to acquire knowledge or rational belief about X. And in the case of moral scientism, X’s not being the domain of science means that the natural sciences cannot replace common-​sense morality as a way of dealing with X or that natural scientific research does not justify the thesis that our moral intuitions and beliefs with regard to X are an illusion.

1.5 Conclusions Let me select what I consider to be the four most important conclusions of this chapter. First, scientism is the thesis that the boundaries of natural science should be expanded in order to include academic disciplines or realms of life that are widely considered not to belong to the realm of science. Second, every adherent and critic of scientism should make clear which variety of scientism she adheres to or criticizes. In doing so, she should specify whether she is talking about (a)  academic or universal scientism; (b) eliminative, methodological, epistemological, ontological, moral, or existential, scientism; (c) full or partial scientism; and (d) in the case of moral and existential scientism: replacement or illusion scientism. Third, the strongest version of scientism one could defend is a conjunction of the following theses: strong full epistemological scientism, strong full ontological scientism, strong full Illusion-​moral scientism, strong Illusion-​ existential scientism, and strong full eliminative scientism. Fourth, when it comes to assessing scientism, it makes sense to put partial epistemological scientism center stage, since it is entailed by most other varieties of scientism. On partial epistemological scientism, in a smaller or larger realm of life that goes significantly beyond what are widely considered to be the borders of the natural sciences, only the natural sciences provide rational belief and knowledge. I have intended throughout this chapter not to say anything for or against scientism or some specific version of scientism. I  have done that elsewhere (Peels 2016, 2017, 2018). What I have said should be compatible with both a defense and a critique of scientism. It seems to me that the conceptual map of scientism that I  have provided in this chapter provides a good starting point providing such a defense or critique. First, it induces every critic and adherent of scientism to make clear which specific variety of scientism she has in mind. Second, it provides insight into which other varieties of scientism

49

A Conceptual Map of Scientism  • 

49

one is committed to defend or criticize in virtue of understanding “scientism” in that particular way.21 Appendix In this Appendix, I defend Figure 1.2, that is, the figure that displays how the varieties of scientism that I distinguished in Figure 1.1 relate to each other.

1.6 Methodological and Eliminative Academic Scientism Remember that, on methodological scientism, all or some academic disciplines different from the natural sciences should adopt the methods of the natural sciences in order to solve the problems of those fields, whereas on eliminative scientism, all or some other academic disciplines are simply reducible to the natural sciences: The problems in those fields are illusory. These are two distinct theses that do not imply each other. Methodological scientism about history as an academic discipline, for instance, says that history should adopt the empirical model of the natural sciences. It does not follow that everything historical is a matter of biology, chemistry, or physics. Methodological scientism does not imply any kind of elimination apart from what one could call a methodological elimination. Also, eliminative scientism does not imply methodological scientism. In fact, they exclude each other. If, say, the humanities should adopt the methods of the natural sciences but remain distinct academic disciplines, as methodological scientism implies, then apparently they should not be eliminated. Hence, methodological and eliminative scientism are mutually exclusive.

1.7 Academic and Universal Scientism Clearly, no version of academic scientism implies any kind of universal scientism, for, by definition, academic scientism is restricted to the academy, whereas universal scientism is not. What about the other way around? One

21. For their helpful comments on earlier versions of this chapter, I would like to thank Jan Boersema, Leon de Bruin, Henk de Regt, Jeroen de Ridder, Jeroen Hopster, Judith Jansen, Peter Kirschenmann, Arthur Rob, Stefan Roski, Emanuel Rutten, Jeroen Smid, Hein van den Berg, Gijsbert van den Brink, and René van Woudenberg. This publication was made possible through the support of a grant from the Templeton World Charity Foundation. The opinions expressed in this publication are those of the author and do not necessarily reflect the views of the Templeton World Charity Foundation.

50

5 0   •  R i k   P e e l s

might think universal scientism implies academic scientism by definition, because, as I said, universal scientism applies both within and outside of the academy. We should not forget, though, that academic scientism is a claim about the method of academic disciplines or whether particular academic discipline should be eliminated, whereas universal scientism is a claim about what we can know, what exists, moral values, or meaning. It cannot be true by definition, then, that universal scientism implies academic scientism. Yet, I do think that each variety of universal scientism implies some variety of academic scientism, at least in conjunction with one or two plausible principles. Take partial epistemological scientism. It claims that all possible knowl­edge about some specific topic that is not normally considered to belong to the realm of natural science, is to be provided by natural science. Such realms could be those of the supernatural, belief and desire attributions, and morality. But it seems right to say that theology, psychology, and ethics also strive for knowledge about respectively God, people’s beliefs and desires, and good and evil. That version of epistemological scientism that says that there is no knowledge to be had in these areas implies eliminative scientism, whereas that version that says that such knowledge is possible, but that only the methods of the natural sciences can deliver such knowledge, implies methodological scientism. If partial epistemological scientism has these implications, then, clearly, so does full epistemological scientism. Does ontological scientism also imply some kind of academic scientism? It does, at least in conjunction with some plausible background assumptions. On full ontological scientism, only the material cosmos exists that can, in principle, be investigated by science. Consciousness, objective good and evil, beliefs, and so forth, do not exist. That would mean that academic disciplines like philosophical ethics, theology, and psychology need to be given up, for the object of their research would not exist. Hence, full ontological scientism implies eliminative scientism. Partial ontological scientism, such as the claim that humans are nothing but molecules, implies partial eliminative scientism, for example, eliminative scientism about psychology or philosophical anthropology. Finally, let us consider moral and existential scientism. If moral scientism says that traditional ethics and common-​sense morality should be replaced with a scientific ethics (its R-​variety), then ethics should adopt the methods of the natural sciences. And, mutatis mutandis, the same applies to existential scientism. On the R-​variety of moral and existential scientism, moral and

51

A Conceptual Map of Scientism  • 

51

existential questions are still thought to make sense and, therefore, we should not give up ethics and, say, theology altogether—​which would amount to eliminative scientism. Rather, these academic disciplines should adopt the methods of the natural sciences, which means that they imply methodological scientism. However, even the full moral and existential scientism on the R-​variety imply at most partial methodological scientism, since they do not imply that all academic disciplines different from natural science should adopt the methods of natural science. What about the I-​variety of moral and existential scientism? On this variety, morality is an illusion, and meaning and purpose or God is an illusion. The I-​variety of moral scientism clearly implies partial eliminative academic scientism, for if morality is an illusion, then, it seems, we should give up ethics. On the I-​variety of existential scientism, ultimate meaning and purpose or God or all of these are an illusion. The more of these are an illusion, the more academic disciplines (philosophical ethics, theology) should be given up.

1.8 The Varieties of Universal Scientism Finally, let us turn toward the relations that hold between the varieties of universal scientism. Let us start with epistemological scientism. Clearly, from the thesis that we can know only things by way of natural science, nothing follows about what does or does not exist. It may be that, even if all we can know is the product of natural science, there exist things that we cannot know anything about. Nor does it entail moral or existential scientism, not even on full epistemological scientism. From the fact that all knowledge is to be delivered through natural science, it does not follow that morality or God is an illusion, nor that science should provide us with a morality or that it should replace traditional religions. One might think, for instance, that science is unable to provide answers to the big questions of life and that we need religions for that, even though they do not provide us with knowledge, but only with helpful answers to live with. Hence, epistemological scientism, whether on its full or partial variety, entails neither ontological, nor moral, nor existential scientism. What does ontological scientism entail? According to Stenmark, ontological scientism entails epistemological scientism. For, if the only things that exist are the ones science can in principle discover or the ones that play a role in our

52

5 2   •  R i k   P e e l s

scientific theories, then the only kind of knowledge we can have is scientific knowledge. Here is how he motivates this claim: if something cannot be discovered by science, then we cannot know anything about it either. If, for instance, God cannot be discovered by scientific means, it follows that we cannot know anything about God. We can only know something about people’s thoughts about God, because these thoughts are, presumably, real, even though the intended object of these beliefs would not exist. (Stenmark 1991: 18, 24) This line of reasoning seems somewhat misleading to me. For, the claim that (in some realm) only those things exist that can in principle be investigated by natural science, is perfectly compatible with the claim that we can acquire knowledge about those entities by nonscientific means. Ontological scientism implies at most partial epistemological scientism, the idea that in some realms of life only natural science can provide us with knowledge. That will depend, however, on how much one takes the natural sciences to acknowledge to exist. There is good reason to think that the natural sciences do not assume or imply certain things to exist that in daily life many of us do believe to exist, such as consciousness, objective good and evil, and God. Given this plausible assumption, we can say that ontological scientism, even full ontological scientism, implies at most partial epistemological scientism. If partial ontological scientism about certain areas is correct, then we can have no knowledge about God or consciousness, apart, of course, from knowledge that they do not exist. Full ontological scientism also implies the I-​varieties of moral and existential scientism. If only those things exist that can in principle be investigated by natural science, then, given that natural science does not seem to admit the existence of God, objective meaning, or good and evil, it would follow that morality, God, and objective meaning are illusions. As to partial ontological scientism, it seems that as such it does not imply moral or existential scientism. Only those particular versions of it do so that say that there are no objective moral values or that God and objective meaning do not exist, that is, those specific versions that are more or less identical to moral and existential scientism. Let us turn to moral scientism. Both its R-​and I-​varieties imply partial epistemological scientism. If natural science should replace common-​sense ethics, then that is presumably because common-​sense ethics does not lead to knowledge. And if morality is an illusion, then nothing can be known about it (again, except for such trivial facts as that it is an illusion). As we

53

A Conceptual Map of Scientism  • 

53

saw, moral scientism on its I-​variety implies partial ontological scientism. This does not seem to be the case on its R-​variety: If science or the methods of the natural sciences should replace traditional ethics, then it is not clear that anything follows about what exists. One might think, for instance, that, for all we know, there are objective moral values, but that we need scientifically construed ethics to know anything about them. Finally, let us consider existential scientism. If there is no God or if there is no objective meaning, as the I-​variety says, then we cannot know anything nontrivial about them, so that both partial ontological scientism and partial epistemological scientism are true. If the R-​variety is true, the natural sciences should replace traditional religions and secular ideologies. It would be implausible to claim such a thing while maintaining that knowledge about God through common sense or other mechanisms that are built in human beings is possible.22 Thus, existential scientism on its R-​variety implies partial epistemological scientism. It does not imply partial ontological scientism, though, for one might think that, for all we know, God exists, but that religions are useless and should be replaced with science. Does existential scientism imply moral scientism? Stenmark thinks it does, for, “[r]‌eligions and world views are in general taken to include some ideas about how we should live and what a good human life is” (Stenmark 2001: 19). I think Stenmark is right that most religions also encompass certain ideas about what the good life is. Yet, it does not follow that existential scientism implies moral scientism. For, one might think that science should replace traditional religions, with their doctrines about God and life after death, but that we still need common-​sense morality. And one might think that God does not exist and that there is no ultimate purpose in life, but that there are still objective facts about what is morally right and wrong. Hence, existential scientism, neither on its R-​variety nor on its I-​variety, implies moral scientism.

References Atkins, Peter W. 1995. “Science as Truth.” History of the Human Sciences 8(2): 97–​102. Barbour, Ian G. 1990. Religion in an Age of Science:  The Gifford Lectures 1989–​1991. Volume 1. New York: SCM Press.

22. Here, we should think especially of the sensus divinitatis, a mechanism, that, according to Alvin Plantinga, God, if he exists, is likely to have implanted in us, in order for us to know him. See especially Plantinga (2000).

54

5 4   •  R i k   P e e l s

Barrett, Justin M. 2012. Born Believers:  The Science of Children’s Religious Belief. New York: Free Press. Carnap, Rudolf. 1967. The Logical Structure of the World:  And Pseudoproblems in Philosophy. Translated by Rolf A. George. Chicago:  Open Court. Originally published in Der logische Aufbau der Welt:  Scheinprobleme in der Philosophie (Hamburg: Felix Meiner Verlag, 1961). Churchland, Patricia S. 1987. “Epistemology in the Age of Neuroscience.” Journal of Philosophy 84(10): 544–​553. Churchland, Patricia S. 2011. Braintrust:  What Neuroscience Tells Us about Morality. Princeton, NJ: Princeton University Press. Crick, Francis. 1966. Of Molecules and Men. Seattle:  University of Washington Press. Crick, Francis. 1994. The Astonishing Hypothesis:  The Scientific Search for the Soul. London: Simon & Schuster. Dawkins, Richard. 1989. The Selfish Gene, 2nd edition, original edition 1976. Oxford: Oxford University Press. Haack, Susan. 1995. Evidence and Inquiry:  Towards Reconstruction in Epistemology. Oxford: Blackwell. Haack, Susan. 2007. Defending Science within Reason: Between Scientism and Cynicism. Amherst, NY: Prometheus Books. Harris, Sam. 2010. The Moral Landscape: How Science Can Determine Human Values. New York: Free Press. Hawking, Stephen and Leonard Mlodinow. 2010. The Grand Design. New  York:  Bantam Books. Hayek, Friedrich A. 1979. The Counter-​Revolution of Science:  Studies on the Abuse of Reason. Indianapolis, IN: Liberty Fund. Hughes, Austin L. 2012. “The Folly of Scientism.” The New Atlantis 37(Fall): 32–​50. http://​www.thenewatlantis.com/​publications/​the-​folly-​of-​scientism Kelemen, Deborah. 2004. “Are Children ‘Intuitive Theists’? Reasoning about Purpose and Design in Nature.” Psychological Science 15(5): 295–​301. Midgley, Mary. 1992. Science as Salvation:  A Modern Myth and Its Meaning. London: Routledge. Neurath, Otto. 1987 (1932). “Unified Science and Psychology.” In Unified Science, edited by Brian McGuinness, 1–​23. Dordrecht: Kluwer. Nielsen, Kai. 1997. “Naturalistic Explanations of Religion.” Studies in Religion 26(4): 441–​4 66. Peacocke, Arthur. 1993. Theology for a Scientific Age: Being and Becoming—​Natural and Divine. Oxford: Blackwell. Peels, Rik. 2016. “The Empirical Case against Introspection.” Philosophical Studies 173(9): 2461–​2485.

5

A Conceptual Map of Scientism  • 

55

Peels, Rik. 2017. “Ten Reasons to Embrace Scientism.” Studies in History and Philosophy of Science A 63: 11–​21. Peels, Rik. 2018. “The Fundamental Argument against Scientism.” In Science Unlimited? The Challenges of Scientism, edited by Maarten Boudry and Massimo Pigliucci, 165–​184. Chicago: University of Chicago Press. Peterson, Michael, William Hasker, Bruce Reichenbach, and David Basinger. 1991. Reason and Religious Belief:  An Introduction to the Philosophy of Religion. New York: Oxford University Press. Plantinga, Alvin. 2000. Warranted Christian Belief. New  York:  Oxford University Press. Provine, William. 1988. “Evolution and the Foundations of Ethics.” Marine Biology Laboratory Science 3: 25–​29. Quine, Willard V.O. 1992. “Structure and Nature.” Journal of Philosophy 89(1): 5–​9. Radnitzky, Gerard. 1978. “The Boundaries of Science and Technology.” In The Search for Absolute Values in a Changing World, Vol. II, Proceedings of the Sixth International Conference on the Unity of the Sciences, 1007–​1036. New York: The International Cultural Foundation Press. Rescher, Nicholas. 1999. The Limits of Science. Original edition 1984. Pittsburgh, PA: University of Pittsburgh Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality: Enjoying Life without Illusions. New York: W.W. Norton. Ross, Don, James Ladyman, and David Spurrett. 2007. “In Defence of Scientism.” In Every Thing Must Go: Metaphysics Naturalized, James Ladyman, Don Ross, David Spurrett, and John Collier, 1–​65. Oxford: Oxford University Press. Ruse, Michael. 1979. Sociobiology: Sense or Nonsense? Dordrecht: Reidel. Russell, Bertrand. 1946. History of Western Philosophy, and its Connection with Political and Social Circumstances from the Earliest Times to the Present Day. London: Allen and Unwin. Russell, Bertrand. 1967. Why I Am Not a Christian and Other Essays on Religion and Related Subjects. London: Unwin Books. Sagan, Carl. 2002. Cosmos. Original edition 1980. New York: Random House. Sellars, Wilfrid. 1963. “Empiricism and the Philosophy of Mind.” In Science, Perception and Reality, 127–​196. London: Routledge. Snow, Charles P. 1972. The Two Cultures: And a Second Look. Original edition 1959. Cambridge: Cambridge University Press. Sorell, Tom. 1991. Scientism:  Philosophy and the Infatuation with Science. London:  Routledge. Stenmark, Mikael. 2001. Scientism: Science, Ethics and Religion. Aldershot: Ashgate. Stevenson, Leslie and Henry Byerly. 1995. The Many Faces of Science: An Introduction to Scientists, Values, and Society. Boulder, CO: Westview Press.

56

5 6   •  R i k   P e e l s

Stich, Stephen P. 1983. From Folk Psychology to Cognitive Science:  The Case against Belief. Cambridge, MA: MIT Press. Trigg, Roger. 1993. Rationality and Science: Can Science Explain Everything? Oxford:  Blackwell. Van Eyghen, Hans, Rik Peels, and Gijsbert van den Brink, eds. 2019. New Developments in the Cognitive Science of Religion: The Rationality of Religious Belief. Dordrecht: Springer. Wilson, Edward O. 1975. Sociobiology: The New Synthesis. Cambridge, MA: Harvard University Press. Wilson, Edward O. 1978. On Human Nature. Cambridge, MA: Harvard University Press.

57

2

SCIENTISM AND ITS RIVALS

Mikael Stenmark

2.1 Introduction What is scientism and, no less important, where and why does it differ from its rivals? The second aspect is crucial because, in assessing scientism, we also need to identify its rivals and the borderlines and gray zones between scientism and these other options ​if we reject one we need to know what alternatives there are and where possible overlaps are. In this chapter I  shall offer answers to these questions and distinguish between different types of scientism. I shall also suggest that liberal naturalism, humanism, social constructionism, religious naturalism, and theism are best understood as rivals to scientism, although that does not mean that they are on all accounts necessarily incompatible with scientism. Again there are borderlines and gray zones between scientism and its rivals. So what is scientism? The Online Medical Dictionary defines scientism as the “concept that other areas besides science can and should be studied in a scientific manner, for example the humanities” (Anonymous 1997). Understood in this way, scientism is a standpoint within academic disputes about whether or not the human and social sciences—​such disciplines as psychology, sociology, anthropology, history, philosophy, literary studies, and religious studies—​are genuinely autonomous and legitimate studies in their own right, like physics, chemistry, and biology. If one tends to answer “no” to this question then one embraces scientism. But the notion is also used in disputes about science and “the rest,” that is, arguments concerning the relationship between science and non-​academic accomplishments, activities, and quests.

58

5 8   • M i k a e l

Stenmark

In particular, the dispute is about whether there exists any genuine knowledge outside of science or whether science provides the only good and legitimate way to arrive at reasonable beliefs. If one tends to answer “yes” to this question about the limits of knowledge and rationality, then one is a spokesperson for scientism. I suggest that we bind these different senses of scientism together in one notion. We can do that if we define scientism as the view that states that the boundaries of science could and should be expanded in such a way that something that has not previously been understood as science can now become or be transformed into science (or if that is after all not possible it must be rejected). Scientism is the view that everything eventually, or as much as possible, could and should be understood in terms of the natural sciences.1 Thus a possible synonym for scientism is “scientific expansionism.” How exactly should the boundaries of science be expanded, what more precisely is to be included within science, and what consequences are there of such expansion? Though these are issues on which there is disagreement, the basic idea is that science can answer many more questions (or different types of questions) than we have previously thought possible. In its most ambitious form, scientism can be defined as the view that science has no real boundaries, that is, eventually it will answer all empirical, theoretical, practical, moral, and existential questions. Science will in due time solve all genuine problems encountered by humankind. However, some proponents of scientism are less ambitious than others in their extension of the boundaries of science. For this reason I propose that we treat people’s scientistic stance as a matter of degree so that we can talk about weaker and stronger forms of scientism. Scientific restrictionism would on this general level be the rival to scientism or scientific expansionism. Restrictionists would argue that the humanities are fine as they are, because questions within these disciplines presuppose methods and theories that can handle values, reasons, meanings, and intentions; and that the methods and theories of the natural sciences cannot capture these features of reality and are thus inappropriate or of limited use in such contexts. Or they might maintain that the scope of what we can know is not limited to, nor can be reduced to, scientific knowledge. On this view, the humanities give us, among other things, hermeneutical

1. For a unifying definition of scientism, see also section 1.4 of Rik Peels’s contribution to this volume.

59

Scientism and Its Rivals  • 

59

knowledge—​knowledge of how to interpret texts—​not accessible to the methods of the natural sciences. The world is taken to be bigger than the world of the natural sciences by such protagonists, who also believe that we can obtain knowledge about this bigger world that cannot be reduced to scientific knowledge. I would venture to say that the received or “official” view of science is a form of scientific restrictionism, so scientism is in this sense a controversial view. I would nevertheless suggest that we should not use the term “scientism” as a pejorative label to indicate automatically an improper use of science, but more as a descriptive but contested term like “feminism” or “naturalism,” “atheism” or “theism.” After all, science is an evolving enterprise and we cannot therefore say a priori that the proposed expansions are illegitimate or misguided. We ought to take scientistic claims seriously because science has in the past managed to deal with issues that had previously been thought to be beyond its scope. Therefore, the issue about the limits of science cannot be settled once and for all. Instead we need to look again and again at the particular claims of expansion of the scientific domain made by proponents of scientism. Moreover, there are people who say at the outset that they are scientistic in their thinking: people whom we might call “scientizers.” Michael Shermer simply embraces the view of scientism without considering it to be a pejorative term: We show deference to our leaders; pay respect to our elders and follow the dictates of our shamans; this being the Age of Science, it is scientism’s shamans who command our veneration. . . . [B]‌ecause of language we are also storytelling, mythmaking primates, with scientism as the foundational stratum of our story and scientists as the premier mythmakers of our time. (Shermer 2002) Both Alex Rosenberg (2011: 6) and Steven Pinker (2013) want to take the term away from its opponents and appropriate the pejorative for a position they are prepared to defend. It is the overwhelming intellectual and practical successes of science that have led some people to think that there are no real limits to the competence of science, no limits to what can be achieved in the name of science. There is nothing outside the domain of science, nor is there any area of human life to which science cannot successfully be applied. A  scientific account of anything and everything constitutes the full story of the universe and its inhabitants. Or, if there are limits to the scientific enterprise, the idea is that, at least, science sets

60

6 0   • M i k a e l

Stenmark

the boundaries for what we humans can ever know about reality. This is the view of scientism.

2.2 Internal Scientism In what follows I shall distinguish between some of the different forms that scientism can take and will examine how these are related to one another. Or, to put it more exactly, I shall identify four scientistic theses, point out some of their consequences, and try to determine their interrelationships. A number of scientists have recently argued that the new Darwinian synthesis has a much greater scope of application than many have previously thought. They are ready to apply evolutionary theory to all aspects of human existence and to develop a Darwinian social and human science. We thus have the, by now classical, statement of Edward O. Wilson: “It may not be too much to say that sociology and the other social sciences, as well as the humanities, are the last branches of biology waiting to be included in the Modern Synthesis” (Wilson 1975:  4). Others hold similar ideas. Daniel C. Dennett writes that Darwin’s “dangerous idea,” that is, evolution by natural selection, bears “an unmistakable likeness to universal acid:  it eats through just about every traditional concept, and leaves in its wake a revolutionized world-​view, with most of the old landmarks still recognizable, but transformed in fundamental ways” (Dennett 1995: 63). Darwin’s dangerous idea is reductionism incarnate, promising to unite and explain just about everything in one magnificent vision. Richard D. Alexander talks about the most recent discoveries in evolutionary biology as the greatest intellectual revolution of the 20th century. He claims (just like Dennett) that these insights will have a profound impact on our self-​view—​to such an extent that “we will have to start all over again to describe and understand ourselves, in terms alien to our intuitions” (Alexander 1987:  3). Richard Dawkins is equally, if not even more, optimistic when it comes to what modern biology can deliver. He claims that we have “no longer . . . to resort to superstition when faced with the deep problems: Is there a meaning to life? What are we for? What is man?” (Dawkins 1989: 1). According to him, science and in particular biology is capable of dealing successfully with all of these questions. Sometimes, however, the transformation of academic disciplines does not stop here—​with the attempt to develop a Darwinian social and human science—​but continues even within the natural sciences themselves. The idea is not only that the humanities and social sciences can be replaced by biology,

61

Scientism and Its Rivals  • 

61

but that biology can be replaced by or reduced to chemistry, and chemistry to physics. The most famous expression of this form of scientism is probably the statement of the Nobel laureate Ernest Rutherford (1871−1937) that “All science is either physics or stamp collecting.”2 A similar view is also endorsed by Francis Crick, who writes that “eventually one may hope to have the whole of biology ‘explained’ in terms of the level below it, and so on right down to the atomic level. . . . The knowledge we have already makes it highly unlikely that there is anything that cannot be explained by physics and chemistry” (Crick 1966: 14, 98). So one scientistic thesis would roughly say that: (S1) The methods and theories of the natural sciences could and should be extended to the social sciences and the humanities in such a way that these become the central methods and theories used and thus replace or at least marginalize previously used methods or theories considered central to these disciplines. Let us call this form of scientism for academic purposes internal scientism. One advocates internal scientism, as I have already hinted, when one argues that disciplines such as psychology, sociology, anthropology, history, philosophy, and religious studies are not really autonomous and legitimate sciences in their own right like physics, chemistry, and biology. Would someone like Wilson explicitly or implicitly embrace (S1)? It certainly seems that way. Wilson thinks biologists will one day discover a genetically accurate and hence completely fair code of ethics (Wilson 1975: 575). To be able to achieve this goal, philosophers ought to be given a long sabbatical: “Scientists and humanists should consider together the possibility that the time has come for ethics to be removed temporarily from the hands of the philosophers and biologicized” (ibid.: 562). In fact, he thinks that in no other domain of the humanities (than philosophy) is a union with the natural sciences more urgently needed. It is needed because “it is astonishing that the study of ethics has advanced so little since the nineteenth century. The most distinguishing and vital qualities of the human species remains a blank space on the scientific map” (Wilson 1998: 62). Philosophy on this view has no autonomy with respect to the sciences but needs an overhaul by “real” scientists.

2. See http://​en.wikiquote.org/​wiki/​Ernest_​Rutherford.

62

6 2   • M i k a e l

Stenmark

Steven Pinker also seems close to embracing (S1), although he starts by saying that “scientism in the good sense” is not an imperialistic drive to occupy the humanities, despite the fact that his key message is that the humanities needs to be transformed and enlightened by the sciences. He is surprised that what he calls the “writers” in the humanities are not delighted by the new ideas of science, that they resent the intrusion of science into the territories of the humanities, when in fact, “it is just the domain that would seem to be the most in need of an infusion of new ideas” and would in such a case “enjoy more of the explanatory depth of the sciences” (Pinker 2013). It is not that the humanities can offer the sciences anything; the only apparent beneficiaries of the humanities’ consilience with the sciences would be the humanities. Hence Pinker summons the humanities to a process of scientization. The sciences can give explanatory depth to the shallower understanding within the humanities. He wants to transform the nonscientific discourse of the humanities into scientific discourse.

2.3 Epistemic Scientism Probably the best-​known form of scientism is one which expresses a particular idea about the limits of knowledge. It says that the only genuine knowl­ edge about reality is to be found through science, and science alone. The only kind of knowledge we can have is scientific knowledge. A number of scientists and scholars have embraced one version or another of this view. Bertrand Russell maintains that “whatever knowledge is attainable must be attained by scientific methods; and what science cannot discover, mankind cannot know” (Russell 1978: 243). Wilfred Sellars agrees, stating that “science is the measure of all things, of what is that it is, and of what is not that it is not” (Sellars 1963: 173). Rosenberg expresses a similar view and relates it closely to atheism. He writes that: We’ll call the worldview that all us atheists (and even some agnostics) share “scientism.” This is the conviction that the methods of science are the only reliable ways to secure knowledge of anything; that science’s description of the world is correct in its fundamentals; and that when “complete,” what science tells us will not be surprisingly different from what it tells us today. (Rosenberg 2011: 6–​7) So much for the philosophers; scientists can be found who hold the same or similar views. Dawkins tells us: “Science is the only way to understand the real

63

Scientism and Its Rivals  • 

63

world” (Dawkins 1993). R. C. Lewontin moves in the same direction, saying that people ought “to accept a social and intellectual apparatus, Science, as the only begetter of truth” (Lewontin 1997). In his essay, “The Limitless Power of Science,” Peter Atkins advocates the “omnicompetence of science” and believes that “science, with its currently successful pursuit of universal competence . . . should be acknowledged king” (Atkins 1995: 132). Moreover, like Rosenberg connecting scientism to atheism, he writes that “There is no reason to suppose that science cannot deal with every aspect of existence. Only the religious—​among whom I  include not merely the prejudiced but also the underinformed—​hope that there is a dark corner of the physical Universe, or of the universe of experience, that science can never hope to illuminate” (ibid.: 125). Let us call this version or versions of the view epistemic scientism and define it as follows: (S2) The only kind of genuine knowledge we can have is the one provided by the sciences. Science, modeled on the natural sciences, is the only source of real knowl­ edge. If one holds this epistemological view, then it is of course not difficult to understand that one would believe that everything, or at least as much as possible, could and should be understood in terms of science—​because what we cannot understand and explain in terms of science is something that we cannot know anything about at all. Notice that this is not the view (a) that science is the paradigm example of knowledge or rationality:  the standpoint that privileges science when it comes to knowledge and truth. On such a view science provides us with the most reliable path to knowledge and rational belief: Science is the royal road to truth. It is definitely not the view (b) that we should respect science within its domain or area of expertise. View (b) is weaker than view (a), because even if one is not ready to privilege science or see it as the best way to obtain knowl­ edge, one would still be ready to respect the results of the sciences. Epistemic scientism is rather the even stronger view (c) that the only genuine knowledge about reality is to be found through science and science alone. Scholars in the humanities—​not yet being transformed by or reduced to the sciences—​might think that they can obtain knowledge; lawyers in the courtroom might think the same; and so might carpenters, chefs, politicians, ethicists, and religious people, but all are wrong; what they claim to be knowl­ edge is not genuine knowledge because only physics, chemistry, and biology

64

6 4   • M i k a e l

Stenmark

(and their subdisciplines) can provide us with that. This stance, for instance, seems to be shared by Dawkins who quotes and agrees with G. G. Simpson, who states that “all attempts to answer that question [What is man?] before 1859 are worthless and . . . we will be better off if we ignore them completely” (Dawkins 1989: 1). But there is a yet stronger epistemological version of scientism, one which focuses on reasonable or rational belief rather than on knowledge. Epistemic scientism denies only that any claim or belief that cannot be scientifically knowable can constitute knowledge. We cannot know anything about reality that transcends the limits of science. Many people, for instance, have religious beliefs. Their truth cannot it seems be scientifically proved; can these people still be rational in accepting these beliefs? Advocates of epistemic scientism could not reasonably deny that. All they are claiming, in fact, is that we cannot know whether these beliefs are true. From this proposition alone, it does not follow that we cannot be rational in accepting them. What is not scientifically knowable might still be rationally believable. Brian Baxter, in developing a Darwinian worldview, seems to acknowledge this when he writes: “Yet for many people, including many religious believers, science is arguably the only form of intellectual endeavour which can produce something worthy of the term ‘knowledge’, even if it is not guaranteed to do so” (Baxter 2007: 3). So he is aware that there are people who would not say that their religious beliefs constitute knowledge. Perhaps they would say that their believing is rather a matter of faith, but at the same time they would accept that if we can obtain knowledge it is by means of science. Nevertheless these forms of scientism are frequently confused because it is not recognized that knowledge and rationality are two distinct epistemic concepts. It is, however, fairly easy to see that the conditions for knowledge and for rationality cannot be the same. In general we think that people 2000 years ago were rational in believing that the earth was flat (their belief satisfying the conditions for rationality), but we would not say that they knew that it was flat (their belief falling short of satisfying the conditions for knowl­ edge). If they knew, it follows that the shape of the earth must have changed since then. Hence the conditions for knowledge and rationality cannot be the same.3 Consequently, one can be rationally entitled to believe things that are not scientifically knowable.

3. See Mikael Stenmark (1995: 216–​225) for a more substantial discussion of the differences between knowledge and rationality.

65

Scientism and Its Rivals  • 

65

We can call this third, even stronger, epistemological version of scientism rationalistic scientism and identify one more scientistic thesis: (S3) We are rationally entitled to believe only what is scientifically justified. An advocate of this thesis claims that a belief is rationally held only when there are good scientific reasons or scientific evidence for its truth. Consequently, everything outside of science is taken as a matter of merely subjective belief, opinion, or superstition. One can therefore fully understand the agenda to strive to incorporate as many other (of the worthwhile) areas of human life as possible within the sciences, so that not only acquisition of knowledge but also rational consideration can be attained in these fields as well. Who holds such a view? It is possible to interpret Paul Boghossian as giving an expression of (S3) with his statement: “We take science to be the only good way to arrive at reasonable beliefs about what is true, at least in the realm of the purely factual. Hence, we defer to science” (Boghossian 2006:  4). He might, however, be saying that even if science is the only good way to arrive at reasonable beliefs, it is not strictly irrational to arrive at beliefs by other routes. He would of course not then call them reasonable beliefs, but neither would he necessarily call them unreasonable beliefs. Notice also that he restricts his view to factual beliefs, thus making it possible to obtain reasonable beliefs in, for instance, ethics—​if these beliefs are understood to be nonfactual—​in an extra-​scientific, say philosophical or common-​sense, way. Philip Kitcher, at least in his discussion of science and religion, seems rather closer to fully endorsing (S3).4 He maintains that “scientific inquiry sets the standards for the acceptability of beliefs. The heart of the conflict between science and religion is a debate about this thesis” (Kitcher 2008: 11). So I take it that his idea is that only those beliefs—​be they religious or of any other kind—​which satisfy the scientific standards of rationality are acceptable. Notice though that it is not easy to be consistent in one’s thinking if one embraces (S3). For instance, how did Boghossian arrive at those beliefs that he expresses in his book Fear of Knowledge, written against relativism and social constructivism? He concludes by maintaining that “we have examined three distinct arguments [constructivism about facts, about justification and about rational explanation] for the claim that we can never explain belief by appeal

4. But apparently not in other cases: See, for instance, Kitcher (2012).

6

6 6   • M i k a e l

Stenmark

to our epistemic reasons alone; and we have found grounds for rejecting each and every one of them” (Boghossian 2006:  128). Are these grounds scientific? The answer is “no.” There is no appeal to merely physics, chemistry, biology, or the like; rather he develops a fairly convincing philosophical case against these arguments. But if, as he says at the beginning of the book, we take science to be the only good way to arrive at reasonable beliefs about what is true, then his beliefs about the invalidity of social constructivism cannot be reasonable because he is not justified in believing that these claims of social constructivists are false. One belief that Kitcher apparently thinks is true, or at least reasonable, is that many religious people in America embrace a hybrid epistemology that says roughly that, as a true religious believer, one should embrace the result of the sciences when they are irrelevant to religious issues, but reject them when they go against one’s faith and holy scripture (Kitcher 2008: 11). What is the status of this belief of Kitcher’s? Is it the outcome of scientific inquiry? At least, it is not obvious that science can discover epistemologies since they are normative entities. So it is not that epistemologies are too small or too far away for physics, chemistry, biology, or any of their extensions to be able discover and evaluate them; it is simply because they are outcomes of a different sort of inquiry, a philosophical inquiry. Kitcher’s view might be reasonable, but it fails to satisfy the scientistic conditions for either knowledge (S2) or for rationality (S3). He also tells us that “liberal democratic theory is largely dedicated to exploring how people with very different conceptions of what is valuable, divergent attitudes towards the ways in which worthwhile human lives should be structured, can find ways to agree on common policies and institutions” (ibid.: 12). But there is nothing in his writings that suggests how the sciences can set the standards for the acceptability of theories of this sort. This, to me at least, is not surprising because I do not think that Kitcher or for that matter Boghossian really believe that their philosophical inquiry must meet the same standards as the sciences; it suffices if they meet good philosophical or humanistic standards. Perhaps they might think that philosophy cannot provide us with knowledge, but most of the time they write as though it can at least provide us with reasonable or rational beliefs. Or maybe an atheist such as Kitcher, and not just the religious believers he criticizes, would embrace a hybrid epistemology? This hybrid secularist epistemology would say, roughly, that, as a true atheist, in discussions about religious beliefs one should assume that scientific inquiry sets the standards for the acceptability of beliefs, but when discussing beliefs and theories in the

67

Scientism and Its Rivals  • 

67

humanities—​or in any other nonscientific area of human life one regards as worthwhile—​one might reject that idea and insist that humanistic inquiry sets the standards for the acceptability of beliefs. Express a scientistic epistemology when you criticize religious belief, but adopt a non-​scientistic epistemology otherwise. Anyway, for the reasons given above, I  shall not consider thesis (S3) in what follows. So, setting aside (S3), in what way is (S2) to be related to (S1)? If only science can provide us with genuine knowledge, then there seem to be two different ways in which we could think about the humanities. One option would be to maintain that since the only kind of knowledge we can have is the one provided by the sciences, the humanities cannot give us any knowledge at all, and this ought to be taken as an argument against the humanities. Rosenberg bluntly expresses this view when he writes: There is only one way to acquire knowledge, and science’s way is it. The research program this “ideology” imposes has no room for purpose, for meaning, for value, or for stories. It cannot therefore accommodate the humanities as disciplines of inquiry, domains of knowledge. . . . the humanities are a scientific dead end . . . When it comes to real understanding, the humanities are nothing we have to take seriously, except as symptoms. (Rosenberg 2011: 306–​307). Perhaps the humanities are important after all, but in other ways, perhaps as a kind of academic entertainment, and for this reason might still be considered a part of the academy. Or then again perhaps not, and these disciplines should be marginalized or not considered to be a proper part of the academy anymore. Or, to approach the matter from a pragmatic angle, if science yields all the real knowledge that there is, then a self-​respecting academic discipline had better make sure that it describes itself as a science, or is understood by the real sciences to be a science, or at least to have the potential to become a science. Perhaps Ian Hutchinson is right that the outbreak of “sciences” in academic description is in part a reflection of scientism at work (Hutchinson 2011: 16). Hence, the second option would be to argue that the humanities could potentially be transformed by the methods and theories of the sciences to such an extent that the humanities might eventually become producers of knowledge. But one can also embrace (S1) without embracing (S2). In such a case one would maintain that the natural sciences are the paradigm that all other

68

6 8   • M i k a e l

Stenmark

academic disciplines must mirror, but still leave open the question of what to think about knowledge claims made outside the jurisdiction of the sciences. Your only concern in those circumstances would be what—​of all activities that humans can be involved in—​should count as science and what characterizes scientific knowledge. Hence (S1) does not entail (S2). But, on the other hand, the acceptance of (S1) seems naturally to lead to an acceptance of (S2) as well. If one thinks that the humanities are epistemologically inferior to the natural sciences, wouldn’t one also be inclined to think that what is not even a proper part of the academy is epistemologically dubious, at the very best? This is likely perhaps, but not necessarily so.

2.4 Ontological Scientism A more ambitious form of scientism than (S2) does not merely state that the only reality that we can know anything about is the one that science has access to. It maintains further that only the reality science can discovery now or in the future exists, and nothing else. Hence, scientism can involve a claim about what kind of things exists “out there.” Let us call this version of scientism ontological scientism and thus identify one more key thesis: (S4) The only things that exist are the ones that the sciences can discover. Reality is what science says it is. Only entities, causes, or processes that science deals with are real, full stop. Ontological scientism thus entails epistemic scientism because we could not know anything about what does not exist. We cannot scientifically know something about a reality to which science does not have access: simply because there is no such reality. But epistemic scientism does not entail ontological scientism. If science only sets the limits for what we can know then it is possible that there are nonscientific dimensions to reality. Epistemic scientism merely entails that we cannot know anything about such aspects of reality. One way of stating ontological scientism is to maintain that nothing but atoms or what they are made of ultimately exist in the world. This is the idea that the only entities and causes in the world are material objects. Wilson calls this view scientific materialism or naturalism (Wilson 1978: 201). Carl Sagan writes: I am a collection of water, calcium and organic molecules called Carl Sagan. You are a collection of almost identical molecules with

69

Scientism and Its Rivals  • 

69

a different collective label. But is that all? Is there nothing in here but molecules? Some people find this idea somehow demeaning to human dignity. For myself, I  find it elevating that our universe permits the evolution of molecular machines as intricate and subtle as we. But the essence of life is not so much the atoms and simple molecules that make us up as the way in which they are put together. (Sagan 1980: 105) Sagan apparently thinks that science has shown us that the only things that exist are material objects and their interactions. We are consequently merely “molecular machines” that are not essentially different from artifacts (i.e., machines). Sagan further claims that “the Cosmos is all that is or ever was or ever will be” (ibid.: 1). All this, Sagan thinks, is scientifically knowable, not just when the scientific project will be fully developed or completed, but right here and now. Crick calls these ideas when applied to human beings, the “Astonishing Hypothesis.” He maintains: The Astonishing Hypothesis is that “You,” your joys and your sorrows, your memories and your ambitions, your sense of identity and free will, are in fact no more than the behavior of a vast assembly of nerve cells and their associated molecules. As Lewis Carroll’s Alice might have phrased it:  “You’re nothing but a pack of neurons.” This hypothesis is so alien to the ideas of most people alive today that it can truly be called astonishing. (Crick 1994: 3) According to Crick, most people’s ideas have, unfortunately, been shaped by prescientific illusions of religion, but only science in the long run can free us from the superstitions of our ancestors. Philosophers have expressed similar views. D. M. Armstrong defends the thesis that “the world contains nothing but the entities recognized by physics” (Armstrong 1980: 156). Hartry Field thinks that we should accept as natural only that which in principle is reducible to physics. He writes: “When faced with a body of doctrines . . . that we are convinced can have no physical foundation, we tend to reject that body of doctrine” (Field 1992: 271). And Paul Thagard asks: “What is reality? My answer will be that we should judge reality to consist of those things and processes identified by well-​established fields of science using theories backed by evidence drawn from systematic observations and experiments” (Thagard 2010: 8).

70

7 0   • M i k a e l

Stenmark

Obviously one could embrace (S1) without accepting (S4). Science then sets the limits for what counts as appropriate academic disciplines but not for what exists. If, on the other hand, one embraces (S4), then (S1) is a prerequisite.

2.5 Rivals to Scientism Let us call the conjunction of the epistemic and ontological component of scientism, that is (S2) and (S4), “core scientism” and ask what its main rivals in the West are. The material surveyed so far might give the impression that religion is the rival to scientism, but that is not so. Another main rival, which has been given many names, is philosophical or liberal naturalism, nonscientific or non-​scientistic naturalism, or soft naturalism. Two of the advocates of liberal naturalism, Mario De Caro and David MacArthur, identify scientism with one of two themes of scientific naturalism (the second being a particular view of philosophy as a non-​autonomous discipline, thus corresponding to S1),5 but argue instead for a “non-​reductive form of naturalism and a more inclusive conception of nature than provided by the natural sciences” (De Caro and MacArthur 2004: 1). So, in their terminology, scientism and scientific naturalism are more or less the same thing. De Caro and MacArthur maintain that we should extend the notion of nature beyond scientific nature to fully include the various aspects and normative dimensions of human nature. Liberal naturalists share the conviction that all attempts to reduce, eliminate, or reconceive intentionality, agency, freedom, meaning, rationality, personal identity, and the like, to what is detectible by the natural sciences entirely miss the kind of importance these phenomena have in our lives and experiences. Nevertheless, they identify their view as being within the family of naturalisms. Liberal naturalism cannot, on their account, be theistic, but shares with the other naturalistic outlooks the rejection of supernatural entities such as gods, the Judaeo-​Christian God, demons, souls, spirits, and ghosts. Kai Nielsen argues in a similar way for philosophical naturalism. He takes naturalism to deny that there are any spiritual or supernatural realities. But he emphasizes that naturalism need not be “reductionistic (claiming that all talk of the mental can be translated into purely physicalist terms) or scientistic 5.  They write that “the first theme is a commitment to a scientism that says not only that modern (or post-​seventeenth-​century) natural science provides a true picture of nature but, more contentiously, that it is the only true picture” (De Caro and MacArthur 2004: 4).

71

Scientism and Its Rivals  • 

71

(claiming that what science cannot tell us humankind cannot know)” (Nielsen 1997: 440). Instead, Nielsen believes that the more plausible forms of naturalism are neither across-​the-​board reductionistic nor scientistic, but are nonscientific or philosophical. Thomas Nagel draws the line between scientific reductionism and various forms of (naturalistic or theistic) anti-​reductionism: On the one side there is the hope that everything can be accounted for at the most basic level by the physical sciences, extended to include biology. On the other side there are doubts about whether the reality of such features of our world as consciousness, intentionality, meaning, purpose, thought, and value can be accommodated in a universe consisting at the most basic level only of physical facts—​facts, however sophisticated, of the kind revealed by the physical sciences. (Nagel 2012: 13). He clearly sides with the second camp, going as far as maintaining that naturalists—​and not just theists—​should think that mind, meaning, and value are as fundamental as matter and space-​time in an account of what there is. Nagel maintains that these things that contemporary science does not help us understand, call for a more uncompromisingly mentalistic or even normative form of understanding. Liberal or philosophical naturalism thus involves both a more inclusive conception of nature and a more inclusive conception of knowledge than are provided by the natural sciences and thus entails a rejection of core scientism. We can therefore identify two main theses of liberal naturalism and formulate them in contrast to core scientism: (LN1) The notion of nature must be extended beyond scientific nature in order to fully capture social reality and normative dimensions of human life. (LN2) There are forms of knowledge other than the one provided by the natural sciences, such as the forms of knowledge that we can find in the social sciences, the humanities, jurisprudence, and in everyday life. So (LN1) says that there are different modes of being or aspects of nature, some detectable by the natural sciences; others by disciplines such as psychology, sociology, anthropology, history, philosophy, literary studies, and religious studies; while others do not require the existence of any academic

72

7 2   • M i k a e l

Stenmark

disciple at all since they are detectable by human reason in everyday life. These ways of being may or may not supervene on scientific nature, but they are not reducible to the latter and are of utmost importance in order to understand nature in the form of human life and experience. Moreover, (LN2) says that we can know things and be rationally entitled to believe things about these aspects or dimensions of human life. There are things that exist or there are features of nature that the sciences cannot fully discover and consequently cannot know much about. So there are different ways of knowing things. There are other sources of knowledge than those provided by the sciences. Liberal naturalism also, of course, presupposes the acceptance of generic naturalism, which consists of at least two additional theses: (N1) There is nothing beyond or besides nature, and consequently everything that exists is a part of nature. (N2) There is no personal God or anything like God, nor are there any nonnatural entities such as ghosts, spirits, or immaterial human souls. Thesis (N1) is of course notoriously difficult to define in a precise way, and it is often explicated, in a negative way, in terms of (N2). If we focus on (N2) it is worth pointing out that it is more ambitious than atheism. Generic naturalism is stronger than atheism in the sense that as a naturalist you do not merely deny that God exists, you also reject belief in disembodied souls, spirits, or ghosts. Quentin Smith maintains further that other “examples of hypothesized supernatural realities that govern or create in some sense the universe [whose existence the naturalists deny] are the governing mind posited by the Stoics or the ‘Absolute I’ posited by the early Fichte” (Smith 2001: 202). We could perhaps also add to this list Plato’s Idea of the Good, Aristotle’s Unmoved Mover, or Hegel’s Absolute. The existence of all nonnatural entities or realities is denied by the liberal naturalists. Hence it is misguided to maintain, as we have seen that Rosenberg does, that scientism is the worldview that all atheists (and even some agnostics) share: Nagel, Nielsen, and many other liberal naturalists are atheists and explicitly reject scientism. Massimo Pigliucci, for instance, thinks that “what really characterizes the New Atheism [and he believes that Rosenberg belongs to this camp], as distinct from previous versions of atheism, is its marked turn toward scientism,” but he maintains both that scientism is philosophically unsound and that it does a disservice to atheism (Pigliucci 2013: 151). In the literature, scientism and scientific naturalism are almost always identified as identical positions. However, I think that we should be aware

73

Scientism and Its Rivals  • 

73

that there are problems in making this identification. Scientistic thinkers typically take (S2) and (S4) to entail or justify (N2); so they are atheists. Science has not discovered God or anything that indicates that God exists, and since science sets the boundaries for what we can know and for what exists, there are no good reasons to believe that God exists. There are a number of problems with this argument, but let me address just one of them here. Let us assume that science has not discovered God or anything that indicates that God exists. But from this, even if we add that science sets the limits for what we can know, it is merely agnosticism that follows, not atheism. Suppose that science has not discovered any intelligent extraterrestrial beings or anything that indicates that they exist. But if that is so, the rational consequence is to suspend judgment about their existence, not to disbelieve in them. Correspondingly, one could embrace scientism but be an agnostic and thus reject atheism in the form of scientific naturalism. The examples of liberal naturalists I have given have all been so-​called analytical philosophers. They typically have a high view of the natural sciences and take its results to be of significant interest for their own discipline, and thus see themselves as naturalists even if they do not, like scientizers, believe in the omni-​relevance of these disciplines. But this, arguably, distinguishes them from mainstream scholars of the humanities—​including most of the so-​ called continental philosophers and analytical philosophers such as Bernard Williams.6 I shall call the view these scholars appear to embrace humanism. It constitutes yet another rival to scientism. Humanism emphasizes the value, dignity, agency, and uniqueness of human being and human life and the essential product of that uniqueness, culture. According to humanists the explanations of the natural sciences can tell us very little or nothing really essential about human culture, whereas the explanations and understandings of the humanities and the social sciences tell us a great deal. They might not say this explicitly, but it is reasonable to identify this as an implicit assumption, since they pay almost no attention to the natural sciences in their scholarly work—​in contrast to liberal naturalists—​ and are instead preoccupied with their own disciplinary theories and material.

6. Williams maintains that one definite contrast to a humanistic conception of philosophy is scientism. He writes that “I have argued that philosophy should get rid of scientistic illusions, that it should not try to behave like an extension of the natural sciences (except in the special cases where that is what it is), that it should think of itself as part of a wider humanistic enterprise of making sense of ourselves and of our activities, and that in order to answer many of its questions it needs to attend to other parts of that enterprise, in particular to history” Williams (2000: 494).

74

74   • M i k a e l

Stenmark

I suggest that when Peter Augustine Lawler writes that “what we make of ourselves owes little or nothing to nature. There is no ‘natural law’ worth heeding, because we alone among the species have a mysterious but real capability to be other than natural beings” (Lawler 2003: 102), this might be taken as a paradigmatic statement of this view. What humanism regards as distinguishing human beings from the rest of the animal world is that we are without much fixed form and consequently have the ability to make of ourselves what we will, or the potential to be molded by the social environment into many different forms in a much shorter time frame than is necessary in the case of other animals. There is something exceptional about us, and to grasp this and what this uniqueness has produced, which is culture, we need methods and theories of a qualitatively different sort than those that the natural sciences can offer. Curtis White maintains that: When hominids became capable of symbols (which is to say, when they became human), they entered upon a new kind of evolution, one that became ever more complex, more self-​knowing, and more independent of biology.  .  .  .  they hallucinated a “parallel” world because, strangely, they could better survive the real world if they first worked out the details symbolically. Eventually, the symbolic world discovered a kind of autonomy. It discovered its own concerns beyond the imperatives of biology and atoms (whatever it is that they want). (White 2013: 77–​78, 79–​80) This symbolic world, this spiritual or extra-​material reality is the focal point of humanism: “the scientist [as scientist] is insensible to the nuance of what-​ it’s-​like to be human, while in art a harmonic shift, an unexpected rhythm, will seem to say so much and so convincingly” (ibid.: 185). Roger Scruton embraces a similar view. He writes that we are able to see each other I-​to-​I, and from this all judgment, all responsibility, all shame, pride and fulfillment arise. We are persons, and personality is of our essence: Hence there are concepts that play an organizing role in our experience but which belong to no scientific theory because they divide the world into the wrong kinds of kind  .  .  .  the kind to which we fundamentally belong is defined through a concept that does not feature in the

75

Scientism and Its Rivals  • 

75

sciences of our nature. Science sees us as objects rather than subjects, and its descriptions of our responses are not descriptions of what we feel. (Scruton 2015: 138) Persons are not describable impersonally. The subject, the person, is therefore in principle unobservable to science, for “if I look for it in the world of objects, I shall never find it. But without my nature as a subject nothing for me is real. If I am to care for my world, then I must first care for this thing, without which I have no world—​the perspective from which my world is seen. That is the message of art, or at least of the art that matters” (ibid.: 137–​138). Thus cultural studies are a distinct realm of human inquiry, which cannot be replaced by science or transformed into science. What all of this could be taken to boil down to is that the basic notion of humanists—​in contrast to naturalists—​is not nature but culture, they strongly emphasize human dignity and uniqueness, and the crucial importance of an I-​to-​I perspective on other humans. So roughly: (H1) Human dignity and human experiences, feelings, actions, intentions, and values should be prioritized over the behavior of objects or physical particles. (H2) Culture goes beyond nature and is of a different kind, and therefore its central features are not detectable by the sciences. (H3) The humanities offer us a different kind of knowledge or understanding than do the sciences, a knowledge that might even be more valuable than what the sciences provide us with, since this knowledge arises from our unique capacity of seeing each other I-​to-​I, as persons and not as objects. A humanist could be an atheist and embrace thesis (N2) of generic naturalism but would probably reject (N1), or he or she could be an agnostic or a theist (see below). Although many humanists and social scientists are social constructivists or constructionists, far from all of them are, and I therefore suggest that we should distinguish between humanism and social constructionism or constructivism. Craig Martin expresses the view he holds in the following way: “The basic idea of social constructionism is that we, as humans, make the world what it is for us. The world is not just there for us to find and discover—​ rather, we make the world what it is through our use of language” (Martin

76

7 6   • M i k a e l

Stenmark

2012: 21). Vivien Burr, in a widely used introduction to the view, maintains that within “social constructionism there can be no such thing as an objective fact. All knowledge is derived from looking at the world from some perspective or other, and it is in the service of some interests rather than others” (Burr 2003: 6). Our ways of understanding the world do not come from objective reality but from other people, both past and present. This means that all our ways of understanding are historically and culturally relative and “we should not assume that our ways of understanding are necessarily any better, in terms of being any nearer the truth, than other ways” (ibid.: 4). It is notoriously difficult to define social constructivism, and it also comes in weaker and stronger forms. The core idea, however, seems to be that we construct reality more than we discover it—​not only in the forms of aesthetics, moralities, religions, institutions, nations, taxes, and money (the social or cultural world) but also in the forms of gender, race, mental illness, mountains, animals, genes, cells, black holes, and quarks (the natural or physical world). Therefore our knowledge is socially constructed:  It, just as reality, is constructed by society or groups within society in ways which reflect the human mood of existence, human needs, interests or values and, not the least, power structures. Reality is never given to us as such, but always as conceptualized or as part of a discourse, and how it is conceptualized depends on our needs, interests, or values. Therefore, no sense can be made of the idea that natural reality or nature is a certain way in and of itself. There is no access to reality, even in the most empirical of the sciences, which is not mediated by language. Furthermore, social constructivists are not ready to privilege science when it comes to knowledge and truth. Rather they typically believe that there are many different, but more or less equally valid, ways of knowing the world, with science being just one of them. Richard Rorty, for instance, tells us that he rejects “the idea that some discourses, some parts of the culture, are in closer contact with the world, or fit the world better, than other discourses. If one gives up this idea, then one will view every discourse—​literary criticism, history, physics, chemistry, plumber’s talk—​as on a par, as far as its relation to reality goes” (Rorty 2007: 36). There are many alternative discourses for describing reality, but none is more faithful than another to the way things are in and of themselves, for there is no way things are in and of themselves. Undeniably, some of the theories these discourses generate will be more useful to us (or at least to those who hold power in society) than others, and hence we will accept some but not others. For these reasons I suggest that we consider social constructionism to be a rival to scientism.

7

Scientism and Its Rivals  • 

77

So social constructionists claim, roughly, that: (SC1) There is no way things really are in the world, independent of human discourses, not even in the sciences. (SC2) Scientists, social and human scholars alike give us knowledge by constructing reality in different ways in the service of some interests rather than others. (SC3) Scientific knowledge is no more privileged than any other form of human knowledge. Notice though that social constructionists do not, on this account, deny that something exists independently of human discourses. There would be something there if we did not exist. But this something does not have an inherent structure that scientists or other inquirers try to discover, rather it is structureless: It is not broken up into things or kinds of things. It is we who structure reality into things and kinds of things. Radical forms of social constructivism might actually lead to an inverted scientism. Sandra Harding, for instance, claims that it makes good sense to think of the natural sciences as a subfield of the social sciences. She writes that “we should think about the natural sciences as being inside critical social sciences because the objects-​of-​knowledge—​‘nature, herself ’—​never come to science denuded of the social origins, interests, values, and consequences of their earlier ‘careers’ in social thought” (Harding 1991: 74). So the acceptance of a radical form of social constructivism would turn the traditional hierarchy of academic disciplines upside down. The social sciences, on such an account, are taken to be more fundamental than the natural sciences. What then about scientism and religion? Let me here address religion in just two forms, religious naturalism and theism. Religious naturalism is best conceived as a reaction against both theists who are religious and naturalists who are atheists: The best option is to be a naturalist who is religious. Donald A.  Crosby writes that “religious naturalists find religious meaning, values, and importance solely in nature or in some aspect of the natural order. . . . Nature and its ongoing changes are metaphysically ultimate for religious naturalists” (Crosby 2007: 672). Religious naturalists typically resist scientistic interpretations of the natural sciences, but many of them still privilege the natural sciences when it comes to knowledge about nature or the natural. Religious naturalists maintain that nature in some shape or form is all there is now, ever has been, and ever shall be. If they did not do so they would not be naturalists. So they embrace (N1). Moreover, and as Jerome Stone

78

7 8   • M i k a e l

Stenmark

highlights, a religious naturalist “seeks to explore and encourage religious ways of responding to the world on a completely naturalistic basis without a supreme being or ground of being” (Stone 2008: xi). Thus both non-​religious naturalists and religious naturalists embrace (N2). But nevertheless religious naturalists, in contrast to non-​religious naturalists, maintain that we can find religious meaning, value, and significance in nature or in some aspect of the natural order. On this reading, what distinguish religious naturalists from non-​religious naturalists is: (RN) Religious meaning, value, or significance can be attributed to or found in nature or in some aspect of the natural order. So certain religious attitudes are added to naturalism and it is this extra element that justifies the qualifier “religious” attaching to the word “naturalism.” Although a number of contemporary scientists, philosophers, and theologians advocate religious naturalism, theism is the most frequently discussed religious rival to core scientism and to naturalism. Scientistic and naturalistic thinkers often calls theism “supernaturalism,” but this is actually rather misguided. If naturalism should be part of the name, it is more accurate to call it “prenaturalism” since it contains the idea that God existed before nature:  God is its creator. Anyway, according to theists, there was in the beginning (or has always been) something with mind, namely God, who is, or at least is similar to, a rational, thinking agent; and this God is the creator of everything that exists. This God has created the galaxies, stars, and planets, and all living things, including human beings. Consequently, reality consists of God and all that God has made. All things depend on God as their source of being. God, on the other hand, is self-​existent: has always existed and will always exist. Keith Ward expresses this worldview when he says: If one asks what caused the universe, or why it is as it is, the theistic answer is that God brought the universe into being in order to realise a set of great and distinctive values. If one asks what caused God, the answer is that nothing could bring into being a reality which wholly transcends space-​time and which is necessarily what it is, in its essential nature, which cannot fail to exist, which is self-​existent, the ultimate union of necessary existence and free activity. To fail to grasp such an idea is to fail to grasp what God is. (Ward 1996: 59)

79

Scientism and Its Rivals  • 

79

Moreover, such theists do not think that God simply launched things into existence and thereafter allowed them to persist on their own. Rather, God is the sustainer of all created things. God continually supports things in existence, moment to moment, throughout their existence. If God did not do this, all things would cease to exist. The idea is that just as God creates a thing at the first moment of its existence, God re-​creates it in all subsequent moments at which it exists. If you believe these things, or at least some things of this sort, the view you embrace is theism. Hence one thesis of theism is: (T1) In the beginning was (or has always been) something with mind, namely God, who is or is similar to a rational, thinking agent; and this God is the creator of everything that exists. Theists also maintain that we can have some knowledge of, or at least hold rational beliefs about, who God is and perhaps also about some of God’s plans for the created world. Theists can, however, have different ideas about how we can know or rationally come to believe things about God. To simplify, we could say that their idea is that we can gain epistemic access to God through reason, experience, faith, or revelation. Either way, theists maintain that the scope of knowledge and rational belief goes beyond what science can deliver and this makes theism a rival to core scientism. Hence a second thesis is: ( T2) We can, by reason, experience, faith, or revelation, come (to some extent) to know God, or at least be rationally entitled to believe in God. What makes it a religious view is the acceptance of: ( T3) We should let our lives be guided by God, or at least strive to come into a right relationship with God and to the rest of God’s creation. Theism first becomes an expression of religious belief when it becomes life orienting or gets entwined with certain religious attitudes such as worship or love of God.

2.6 Different Worldviews We have seen that scientism comes in many colors. In its academic internal form, it is best understood as a standpoint within academic disputes about

80

8 0   • M i k a e l

Stenmark

whether or not the human and social sciences—​such disciplines as psychology, sociology, anthropology, history, philosophy, literary studies, and religious studies—​are genuinely autonomous and legitimate studies in their own right, like physics, chemistry, and biology. If one tends to answer “no” to this question then one embraces scientism. But the notion is also used to capture a particular epistemological and ontological standpoint that is of relevance also outside the academy. Scientism in this form contains essentially two claims: (1) The only things and properties that really exist are the ones that the natural sciences can discover, and (2) the only kind of knowledge we can have is the one provided by the natural sciences. If one holds such a view one is an advocate of what I have been calling “core scientism.” I have suggested that the rival to internal scientism is scientific restrictionism, and the rivals to core scientism (or scientific naturalism that would, roughly but only roughly, be the same view) are, at least, liberal naturalism, humanism, social constructionism, religious naturalism, and theism. And, as I said at the beginning of the chapter, it is important to identify the borderlines and the gray zones between these views, or rather worldviews, before we carefully assess their merits, because if we reject one thesis we need to know what the alternatives are. But, and this is important, by saying that they are rivals to scientism I do not mean to say that they have to be strictly speaking incompatible with scientism, only that they contain elements that are in serious tension with its epistemology and ontology or its overall tendency to be deeply suspicious about everything in reality that cannot be described, understood, or explained by the natural sciences.

References Alexander, Richard D. 1987. The Biology of Moral Systems. New York: Aldine. Anonymous. 1997. “Scientism.” In Online Medical Dictionary. http://​www.mondofacto. com/​facts/​dictionary?scientism. Armstrong, D.M. 1980. “Naturalism, Materialism, and First Philosophy.” The Nature of Mind and Other Essays. St. Lucia: University of Queensland Press. Atkins, Peter. 1995. “The Limitless Power of Science.” In Nature’s Imagination. The Frontiers of Scientific Vision, edited by John Cornwell, 123–​125. Oxford:  Oxford University Press. Baxter, Brian. 2007. A Darwinian Worldview. Aldershot: Ashgate. Boghossian, Paul. 2006. Fear of Knowledge. Oxford: Oxford University Press. Burr, Vivien. 2003. Social Constructionism, 2nd edition. London: Routledge. Caro, Mario de, and David MacArthur (eds.). 2004. Naturalism in Question. Cambridge, MA: Harvard University Press.

81

Scientism and Its Rivals  • 

81

Crick, Francis. 1966. Of Molecules and Men. Seattle: University of Washington Press. Crick, Francis. 1994. The Astonishing Hypothesis:  The Scientific Search for the Soul. New York: Charles Scribner’s Sons. Crosby, Donald A. 2007. “Religious Naturalism.” In The Routledge Companion to Philosophy of Religion, edited by Chad Meister and Paul Copan, 1145–​1162. London: Routledge. Dawkins, Richard. 1989. The Selfish Gene. Oxford: Oxford University Press. Dawkins, Richard. 1993. “Thoughts for the Millennium.” In Microsoft Encarta Encyclopedia 2000. Dennett, Daniel C. 1995. Darwin’s Dangerous Idea. London: Penguin Books. Field, Hartry. 1992. “Physicalism.” In Inference, Explanations, and Other Frustrations, edited by John Earman, 271–​291. Berkeley: University of California Press. Harding, Sandra. 1991. Whose Science? Whose Knowledge? Milton Keynes:  Open University Press. Hutchinson, Ian. 2011. Monopolizing Knowledge. Belmont, MA: Fias. Kitcher, Philip. 2008. “Science, Religion, and Democracy.” Episteme 5(1): 1–​18. Kitcher, Philip. 2012. “The Trouble with Scientism:  Why History and the Humanities Are Also a Form of Knowledge.” New Republic, May 4, 2012. http://​ w ww.newrepublic.com/​ a rticle/​ b ooks- ​ a nd- ​ a rts/ ​ m agazine/ ​ 103086/​ scientism-​humanities-​knowledge-​theory-​everything-​arts-​science. Lawler, Peter Augustine. 2003. “The Rise and Fall of Sociobiology.” The New Atlantis Spring 2003. Lewontin, Richard. 1997. “Billions and Billions of Demons: A Review of Carl Sagan’s Book The Demon-​Haunted World.” The New York Review of Books, January 9, 1997. Martin, Craig. 2012. A Critical Introduction to the Study of Religion. Sheffield: Equinox. Nagel, Thomas. 2012. Mind and Cosmos. Oxford: Oxford University Press. Nielsen, Kai. 1997. “Naturalistic Explanations of Religion.” Studies in Religion 26(4): 441–​4 66. Pigliucci, Massimo. 2013. “New Atheism and the Scientistic Turn in the Atheism Movement.” Midwest Studies in Philosophy 37(1): 142–​153. Pinker, Steven. 2013. “Science Is Not Your Enemy.” New Republic, August 6, 2013. http://​www.newrepublic.com/​article/​114127/​science-​not-​enemy-​humanities. Rorty, Richard. 2007. “Main Statement of Richard Rorty.” In What’s the Use of Truth?, edited by Richard Rorty and Pascal Engel, 31–​45. New  York:  Columbia University Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality. New  York:  W.W. Norton & Company. Russell, Bertrand. 1978. Why I Am Not a Christian. London: Unwin Paperbacks. Sagan, Carl. 1980. Cosmos. New York: Ballantine Books. Scruton, Roger. 2015. “Scientism and the Humanities.” In Scientism:  The New Orthodoxy, edited by Richard N. Williams and Daniel N. Robinson, 131–​146. London: Bloomsbury.

82

8 2   • M i k a e l

Stenmark

Sellars, Wilfred. 1963. “Empiricism and the Philosophy of Mind.” In Science, Perception and Reality. London: Routledge. Shermer, Michael. 2002. “The Shamans of Scientism.” Scientific American 28(6): 35. Smith, Quentin. 2001. “The Metaphilosophy of Naturalism.” Philo 4(2): 195–​215. Stenmark, Mikael. 1995. Rationality in Science, Religion and Everyday Life. Notre Dame, IN: The University of Notre Dame Press. Stone, Jerome A. 2008. Religious Naturalism Today. New York: SUNY Press. Thagard, Paul. 2010. The Brain and the Meaning of Life. Princeton, NJ:  Princeton University Press. Ward, Keith. 1996. God, Chance and Necessity. Oxford: Oneworld Publications. White, Curtis. 2013. The Science Delusion. Brooklyn, NY: Melville House. Williams, Bernard. 2000. “Philosophy as a Humanistic Discipline.” Philosophy 75(4):  477–​496. Wilson, Edward O. 1975. Sociobiology. Cambridge, MA: Harvard University Press. Wilson, Edward O. 1978. On Human Nature. Cambridge, MA: Harvard University Press. Wilson, Edward O. 1998. “The Biological Basis of Morality.” The Atlantic Monthly 281(4): 53–​70.

83

3

PHILOSOPHICAL CHALLENGES FOR SCIENTISM (AND HOW TO MEET THEM?)

Alex Rosenberg

3.1 Introduction Scientism is sometimes defined as the exaggerated respect for the findings of the natural sciences and the unreasonable extension of their methods to answer questions outside their original domains (Haak 2003; Putnam 1992). Advocates of scientism like me (Rosenberg 2011)  accept this definition subject to the elimination of the qualifications “exaggerated” and “unreasonable.” We hold that science can answer all cognitively significant questions, and that such questions as it cannot answer are in one respect or another pseudo questions, based on mistaken presuppositions. Scientism does not already have the answers to all questions, or even all philosophical questions, and it almost certainly won’t until science is “completed.” Even then it will not be evident that these are the right answers to all these questions since it won’t be evident that science is complete. But that will be no more reason to doubt the answers than it will be to doubt the science. Meanwhile, scientism does have the answers to many of the philosophical questions that have attracted the most interest by philosophers since Plato and that still hold the attention of philosophers and non-​philosophers: questions about the existence of God, the nature of reality, the purpose of the universe, the meaning of life, the basis of morality, the existence of free will, the relation of the mind to the brain, and the character of personal identity through time. Most of these answers are at variance with common sense and received philosophy. Non-​scientistic answers to these questions, especially “naturalistic” ones, seek to reconcile science

84

8 4   • A l e x

Rosenberg

to the “manifest image.” In a tradition that goes back to Hume, naturalists have sought to vindicate compatibilism or soft determinism—​the claim that free will and determinism are not logically contradictory. Unlike Hume, contemporary naturalists have been prepared to infer “is” from “ought.” They have sought to ground moral normativity in Darwinian adaptations. Inspired by Hume they have tried to substitute for numerical personal identity some kind of causal continuity among experiences. Most of all they have sought a distinct causal role for mental states, with intentional content, in the explanation of human behavior that is compatible with physicalism—​the thesis that mental states are nothing but physical states. Scientism views each of these projects as doomed to fail:  Each is either fatally undermined by obvious problems it can’t solve or involves subtle or not so subtle persuasive redefinitions of the problems as to simply change their subjects instead of answering their questions. Scientism is pessimistic or, perhaps better, disenchanted naturalism. Scientism as a philosophy faces two great challenges:  First, how to accommodate mathematics. If numbers are abstract objects with which we can have no causal relations, it is difficult to see how we acquired any mathematical knowledge. But no satisfactory nominalism has yet been provided to enable scientism to deal with this problem. Second, and relatedly, scientism is eliminativist about intentionality—​derived in language and original in thought. This obliges it to provide an account of how it can even express the denial of original intentionality without inconsistently committing itself to derived intentionality in the expression of its denial. It needs to be emphasized that in what follows I  offer a broad-​brush strokes account of the principle theses of scientism and the answers it offers to a range of philosophical questions. I do not pretend to offer closely reasoned arguments for these theses, nor conceptual analysis that underwrites the claims of scientism in epistemology, metaphysics, meta-​ethics, and the philosophies of science in general and the special sciences in particular. (But see Rosenberg 2011 for at least some of the more detailed and sustained arguments.) My aim is to locate the position of scientism among a range of broadly naturalistic philosophical theories, and then to motivate the special concern that scientism should feel regarding mathematical knowledge and its commitment to eliminative materialism. For scientism most of metaphysics is easy. Almost all of it can pretty much be read off of science: Reality is fermions and boson and the aggregations of them governed by the laws of physics. The physical facts fix all other facts at least in this sense: Any other corner of the universe that is fermion by fermion

85

Philosophical Challenges for Scientism  • 

85

and boson for boson arranged exactly as the ones in the neighborhood of the Sun are arranged, will share all other arrangements in common with our region of space, including all other synchronic chemical, biological, psychological, and social facts. It will be for the various special sciences to uncover these facts and explain in detail how they are related. The biological facts—​ including especially the appearance of a “means/​ends” economy in nature will be explained as purely Darwinian processes driven by the second law of thermodynamics. Since the mind is the brain, the biological facts will include the psychological and social ones, and all of the processes in its domain will secure their appearance of teleology from the operation of the process that Darwin discovered. It will also follow from the fact fixing of physics that there is no free will or for that matter a self, soul, or person, which endures numerically but immaterially identical over the periods of time we might suppose ourselves to do so. Since there is nothing that endures numerically identical as a part, proper or not, of our bodies over our lifetimes, there is no material self either, just a highly adaptative illusion of one. Scientism must make equally short work of meta-​ethics: No ethical theory is grounded more firmly than any other, and whichever one humans adopt is a matter of natural and cultural Darwinian selection. Substantive ethics is of course a nonstarter for scientism. The reasons are easily stated: Scientism holds that the only grounds for ethical truths, as for other truths, will have to be found in science. The only such grounds that naturalistic philosophers and scientists have ever found for such claims are in the application of Darwin’s theory to ethical norms as the result of gene/​culture coevolution (Dennett 1995; Harris 2010). If our core ethical norms are to be justified scientistically, it will be evidently be owing to their having been chosen or shaped by natural selection operating on Homo sapiens and their ancestors. The consequences of our core moral norms for survival and reproduction are too great for them not to have been a product of a Darwinian process. Now consider the apparent coincidence between this fact and the “fact” that they are the right moral norms. Science (and scientism) cannot accept coincidences of this magnitude. The fact that our moral norms evolved as adaptations and the fact that they are the “right” moral norms needs to be explained, and the only explainer can be a Darwinian one. Alas, no such explanation is in the offing: It can’t be that Darwinian processes somehow track moral truths, for such processes are notoriously bad at tracking truths as opposed to beliefs that are adaptive. On the other hand, as David Hume argued long ago, no mere facts of the matter, like the evolutionary pedigree of our moral norms can morally justify them:  “ought” never follows from “is.” With these two

86

8 6   • A l e x

Rosenberg

alternatives excluded, the only way left to explain the apparent coincidence that our norms are adaptive and that they seem like the right ones to us, is the conclusion that their appearance of moral rightness is an(other) adaptive illusion. It’s not that there is some other package of moral norms that is right, correct, or justified. None are. Moral norms are local adaptations foisted on us unawares by Mother Nature as solutions to design problems our ancestors faced even before the emergence of Homo sapiens on the African savanna 200,000 years ago or so. Notice that if this argument for “moral nihilism” is right, it also explains why normal humans have no option but to act cooperatively, generously, even sometimes altruistically. We are the products of many millions of years of selection for such dispositions.

3.2 Scientism’s Epistemology, Its Presuppositions and Implications You might suppose that scientism can make equally quick work of epistemology. If we want to understand what knowledge is, surely all we need is to identify the scientific method. The sciences secure knowledge; indeed they are the only human enterprises that do so reliably. Accordingly we should be able to read our account of the nature, scope, and justification of knowl­ edge off the methods they employ to secure knowledge. Scientism is certainly not going to either demand a further philosophical justification for these methods, or allow that one is required. But establishing exactly what that method should be is by no means easy. After all, different sciences perforce employ methods suited to their own domains and not those of others. It’s perfectly clear that for finite epistemic agents like us, for example, the methods of physics—​even if there is one set of methods that are employed in physics—​are not going to secure knowledge in almost any area beyond physical chemistry. They won’t tell us much of what we want to know about the biological domain. It’s not difficult to show that the laws that physics discovers by the use of its methods do prescribe Darwinian processes as the only means by which adaptation can emerge anywhere in the universe (Rosenberg 2011, ­chapters 3 and 4). But it’s doubtful that methods and data from physics would enable creatures with our cognitive and computation limitations to derive Darwin’s theory ab initio. Given human cognitive and computational limitations, and human technological interests, it is unlikely that we can frame a general scientific method that both informatively and correctly describes the methods of all the sciences. The injunction to operate inductively may be correct but it is insufficiently

87

Philosophical Challenges for Scientism  • 

87

informative. Narrower methodological principles such as Ockham’s razor or inference to the best explanation or model building, etc., on the other hand must be applied with discernment, not to say “art.” Science will generate no algorithm for generating the right rules or for when to apply them, except possibly post hoc, when it is too late. Is it enough for scientism to argue that each scientific discipline will build its proprietary epistemology, driven by the nature of its research domain? Each will be subject to the requirement of consistency with the epistemology of more fundamental science—​and we know well the order of fundamentality. And each will help itself to breakthrough methods and instruments offered and epistemically certified by advances in the more fundamental disciplines. Is that all we need to say? Alas no. This is largely because challenges to scientism are almost always epistemic. How do you know scientism is true? How can scientism account for our knowledge of this or that domain of knowl­ edge, usually mathematics and logic; sometimes normative ethics, history, or human affairs generally; occasionally revealed religion. In the latter cases, scientism doesn’t have much trouble writing off the purported domains, as ones in which there is no knowledge, only assertions with a variety of social and psychological functions. There is no cumulation of theory in these disciplines; no secular improvement in the range and precision of prediction; little concurrence about their boundaries, fundamental theories, or even taxonomies. But mathematics and logic cannot be written off so blandly. Surely these are domains in which our claims to knowledge are at least as strong as they are in physics. As we shall see, a defense of scientism against challenges from these domains requires more than promissory notes that epistemology will be able to accommodate them when science is complete. Scientism’s epistemology has to be “naturalistic.” “Naturalistic epistemology” is a term of art from Quine (1969), if not from Dewey (1938), which has been attractively developed in the hands of philosophers such as Hilary Kornblith (2015) for more than a score of years. As Quine pointed out, for the naturalist there can be no first philosophy prior to science that somehow certifies its results as knowledge. We perforce reject the notion, inherited from Descartes and so freighted by Kant’s Critique of Pure Reason, that without foundations in pure reason, science’s results could not claim to be knowl­ edge. Naturalism rejects the demand made by traditional epistemologists, including empiricists, that there are non-​instrumental epistemic norms that scientific knowledge must satisfy. In this rejection of traditional priority of epistemology to science, scientism concurs.

8

8 8   • A l e x

Rosenberg

Naturalistic epistemology holds that knowing is a cognitive state with causes and effects, and more important that when X knows that P there is a causal connection between (the facts that make) P (true) and X’s knowledge of P. The task of epistemology on this view is to determine which causal chains reliably provide true beliefs, and which do not. Since causal inquiry is the domain of science, the task is to be discharged by science. Once these epistemically reliable causal chains are identified, the epistemologist is in a position to certify some beliefs as justified and true and so as knowledge. Scientism is entirely sympathetic to this program. It differs from the epistemology of many naturalists in not seeking to recover most or even much common-​sense knowledge by the application of science. In fact, it holds that science, especially contemporary psychology, has revealed how unreliable are knowledge claims underwritten by what common sense certifies, for example, eyewitness testimony in judicial inquiries (cf. Loftus 2008). But then how does naturalism—​scientistic or otherwise—​respond to the circularity charge: Sciences and only science tells us what knowledge is and that certifies what science tells us as knowledge. Without a prior conception of knowledge and a criterion for when it is achieved, how can science claim for itself the status of knowledge—​justified true belief ? It can’t appeal to itself. Some may say that science doesn’t have to. Science is corrigible and many of its claims will be tentative, most of its claims at the frontiers of knowledge will not be immediately deemed, even by those who uncover them, to be knowledge. Some scientists may even be global skeptics, doubting that we can really be sure of anything in science. Consider how so secure an edifice as Newtonian mechanics was toppled in a few short decades. Think about the incompatibility of our two best physical theories—​quantum mechanics and general relativity? There are good reasons to be modest in our pretensions to knowledge. Some serious scientists (Susskind 2013)  deny science can, do, or should provide knowledge, just practically successful instruments for organizing “experience,” whatever that exactly is. Science provides at most justified belief, where justification is a matter of practical success in attaining our technological aims. Practically justified belief maybe enough for scientists, and be even more congenial for engineers. And even some philosophers might take “justified” as a matter of some property qualified long term, improving predictive success. And surely even those scientists who eschew “knowledge” because they can’t establish the truth of their most well-​justified findings and theories are willing to certify them as justified, or at least more justified than their predecessors.

89

Philosophical Challenges for Scientism  • 

89

The epistemology of scientists who adopt this view, and philosophers who agree with them, of course is not naturalism. It is a species of pragmatism, a view that I for one think incompatible with naturalism since it gives epistemology priority over metaphysics when it even acknowledges the latter’s existence. In fact I am inclined to call this view Protagorianism—​for its adoption of the maxim that “man is the measure of all things—​of what there is, and what there is not.” (Rorty 1979 is an illustration of this kind of pragmatism, following Dewey 1938.) Pragmatism certifies as knowledge whatever works for us, and designates as truths what science, in the limit and when complete, contains by way of theories and findings. This latter claim may be something pragmatism and scientism share. But scientism can’t hold that either now or in the limit scientific truth is whatever works for us. The reason scientism can’t be pragmatic in its epistemology starts with its metaphysics. Science tells us that we are components of the natural realm, indeed latecomers in the scheme of things that goes back 13.8 billion years. The universe wasn’t organized around our needs and abilities, and what works for us is just a set of contingent facts that could have been otherwise. Among the explananda of the sciences is the set of things—​including beliefs (let me have this notion for the nonce) that work for us. Once we have begun discovering things about the universe that work for us, science sets out to explain why these discoveries do so. It’s clear that one explanation for why things work for us that we have to rule out as unilluminating, indeed question begging, is that they work for us because they work for us. If something works for us, enables us to meet our needs and wants, then there has to be an explanation for why it does so, reflecting facts about us and the world that produce the needs and the means to satisfy them. The explanation of why some methods work for us must be a causal explanation, broadly speaking. It must show what facts about reality make the methods we employ to acquire knowledge suitable for doing so. The explanation has to show that the fact our methods work—​for example, have reliable technological application among other things—​is not a coincidence, still less a miracle or accident. That means there have to be some facts, events, processes that operate in reality, and which brought about our pragmatic success. The demand that success be explained is a consequence of science’s epistemology. If the truth of such explanations consists in the fact these explanations work for us (as pragmatism requires), then it turns out that the explanation of why our scientific methods work is that they work. No one should be satisfied with this explanation.

90

9 0   • A l e x

Rosenberg

So our epistemology, our methodology for securing practically useful, predictively reliable beliefs about the world, has to be a set of rules underwritten by a causal explanation of why they regularly and increasingly produce such reliable beliefs. Here’s an example: Consider the methodological rule that in pharmacological and other experiments on people, double-​blind procedures are both required and reliably successful in securing knowledge. The explanation for why this rule works relies on such facts as the existence of a placebo effect on subjects and an “experimenter” effect (Rosenthal and Jacobson 1968) on scientists, leading to expectations biasing observations. These two effects together explain why the double-​blind rule works in pharmacology, or at least they explain why using it is a necessary condition for securing reliable results. It’s a simple example of the general character of methodological, that is, epistemic justification that scientism accepts: The norms of reliable information acquisition (call it knowledge for the moment) are those that take into account the causal processes operating in the universe. They are hypothetical imperatives (if you wish to avoid experimenter expectation effects and Placebo effects, then employ double-​blind experimental methods) whose effectiveness reflects the operation of regularities in their domains of application that can be explained, usually by considerations from more fundamental scientific theories. This will work all the way down to norms of research employed in fundamental physical research, where oddly enough the empirical results are firmly established but the methodological rules lack a deeper explanation, as might be expected at the basement level of science. (The success of the methodological rule, “Assume objective chance operates at the quantum mechanical level” is explained by nothing more than the fact that objective chance operates at the quantum mechanical level.) The epistemology of scientism is given by the norms of scientific methodology, which in turn are justified as heuristic rules, hypothetical imperatives, by scientific theories that explain why they work—​that is, enable us to expand the domains and the precision of our explanations and predictions. These theories are rarely if ever ones uncovered in the domains whose methods they underwrite. For example, the double-​blind method employed in pharmacology is the result of work in social psychology. But what gives us confidence—​epistemic warrant—​that the explanations cum justification for the rules that these theories provide are true, or approximately true, or increasingly approximately true? Here there is a limited role for prediction. Why suppose that a putative explanation is an actual explanation or even part of the correct explanation or on the right track toward an explanation of anything, a methodological norm, regularity in phenomena, a model that

91

Philosophical Challenges for Scientism  • 

91

works or even a theory? Scientism has to reject several alternative accounts of explanation because like a pragmatic epistemology, they ground epistemic warrant on satisfying human needs or wants. These are the erotetic or pragmatic or unifications accounts of explanation. (For an excellent introduction to these accounts, see Woodward 2014.) These accounts of explanation between them explain a great deal about what humans rightly or wrongly take to be explanations, but they reveal that too often what we accept is a matter of the circumstances in which individuals ask why questions, and too often the explanations reflect facts about human cognitive styles and computational limitations. Scientism needs a standard of explanation that will be endorsed by agents of far greater cognitive powers than our current ones. How do we tell whether as an objective matter a putative explanation really explains, really identifies the causes of the explanantia? Or at least whether it provides a better, an improved explanation that supersedes a previously accepted one? Here predictive success plays a modest epistemic role: An explanation needs to have some predictive consequences, and a better explanation must eventually have some new predictive consequences, either improved in precision or in range of phenomena, either by itself or in conjunction with other theories and explanations, either sooner or later. The predictions needed to certify an explanation as good or better may be indirect, even practically inconsequential. Because predictions subject explanations to tests of correctness independent of us, their role is indispensable. Contriving putative explanations that satisfy humans, that reduce the itch of curiosity, that exploit our preconceptions, is easy. The label “just so story” that Gould and Lewontin (1979) employed to stigmatize adaptational explanations appropriately attaches to explanations in all the domains of science. The meretricious appeal of just so stories held science back for a couple of thousand years. The source of many of these stories and the standards that urge that they stand to reason is “common sense.” Science tells scientism to be extremely suspicious of epistemic norms arising from this source, or urged by introspective reason, even endorsed by it as indubitable. (See Hirstein 2005 for an introduction to the scope and degree of introspection’s and common sense’s errors.) Not only is its track record poor, but cognitive neuroscience has shown that most of what conscious experience and reflection tells us about the world, and about rules for providing ourselves with knowledge of the world is unreliable, especially in circumstances where quantitative inference is involved. (For an exploration of one of the most egregious examples of its fallibility, see Weiskranz 2009.)

92

9 2   • A l e x

Rosenberg

Scientism’s epistemology has serious consequences for many disciplines, including the interpretative social sciences and the humanities. The metaphysical explanation for why these disciplines cannot satisfy the weak predictive demands that scientism imposes is to be found in scientism’s commitment to eliminativism about intentionality. Though this is the subject of section 3.5 below, it is worth at least acknowledging the epistemic consequence here. Consider history, for example. Why must scientism reject its claims to knowledge? The contemporary common-​sense belief/​desire theory of mind that stands behind interpretative understanding of the sort that constitutes most history and biography is no better at predicting human action than Homer’s version of the theory was in the eighth century BC. Historical explanations that use folk psychology to identify the considerations causally relevant to their explananda, will be at most right by accident. But the folk psychology that animates these explanations cannot pass reasonable tests of predictive improvement. For example, we have not managed to get beyond Thucydides’s theory as an important predictive resource in international relations theory. Folk psychology wasn’t too bad as a tool for getting human kind from the bottom of the food chain on the African savanna to the top in a matter of a million years or so. But it is not much better than that, as cognitive science and neuroscience between them show. As we shall see in section 3.5, Scientism diagnoses the problems of folk psychological explanation as the result of a mistaken view that the brain represents content, and does so sententially, in neural circuits that bear propositionally expressed thoughts. The assumption that it does so is not a harmless idealization but a significant cause of the predictive and explanatory failures of every theory that makes it It’s no part of scientism’s epistemic obligations to explain why the interpretative disciplines persist, and why humans generally cannot surrender the conviction that they do provide significant increases in knowledge. Answers to these questions require the resources of cognitive neuroscience. But scientism’s admiration for the emotive force, entertainment value, aesthetic power of the great works of interpretative understanding is second to none. But none of these effects on readers, listeners, or viewers require that it constitute knowledge.

3.3 Scientism’s Epistemic Challenge Like every other naturalistic epistemology, and a whole lot of nonnaturalistic ones, scientism has trouble with mathematics. If we take mathematics at face value, its domain of numbers, sets, functions, lines, points, etc., are abstract

93

Philosophical Challenges for Scientism  • 

93

objects; ones that don’t exist in space and time, have no parts, size, shape, mass, or other dimensions associated with concrete particular objects. This of course prohibits us from having any causal relationship with them. Accordingly, on a naturalistic epistemology we cannot have any knowledge of them, and of the relations among them. This conclusion would not just be a fly in the ointment of all naturalistic epistemologies, including scientism’s. It is, in the view of some, sufficient reason to reject Scientism’s epistemology all together, and indeed all naturalistic epistemologies. For there seem powerful reasons to suppose that we have mathematical knowledge, that mathematical knowledge is the most secure knowledge that there is: Many of our mathematical beliefs are absolutely true and more securely justified than any other beliefs we have. If we can’t have mathematical knowledge, it is doubtful whether we have any knowledge and are reduced to skepticism, or at least open to challenges from skeptics. Even if you are not yourself a mathematician, the one fact about your mathematical education that differs from every other discipline you studied, is that in this domain you never had to unlearn anything you were taught. In every other domain or discipline we begin with simplifications, approximations, and idealizations. In mathematics from the beginning you learn propositions that never need correction, qualification, completion, and if your teacher is any good you learn compelling arguments for these propositions. Mathematics is the domain of knowledge par excellence. Our knowledge of mathematical truths and knowledge of the existence of their truth makers is the basis of the most serious objection to scientism. The combination of their apparent necessity, their insulation from any form of causal interaction with us, and their indispensability in science itself, makes mathematical truths the problem from hell for scientism. Of course it’s not just scientism, or more broadly, naturalism that has problems with mathematics. Empiricism long had problems with abstracta, especially numbers. Hume celebrated Berkeley for his “wonderful discovery in the republic of letters,” that what we think of as abstract objects turn out to be just words—​tokens in thought, sound, and inscription. Alas the solution to the problem of mathematical knowledge was not so simple. Empiricism had to account for the certainty, necessity, and universality of mathematical truths—​ the features that explain why you never have to unlearn any of it. Empiricists thought they had a solution to this problem when they claimed mathematical truth to be definitions, and logical consequences of definitions, where logical consequence is also a matter of definition. It took until the 1930s for Gödel to explain why the “logistic” program for solving empiricism’s problem

94

9 4   • A l e x

Rosenberg

about mathematical certainty could never succeed. Gödel himself, like other mathematicians, was driven to insist that numbers are abstract particulars and that since we have knowledge of them, naturalism is false. This is a view shared all the way back to Plato of course. (For an introduction to these issues, and especially the history of the philosophy of mathematics, see Shapiro 2000.) It is easy to say that rationalism can solve these problems. Rationalism is the thesis that the mind has a priori knowledge including knowledge of mathematical truths or at least truths from which mathematics can be erected, constructed, derived. It’s easy to assert rationalism, but mere assertion doesn’t actually effect the construction of mathematical knowledge. We must not confuse rationalism with theories that accord the mind, that is, the brain, innate concepts, and holds that among these are mathematical ones. Innateness is a psychological claim about the origin of ideas, concepts, thoughts, not an epistemological claim about their warrant or a metaphysical claim about the ontological status of their referents. Abstract objects and human thought are causally completely sealed off from one another. Whether our numerical concepts are caused by experience or emerge from hard-​wired neural circuitry doesn’t actually do any work in solving the problem of mathematical knowledge. If there are abstract particular mathematical objects and if we have knowl­ edge of the relations among them, every epistemology taught in the schools is in trouble. This is the basis for a powerful tu toque argument open to scientism in response to its critics. No one has a satisfactory epistemology of mathematics; no one has a satisfactory ontological account of the existence of numbers. On this basis there may not be much reason to prefer other packages of metaphysics and epistemology to scientism’s package. And when we add the arguments for scientism outside of this domain into the scales, the balance tips toward scientism. Well, this may be a powerful response to objections, but it is no solution to the problem. Insofar as scientism lays claim to being an adequate philosophy, it needs to deal with the problem of mathematical knowledge and mathematical objects. Of course if the options facing scientism in doing so were attractive, they’d also be helpful to less radical doctrines such as naturalism. If some version of nominalism were true, as Hume and Berkeley hoped, scientism, naturalism, and empiricism could deny that there are numbers, holding instead that there were only numerals, including ones in our (physically concrete) thoughts and that we can make sense of statements about numbers without quantifying over them, by translating them into statements about numerals. As for the grounds of our knowledge of mathematical truths,

95

Philosophical Challenges for Scientism  • 

95

scientism, naturalism, and empiricism can adopt John Stuart Mill’s views that all the truths we learn about the relationships among numbers are inductive generalizations from instances of these truths we meet with in experience. We could extend this theory to include, besides enumerative induction, inference to the best explanations of the inductively supported truths, and reasonably hope that this form of reasoning will give us all the higher math we want to adopt, all the way up to the four color theorem and the proof of Fermat’s theorem. Naturally Mill’s treatment of mathematical truths won’t give us any of the truths that objectors demand we provide causal grounds to accept. For each of the mathematical truths that they demand be certified as known are necessary truths, and the demand is that besides their truth scientism (naturalism and empiricism) certify that we know their necessary truth, and their necessary nontautological truth for that matter (remember Gödel). As an exponent of scientism I would readily settle for an epistemology that gives me all the truths of mathematics as well grounded without going to the lengths of being necessarily true. Scientism does not take sides on the nature of modality or even its indispensability in a scientistic metaphysics or logic. Another attractive approach to the problem of mathematical knowledge for naturalistic epistemologies also happens to be one that accords especially well with scientism. This approach begins by denying the indispensability of mathematics to science. To this claim it adds the radical claim that mathematical statements are all false, since their truth requires the existence of numbers and other abstracta. However, among the falsehoods there are important, useful, statements, statements that are so to speak true-​in-​the-​story that scientifically useful mathematics tells. Fictionalism about mathematics accepts that “2 + 2 = 4” is, like “Holmes was a consulting detective,” a fiction, a statement that is strictly false, but an important part of a calculating device that makes short work of otherwise insoluble scientific problems. Some proponents of fictionalism about mathematics have separated it from the arguments against the indispensability of mathematics to science. They accept that science really is committed to the existence of numbers and other abstract objects, and therefore it is committed to some falsehoods (cf. Kitcher 1984). They assuage any discomfort about this conclusion by dividing the commitments of science into ones about concrete entities and relations—​ particles, fields, and their relations; and ones about abstracta like numbers. The difference between these commitments is that the former are true (or at least approximately true) and the latter are false (though useful). This is not an option open to scientism, as it would require an epistemic basis other than

96

9 6   • A l e x

Rosenberg

science itself for our ability to distinguish the true commitments of science from the false ones. If scientism is going to accept fictionalism about mathematics it is also going to need to adopt the dispensability for science of its literal truth. For scientism the price of the package of fictionalism and indispensability may be too high. What are the prospects of showing that numbers are dispensable and statements about them are fictitious? Hartry Field (1980) has made a start, showing that we can do Newtonian mechanics without quantifying over numbers. The trouble is it looks like Field’s approach and others like it require that we quantify over sets. Can scientism and other naturalisms accept sets with equanimity? Perhaps some work of David Lewis (1993) can help. Like others, Lewis holds that mathematics, or at least most of it, is reducible to set theory; that numbers can be “generated” from classes; and in particular they can all be generated from the singleton class, with one member. Take any individual thing; there is, besides that thing, the singleton class of which it is a member. Given some basic axioms Lewis develops that have all the obviousness of the first of them (one class is part of another if the first is a subclass of the second), Lewis claims to be able to build up almost all mathematics from these minimal commitments to abstracta. And they are mighty minimal: We start with a perfectly concrete thing, a chair or table or knife or fork, and then recognize the class of things consisting of it and it alone. This singleton class is different from the individual item that is its only member, it is abstract but is not physically distinguishable from its member, and there seems no room for or reason why scientism, or naturalism, or empiricism, should have any qualms about its existence, or qualms about supposing that admitting its existence opens up flood gates to other more suspect kinds of abstracta. All we need to accept is that in having causal access to the individual, we also have causal access to its singleton class, and then to the class of the chair and its singleton class, and so on ad infinitum or at least up to our scientific needs. Is this so hard? If Lewis’s program is as good as his word, and the singleton class is the sole basic kind of abstractum we need to admit in order to have our mathematics and eat our scientism too, then there may be hope for an answer to the opponents’ challenge from mathematics. At any rate, the problem for scientism, and the other naturalisms will be a clear and limited one: find a causal theory of our knowledge of singleton sets, given our complete knowledge of their members. And if no such causally underwritten knowledge is in the offing, we can always try treating the singleton set like Sherlock Holmes.

97

Philosophical Challenges for Scientism  • 

97

3.4 Scientism’s Really Serious Problem: Eliminativism about Intentionality Scientism has to be eliminativist about the propositional attitudes. It has to deny that there are sentences represented in the neural circuitry, that is, it has to deny “original intentionality,” to use Searle’s convenient label. It has to deny that there are thoughts about anything. It doesn’t have to deny there are thoughts. It just has to deny that they are about something, with a “topic” on which they make “comments,” in Dretske’s nice nontechnical terms. It would be nice if scientism didn’t have to be eliminativist about mental content. It would save advocates of scientism a lot of hard work and remove the greatest impediment to its acceptance. But alas it is not to be. Scientism has to deny that thought, brain states, the neural circuitry have intentional content because that’s pretty much what neuroscience forces it to accept. Let’s see why. The mind is the brain and the brain is nothing but neural circuitry, logic gates through which currents flow all the way from where the peripheral nervous system meets the world to the afferent processes that move the body and its parts. Now there are several ways to see why scientism has no choice but to go eliminativist. First way: Think about Watson, the IBM computer that beats human champions at Jeopardy—​the rather peculiar American quiz show. Combine the achievement of the Watson group at IBM with the insight of Searle’s (1980) Chinese Room, and it becomes clear that Watson has only derived intentionality. There is no original intentionality of any amount in all those integrated circuits, just logic gates steering changes in charge distributions. All the rest is our interpretation. It’s we who accord Watson its derived intentionality. Now consider the brain of one of Watson’s software engineers, one of the humans who confer derived intentionality on Watson’s hard drive memory store. The scientist’s brain is, like Watson’s brain, a (much larger) set of integrated circuits, logic gates, steering changes in charge distributions. Where does its intentionality come from? Well, it can’t come from one of the software engineers’ colleagues, family members, or any other human who interacts with the engineer. There lies infinite regress. Some will argue, like Searle (1980), Horgan and Tienson (2002), that it’s the engineers’ consciousness that bears original intentionality, and that allows him or her to confer derived intentionality on Watson. But this won’t do for scientism either (it shouldn’t do for anyone): What’s intrinsically originally intentional about tokens—​ particular mental contents—​ moving through consciousness? The qualitative characters of these images, silent “sounds” and

98

9 8   • A l e x

Rosenberg

other items passing through consciousness are not themselves “about” anything else, nor are they “about” themselves for that matter. One has only to have a particular sensory memory experience of yellowness to see that its qualitative feature has no referential content. The illusion of intentionality is just a matter of the pattern of succession and association of these tokens passing through consciousness together with their bodily causes and learned effects. Think of the tokens passing through the conscious states of the newborn infant. What are they “about”? It’s not merely for philosophical reasons that scientism has to eschew any confidence in the claims of introspection. It is equally owing to what cognitive neuroscience has taught us in recent years about how wrong introspection is (cf. Hirstein 2005; Loftus 2008). The whole history of science is the history of the repeated and recursive reconstruction of theories given initially by introspective reason and common sense and purged of these two by the elimination and replacement of one false claim after the other, in spite of the fact that each of these claims is vouched safe by conscious experience. A second deep pier in the foundations of eliminativism is provided by the molecular biology of learning, and memory. Eric Kandel won the Nobel Prize for showing how classically conditioned learning is realized in the neurology of the sea slug. Roughly, short-​term learning is a matter of changes in the concentration of neurotransmitter molecules and sodium, chlorine, and potassium ions in the synapses; while long-​term learning is the result of a feedback loop from them to the somatic genes in the axons that build new synapses between neurons. There is presumably no intentionality, content, or aboutness involved in classical conditioning of the sea slug. Kandel went on to show that explicit, declarative memory in the mammalian hippocampus—​first in rats and then humans—​is exactly the same in its molecular biology and somatic gene switching. The only difference between what goes on in the sea slug, the rat, and the human is that many more neurons are involved in the rat and even more orders of magnitude and more neurons in the human (cf. Kandel et al. 2000). The scholastics used to say, natura non facit saltus, nature makes no jumps: If there is no original intentionality inside the sea slug’s small number of neurons, there isn’t any inside the rat’s larger number or inside the human’s even larger number of neurons. Aboutness will have to be extrinsic to the brain: Content will have to be broader. Enter teleosemantics—​the program of grounding original intentionality in a Darwinian process of shaping the response of neurons to their afferent stimuli. Darwin showed that all appearance of purpose in nature is really the product of blind variation and natural selection (or less misleadingly,

9

Philosophical Challenges for Scientism  • 

99

environmental filtration). Since behavior, including linguistic behavior, of higher primates like us appears to be highly purposive, the thought that controls it must be highly purposive in character. Accordingly, it has to be the result of, and also to be operated by, Darwinian processes of development and Darwinian mechanisms in the developed brain. Teleosemantics seeks to convert this insight into an analysis of the original intentional content of neural circuits and the derived intentional content of speech and writing. It has to work if any theory can, since for scientism, as for all naturalistic views, Darwinism about purpose is the only game in town. No wonder naturalists like Dan Dennett (1969 but see also Bennett 1976 and then Millikan 1984) were among its first developers. Now teleosemantics gives us a lot: the strategy of according content to neural circuitry by reverse engineering it from the marvelous environmental appropriateness of the behavior the neural circuitry produces. To a first approximation and subject to many subtle qualifications, teleosemantics holds that neural circuits are, by definition, “about” the stimuli to which they have been selected for responding appropriately. This selection will be both phylogenetic—​hard-​wired, and ontogenetic—​learned. The trouble is that we know in advance that teleosemantics will not give us as much of or even the kind of content we need. To begin with there is the holism problem: No behavior, no matter its degree of appropriateness to specific environmental circumstances, even environments constituted by the linguistic performances of others, can discriminate between various belief/​ desire packages that can give rise to the same verbal behavior. (Is the cause a desire to lie, or to be diplomatic or tell a joke or to rhyme for that matter?) Even holding desires or beliefs constant (as though we had a recipe for doing so), won’t enable us to read back from behavior, including the noises coming from people’s mouths or the marks they make on paper, which of an indefinitely large number of slightly different propositions they intend to convey, given the vagaries of individual idiolects. It is for these reasons that Fodor (1990) and others have been able to show that the finely discriminative appropriateness of behavior to its environment (long and short term) is never enough to confer any unique propositional content on the neural states that produce it. The details behind this conclusion are to be found by considering three different problems for teleosemantics:  of which distal object is the specific topic of a neural circuit, what the difference is between mistaken content and disjunctive content, and how natural selection can discriminate nomically coextensive properties. (For further discussion of the distal object problem for teleosemantics, see Dretske 1995; for the disjunction problem see Fodor

01

1 0 0   • A l e x

Rosenberg

1990; and for nomically coextensive properties see Fodor and Piatelli-​ Palmarini 2010.) It won’t do to suggest that neural states needn’t contain unique propositions and can do the work by being “ambiguous” or “ambivalent” about indefinitely many different propositions, since all the propositions in these sets can be united in a disjunctive unique proposition and still not circumvent the problem of distal object, the problem of distinguishing mistakes from disjunctive content, or discriminating coextensive properties in environmentally appropriate behavior. Now, when these problems are added to the Watson argument and Kandel’s discovery, the right inference to draw is not that teleosemantics is a dead end as a theory of how the neural circuitry can have original intentional content. Rather, it reveals that there is no content there, just the neural causes of finely tuned behavior. Scientism urges that we draw the conclusion that original intentional content is an illusion, a myth, a convenient fiction—​ the intentional stance, not a fact about us or the world, since content cannot be fixed by the physical facts. Why is scientism pessimistic about finding a theory of original intentionality that solves these problems? The reason is to be found in its commitment to a Darwinian account of teleology and the recognition that intentionality is at bottom a species of teleology. (See Bennett 1976 for a powerful argument to this conclusion.) If Darwinism is the only game in town when it comes to a causal account of teleology, it is also the only way to account for original intentionality. So the failure of teleosemantics—​ Darwinism about content—​is a powerful reason to go eliminativist. One alternative not worth canvassing for scientism is the notion that original intentionality is in public language, speech, and writing, from which the intentionality of thought is derived. What after all could tokens of speech or inscription be but just acoustical disturbances in the air or groves or stains on clay, animal hides, and eventually paper. These physical things can no more have aboutness intrinsically than Watson’s integrated circuits. Their aboutness, their intentionality arises from the roles they play in the thought-​controlled interaction of human beings. Doubtless, as a matter of genealogy, once it emerged in public language—​nouns, verbs, adjectives, and adverbs, came back into the mind to improve thinking. (We are, after all, in Denett’s (1995) terms, Gregorian creatures, not just Popperian ones.) But the mental versions of word-​and sentence-​tokens are also just tokens flitting across consciousness with no intrinsic intentionality either. It is the web of connections between these tokens in conscious thought and with the spoken/​written public tokens of language that builds up the illusion of original intentionality, along with the apparatus that enables us to articulate that illusion. There is no language

10

Philosophical Challenges for Scientism  • 

101

without prior cognitive agents and all language needs to get started is cognitive agents’ using gestures and noises to facilitate cooperation. Intentionality can’t get its start anywhere but in the mind, and it stops the moment minds cease to exist. So, what is the problem for scientism? Well, no original intentionality, no derived intentionality. No derived intentionality, no meaning, in particular no linguistic meaning, no semantic values, no truth or falsity. It turns out that scientism’s problem of mathematical truth is the tip of the iceberg, or small beer, just a fly in the ointment, a mere technicality compared to the more serious problem that scientism cannot even express itself as a set of claims about anything—​us, reality, ethics, etc. That’s because there is no such thing as about. Scientism refutes itself out of its own mouth in a sort of pragmatic contradiction, something like “I believe that there are no beliefs.” What can scientism do about this problem? Well, to begin with, it should not be treated as a counsel of despair, but rather as a problem in the research program of scientism, indeed of neuroscience. And there are some promising lines of inquiry here. Go back to neuroscience and what it is teaching us about the nature of cognition. Neuroscientists use the word “representation” to identify the neural circuits’ encoding of inputs from the peripheral nervous system in, for example, the lateral geniculate nucleus or the visual cortex. But it is clear to the philosophically sophisticated reader that these scientists use the word “representation” without according it any commitment to intentional content. In fact, there is an explicit commitment to describing neural representations in terms of structures of neural axonal discharges that are physically isomorphic to the inputs that cause them. (For an entire textbook of examples, see Purves et al. 2012.) Suppose that this way of understanding representation in the brain is preserved in the long-​term course of research providing an understanding of how the brain processes and stores information. Then there will be considerable vindication for the mind/​brain as a neural network whose physical structure is identical to the aspects of its environment that it tracks and where its representations of these features consist in this physical isomorphism. Put crudely, the representation that motion is governed by Newton’s laws consists in some data structure in the brain that is isomorphic to the facts about the world that (approximately) realize the relationships Newton’s laws record. It is important to emphasize that this isomorphism between mind/​ brain and world is not a matter of some relationship between reality and a map of reality stored in the brain. Maps require interpretation if they are to be about what they map, and both eliminativism and neuroscience share a

012

1 0 2   • A l e x

Rosenberg

commitment to explaining the appearance of aboutness by purely physical relationships between informational states in the brain and what they “represent.” If we are going to explain intentionality we can’t sneak it in, not even a little bit. If we are going to explain away intentionality, the same thing goes. For this reason the mind/​brain-​to-​world relationship must be a matter of physical isomorphism—​sameness of form, outline, structure that doesn’t require interpretation. If this approach pans out, the next task is employing it to provide some alternative for semantic evaluability, to substitute for saying that sentences or statements are true or false. Then we can hope to salvage enough semantic evaluability to ward off the suspicion that scientism cannot even be thought. Exploiting structural isomorphism to build something like a concept of truth exploits the correspondence theory of course: P is true if and only if P corresponds to facts about the world. Everyone knows the difficulties that attend this formulation. Many of them are circumvented by the formula that “representations”—​charges at nodes in the neural circuitry—​can, to greater or lesser degrees be physically similar to distributions of matter and fields of energy in the environment. Which sets of charges are more isomorphic than others? The ones that work together with the motivational components of the brain to meet needs and avoid harms, increase efficiency and reliability in these tasks, and otherwise to shape behavior in ways that make it more environmentally appropriate for creatures subject to natural selection. One has to be careful in formulating the isomorphism substitute for truth to avoid smuggling intentionality back in. For example, you can’t employ “wants” along with (biologically, neurologically identifiable) needs to specify which neural structures are more physically similar to distributions in the world. You cannot succumb to the temptation to treat the success of some distributions of charges in the brain as constituting their greater accuracy than others. That would be pragmatism. Success is an indicator, not constitutive of accuracy. The substitute for truth is not success but physical isomorphism. From this notion we can seek to build substitutes for the full truth about reality; for full truth about aspects, components, regions of reality; as well as approximate truth, substitutes for more or less approximate truth; and substitutes for degrees of falsity. More successful brain structures, ones more physically similar to distributions in the world, will get expressed in the imprecise, illusion-​ridden, but highly convenient medium of speech—​silently in the mind, and noisily on the tongue.

013

Philosophical Challenges for Scientism  • 

103

One attraction of this approach is its friendliness to mathematics and its ability to accommodate so much of what we want by way of a fictionalist approach to mathematical statements. Physical isomorphism is often (perhaps always) a matter of sameness of mathematical structure. Insofar as we acquire physical understanding of the world, that understanding is expressed in equations and formulae. It is tempting to suggest that these mathematical structures are “represented” (in the neuroscientist’s nonintentional “sense” of “represented”) in the brains of the scientists who understand and employ them as accessible isomorphisms, identical in structure, to aspects of reality, aspects we can independently “identify” by observing the behavior (including the linguistic behavior) of the scientists and their coworkers. Meanwhile, when we express these structures in sentences that appear to express our commitment to their truth, we do not utter truths (since there are no truths of any kind really), we offer utterances and inscriptions that are part of a practice, like chess, for example, but one in which “good” moves have much greater payoffs. It will be relatively natural to see how all the mathematics we “want” can be developed simply by the recursive rearrangement of mathematical structures into one set of neural circuits by the operation of other neural circuits. And the systematicity of these processes may well be ones that make for the systematicity of language, not just the language in which we express mathematics, but the language we use for so many other transactions in life. Of course any two physical systems may share many different mathematical structures in common, and a given mathematical structure represented in the brain may be isomorphic with many different structures in reality. But we cannot discriminate the one it is intended to represent, or that it is supposed to be true “of.” These locutions are heavy with just the intentionality that scientism denies itself. Here is a problem of underdetermination or holism that scientism shares with intentionality-​dependent theories of mind. Here again, we can only invoke pragmatic criteria for discriminating successful structural representations—​our substitute for true ones, from unsuccessful ones—​the ones we prescientistically would call the false ones. Scientism needs to find an alternative to communicating its claims by semantically evaluable means if neuroscience does so. As is often the case in science, disciplines advance even when their serious foundational problems go unnoticed. So it is with the sciences of the brain, helping themselves to notions like “representation” that they really don’t have any right to, while making significant progress in coming to grips with other more tractable problems.

014

1 0 4   • A l e x

Rosenberg

Meanwhile it is worth noticing that many of the conclusions scientism advances that are most threatening to the “manifest image,” the common-​ sense understanding of ourselves, ultimately do not hinge on the denial of intentionality’s foundations:  the denials of free will, of the existence of an enduring self, of any real teleology anywhere in nature or human history, or even in individual human action. The domains of human affairs in which eliminativism makes a difference are narrative history, the literary humanities, hermeneutics, phenomenology, psychoanalysis, and pop psychology. The claims of all these disciplines will turn out not even to be false, if there is no such thing as meaning at all. How can these theses of scientism be understood without being expressed in ways that require intentionality, the meaningfulness of thought and speech in terms of content and aboutness? Well, roughly, scientism turns out to be something like this: Once human brains begin to register neural circuitry isomorphic with all or sufficiently large regions of reality, it will be able to determine from these neural structures alone (and without further search for new structures isomorphic with reality—​new “truths”), that the reality to which they are isomorphic contains no structures even resembling those required by the manifest image, common sense, religion, or other wishful thinking. Of course, as David Hume advised, we’ll still be able to speak with the vulgar, while we think with the learned.

References Bennett, Jonathan. 1976. Linguistic Behavior. Cambridge: Cambridge University Press. Dennett, Daniel. 1969. Content and Consciousness. London: Routledge. Dennett, Daniel. 1995. Darwin’s Dangerous Idea. New York: Simon and Schuster. Dewey, John. 1938. Logic:  The Theory of Inquiry. New  York:  Holt, Rinehart and Winston. Dretske, Fred. 1995. Naturalizing the Mind. Cambridge, MA: MIT Press. Field, Hartry. 1980. Science Without Numbers. Princeton, NJ: Princeton University Press. Fodor, Jerry 1990. A Theory of Content and Other Essays. Cambridge, MA: MIT Press. Fodor, Jerry and Massimo Piattelli-​ Palmarini. 2010. What Darwin Got Wrong. New York: Farrar, Straus and Giroux. Gould, Stephen J. and Richard Lewontin. 1979. “The Spandrels of San Marco and the Panglossian Paradigm.” Proceedings of the Royal Society B 205: 581–​598. Haack, Susan. 2003. Defending Science Within Reason: Between Scientism and Cynicism. Amherst, NY: Prometheus Books.

015

Philosophical Challenges for Scientism  • 

105

Harris, Sam. 2010. The Moral Landscape: How Science Can Determine Human Values. New York: Simon and Schuster. Hirstein, William. 2005. Brain Fiction: Self-​Deception and the Riddle of Confabulation. Cambridge, MA: MIT Press. Horgan, Terry and John Tienson. 2002. “The Intentionality of Phenomenology and The Phenomenology of Intentionality.” In The Philosophy of Mind, edited by David Chalmers, 520–​532. New York: Oxford University Press. Kandel, Eric, James Schwartz, and Thomas Jessell. 2000. Principles of Neural Science, 4th ed. New York: McGraw Hill. Kitcher, Philip. 1984. The Nature of Mathematical Knowledge. Oxford:  Oxford University Press. Kornblith, Hillary. 2015. A Naturalistic Epistemology: Selected Papers. New York: Oxford University Press. Lewis, David. 1993. “Mathematics Is Megethology.” Philosophia Mathematica 3: 3–​23. Loftus, Elizabeth. 2008. Eyewitness Testimony: Civil & Criminal, 4th ed. Charlottesville, VA: Lexis Law Publishing. Millikan, Ruth. 1984. Language, Thought and Other Biological Categories. Cambridge, MA: Bradford Books. Purves, Dale, George Augustine, David Fitzpatrick, William Hall, Anthony-​Samuel LaMantia, and Leonard E. White. 2012. Neuroscience, 5th ed. Sunderland, MA: Sinauer. Putnam, Hilary. 1992. Renewing Philosophy. Cambridge, MA: Harvard University Press. Quine, Willard V. O. 1969. Ontological Relativity and Other Essays. New York: Columbia University Press. Rorty, Richard. 1979. Philosophy and the Mirror of Nature. Princeton, NJ: Princeton University Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality: Enjoying Life Without Illusions. New York: W.W. Norton. Rosenthal, Robert and Lenore Jacobson. 1968. Pygmalion in the Classroom. New York: Holt, Rinehart & Winston. Searle, John. 1980. “Minds, Brains, and Programs.” Brain and Behavioral Science 3(3): 417–​457. Shapiro, Stephen. 2000. Thinking about Mathematics: The Philosophy of Mathematics. New York: Oxford University Press. Susskind, Leonard. 2013. The Theoretical Minimum. New York: Basic Books. Weiskranz, Lawrence. 2009. Blindsight. New York: Oxford University Press. Woodward, James. 2014. “Scientific Explanation.” In The Stanford Encyclopedia of Philosophy (Winter 2014 Edition), edited by Edward N. Zalta. http://​plato.stanford.edu/​archives/​win2014/​entries/​scientific-​explanation/​.

016

4

SCIENTISM WITH A HUMANE FACE

James Ladyman

4.1 Scientism: Sin or Salvation? Scientism is usually thought of as sinful but it can be redeemed for our salvation. All that is required to arrive at a worthy form of scientism is application of the heavenly virtues of charity, humility, and temperance. Scientism should not be dogmatic or uncritical, nor should it ignore the actual limitations of current scientific knowledge. There are other modes of inquiry that deserve epistemic respect, and scientists should not be deferred to about matters beyond their expertise.1 However, limits should not be placed on what science can study and we cannot say in advance what the limits of future science will be. Where it conflicts with common sense, religion, and tradition, science should be regarded as authoritative for the purposes of education and public policy as well as objective inquiry; and scientific knowledge is even relevant to moral and political deliberation.2 This is the core of scientism.3 Section 4.3 of this chapter elaborates a way of thinking of scientism as a stance (in the sense of Bas van Fraassen explained in the next

1.  As discussed further below, to say that an activity deserves epistemic respect means among, other things, that it makes a contribution to knowledge. 2.  “Science” should be taken to mean well-​established science, not the latest hypotheses. Of course, there are borderline cases and they should be taken on a case-​by-​case basis. 3.  There are many forms of scientism and many ways of categorizing them, as explained by Rik Peels in his chapter in this volume. The core of scientism is epistemological and methodological. The humane scientism defended here does not match exactly any of the varieties identified in the literature, but it is has something in common with many if not all of them.

017

Scientism with a Humane Face  • 

107

section) characterized in terms of positive and negative components. Section 4.4 characterizes and defends a humane form of scientism worthy of further development. The term “scientism” is standardly used pejoratively to refer to: (a) The misapplication of scientific methods outside of appropriate domains;4 (b) excessive faith in science’s ability to replace existing forms of inquiry; (c) disdain for the contribution of the arts and humanities to our knowledge and self-​understanding; and (d) extreme forms of scientific realism and reductionism about current science, and internalism about the history of science.5 Scientism is usually portrayed as a kind of scientific imperialism that urges scientists to conquer new domains. For example, neuroscience, cognitive science, and Darwinism being used in literary criticism might be so regarded. However, the metaphor of conquest is not apt because the scientific study of a new domain need not involve conflict with existing forms of inquiry, even if some enthusiasts (often popularizers in evolutionary psychology or neuroscience) overreach by claiming so. The term “scientism” was introduced to criticize those who advocated that human beings and society could be studied with scientific methods. It was largely (a) that was at issue then as now. The charge of scientism was first brought against those who had the temerity to argue that there is something to be learned from what we now call the behavioral sciences.6 Despite the tradition of philosophers of social science arguing against naturalism, the view that there is no knowledge at

4. Susan Haack (2003) makes this number 5 among six features she associates with scientism (the last of which is (c) below, and more of which are discussed in section 4.3). It is the most important since, fundamentally, the debate about scientism is about whether or not we should set limits to scientific inquiry, or so this chapter argues. Haack is right that the way the term “scientism” is standardly used makes it definitively a bad thing. However, here it is argued that it can be reclaimed and used positively because it was originally deployed to criticize a scientific expansion that proved successful. 5. In debates about the historiography of science, “internalism” is taken to refer to the idea that scientific theory change can be explained with reference to arguments, evidence, and experimental results; and without reference to economic, political, psychological, and social factors. 6. Some, such as Hayek (1979), use the term “scientism” to criticize particular forms of putative social science that they regard as pseudoscientific. Misapplication of scientific methods is always bad science, and the superficial imitation of science and unwarranted appeals to scientific status are not scientism but pseudoscience.

018

1 0 8   • 

James Ladyman

all to be had in economics, scientific psychology, and so on, is implausible in the extreme, and current critics of scientism do not defend it. Instead, the boundary of what cannot be studied scientifically is standardly placed at consciousness or experience. Scientism that denies limits to science need not support (b)–​(d). To reclaim scientism as a badge of honor, and to advocate the unlimited application and/​or further development of scientific methods, does not require regarding scientific knowledge of human beings as replacing all the other ways people study their humanity and culture. The scientistic cause need not devalue and threaten the arts and humanities as in (b) and (c). Indeed, far from replacing them, science has of course given the arts and humanities much more to study; since, for example, one can now study literary form in scientific writing. Moreover, the arts and humanities have to some extent abandoned traditional presuppositions about human beings and the world, for example, essentialist ideas about gender and race, and geocentric ideas about cosmology that science has refuted. To a significant extent we now have an arts and humanities that reflects and is indebted to our scientific knowledge of the world. However, these developments do not undermine the fundamental goals of humanistic study, and indeed any sensible humanist does not want to be in the grip of falsehoods. The appreciation of facts is not exclusively a trait of scientists. The arts and humanities have evolved with the advancement of science, and rightly so. It would be absurd for literary critics, historians, or philosophers to continue to think in the terms of the kinds of theories we had about ourselves before we acquired our scientific understanding of our bodies, our brains, and our evolution. Insofar as charge (a) presupposes that there are a priori limits on what can be scientifically studied, it begs the question against scientism. There may be good arguments for limits to science, but putative examples have been repeatedly refuted by the progress of science so far, and the prospects for further extensions of science into new domains are good. For example, the study of the brain and consciousness is at an early stage, and big data and supercomputing are opening up new possibilities. However, how successful they will ultimately be remains to be seen, and in both domains there is a good deal of pseudoscience or at least overreach. The defender of scientism does not say in advance how and where the application of scientific methods will be successful, or what kind and extent of knowledge is possible. Often the questions science ends up answering are not the ones with which we began, since the scientific study of the world teaches us how to think about it better.

019

Scientism with a Humane Face  • 

109

Science also teaches us how to do better science. These arguments are made in more detail in section 4.3. As for (b), clearly, scientific hubris is as bad as any other kind.7 Scientism must guard against an uncritical attitude to science. It must frankly acknowledge how much remains speculative or unknown, and avoid complacency about the extent to which current science is free from economic, ideological, and political influence. However, such self-​critical vigilance is no ad hoc modification of scientism. Rather, it is compatible with the scientistic spirit given voice in this chapter, which is arguably just the spirit of science itself (about which more is said in section 4.3). In relation to (d), there is a scientistic view that all scientific theory change is rational and independent of cultural, historical, psychological, social, and political factors, and that our best scientific theories are entirely and literally true. However, arguably that is undermined by the scientific study of science. Any defender of scientism who reflects with a scientific attitude on the history and philosophy of science can reasonably be expected to have an appropriately circumspect view of science, in contradiction of (d), and again in keeping with the scientistic spirit. Similarly, strong forms of reductionism are arguably not scientifically defensible and need play no part in scientism (see Ladyman and Ross 2007: chs. 1–​2) for a discussion of reductionism and scientific realism and antirealism). The issue of whether scientism must involve some form of materialism or physicalism is taken up in section 4.3. Scientism can be seen as a struggle for science’s self-​determination in seeking to liberate territory from the forces of superstition and the supernatural, and denying non-​scientists the power to stipulate what can and cannot be studied scientifically. Disdain for the arts and humanities (c) is not a core component of scientism, but it does regard beliefs about human nature and the world based on common sense, intuition, religion, and other forms of tradition, as having no privileged claim to truth over rival beliefs based on differing intuitions or traditions, and no immunity to revision in the light of scientific knowledge.8 It also denies that there are any domains of inquiry that are, in principle, off limits for science, while accepting that actual science and

7. Haack (2003) defines scientism as “a too uncritically deferential attitude toward science.” 8.  Common-​sense beliefs are of course often true and confirmed by more systematic study. Likewise traditional belief systems embody much knowledge. That is not the issue here. Rather the point is that if scientific study determines that a folk remedy is inefficacious, the fact that its use is entrenched in a culture cuts no epistemic ice.

10

1 1 0   • 

James Ladyman

scientists are imperfect and limited. (The relationship between science and value is discussed in section 4.4.) In philosophy scientism is similarly a struggle for liberation, but this time from common sense, intuition, and tradition. Ladyman and Ross (2007) argue for a rebellion against the dominant concepts and methods of analytic metaphysics and their replacement by naturalized metaphysics. They articulate and defend a form of scientism in metaphysics that synthesizes empiricism and materialism as characterized by Van Fraassen, adopting his notion of a philosophical “stance,” as briefly explained in the next section. They say little about scientism outside of philosophy, and are much more circumspect than those advocating strong generic forms of scientism such as Alex Rosenberg (2011). Notably, unlike the latter they do not advocate physicalism.9 Brown and Ladyman (2009) argue for a weak form of physicalism, and argue that, like empiricism and materialism, it has positive and negative components. In what follows I build on all the above work to defend “the scientistic stance.” I argue that scientism, like empiricism and materialism, is so fundamental to the philosophical views of its adherents that it is arguably best considered as a stance or orientation rather than as a particular doctrine. So construed, various positive and negative components characterize stronger and weaker forms of scientism. I  argue that the scientistic spirit should be identified with the intellectual qualities that many scientists and philosophers of science have valued most in science, and so ought to be similarly valued. It is compatible with a humane form of scientism that avoids the excesses of (b)–​(d) above.

4.2 Van Fraassen on Stances The motivation for the idea that some venerable philosophical positions should be recast as stances is the implausibility of the alternative:  namely, understanding them as doctrines.10 Van Fraassen characterizes the latter view as implicitly adopting what he calls “Principle Zero”:  For any philosophical position X, there is a statement X+, such that to adopt X is to believe

9. This chapter is compatible with everything that Ladyman and Ross (2007) says but it says things about scientism more broadly with which Don Ross may not agree, or that he may think are too weak. Note that everything that follows is compatible with denying the very strong form of scientism that Rosenberg defends, which tends toward including (c). Ladyman (2011) develops the scientific stance in relation to metaphysics. 10. What follows is a brief exposition of ideas in The Empirical Stance (Van Fraassen 2002).

1

Scientism with a Humane Face  • 

111

X+ (2002: 41). While it is interesting and important to explore the various different doctrines associated with philosophical positions such as empiricism and materialism, to reduce them to such doctrines does not do justice to the depth and breadth of their influence, nor to what it is to be an empiricist or a materialist. To have such a stance is to engage in certain forms of life and not others; to have various attitudes, commitments, and values; and to adhere to certain norms. The concept of a stance is a much richer one than the concept of a belief or doctrine. The appraisal of beliefs takes place from the point of view of a stance, and the methods and background beliefs associated with different stances may differ. Van Fraassen’s epistemology is voluntarist in the sense that there is no ultimate source of epistemic authority, and we have a choice about upon what to rely. Stances do not admit of neutral adjudication or justification. However, a stance may lead to one’s life going more or less well by one’s own lights and one may adhere to a stance, or abandon it and adopt another one accordingly. In the case of empiricism, X+ is the claim that all knowledge comes from experience, with the corollary that there is no a priori knowledge. Van Fraassen points out, however, that it is implausible to claim that we know by experience that all knowledge comes from experience. On the other hand, the claim that there is a priori knowledge seems to be ruled out of empirical confirmation a priori by empiricists. These principles of empiricism cannot be treated as falling within their own scope on pain of inconsistency. The attempt to grapple with these problems under the grip of Principle Zero leads to the dichotomy between “naturalized” and “transcendent” empiricism, where the former takes empiricism to be empirically justified, and the latter takes it to be somehow uniquely lacking the need for empirical justification. Van Fraassen takes it that neither of these two positions is acceptable and so concludes that Principle Zero is false, and that empiricism is best understood as a “stance.” He goes on to present materialism as a stance too. Stances have positive and negative doxastic and methodological components. In the case of empiricism, the positive components are to take one’s knowledge from experience, and to regard any proposition as in principle subject to empirical consideration. The negative components of empiricism are the view that there is no substantive a priori knowledge, and that rationalist methods do not deliver knowledge of concrete reality.11 11.  The restriction to “substantive” knowledge is intended to allow the empiricist to concede that there is analytic a priori knowledge of a sort, for example, knowledge that a vixen is a female fox had by someone who has no experience of foxes. Empiricists of course have the

12

1 1 2   • 

James Ladyman

Explanation should not be by posit. Indeed the demand for explanation must at some point be rejected, and many empiricists disdain explanation in terms of unobservable entities, properties, and processes. As characterized by Van Fraassen, materialism has much in common with scientism insofar as among its positive commitments are that we should defer to science about what exists, and that science is complete in the sense that there is nothing real that it does not describe. The materialist also thinks science tells us that everything that exists is material. Among its negative components are the attitude that there is nothing supernatural, and the value that science should not posit sui generis intentional, mental, or spiritual entities or properties to explain human behavior; and intentional, mental, or spiritual phenomena such as they are. The problems with empiricism include the role that a priori thought plays in science, how to deal with the problem of induction and confirmation of theory by evidence, and the importance in science of the search for explanations of phenomena in addition to regularities that subsume them, and the success of positing unobservables such as fields and particles. The problems with materialism are that we need to know what exactly science is for us to be able to defer to it; that current science is nowhere near complete; and finally and fatally, that science tells us that not everything is material, indeed arguably it tells us that nothing is.12 In both cases, and in the case of physicalism (about which more below), the positive components are harder to defend than the negative ones. For example, the view that all knowledge comes from experience is harder to defend than skepticism about the pretensions of those who claim a priori knowledge of reality, and skepticism about the pretensions of those who claim to know of the existence of immaterial things retains its appeal even if matter is not what it was thought to be. Arguably, materialism can be revised to physicalism in a way that retains its core negative commitments (Brown and Ladyman 2009). In the next section it is argued that scientism is best thought of as a stance, and that as such it synthesizes elements of empiricism, materialism, and physicalism.

problem that rationalist methods seem to give knowledge of mathematics. Hence the restriction to “concrete” reality in my formulation may be too concessive. 12.  Science tells us not everything is material because the ontologies of the special sciences include such things as mating strategies and markets. Physics tells us that ordinary matter is mostly empty space and that subatomic particles are very different from motes of dust. See Ladyman and Ross (2007).

13

Scientism with a Humane Face  • 

113

4.3 The Scientistic Spirit If there were a Principle Zero for scientism it would be a doctrine such as that all knowledge comes from science. So understood it would face the same dilemma as empiricism in answering the question as to how we know scientism is correct. Furthermore, the idea that all knowledge comes from science is not plausible because we have plenty of other sources of knowledge including memory, perception, and testimony. All of these are fallible, of course, but so is science. Both the idea that science tells us that all knowledge comes from science, and the idea that the latter claim is uniquely self-​justifying can make scientism so characterized seem idiotic and dogmatic, but neither is part of what we should take the position to be. It is more appropriate to think of scientism as a stance because it too is richer than doctrines like the above and it involves norms, values, commitments, and forms of life.13 It is helpful in this regard to distinguish between the positive and negative components of scientism, by analogy with the positive and negative components of the other aforementioned stances. The core positive commitment of scientism is that there are no domains of inquiry that are in principle off limits for science. Everything real can in principle be investigated by scientific methods and no limits should be placed on what science can study.14 It follows that we should not believe in what is claimed to exist but posited to be inaccessible to science in principle. What can be studied by science is what we take to be the natural world. In particular, we ourselves and our cultures and societies are part of nature. This all clearly has much in common with aspects of materialism as characterized by Van Fraassen (and very much echoes Wilfrid Sellars). In common with empiricism, scientism holds that science based on experience is the best way to acquire knowledge about the world. Scientific culture and its methods are supremely reliable and self-​correcting.15 It is important to add the commitment 13. A lot more could be said about all of these in relation to all the stances mentioned in this chapter, especially about the associated forms of life. The commitments, norms, and values of the scientific stance are like everything in science, subject to self-​critical scrutiny and revision and this makes it very different from other stances. 14. Of course making this precise requires a specification of what science and scientific methods are, about which more below. As a first approximation, take them to be given by ostension of the actual scientific community (as Ladyman and Ross do in formulating their “Principle of Naturalistic Closure,” which is critiqued in René van Woudenberg’s contribution to this volume). Note that scientism can also study itself from its own point of view. 15. There are numerous cases from recent science, such as that of that of the expansion of the universe, where the relevant scientific community has corrected its own errors in response to

14

1 1 4   • 

James Ladyman

that science be regarded as authoritative for the purposes of education and public policy. Scientific knowledge is even relevant to moral and political deliberation, as argued in the next section. A further positive commitment is to science replacing all other forms of inquiry. As discussed in the introduction, the latter is not necessary for scientism. In the next section it is argued that it should not be adopted. The core negative commitment is that, as said above, beliefs about human nature and the world based on common sense, intuition, and religion and other forms of tradition have no claim to the truth over rival beliefs based on differing intuitions or traditions, and no immunity to revision in the light of scientific knowledge. This can be thought of as subsuming empiricism’s rejection of a priori knowledge, and materialism’s repudiation of the supernatural. According to scientism, common sense, intuition, and religious and other traditions have no epistemic authority over science. There is no systematic institutional source of knowledge of the objective nature of the world that trumps science.16 A further negative component is disdain for the arts and humanities, and all other forms of humanistic inquiry. Again, as discussed in the introduction, this is not necessary for scientism. In the next section it is argued that it should be repudiated as not in keeping with the scientistic spirit. Scientism is popularly associated with many other ideas. As noted in the introduction in (d), these include unrealistic ideas about how science itself works, as well as strong forms of materialism and reductionism (about which more below). Scientism is also associated with various forms of ethical and political views, some of which are discussed in the last section. However, none of these are part of the core idea of scientism, and many of them conflict with its spirit, as argued below. As with empiricism and materialism, the positive components are harder to defend than the negative ones. For example, the claim that science tells us about the fundamental nature of space and time is much less defensible than the claim that we cannot learn about them by any other means. Similarly, the claim that science tells us about everything is less plausible than the

new evidence without external pressure being brought to bear. The replication controversy in psychology and the disputes about the evidence base in behavioral economics are examples where the process has not yet produced answers and may show some orthodox science to be badly flawed in some respects, which often is the case. 16. Journalism and the law, and disciplines such as archaeology and history, in so far as they establish facts about objective reality, rely on scientific methods and proto-​scientific reasoning.

15

Scientism with a Humane Face  • 

115

repudiation of putative realms of being that are posited by religious and other traditions. The best reason to adopt the core scientism characterized above is the track record of science compared to other forms of inquiry. Without science we would have a tiny fraction of the knowledge that we have, and we would still be in the grip of various dogmas about life and the planet depending on our particular cultural background. Religions and traditional forms of knowledge agree on little with regard to the objective nature of reality, and disagree substantially about the nature of the human body, mind, and soul (if any). Many religions also have internal doctrinal disputes about fundamental metaphysical matters, such as the doctrines of transubstantiation in Christianity and reincarnation in Buddhism. Science, by contrast, is a uniquely universal form of culture. At the time of writing the largest radio telescope in the world is being built in China to test the same theories, by the same kinds of methods, as the largest optical telescope in Chile. The powerful particle collider, also being built in China, uses the same physics as CERN in Geneva. Science is uniquely successful as a form of epistemic inquiry, and the track record of attempts to curtail its scope is poor. In the next section more is said in favor of scientism. It is ironic indeed that some defenders of scientism lend it the air of dogmatism and a lack of epistemic humility when science itself is so often characterized by scientists echoing Karl Popper, as involving fallibilism and a constantly critical attitude to one’s own beliefs (see, for example, Deutsch 2011). Of course, much has been said in response to Popper, echoing Thomas Kuhn (1962) about how everyday scientific activity does not involve the scientists involved questioning their fundamental theories at all. Indeed, science is now so vast and specialized that the average scientist must accept and learn how to use a huge amount of theory that they may only dimly understand. However, Popper’s point can be taken to apply to the scientific community as a whole, in which case one does indeed find that even the most cherished assumptions are forever under scrutiny and subject to refinement and further testing. For example, the equivalence of gravitational and inertial mass has been known since the 17th century and tested repeatedly ever since, but physicists still devise new ways of verifying it to ever-​greater precision. However, there is a division of labor in such respects, and the vast majority of physicists never question the equivalence principle in their daily work. Fallibilism is integral to the scientistic spirit, and it is entirely in accord with it to say that science is the worst source of knowledge about the world apart from all the rest. The history of science teaches us that even cherished laws may be subject to revision. For example, the “fundamental dogma” of

16

1 1 6   • 

James Ladyman

molecular biology, according to which information can pass from the genotype to the phenotype but never vice versa is not entirely correct. The denial of Lamarkian inheritance (or “inheritance of acquired characteristics”) that the fundamental dogma expresses was central to the synthesis of evolutionary biology with molecular genetics, but the revised understanding we now have through epigenetics was not developed by creationists or intelligent design theorists but by biologists themselves. Similarly, it was physicists striving for scope and accuracy, and being prepared to challenge their own principles, who brought about revolutions in physics that led to profound departures from the orthodoxy of Newtonian science. Three objections to the above are as follows: (1) Scientism is empty without a definition of science; (2) science cannot tell us about right and wrong, but moral norms and values are part of reality, so scientism is refuted; and (3) science cannot tell us about the first-​person perspective, but the latter is real, so scientism is refuted. Objection (1) raises an analogue of Hempel’s dilemma for physicalism for the positive doxastic components of scientism. The problem is that in assertions such as “scientific methods can be extended to any domain,” the scope of quantification can be read two ways. Either, for any domain, there is a scientific method for studying it; or, there is a scientific method such that any domain falls within its scope. The first reading is obviously weaker and allows for the fact that the methods of science have developed as science has been extended to new domains. These methods, and indeed the domains themselves, are integrated to varying extents (as argued below). Nonetheless, it would be rash to bet that the current range of scientific methods won’t change as science continues to develop and to study new domains, as well as continuing to study itself. Hence, the promise of the scientific method for every domain might seem to be either false of current scientific methods, or only trivially true, if it is taken to mean that some scientific method can be found, and if no constraints are put on what counts as a scientific method. The nature of the latter is of course highly contested. It is unreasonable to demand that the advocate of scientism give a full account of the scientific method but they must have something to say about it. Haack regards what she characterizes as “preoccupations” with the demarcation of science from nonscience, and with identifying the scientific method and regards, as characteristics of scientism. She thereby presupposes that the study and

17

Scientism with a Humane Face  • 

117

articulation of scientific methodology is unnecessary. However, the demarcation of science from pseudoscience is, arguably at least, more important than ever, because there is a lot of material that adopts the superficial form of good science but violates standards of rigor that are required for reliable inferences to be drawn from data. Even the most expert scientists must guard against misapplications of statistics, because there are so many subtle traps into which to fall. Continual scrutiny of methodology is integral to science’s capacity to produce reliable knowledge and to correct its own errors. Given that science has such an important role in the law, medicine, and public policy, the demarcation of science from pseudoscience is vital. While we may not be able to say precisely what is and is not scientific, we have learned a great deal about the difference between good and bad ways of reasoning inductively. Statistics is a science that barely existed before the 20th century. Despite the sophistication and variety of techniques current science uses to study the world it is unlikely to be apt to understand everything. Science evolves new methods as it extends into new domains, so the right way to understand the optimism of scientism is in terms of the indefinite extendability of current science rather than its all-​encompassing nature. One thing we have learned from the history of science is that science adapts and evolves. The social sciences don’t use the same measuring devices as physics. It is reasonable to conjecture it will continue to do so. Hence scientism too must continually evolve along with science itself. For example, the ideas now used to describe genetics and the brain incorporate ideas of function and information that could have had no place in the scientism of the past because they had no place in the science of the past. It took Darwin’s theory of evolution by natural selection to transform what could count as scientific by redeeming teleological reasoning. Scientism must continually evolve as science evolves. However, there is a common core of both scientific theories and methods. The whole of natural science is based on the taxonomy of the periodic table of atomic structure, the fundamental forces and particles of the Standard Model, and the common system of scientific units and fundamental dimensions. These theories and their experimental technologies are applied to everything from the early and distant universe to the human brain.17 The use of logic and mathematics is common to all the sciences, and all scientists recognize the

17. While there is no doubt that these theories are only approximations that will be refined by future developments, they, like Newtonian mechanics and ray optics, will continue to have their domain of applicability, and should be expected to be explained as a limit in future science, in accordance with the “correspondence principle” (see Ladyman and Ross 2007: ch. 2).

18

1 1 8   • 

James Ladyman

need to consider competing explanations, to look for sources of error in their experiments, to make precise measurements, and so on. The idea that empirical testing is the ultimate source of epistemic authority in science is pretty much universal. Extensions of the scientific method must be integrated into the existing edifice, and of course old methods are often used to assess the reliability of measurement devices and background theories. The integration of methods and theories is essential to the success of current science, which is often highly multidisciplinary and hugely collaborative. In sum, (1) can be met by characterizing science in evolutionary terms as whatever evolves from current science. In any case, the arguments that matter in practice are those concerning extending current science into new domains in which case the extension of the term “scientific methods” is clear (though again it is routine in science for there to be new methods and modifications of existing ones). As for the relationship between scientism and physicalism, it must be noted that the successful application of the physics of particles and fields to chemistry and biology has made science integrated and unified, and at least provides prima facie grounds for some kind of asymmetric supervenience of everything on the physical. Whether or not this vindicates strong forms of reductionism it certainly means that there is something to the idea that every putatively nonphysical thing, such as a mind, somehow depends on the physical stuff associated with it. For some, of course, all this is very much understated and we have good reason to be more or less eliminative reductionists about everything except fundamental physics. Scientism as understood here, however, is definitely not committed to such views, nor to materialism. Indeed, far from reducing everything to matter in motion, as mentioned above, physics itself has refuted materialism and in its place we have at best a more vague kind of physicalism. Nonetheless, atomism did triumph in chemistry and led us to the more complex conception of the atomic world that we now have.18 Sui generis chemical and vital forces were not fecund for the development of explanatorily and predictively detailed and successful theories, and the pursuit of unity by building bridges between disciplines is one of the hallmarks of current science. However, even strong reductionists must concede that when it comes to epistemology and methodology, simple reductions seem unavailable. The best we can say is that science does not advance by positing new kinds of physical entities, properties, or processes solely to account for

18. The price of success was the discovery that atoms are not in fact “atomic” in the original sense of indivisible and mechanical.

19

Scientism with a Humane Face  • 

119

biological or psychological phenomena. This is an inductive hypothesis that may ultimately turn out to be false, but it is seems to be a warranted generalization of the lessons learned when the programs of sui generis chemical, physiological, and vital forces failed (Brown and Ladyman 2009). In respect of objection (2), it is no part of the core of scientism as understood here to say that scientific knowledge includes what is right and wrong, nor to say that science tells us there is no right and wrong. However, conceding that moral norms are part of reality does not refute the scientism characterized here, since nothing that has been said about the latter so far entails denying the former. Those who adopt the scientistic stance may hold some form of the fact/​value distinction and argue that science is completely silent about the moral norms we should adopt, and hence about all or part of ethics and politics. However, they may also insist that we should be scientifically informed about the biological and social facts that bear on the practicality of ethical and political policy proposals. They may also argue that advertised sources of moral knowledge need not command our assent. Scientism as characterized above is nonetheless compatible with allowing that different sources of beliefs about various matters, and especially ethics, may legitimately play an important role in social and public policy. If science conflicts with those sources of belief, that can only be because the relevant science establishes facts that they deny, and in those circumstances it is hardly appropriate to criticize the new scientific knowledge for correcting error. In the next section it is argued that this has indeed happened with ethics and the human sciences, and a good thing it is too. In any case, adhering to epistemic norms and values is already part of scientism as characterized here, since science relies upon them. Accepting moral norms and values is not relevantly different, so, while so doing is not part of the core of scientism, it is compatible with it. Objection (3) is an interesting challenge. It is closely related to the idea of the hard problem of consciousness. In reply, note that science has already at least told us some things about the first-​person perspective, by distinguishing different forms of consciousness and experience, and revealing the existence of blindsight and other phenomena that shatter a priori presuppositions about perception. If we characterize science in terms of what can be understood from the third-​person perspective, then it is true by definition that it cannot tell us about the first-​person perspective. However, even if there are ineffable truths about the latter this is hardly a threat to scientism, which is intended to take a stand against taking anything other than science as a source of knowledge.

210

1 2 0   • 

James Ladyman

The rest of this chapter defends a form of scientism that has (i) the core positive and negative components discussed above, but not stronger ones; and (ii) the epistemic charity, humility, and temperance that is in keeping with the spirit of science discussed above. The resulting form of scientism is cognizant of the limitations of current scientific knowledge, and the ways in which seemingly supremely well-​confirmed theories have turned out to be wrong in significant respects. Furthermore, a realistic grasp of the depth and breadth of current science, and its conceptual and mathematical sophistication, means that we should be very skeptical about grand metaphysical claims that are alleged to follow from theories, especially if the latter are speculative and have not enjoyed their own predictive success. Not everything that comes out of the mouth or pen of a scientist is science, and scientists discussing other specialisms, philosophy, and indeed scientism may be particularly liable to what Ken Waters (2017) calls “science overreach.” Scientism demands eternal vigilance and perpetual critique. Scientists should not be deferred to about matters beyond their expertise, and there are other modes of inquiry that deserve epistemic respect, such as history and law. The next section outlines a humane scientistic stance.

4.4 Humane Scientism Scientism began with the extension of science to the study of human beings and societies. As Richard Olson says, “scientism” is “the transfer of ideas, practices, attitudes and methodologies from the context of the study of the natural world (which was assumed to be independent of human needs and expectations) into the study of humans and social institutions” (Olson 2008: 1). Whether the critics of scientism like it or not, this has been done successfully. In response, they make much of the importance of humanistic values as if the pretensions of science to tell us about ourselves were a threat to our humane treatment of one another. It is worth reflecting on how our ethics have also evolved along with the scientific image of people. Ironically, there is certainly a correlation between the development of the naturalistic understanding of human beings, and the development of the idea of universal human rights and the abolition of forms of cruel punishments and various forms of ill treatment. This is most clearly seen in the case of atypical human psychology. The identification of conditions such as autism, dyslexia, and dyspraxia has led to a much better understanding of the normal range of human variation, and a much better way of treating children as they develop. Ideas of naturalness in cultural and religious traditions are the alternative to the

12

Scientism with a Humane Face  • 

121

scientific understanding and they are associated with a punitive attitude to deviation from the norm. The treatment of the mentally ill leaves much to be desired, and pharmacology arguably determines taxonomy and treatment in a scientifically dubious way. Compared to the “madhouses” of recent history, however, we have come a long way. If we care about human well-​being, we should note that what science has taught us about ourselves and society has had a largely positive impact on our treatment of people that do not fit the prescientific conceptions of how they are meant to be. The claim that we would be better off without the scientific understanding of human behavior, character, and cognition seems completely absurd once it is made explicit. However, it is the logical consequence of the rejection of the core scientism argued for above. Critics of scientism should be careful for what they wish, and careful not to take for granted the understanding we have of ourselves as natural beings. In the absence of science, we know that often what fills the void is prejudice and superstition. This why it is important that scientism decrees that beliefs about ourselves based on common sense, intuition, religion, and other forms of tradition, have no prima facie claim to the truth over those of other traditions, and certainly no immunity to revision in the light of scientific knowledge. If science establishes facts that inherited belief systems deny, it is hardly appropriate to criticize the new scientific knowl­ edge for correcting error. In his book on scientism Tom Sorell (1991) discusses “philosophy and the infatuation with science” as if this was something aberrant, but if we think of science as the human institution that gives us reliable knowledge of objective reality then it is quite right for philosophers to be infatuated with it. The sciences are the progeny of philosophy and philosophers are right to be appreciative of them. The problem is rather that too often discussions in philosophy presuppose accounts of cognition or physical reality that are uninformed by and even incompatible with what science is telling us. Similarly, rather than science being culturally dominant and informing decision-​making too much, governments are much more often beholden to other interest groups and scientific evidence is ignored. For example, in the United Kingdom, bishops sit in the House of Lords but there are no ex officio representatives of science, and very few politicians have been scientists. Scientism might be associated with cold and uncaring approaches to decision-​making, reducing individual people to units of economic output, and neglecting the value of public and social goods. However, there is nothing about the core values of scientism that encourages these attitudes, and they arguably conflict with it. Science tells us that every individual human being is

21

1 2 2   • 

James Ladyman

born utterly incompetent and remains unable to fend for itself for years as a result of the brain having the flexibility for the child to be taught the cognitive content of whatever culture into which it is born.19 Sociality is essential to the unique features of our cognition, and collective endeavor is essential to our flourishing. Hence, scientific knowledge of our nature does not support individualism, and there is no reason to associate clinical impartiality or the determination to be objective with a lack of compassion. Current scientific knowledge is the product of a social history of experiment, theorizing and application, interacting and refining each other through the constant dialogue of people questioning and correcting each other. Of course, scientists too are susceptible to confirmation bias; some even cheat and lie, distort and dissemble, and abuse power just as people do in other walks of life. More prosaically sometimes people are marginalized unfairly or are ignored because they lack status. However, the scientific community functions to produce reliable beliefs, albeit imperfectly, because of shared commitments to knowledge, understanding, and truth that are enacted every day when people exchange and explain ideas to each other, and debate with the common goal of reaching a better understanding of the world.20 Any association of science with rampant individualism and selfishness would be wholly unwarranted. Current science is collaborative and fundamentally reliant on epistemic respect and trust between those among whom scientific labor is divided. The culture of science also essentially involves questioning the calculations, methods, or beliefs of one’s collaborators on occasion. Science is not a fixed set of beliefs but rather a dynamic network. It is sustained by checking and double-​checking, and the relentless search for greater accuracy and precision, both in measurements and models. Science is social knowing based on testimony, but it is also based on the actuality and possibility of third-​party scrutiny. In many cases the scientific knowledge we have is the product of the collaborative efforts of many individuals over many years, each playing their part in accumulating data and correcting error.21

19. I am not here taking a stand on the nativism debate, but pointing out that human infants remain wholly dependent upon caregivers for a much longer period of time than the young of other species. 20.  This does not presuppose scientific realism since even empiricists like Bas van Fraassen accept that science tells us the (approximate) truth about the phenomena. 21.  Scientific papers are now almost always co-​authored and in many cases co-​authored by dozens of individuals.

213

Scientism with a Humane Face  • 

123

In the previous section it was pointed out that adherents of scientism may also hold some form of the fact/​value distinction and argue that science is silent about the moral norms we should adopt, and hence about all or part of ethics and politics. Hence, they may completely agree with Haack when she says, “results from the sciences can give us information about the relation of means to ends, but cannot by themselves tell us what ends are desirable” (2003: 45). However, we should be scientifically informed about the biological and social facts that bear on the practicality of ethical and political policy proposals. In this minimal sense, science is relevant to moral deliberation, but in some cases it may do more and transform how we think about the issues. To many critics of scientism, the idea that we are natural beings subject to the causal nexus of the physical world seems somehow to reduce our humanity. However, we owe it to those afflicted with organic problems such as the brain tissue degeneration that occurs with dementia to understand the biological processes involved as well as possible. Better scientific knowledge has been accompanied by demands for more compassionate treatment of dementia sufferers and their caregivers. Science cannot tell us what our ends should be, but it transforms our conception of ourselves, the world, and what is possible for us. Science clearly cannot tell us what we should value, but given that we do value individual people extremely highly, it can help us enact our beneficence. Scientific medicine is obviously the most striking example of the miraculous humanitarian intervention that is made possible by science and technology; others include mitigation of famine and natural disasters. Liberals accept that there are competing conceptions of the good and that our political institutions should accommodate them, not favor any one of them in particular. Hence they may be construed as arguing that since we do not know what a good human life is, different beliefs about the matter are politically legitimate. The extent to which economics and political science incorporate norms and values varies, but when it comes to matters such as predictions about the effect of policy proposals, there is nothing comparable to the precision and predictive success of the physical sciences, and in some cases there may be little that can be said that is objective. Analogously, scientism may be understood as making science authoritative in respect of knowledge while allowing that different sources of beliefs about various matters, especially ethics, may legitimately play an important role. It is also consistent with scientism as characterized above that there are legitimate sources of belief and knowledge other than science. To say that an activity or group merits epistemic respect means, minimally, that it is a seat of rational inquiry, is methodologically

214

1 2 4   • 

James Ladyman

reliable to some extent, and makes a contribution to knowledge. There are many such examples including legal inquiry, folk local and natural history, academic history, and so on. Humane scientism must take account of the hugely varied methods of the sciences and the rich role of experiment and practice, as well as theory and modeling. The previous section argues that scientism does not require either materialism or reductionism, although arguably it does depend on the integration and unity of science that supports a weak form of physicalism. Scientism does not require the denial of the first-​person perspective, nor need devalue the explorations of the latter in the arts and humanities. Scientism does not demand belief in the omniscience or omnicompetence of science and the replacement of all other forms of inquiry, including the traditional modes of study in the arts and humanities. However, it does deny that science has any particular limits in principle, in the sense that there are no domains to which scientific methods and theories cannot be applied. On what grounds could it be decided that some domain is out of reach of science? Haack is right that there are various rhetorical weaknesses with scientism as hitherto formulated. It is associated with an uncritical attitude to science, and the use of the word “scientist” and its cognates as epistemic honorifics is sometimes suggestive of a tribal mentality There is too much deference to scientists outside of their areas of expertise in the media, and a lack of appreciation of how specialized science now is, and how easy it is for a scientist to know nothing about even the basics of other disciplines (and even other parts of their own disciplines) especially when it comes to pronouncing on conceptual or philosophical issues. Heather Douglas (2009) and Philip Kitcher (2011) argue that we should defer to scientists’ judgments in their areas of expertise, but not on questions of public policy. Again, this seems uncontroversial, and required by scientism, as we should only ever defer to experts in their domains of expertise. Some scientists become experts in the relevant area of public policy, but many do not; and being good at, for example, designing nuclear warheads, obviously does not ipso facto make one good at deciding whether to build them and if or when to use them. Scientism must not be pseudoscientism. Humane scientism must be especially careful to insist that much of the use of quantification and metrics outside scientific research in, for example, corporate or government documents and procedures is pseudoscience, bearing none of the epistemic hallmarks of true science, because it involves only the superficial guise of measurement, but lacks features such as calibration, precision, and empirical reliability. This happens a great deal with so-​called evidence-​based decision-​making,

215

Scientism with a Humane Face  • 

125

which in fact consists in the manipulation of numbers which disguise massive value judgments; judgments that may have been made on the most dubious of grounds, about what categories should be used, how they should be aggregated, and how the scores assigned by different scorers should be combined. The fact that the methods in these domains are in constant flux is a sure sign that something is wrong. Among current sciences there are forms of highly theoretical reasoning, at one extreme, and forms of experiment and data gathering with little in the way of theory at the other. Mathematical manipulation may be used to provide the superficial appearance of science and statistics can be deployed in ways that even the cognoscenti may not (at first at least) realize are bogus. Trust in science must be differentiated and proportionate. Established science that has been the subject of ever more rigorous testing over decades or centuries, is obviously not on a par with the latest speculations about string theory or the multiverse. Many scientists are very good at what they do, but less good at describing it, and many say very confused things when they come to discuss the bigger picture or how science works. Accordingly, scientism does not involve blind faith, dogmatism, or uncritical adherence to some imagined complete corpus of scientific knowledge. Indeed, science thrives on the relentless search for and reduction of error and imprecision within its own theories, models, and data. In sum, humane scientism takes science to be authoritative in respect of objective knowledge, including about human beings and society. It recognizes no limits to science in principle, but is also antithetical to scientific hubris and hype. However, humane scientism holds the best of the arts and humanities in high esteem and recognizes the role that culture and custom, and religion and tradition, play in good human life.22

References Brown, Robert and James Ladyman. 2009. “Physicalism, Supervenience and the Fundamental Level.” The Philosophical Quarterly 59: 20–​38. Deutsch, David. 2011. The Beginning of Infinity:  Explanations That Transform the World. London: Viking Adult.

22. Thanks to the organizers and the other contributors to the Amsterdam conference, from whom I learned a lot about scientism. Special thanks are due to Alex Rosenberg, Susan Haack, and Don Ross. I am extremely grateful to Jeroen de Ridder, Rik Peels, Damian Veal, and René van Woudenberg for detailed comments on a draft.

216

1 2 6   • 

James Ladyman

Douglas, Heather. 2009. Science, Policy, and the Value-​ Free Ideal. Pittsburgh, PA: University of Pittsburgh Press. Haack, Susan. 2003. Defending Science—​ Within Reason:  Between Scientism and Cynicism. Amherst, NY: Prometheus Books. Hayek, Friedrich A. 1979. The Counter-​Revolution of Science:  Studies on the Abuse of Reason. Indianapolis, IN: Liberty Fund. Kitcher, Philip. 2011. Science in a Democratic Society. Amherst, NY: Prometheus Books. Kuhn, Thomas. 1962. The Structure of Scientific Revolutions. Chicago:  University of Chicago Press. Ladyman, James. 2011. “The Scientistic Stance: The Empirical and Materialist Stances Reconciled.” Synthese 178(1): 87–​98. Ladyman, James and Don Ross, with David Spurrett and John Collier. 2007. Every Thing Must Go: Metaphysics Naturalized. Oxford: Oxford University Press. Olson, Richard G. 2008. Science and Scientism in Nineteenth-​ Century Europe. Urbana: University of Illinois Press. Rosenberg, Alexander. 2011. The Atheist’s Guide to Reality:  Enjoying Life without Illusions. New York: W.W. Norton. Sorrell, Tom. 1991. Scientism:  Philosophy and the Infatuation with Science. London:  Routledge. Van Fraassen, Bas C. 2002. The Empirical Stance. New Haven, CT: Yale University Press. Waters, C. Kenneth. 2017. “No General Structure.” In Metaphysics and the Philosophy of Science:  New Essays, edited by Matthew Slater and Zanja Yudell, 81–​107. New York: Oxford University Press.

217

5

P H I L O S O P H Y, S C I E N C E , A N D COMMON SENSE

Hilary Kornblith

Wilfrid Sellars (1963) articulated a view of philosophy that many find congenial. Science presents us with a certain account of what the world is like, but this account is different, in fundamental respects, from what Sellars referred to as “the manifest image”: “the framework in terms of which man came to be aware of himself as man-​in-​the-​world” (ibid.:  6). As Sellars saw it, some seek to reconcile these images, by giving either the scientific image some sort of primacy, or, alternatively, by giving the manifest image primacy, or, instead, treating the two images as co-​equal. On Sellars’s view, “. . . the conceptual framework of persons [which is central to the manifest image] is not something to be reconciled with the scientific image, but rather something to be joined to it” (ibid.: 40). My own view on these matters is quite different: We should endorse features of the manifest image only to the extent that they are part of the scientific image. Many will, no doubt, see such a view as far too deferential to science. Among the many alleged shortcomings of the view I favor, it will be claimed, is that it leaves no place for philosophy. In this chapter, I will focus on the way in which these issues play out in epistemology. Apparent conflicts arise between the scientific image and the manifest image of doxastic deliberation. We often form beliefs unreflectively, but we also stop, at times, to deliberate about what to believe, and the reflective activity in which we engage during these moments is central to much of epistemological theorizing. We may view this activity from the first-​person perspective, that is, from the perspective of the deliberator him-​or herself. But we may also view this activity from the third-​person perspective,

218

1 2 8   • H i l a r y

Kornblith

as investigators do in the cognitive sciences. This contrast provides a case study in the conflict between the manifest and scientific images. I will argue that the only reasonable way to deal with this conflict is to treat the manifest image of doxastic deliberation as nothing more than an illusion. This does not, thereby, leave no room for epistemology, nor does it present us with an unacceptable or unappealing1 picture of our place in the world. I hope, by way of this case study, to provide some reason for favoring my preferred account of the relationship between the scientific and the manifest image. In Section 5.1, I describe the phenomenology of doxastic decision, and I  isolate a number of its distinctive features. Doxastic deliberation is here described from the first-​person point of view. In Section 5.2, I present a scientific account of doxastic decision, and I isolate a number of its distinctive features. Doxastic deliberation is here described from a third-​person point of view. These different perspectives on deliberation seem, on their face, to be incompatible. It is not merely that these perspectives look at a certain phenomenon in different ways or from different angles. Rather, these perspectives seem to offer rival accounts of a single phenomenon; if taken at face value, they cannot both be correct. In Section 5.3, I  examine two attempts to explain away the appearance of incompatibility, and I argue that both of them fail. This, of course, forces us to choose. We can, of course, reject both perspectives, and endorse some third alternative, but we cannot accept both the first-​and third-​person accounts of doxastic deliberation as literally true; they are, just as they initially appeared, inconsistent with one another. In Section 5.4, I argue that the only reasonable position here is to accept the scientific perspective on deliberation, and thereby reject as illusory the first-​person view of deliberation. Finally, in Section 5.5, I examine the implications of such a view for epistemology and for philosophical methodology more broadly.

5.1 Christine Korsgaard describes the phenomenology of decision in a famous passage in The Sources of Normativity:

1. Of course, the test of a theory has nothing to do with how appealing or unappealing it may be. At the same time, the scientific picture has often been portrayed as one which is somehow alienating, and it is at least worth pointing out that it need not be seen in that way at all. For one important example of the suggestion that the scientific picture is, in this way, alienating, see Strawson (2008).

219

Philosophy, Science, and Common Sense  • 

129

A lower animal’s attention is fixed on the world. Its perceptions are its beliefs and its desires are its will. It is engaged in conscious activities, but it is not conscious of them. That is, they are not the objects of its attention. But we human animals turn our attention on to our perceptions and desires themselves, on to our own mental activities, and we are conscious of them. That is why we can think about them. And this sets us a problem no other animal has. It is the problem of the normative. For our capacity to turn our attention on to our own mental activities is also a capacity to distance ourselves from them, and to call them into question. I perceive, and I find myself with a powerful impulse to believe. But I back up and bring that impulse into view and then I have a certain distance. Now the impulse doesn’t dominate me and now I have a problem. Shall I believe? Is this perception really a reason to believe? I desire and I find myself with a powerful impulse to act. But I back up and bring that impulse into view and then I have a certain distance. Now the impulse doesn’t dominate me and now I have a problem. Shall I act? Is this desire really a reason to act? The reflective mind cannot settle for perception and desire, not as such. It needs a reason. (Korsgaard 1996: 92–​93) Korsgaard very nicely captures the phenomenology of deliberation in this passage. Most of the time, we act without self-​consciously deliberating about what to do. The alarm clock rings in the morning and I turn it off and promptly get out of bed. I don’t, at least typically, lie there thinking about whether I would like to stay in bed longer. I simply get up without any further thought. I shower and get dressed, again, without self-​conscious deliberation. I don’t think about whether I really want to do these things. I just do them. These actions are nonetheless intentional. They are my doings, even if I do not deliberate about them. But I can deliberate about what to do. After a long week of rainy days, I wake to see that it is a beautiful sunny day. I think about getting out in the sunshine to enjoy the day, and then I remember that I have papers to grade. I find myself pulled in two different directions. What should I do? I stop to deliberate: Should I stay home and grade papers, or should I go outside and enjoy a rare sunny day? As Korsgaard notes, in such moments of deliberation, we feel at a certain distance from our desires. I have the desire to enjoy the day, and I also have the desire to attend to my responsibilities and grade papers. I can survey my desires, and make a choice as to which will guide my action. When I act

310

1 3 0   • H i l a r y

Kornblith

unselfconsciously, my desires move me without my playing any active role in choosing among them.2 But when I deliberate, I temporarily put a hold on my desires, intervening in the causal process by which they would otherwise produce action, much as I may step on the clutch and temporarily disconnect the engine of my car from the drivetrain. In stopping to deliberate, I engage the clutch in my mental mechanism while I decide which of my desires should be connected to the drivetrain. When I decide to take the day off and enjoy the sunshine, I shift my desire for a holiday into gear and I re-​engage the mental mechanism so as to produce action. It is much the same with belief. Most of my beliefs are formed unselfconsciously, without the need for deliberation. I thought I had left my car keys on the kitchen table, but when I see them on the counter next to the toaster, I don’t stop to think about whether my visual evidence really gives me adequate reason to abandon my former belief. Seeing the keys next to the toaster produces belief change without any need for my intervention in the process. But it isn’t always so. I look out the window and I see a hawk at the top of a distant tree, and I think, it’s a red-​tailed hawk. But then I remember that a Swainson’s hawk has recently been seen in the neighborhood, and I stop to think: What are the distinguishing marks of these two birds? Do I really have good reason to believe that it is the red-​tail rather than the Swainson’s? When I first saw the bird, I was brought to believe that I was looking at the red-​tail. The belief was simply produced in me without my intervention. But when I remembered about the Swainson’s hawk in the area, I stepped back from my initial belief and disengaged. I put in the clutch, as it were; I intervened in the mental mechanism that had produced my belief, and so, at least for the moment, suspended judgment. I stop to scrutinize my evidence and form a judgment about its strength. Yes, indeed, I conclude, it lacks the distinctive marks of the Swainson’s, and given the distance I am from the bird, and the angle I’m viewing it from, I would see those marks if they were there. And now, just as in the case of action, I re-​engage my mental mechanism, in this case, allowing the belief about the red-​tail to be formed once again, or, perhaps, to remain in place. Let us focus on the case of doxastic deliberation. As viewed from the first-​person perspective, it has a number of distinctive features. While belief

2. It is thus that Harry Frankfurt (1988: 21) says of such cases that the agent is “a helpless bystander to the forces that move him.”

13

Philosophy, Science, and Common Sense  • 

131

unselfconsciously arrived at is obviously produced by some sort of mental mechanism, cases of deliberation seem to illustrate our ability to intervene in the operation of such mechanisms. We stand outside the sphere in which those mechanisms operate and put them temporarily on hold while we deliberate. When we come to a conclusion about which belief our reasons favor, we grab ahold of the levers of the mechanism, as it were, and engineer the production of the belief that reason dictates. Then we let go of the mechanism, and allow it to operate, once again, as it normally does, without our intervention.3 Our ability to reflect on our beliefs, something which other animals are incapable of, thus makes reflective belief acquisition utterly different in kind than the merely mechanistic production of belief that goes on in other animals, or in us at times when we are unreflective.4 There is more to it than this, of course. Our motivation in scrutinizing our reasons for belief, when we deliberate, is to provide a check on the mechanisms by which our beliefs are produced when we form beliefs unreflectively. Such mechanisms may be, at times, quite reliable in producing true beliefs, but they are not uniformly so. By stopping to reflect on the strength of our reasons, we do not simply cross our fingers and trust to luck that our belief-​producing mechanisms are operating reliably. Instead, by taking control of our belief acquisition and assuring that our beliefs are appropriately responsive to our reasons, we thereby improve our reliability. Deliberating about what we ought to believe thus helps us to locate errors that we might otherwise make, thereby improving the quality of our epistemic performance.5 When we form beliefs unreflectively, the mechanisms by which our beliefs are produced operate outside of consciousness. But when we stop to reflect, and form beliefs as a product of deliberation, the processes by which our beliefs are produced are directly available to consciousness. Indeed, our understanding of these processes is not comparable to the introspective

3. Thus, for example, Paul Boghossian (2016) contrasts what goes on in adult human beings with the information processing that goes on in other animals as a difference between “reasoning and mere mechanism.” 4.  Similarly, for Frankfurt (1988), there is a fundamental difference between adult human beings, who are capable of reflecting on their desires and choosing which of them they wish to be effective in producing action; and other sorts of animals, who are not. The former are persons, capable of freedom of the will, on Frankfurt’s view, while the latter are only capable of free action. 5. Thus, Ernest Sosa (1991: 240) claims that belief reflectively arrived at has “a better chance of being right” than unreflectively produced belief.

312

1 3 2   • H i l a r y

Kornblith

recognition of antecedently existing mental states, such as when we notice that we are experiencing a certain emotion. Rather, we know what processes are producing our beliefs, when we deliberate, because, in such situations, we are the author of our beliefs.6 In deliberation, we take control of our belief acquisition, and we know how our beliefs are produced because we know what we are doing. Similarly, our reasons for belief are wholly transparent to us when we form beliefs as a result of deliberation, in sharp contrast to cases of unreflective belief acquisition. When I form beliefs unreflectively, I may find myself at a loss if you ask me what my reasons are for holding a particular belief. But when my belief is formed as a result of deliberation, I am ideally placed to answer the question as to my reasons for belief. I know what my reasons for belief are because I chose to form the belief on the basis of a particular set of reasons. Thus, the first-​person perspective on belief acquisition as a product of doxastic deliberation reveals that it has four distinctive features:  (1) Deliberation allows us to stand outside of the mechanistic processes, which would otherwise produce belief, allowing us to directly intervene in belief production; (2) by intervening in this way, we provide an extra check on the processes by which our beliefs are produced, thereby assuring that our beliefs are appropriately responsive to our reasons, and making it more likely that the resulting beliefs are true; (3) when we form beliefs as a result of deliberation, we are directly acquainted with the processes by which our beliefs are formed; and (4) because we are the author of our beliefs when we form them as a result of deliberation, the reasons for which we hold our beliefs are fully transparent to us. This view of self-​conscious belief acquisition is part of what Sellars referred to as “the manifest image.” It is our common-​sense picture of how such belief acquisition fits in, or fails to fit in, to the wider world. This picture of belief acquisition plays a particularly central role in the work of those philosophers who, like Sellars, would have us draw a sharp distinction between “the space of reasons” and “the space of causes.” And it is the apparent clash between this picture of deliberative belief acquisition, and a scientific picture of that same phenomenon, which makes Sellars worry about how we might reconcile, or whether we should reconcile, the manifest and the scientific images of

6. This is the central focus of Moran (2001).

31

Philosophy, Science, and Common Sense  • 

133

humankind.7 We thus need to turn to a scientific account of self-​consciously formed belief to understand how this problem arises.

5.2 There are many different sciences that are relevant if we wish to understand belief acquisition. We may, for example, look at work on the structure of the various sense organs. Or we may, instead, look at work at the neural level examining how signals from the sense organs are processed by the brain. We may, as well, take an evolutionary perspective on the large-​scale structure of belief-​acquisition processes. Here, however, I  will focus on psychological investigations of belief acquisition. It is no challenge at all to the manifest image that there are a host of psychological processes that go to work in us when we unselfconsciously form beliefs. We have no first-​person perspective at all on unselfconsciously arrived-​at belief,8 so there is no danger of conflict between whatever picture of such belief acquisition science may offer and the manifest image of that phenomenon. Since there is no manifest image of unselfconscious belief acquisition, there can be no conflict between the scientific image and the manifest image here. More than this, it is not terribly surprising that the picture of unselfconscious belief acquisition, which psychology offers us is largely, in a word, mechanistic. Belief acquisition is a complicated affair, and although there seem to be input systems that are largely modular (see, e.g., Fodor 1983), belief fixation itself seems deeply dependent, not only on external stimuli, but on the content of antecedent beliefs. A good deal of cognitive psychology is devoted to understanding exactly how various cognitive subsystems involved in unselfconscious belief acquisition—​such as perception, memory, language acquisition, and use, and so on—​function and interact. At the same time, however, the approach that cognitive psychologists take to the acquisition of belief as a product of deliberation is no less mechanistic. Thus, for example, recent work on System 1 and System 2 treat the

7. Much the same set of issues has long been the focus of Thomas Nagel’s work, although he conceptualizes this, not as a clash between the scientific and the manifest image, but between subjective and objective views of various phenomena (see Nagel 1979, 1986). 8. This is not to deny that, of course, we do have some views about the origins of such beliefs. It is just that these views do not derive from the first-​person appearances of unconscious processes since such processes do not give rise to any first-​person appearances.

314

1 3 4   • H i l a r y

Kornblith

self-​conscious System 2 processes no less mechanistically than the unselfconscious System 1 processes (Evans 2010; Evans and Frankish 2009; Kahneman 2011). The kinds of rules that govern System 1 thinking look quite different from those which govern System 2, but each of these systems is thoroughly rule governed in its operation.9 Indeed, from the perspective of contemporary cognitive science, anything short of an explanation of the operation of these systems in terms of the operation of rules is, in virtue of that very fact, not a full explanation of their operation at all. There are elaborate rule-​governed mechanisms that operate so as to produce beliefs; some of them are entirely unavailable to consciousness, while others operate, at least in part, at the conscious level. Thus, belief acquisition, whether it involves conscious processing or nothing but unconscious processing, is entirely mechanistic. Rule-​ governed processes are the only kinds of processes to be found here. An examination of the processes that operate below the level of consciousness reveals that, while many of them are extremely reliable, there are many, as well, which are quite unreliable. There has been a great deal of investigation of these psychological processes, and much work has been done to understand how reliably they operate in real-​world conditions. When psychologists investigate the reliability of these belief-​producing mechanisms, they are not interested in whether they would tend to produce true beliefs in all possible worlds. Rather, they are interested in the extent to which they tend to produce true beliefs in worlds very much like our own. The fact that our perceptual mechanisms, for example, would simply melt at the temperatures found in the daytime on Venus, and thus fail to operate there at all, does not count against the kind of reliability about which perceptual psychologists are concerned. Similarly, inferential processes may operate in ways that are also tuned to local features of the environment, and thereby count as extremely reliable for all the purposes that psychologists are interested in investigating.10 The fact that such processes might not be at all reliable in worlds quite different from our own is of no concern to psychologists. Just as the processes that operate below the level of consciousness include both reliable and unreliable processes, the processes that operate at least in part at the conscious level are not uniformly reliable. For this reason, the

9. This is not meant to suggest, however, that System 1 and System 2 do not interact. For a discussion of their interaction, see Kornblith (2012: section 5.3). 10. I have discussed the ways in which inferential processes might be tuned to features of the environment in Kornblith (1993).

315

Philosophy, Science, and Common Sense  • 

135

question of whether we are more reliable when we self-​consciously scrutinize our reasons for belief, as we do when we deliberate, has no simple answer. The phenomenon of confabulation illustrates an important range of cases in which self-​conscious deliberation may fail to produce any changes in the content of what we believe.11 In cases in which a belief is formed as a product of non-​rational or irrational influences, subjects may stop to reflect on that belief and privately question whether it is appropriate to go on believing as they already do. In such cases, a number of different psychological processes will go to work that are likely to convince subjects that continuing to believe is the rational thing to do. Subjects who ask themselves why they formed a target belief are likely to come to believe that they had good reasons for believing it in the first place (Nisbett and Wilson 1977). When subjects stop to consider the evidence that they had for and against a preexisting belief, whether it was rationally formed initially or not, evidence that counts in favor of it will more readily be brought to mind than evidence that counts against it, thereby reconfirming, once again, that they should continue to believe as they already do (Lord et al. 1979). The upshot of deliberation, in a very wide range of cases, is that subjects will go on believing as they did before, whether or not their belief was reliably formed initially, and whether or not it was based on good reasons. While deliberating in these cases does not result in greater reliability, it does result in greater confidence. In another important range of cases, subjects will actually decrease their reliability as a result of self-​conscious scrutiny of their beliefs (Halberstadt and Wilson 2008). Just as consciously attending to a bit of behavior, such as tying one’s shoes, may interfere with one’s smooth and accurate performance of the task, the same is true in an important range of cases for accurate reasoning. This is not to deny that stopping to reflect on the epistemic status of one’s beliefs may also, at times, improve one’s accuracy. The point here is simply that it does not in fact produce this result nearly as often as it seems to. It is certainly true that deliberating about what to believe may convince subjects that they are directly acquainted with the processes by which their beliefs are produced. In the case of antecedently existing beliefs that a subject comes to question, subjects will confidently claim to know what their reasons were for coming to believe as they did. But they are not at all accurate in these judgments (Nisbett and Wilson 1977). Subjects who consider taking on new beliefs, and who evaluate the evidence for and against such new claims, will

11. I have discussed this in further detail in Kornblith (1998, 2002: ­chapter 4, 2012: ­chapter 1).

316

1 3 6   • H i l a r y

Kornblith

confidently report that their newly acquired beliefs are based on a small body of evidence that they are able to fully articulate, since they just, it seems, went through the reasoning in deliberation and are thus directly acquainted with the source of their new belief. But this is largely a product of the limitations of working memory, rather than an accurate report of the inferences that actually went on in them (Miller 1975). Worse still, as Ziva Kunda remarks: Our judgments, feelings, and behaviors can be influenced by factors that we have never been aware of and have only been exposed to subliminally, by factors that we were aware of at one time but can no longer recall, and by factors that we can still recall but whose influence we are unaware of. (Ziva 1999: 308) We are just not very good at accounting for the various reasons for which we form our beliefs, whether they are formed unselfconsciously, or under conditions of deliberation. The picture that emerges from this body of research thus seems to stand in sharp conflict with the view we have of ourselves when we deliberate. In light of this work, let us review each of the four features identified in the previous section which the first-​person perspective would have us attribute to our doxastic deliberations. As viewed from the first-​person perspective, the act of deliberation seems to allow us to stand outside the various mechanistic processes by which our beliefs are produced and, in light of our deliberation, intervene in the operation of those processes. But the picture that psychological research offers leaves us no such place to stand. There is no place for the self to occupy outside of the processes by which our beliefs are formed. When we form beliefs unselfconsciously, our beliefs are produced by way of mechanistic processes. But work in the cognitive sciences gives an equally mechanistic account of how it is that beliefs are formed under conditions of deliberation. Just as we can ask, about unselfconscious belief acquisition, what the processes are that go on in us, leading to the production of belief, we may ask the very same question about the processes going on in us under conditions of deliberation. The account that the cognitive sciences give of these latter processes seems to be fundamentally at odds with the picture we are presented with from the first-​person perspective. It is not just that the perspective of deliberators on their own mental processes is inevitably partial, and thereby leaves out crucially relevant features of the manner in which their own beliefs are formed. In addition, as the phenomenon of confabulation illustrates, the partial

317

Philosophy, Science, and Common Sense  • 

137

picture that the first-​person perspective presents on belief acquisition is not even fully correct as far as it goes. The appearance of being able to intervene in the operation of mechanistic processes thus seems, in the light of experimental work, to be nothing more than a manifestation of the illusion of control (Langer 1975). When we scrutinize already existing beliefs in deliberation, the deliberator is under the impression that these beliefs are being subjected to an extra layer of scrutiny, thereby assuring that these beliefs are appropriately responsive to reason, and thereby improving the reliability of the overall belief acquisition and retention process. But the experimental work on the processes that go on during deliberation show that this extraordinarily optimistic view of the deliberative process is not at all the typical case, let alone the invariable result of deliberation. Deliberation is often ephiphenomenal with respect to the fixation of belief. It often provides no additional check whatsoever on preexisting beliefs, but instead serves only to provide the subject with additional confidence that beliefs already formed were fully in order epistemically. Worse still, the process of deliberation may actually decrease reliability. The various factors that are responsible for belief fixation, both rational and irrational, need not reveal themselves even to the conscientious deliberator, and thus need not be available for deliberative scrutiny. Deliberation is not remotely as efficacious in producing good epistemic effects as it seems to be from the perspective of the deliberator. For similar reasons, the fact that the deliberator is under the impression that the processes by which beliefs are formed, as a product of the act of deliberation, are directly available to introspective scrutiny seems to show yet another respect in which the perspective of the deliberator gives rise to a thoroughly inaccurate view of what is going on in the deliberative process. Experimental work shows here that there is much more going on psychologically than is available to the introspective gaze of the deliberator, and that the deliberator’s account of the causally relevant factors of the deliberative process is not even accurate as far as it goes. There is an illusion of first-​person access to the deliberative process, and the experimental work seems to show just how large the gap is between the facts and the view of the deliberative subject. By the same token, the deliberator’s impression that the process of deliberation allows for direct access to the reasons for which one holds those beliefs that one arrives at as a result of the deliberative process is also illusory. One may hold beliefs for reasons of which one is unaware, even under conditions of deliberation. Similarly, the reasons for which one believes one

318

1 3 8   • H i l a r y

Kornblith

holds a particular belief may, in fact, be no part of one’s reason for belief at all, even under conditions of deliberation. One’s apparent authorship of one’s beliefs, when deliberating, does not, in fact, guarantee access to one’s reasons for belief. Thus, the four features that we identified as distinctive of the first-​person view of deliberation each seem to be in direct conflict with the results of experimental work on the deliberative process. If taken at face value, these two perspectives lead to contradictory results. This is simply one illustration of the clash between the manifest image and the scientific image. The question now arises as to whether we should take this conflict at face value. Before turning to that issue, let me address one concern that may arise about the experimental work that I  have cited here.12 I  have attempted to characterize some central features of the deliberative process, and to question the reliability of a first-​person understanding of these features, on the basis of a variety of psychological experiments. There are a number of questions that may reasonably be raised whenever one engages in such an enterprise. If experiments are performed in the laboratory under controlled conditions, one may reasonably wonder to what extent these conditions shed light on the phenomena under study as they occur in ordinary life, outside the laboratory. On the other hand, if one makes use of careful field observations of ordinary life outside the laboratory, one may reasonably wonder whether a more careful controlled experiment would yield the same result. More than this, any single paper, or, for that matter, any single book, will inevitably draw on a small sample of studies, and we are all familiar with the fact that individual studies may contain results that turn out not to be reproducible, or which simply fail to generalize in the way in which the author who draws on those studies argues that they will.13 These are all reasonable concerns. At the same time, it would be a mistake to use these very general concerns to support a broad skepticism about experimental work in psychology, or to encourage a reliance on first-​person judgment in the place of experimental evidence. The details matter. I have tried to draw only upon work that has been shown to produce robust results, yielding similar upshot in a wide range of

12. The editors of this volume raised this concern, but it is worth noting that it is one which I have often been confronted with in discussion, as have many others who attempt to draw on experimental work in psychology in discussing issues in epistemology, philosophy of mind, and action theory. 13. For an important discussion of concerns about reproducibility, see a paper authored by the Open Science Collaboration (2015: aac4716).

319

Philosophy, Science, and Common Sense  • 

139

different circumstances. Given this robustness, these conclusions will not be upset should some single study prove to be mistaken, much as one must acknowledge that scientific work of any sort is always subject to the possibility of correction in the light of future work. We must follow the evidence where it leads, while simultaneously acknowledging that new evidence may always lead us to conclusions we had not anticipated. My intention here is to give an accurate, if fairly broad-​brush, account of what our best current theories suggest.14

5.3 Here I examine two different ideas about how the appearance of contradiction between the scientific image and the manifest image of doxastic deliberation might be resolved. The first focuses on the deliberator’s impression of standing outside of the processes by which beliefs are influenced and produced, and the apparent ability, under conditions of deliberation, to intervene in these processes. Some will argue that my account of the first-​person perspective here is overly metaphysical, and that the first-​person perspective should be understood far more innocently than I have suggested. The second idea involves the suggestion that I  have failed to distinguish between the reasons for which a person holds a belief, and the mere causes of believing. Once one appreciates the force of this distinction, it will be argued, the appearance of conflict between the scientific image and the manifest image of deliberation disappears. Let us examine each of these suggestions in turn. Korsgaard argues that there is no conflict between the scientific picture of deliberation and the view one has in the act of deliberation, at least when it comes to questions about freedom. The freedom discovered in reflection is not a theoretical property which can also be seen by scientists considering the agent’s deliberations third-​personally and from the outside. It is from within the deliberative perspective that we see our desires as providing suggestions which we may take or leave. You will say that this means that our freedom is not “real” only if you have defined the “real” as what can be identified by scientists looking at things third-​personally and from outside. (Korsgaard 1996: 96)

14. John Doris (2015: 44–​49) confronts this very worry in greater detail.

410

1 4 0   • H i l a r y

Kornblith

Now when I  described how deliberation appears from the perspective of the deliberator, I purposely avoided talking about freedom. I did, however, following Korsgaard, speak of how deliberators feel at a distance from their beliefs or desires, and how it seems to them that they may intervene in the mechanistic processes by which their beliefs or their actions are produced. I  avoided talking about freedom here because the notion of free will is so controversial, and, depending on whether one is a compatibilist or an incompatibilist about free will, one may have very different ideas about the bearing of scientific descriptions of the deliberative process on questions about the agency of the deliberator. I do not believe that the upshot of these investigations is a matter of definition, let alone a matter of how one defines the term “real.” But it is better to avoid these issues entirely by sidestepping questions about freedom. We may ask, instead, as I did, whether deliberators are able to stand outside the mechanistic processes that would otherwise produce their beliefs and actions, as it seems from the first-​person perspective, or, instead, whether that is impossible, as the scientific account of deliberation would have it. Korsgaard argues that we may see how her preferred account of deliberation is insulated from the possibility of scientific disconfirmation if we understand just what freedom amounts to. . . . we do not need the concept of “freedom” in the first instance because it is required for giving scientific explanations of what people do, but rather to describe the condition in which we find ourselves when we reflect on what we do. But that doesn’t mean that I am claiming that our experience of our freedom is scientifically inexplicable. I am claiming that it is to be explained in terms of the structure of reflective consciousness, not as the (possibly delusory) perception of a theoretical or metaphysical property of the self. (Korsgaard 1996: 96–​97) But even if we allow Korsgaard to insulate talk of freedom from conflict with scientific results in this way,15 we may still surely ask about whether deliberators genuinely do stand outside of the mechanistic processes by which

15. Of course, I don’t believe that we should allow this. If the claim that we are free is meant, as Korsgaard says, to “describe the condition in which we find ourselves when we reflect on what to do,” then the question must inevitably arise as to whether this description is correct. Merely insisting that talk of freedom does not involve a theoretical or metaphysical property does not guarantee that our attempt to describe our situation is automatically correct.

41

Philosophy, Science, and Common Sense  • 

141

their beliefs and actions are produced, intervening in the causal processes that result in action and belief, as it seems from the deliberative perspective. The apparent conflict between a scientific account of deliberation and the first-​ person perspective of the deliberator thus remains. One might, instead, attempt to deny that there is any real conflict between the perspective of the deliberator and the scientific account of deliberation by drawing on the distinction between the reasons for which a person holds a belief and the (mere) causes of believing. As has long been pointed out, not every cause of a belief is a reason for belief, even when the cause is itself a belief. If John has been so traumatized by his experiences in war that the mere mention of war causes him to believe that he is in danger, this does not show that his belief that the word “war” has been mentioned is a reason for his belief that he is in danger. The point here is not that this is not a good reason for the belief; it is not any kind of reason at all. Everyone will need to make the distinction between reasons and mere causes, but with this distinction in place, it may seem that one is well on the way to disarming the apparent conflict between the scientific account of deliberation and the view of deliberation that is presented by the first-​person perspective. The scientific account, it may be said, tells us what the causes of our beliefs happen to be; the first-​person perspective, on the other hand, offers us a view of our reasons for belief. The two accounts do not conflict, on this way of thinking about it, because they are accounts of different subject matters. It is only by ignoring the distinction between reasons and causes that the two perspectives even seem to conflict. The first thing to point out about this move is that, even if it were to succeed in addressing some of the apparent conflict between the scientific perspective and the first-​person perspective, it cannot possibly address all of the conflict. When we deliberate, we not only seem to be acquainted with our reasons for belief; we also seem to know a great deal about the causes of our beliefs. When deliberating about whether to go on holding an existing belief, subjects typically come to have quite strong convictions about how those beliefs came about (Nisbett and Wilson 1977). Subjects have views, under conditions of deliberation, about what factors have influenced them, and what factors have not. The experimental work on deliberation tells us that many of these convictions are mistaken. None of this need concern reasons for belief; these are beliefs about the causes of our beliefs. Similarly, when subjects deliberate about whether to take on a new belief, they also form judgments about the effects of their deliberation on belief acquisition and retention. They have views about which factors are causally

412

1 4 2   • H i l a r y

Kornblith

relevant in producing their new beliefs and which are not. Again, these are claims about the causes of belief, and thus clearly addressed to the very same subject matter as the empirical investigations of deliberation. The first-​person perspective on deliberation, though it does, to be sure, include views about a subject’s reasons for belief, involves commitment to various claims about the causes of belief (and action) as well. One cannot avoid a clash between the scientific and the first-​person perspective by insisting that they concern different subject matters. It is worth pointing out, as well, that this sort of move cannot succeed even apart from these points. Issues about reasons for belief are not nearly so independent from causal matters as this attempt at reconciliation would have it. The claim that a subject holds a belief for a certain reason can be undermined by various discoveries about its causal origins, as the reaction of subjects to the various experimental results reported here reveals. We find the results of work on implicit bias (e.g., Pronin et al. 2002) so disturbing, for example, not only because it reveals facts about the causal source of our beliefs about which we were previously mistaken, but because it clearly shows that our views about the reasons for which we held our beliefs were mistaken as well. Even in cases where deliberators form beliefs about their own reasons, these claims are not immune to conflict with claims about the causes of those beliefs. Attempts to immunize the first-​person perspective on deliberation from the possibility of conflict with the scientific perspective are thus uniformly unsuccessful.

5.4 Since the conflict between the scientific and the manifest image of deliberation cannot be avoided, how should we proceed? Should we reject the scientific image and simply endorse the manifest image? Or should we, instead, endorse the scientific image and reject the manifest image? Or, perhaps, finally, should we reject them both and adopt some third alternative? It is useful, I  believe, to consider the many ways in which scientific views have, as science has advanced, regularly come into conflict with common sense. The history of science is full of discoveries that were initially resisted or rejected precisely because they clashed, in so striking a manner, with antecedently held commonsensical views. The fact that the earth circles the sun rather than vice versa; the germ theory of disease; the fact that human beings are descended from other primates—​all of these discoveries faced substantial initial opposition as a result of their conflict

413

Philosophy, Science, and Common Sense  • 

143

with what was regarded at the time as common sense. It is characteristic of science, of course, that it regularly comes into conflict with antecedently held views, many of which are deeply ingrained, and many of which, as well, carry substantial emotional attachments. It is one of the wonderful benefits of a scientific understanding of the world that it can revise and correct such views. At this late date in the history of scientific investigation, no one can reasonably suppose that when science conflicts with antecedently held common-​sense, common sense deserves any special deference, let alone that science should bow to common sense in the face of such conflicts. Such an approach would amount to a simple rejection of science in favor of ignorance and superstition. If the features of the manifest image that conflict with science are not simply to be rejected as mistake or illusion, some special feature of that image must be identified that distinguishes it in relevant ways from other prescientific views. In the case at hand, our understanding of deliberation, it is difficult to see what such a feature might be. As the cognitive sciences have progressed, we have made many remarkable discoveries about the ways in which our minds work. Many of these pose no special threat to the manifest image. There are other cases that undermine some easily overturned bit of prior common sense, and there are still others that threaten more central features of the manifest image itself. Deliberation, it seems, may be an intermediate case. The discoveries that have been made about deliberation certainly show it to be quite different than we had antecedently thought, but nothing here threatens to overturn our conception of ourselves as subjects with beliefs and desires, for example, as some16 have suggested a scientific account of cognition, or of the brain, might lead to. I have no sympathy whatever for eliminationist approaches to the mind, but this is not because I think that eliminationism is in principle untenable. Once one allows that what we should believe about the world around us, and about ourselves, is simply determined by the best available empirical theory, there is just no saying in advance where scientific progress may lead us. Small changes in antecedently existing common sense are allowed by everyone; larger changes that overturn important features of our self-​conception, as were the product, for example, of Darwinism, are very much of a piece. Some of the changes that science forces on us are easily accepted, while others

16. The Churchlands have, of course, been a driving force here. Classics of eliminationism include Churchland (1981) and Churchland (1989).

41

1 4 4   • H i l a r y

Kornblith

are more threatening, but none of this, it seems, has anything whatever to do with the epistemic standing of science. In the face of conflicts between science and common sense, science must inevitably win out. When the changes that scientific progress leads to are emotionally difficult to accept, it seems we just need to get used to them, and threats to the manifest image, whether small or large, are exactly of that sort. The conflict between the scientific image and the manifest image of deliberation may thus only be resolved by recognizing that in this case as in so many others, science has allowed us to revise and correct our earlier views. Just how deeply this will cut into the manifest image remains to be seen as our scientific understanding of the mind progresses.17

5.5 I have chosen doxastic deliberation as a case study here for a number of reasons. It nicely illustrates the ways in which the scientific image and the manifest image may come into conflict. It shows as well, as I  have argued, how the manifest image must give way in such cases. Attempts to deny the incompatibility of the scientific and manifest images, or to assign some special weight to prescientific views, are simply misguided. In cases of conflict, such as this one, the manifest image—​our prescientific view of ourselves—​must simply be abandoned in the face of scientific progress. But where does all of this leave philosophical thinking? Epistemologists who are concerned with the nature of knowledge and justification have often—​ perhaps typically—​ begun their theorizing by 18 focusing on doxastic deliberation. We see this not only in Descartes’s Meditations and Chisholm’s Theory of Knowledge (1989); it is, to be sure, an utterly natural way of raising epistemological questions. More than this, when epistemologists think about such deliberation, they tend to think about it from the first-​person perspective. This does far more than allow

17. I have focused here on conflicts between the scientific image and the manifest image because this is the issue that arises with respect to doxastic deliberation, but I  recognize that the position I  articulated at the beginning of this chapter is broader than that:  I reject any features of the manifest image that are not themselves part of the scientific image. I endorse this stronger view simply because I reject the idea that there are any extra-​scientific sources of evidence about the world: Accepting features of the manifest image that are not a part of the scientific image involves going beyond our available evidence. I have discussed the reasons for this in more detail in Kornblith (1994). 18. Ernest Sosa (2015) makes much of this.

415

Philosophy, Science, and Common Sense  • 

145

us to raise the question, “What should I  believe?” It has us address that question by taking the first-​person perspective on deliberation at face value. Instead of raising the question of what actually goes on when we deliberate, this approach simply assumes that the first-​person perspective gives an accurate picture of the deliberative process. If what I have argued here is even roughly right, then such an approach is radically mistaken. It puts epistemological theorizing on the wrong path from the very first step. We are interested in understanding a certain phenomenon—​the process of deliberation—​because we believe that it might help to illuminate the broader issues of justification and knowledge. All of that is well and good, but when the phenomenon we seek to understand is one that has been studied extensively by the cognitive sciences, and when scientific advances have begun to displace our prescientific understanding of some phenomenon, we cannot simply turn our backs on the ways in which that understanding has supplanted our common-​sense views. A philosophical methodology that would have us study these phenomena from the armchair, however, would do precisely that. One important lesson to be learned here, then, is that we must allow our philosophical theorizing to be informed by our best available scientific theories. An armchair approach to philosophical theorizing flies in the face of that important lesson. One might worry that the views defended here are somehow self-​ undermining.19 The thought goes something like this. I  have presented arguments for the view that the results of reflection cannot simply be accepted at face value, and yet, in presenting these arguments, I  hope to convince readers to accept my conclusion on the basis of the very sort of reflection I seek to challenge. This suggestion, however, misunderstands the position I seek to defend here. I have argued that reflection, as it presents itself from the first-​person perspective, gives a misleading and often inaccurate view of the very processes that go on in us when we stop to reflect. Doxastic deliberation, I  have argued, is, in fact, a much more complicated process than it appears to be from the first-​person perspective. The arguments I offer for this conclusion are based on experimental evidence, however, rather than first-​person reports alone. And I  have argued that when there is a conflict between the first-​person perspective and the best available theorizing based on experimental results, we need to go with the

19.  This worry has been suggested to me by the editors of this volume, and also in Tweedt (2015).

416

1 4 6   • H i l a r y

Kornblith

empirical theorizing rather than our own first-​person judgments unaided by experimental input. There is no contradiction in such a position, nor is it in any way self-​undermining. The position I have argued for is not that any second-​order belief is automatically suspect. Rather, I have argued that beliefs about the mental states and processes involved in deliberation derived exclusively from our first-​person perspective on them, independent of experimental input, are suspect. The arguments I have presented do not encourage the reader to form conclusions on the basis of the very processes whose reliability I challenge, since these arguments depend on experimental input in precisely the way I argue provides the needed corrective to unaided first-​personal investigations. The kinds of questions that philosophers ask are often quite different from those that occupy scientists qua scientists. Many of the questions philosophers ask, for example, are normative questions, even in epistemology, while much of the focus of psychological theorizing is more straightforwardly descriptive. But this does not make scientific work irrelevant to philosophical thought. Our normative questions, after all, concern various phenomena in the world. In the case of epistemology, we are interested, inter alia, in a variety of normative questions about doxastic deliberation, about belief, about knowledge, about memory, and so on. Psychologists have studied these phenomena extensively. We cannot expect to understand the normative issues concerning these phenomena if we have a mistaken view about what these phenomena amount to. Only science can provide the accurate understanding that is needed as input to philosophical theorizing. The relationship between scientific investigation and philosophical theorizing, however, is not so simple, I believe, as science providing the understanding of the phenomena about which philosophy has its own distinctive questions. This way of conceiving of the relationship between science and philosophy makes it sound as if the philosophical questions themselves remain unchanged in the face of the improved understanding of various phenomena that science may offer. It should be clear, however, that the questions that philosophy may have about various phenomena may themselves need to be revised in the face of better understanding of the phenomena themselves. When our views about what goes on when we deliberate are revised in the light of work in the cognitive sciences, the philosophical questions we have about deliberation are inevitably changed as well. Science not only helps us to address the philosophical questions that we had before we became acquainted with scientific advances; it helps us to revise the philosophical questions we ask in light of the better understanding of various phenomena that science provides.

417

Philosophy, Science, and Common Sense  • 

147

This way of viewing the relationship between philosophy and science is not, I recognize, uncontroversial. If the case study of deliberation that I have chosen here is at all representative, however, this approach has much to recommend it.

References Boghossian, Paul. 2016. “Reasoning and Reflection:  A Reply to Kornblith.” Analysis 76(1): 41–​54. Chisholm, Roderick. 1989. Theory of Knowledge, 3rd ed. Upper Saddle River, NJ: Prentice Hall. Churchland, Patricia. 1989. Neurophilosophy. Cambridge, MA: MIT Press. Churchland, Paul. 1981. “Eliminative Materialism and the Propositional Attitudes.” Journal of Philosophy 78: 67–​90. Doris, John. 2015. Talking to Our Selves:  Reflection, Ignorance, and Agency. Oxford: Oxford University Press. Evans, Jonathan St.B.T. 2010. Thinking Twice: Two Minds in One Brain. Oxford: Oxford University Press. Evans, Jonathan St.B.T. and Keith Frankish (eds). 2009. In Two Minds: Dual Processes and Beyond. Oxford: Oxford University Press. Fodor, Jerry. 1983. The Modularity of Mind. Cambridge, MA: MIT Press. Frankfurt, Harry. 1988. “Freedom of the Will and the Concept of a Person.” In The Importance of What We Care About, 11–​24. Cambridge: Cambridge University Press. Halberstadt, Jamin and Timothy Wilson. 2008. “Reflections on Conscious Reflection: Mechanisms of Impairment by Reasons Analysis.” In Reasoning: Studies of Human Inference and its Foundations, edited by Jonathan Adler and Lance Rips, 548–​565. Cambridge: Cambridge University Press. Kahneman, Daniel. 2011. Thinking, Fast and Slow. New York: Farrar, Straus and Giroux. Kornblith, Hilary. 1993. Inductive Inference and its Natural Ground. Cambridge, MA: MIT Press. Kornblith, Hilary. 1994. “Naturalism:  Both Metaphysical and Epistemological.” Midwest Studies in Philosophy 19: 39–​52. Kornblith, Hilary. 1998. “Introspection and Misdirection.” Australasian Journal of Philosophy 76: 48–​60. Kornblith, Hilary. 2002. Knowledge and Its Place in Nature. Oxford:  Oxford University Press. Kornblith, Hilary. 2012. On Reflection. Oxford: Oxford University Press. Korsgaard, Christine. 1996. The Sources of Normativity. Cambridge:  Cambridge University Press. Kunda, Ziva. 1999. Social Cognition:  Making Sense of People. Cambridge, MA: MIT Press.

418

1 4 8   • H i l a r y

Kornblith

Langer, Ellen J. 1975. “The Illusion of Control.” Journal of Personality and Social Psychology 32: 311–​328. Lord, Charles, Lee Ross, and Mark Lepper. 1979. “Biased Assimilation and Attitude Polarization: The Effects of Prior Theories on Subsequently Considered Evidence.” Journal of Personality and Social Psychology 34: 2098–​2109. Miller, George. 1975. “The Magical Number Seven Plus or Minus Two: Some Limits on our Capacity for Processing Information.” In The Psychology of Communication, 14–​4 4. New York: Basic Books. Moran, Richard. 2001. Authority and Estrangement:  An Essay on Self-​Knowledge. Princeton, NJ: Princeton University Press. Nagel, Thomas. 1979. Mortal Questions. Cambridge: Cambridge University Press. Nagel, Thomas. 1986. The View from Nowhere. New York: Oxford University Press. Nisbett, Richard and Timothy Wilson. 1977. “Telling More than We Can Know: Verbal Reports on Mental Processes.” Psychological Review 84: 231–​259. Open Science Collaboration. 2015. “Estimating the Reproducibility of Psychological Science.” Science 349: aac4716. Pronin, Emily, Daniel Lin, and Lee Ross. 2002. “The Bias Blind Spot: Perception of Bias in Self versus Others.” Personality and Social Psychology Bulletin 28: 369–​381. Sellars, Wilfrid. 1963. “Philosophy and the Scientific Image of Man.” In Frontiers of Science and Philosophy, edited by Robert Colodny, 35–​ 78. Pittsburgh, PA: University of Pittsburgh Press. Sosa, Ernest. 1991. “Knowledge and Intellectual Virtue.” In Knowledge in Perspective:  Selected Essays in Epistemology, 225–​244. Cambridge:  Cambridge University Press. Sosa, Ernest. 2015. Judgment and Agency. Oxford: Oxford University Press. Strawson, Peter. 2008. “Freedom and Resentment.” In Freedom and Resentment and Other Essays, 1–​28. London: Routledge. Tweedt, Chris. 2015. “Review of Hilary Kornblith, On Reflection.” Journal of Moral Philosophy 12(5): 656–​659.

419

6

I S S C I E N T I S M E P I S T E M I C A L LY V I C I O U S ?

Ian James Kidd

6.1 Introduction Many critics agree that scientism is a bad thing and there is no shortage of proposals for why different forms of scientism are objectionable.1 John Dupré has argued that inflated estimations of the explanatory scope of science prevent us from adopting a properly “pluralistic view of epistemic excellence” (2001: 115). Bernard Williams judged that an overactive zeal for science was liable to encourage “misunderstanding of the relations between philosophy and the natural sciences” (2006: 182). Other philosophical critics of scientism, such as Susan Haack (2000) and Mary Midgley (2001), offer other possibilities.2 These forms of philosophical anti-​ scientism can be grounded in a very diverse range of concerns—​ philosophical and practical, social and spiritual—​which is itself a reflection of the diversity of forms of scientism, identified by scholars such as Rik Peels (this volume) and Mikael Stenmark (2001). My concern in this chapter is a very general style of criticism of scientism and not the more specific ones just described. Specifically, I explore the criticism that scientism is epistemically vicious. An epistemic vice is a negative epistemic character trait such as dogmatism or intolerance, to be contrasted with epistemic virtues—​those praiseworthy “excellences” of epistemic character such as curiosity

1. A recent set of papers exploring this question is Clarke and Walsh (2011), Kidd (2011), Maki (2011), and Clarke and Walsh (2013). 2. I give a fuller account of Midgley’s anti-​scientism in Kidd (2015).

510

1 5 0   • I a n

James Kidd

or humility.3 Many critics of scientism have naturally adopted a vocabulary of vices—​they speak of the “arrogance” of the typical scientistic “assumption of the absolute superiority of the scientific method” (Dupré 2001: 116), of the “dogmatic” confidence that science can successfully take over the tasks of philosophy and other humanistic disciplines (Midgley 2001: 19), and of the “envy” of science that Susan Haack attributes to scientistic philosophers (2000: 200). Alongside such explicit charges are more implicit appeals to the perceived viciousness of scientism—​the economist F. A. Hayek’s criticism of scientistic economists for their “desire to organise the work of others” by rigidly imposing scientific methodologies upon them (1952: 123) or the philosopher of science Paul Feyerabend’s castigation of the “frozen mental attitude” of those who succumb to exaggerated “myths” about science (1993: 167). The practice of charging scientism with epistemic vices is important for at least three reasons. The first is that it is a local instance of the more gen­ eral critical practice of charging individual or collective epistemic agents with vices—​one that is widely used in modern cultures and which I  call “vice-​charging” (Kidd 2016). Although philosophers have largely neglected it, this and other forms of “agent-​based appraisal” are robustly defended by argumentation theorists, typically by appeal to virtue epistemology.4 This practice can be and often is directed at all sorts of philosophical and political stances and convictions but the focus of this chapter, as of this volume, is upon scientism—​so, for instance, many religious or philosophical stances may be vulnerable to vice charges.5 Second, recent commentators have argued that scientism encourages a perception of scientists as being vicious that contributes to an erosion of the authority of science. Philip Kitcher recently warned that challenges to the authority of science in modern societies is partly due to a culturally pervasive scientism that promotes an “image of scientists as over-​ambitious and arrogant” (2011: 15). And the third reason for taking seriously charges of vice against scientism is that it resonates within a long tradition that invests the authority of science in the virtues or qualities of its practitioners—​an enduring perception of relations between the “authority of knowledge and the character of knowers” that shapes normative conceptions of the “scientific 3. Two excellent studies in vice epistemology are Battaly (2014) and Cassam (2016). 4. “Agent-​based appraisal” is defended by Aberdein (2014), Battaly (2010), and Kidd (2016). 5. Cooper (2002) argues that forms of philosophical humanism and absolutism (such as scientific realism) are guilty of the vice of “hubris,” while charges of arrogance and other epistemic vices against doctrines of religious exclusivism are very common.

15

Is Scientism Epistemically Vicious?  • 

151

self ” (Shapin 2008: xvi). If so, then charges of vice are not merely rhetorical, but can go deep.

6.2 Scientism, Stance, and Vice I start by offering a characterization of scientism as a stance in the sense described by Bas van Fraassen—​an “attitude, commitment, approach, a cluster of such” that may also include certain beliefs (2002: 47). In his book, The Empirical Stance, Van Fraassen uses it to characterize empiricism, which cannot, he argues, consist of a metaphysical position adopted on the basis of something other than experience: that violates the empiricists’ reliance on experience. Nor can it consist in an unwavering acceptance of certain empirical propositions:  that violates the empiricists’ commitment to the contestability of all empirical claims. Van Fraassen argues that it is therefore better to understand empiricism as a stance—​a cluster of commitments and dispositions that guides inquiry, characterized by, for instance, commitments to “calling us back to experience” and an attitude of “admiration for science” (2002: 47).6 I will characterize scientism as a stance, taking my cue from James Ladyman (2011), but with two provisos. The first is that there is a plurality of scientistic stances, rather than one single scientistic stance, and they differ according to their components. Ladyman characterizes his scientistic stance as incorporating an attitude of “hostility” to nonnaturalistic metaphysics, but also of respect for its naturalized variants (2011: 97). But that attitude of tempered respect is absent from other scientistic stances whose attitudes to metaphysics, or to philosophy tout court, are ones of hostility, scorn, or derision—​as in Steven Hawking and Leonard Mlodinow’s (2010) recent judgment that philosophy is “dead” because it has not “kept up” with scientific developments.7 The second proviso is that a scientistic stance can change its components over time in response to different sorts of factors. Perhaps its attitudes toward philosophy harden or its commitment to secular values becomes more pronounced. So stances are plural and changeable, even if some are more conservative and resistant to change than others—​a point explored by Ratcliffe (2011) and Rowbottom and Bueno (2011).

6. The ontology of stances is explored further by Rowbottom and Bueno (2011) and Teller (2004). 7. I criticize Hawking’s judgment in Kidd (2011).

512

1 5 2   • I a n

James Kidd

My suggestion is that scientistic stances can  incorporate attitudes and dispositions that are epistemically vicious, with the consequence that an agent who adopts that stance will be led to conduct themselves epistemically in recognizably vicious ways. Stances might be vicious in two related ways. First, if a stance is vicious then one must be at least latently or incipiently vicious to find it an attractive or compelling option. The availability of certain types of stance could encourage certain vicious dispositions by providing examples of people who adopt that stance, exemplifying those ways of thinking, feeling, and acting. If those people are authoritative and enjoy markers of esteem, then that can, however inadvertently, serve to legitimate the vicious stance. Second, a stance is vicious if the price of adopting and employing it is that one is led to think, feel, and act in vicious ways—​a stance is vicious if one must be vicious to adopt it. The vices need not be epistemic, of course: Racism is an example of an ethically vicious stance and snobbery of an aesthetically vicious stance.8 The real test of the vicious nature of a stance will be careful scrutiny of the epistemic conduct of agents who adopt those stances—​of the attitudes and beliefs and commitments of scientistic persons.9 An abstract analysis of the vicious components of a stance is not in itself enough to complete a charge of vice—​one needs to give instances of vicious conduct that can be attributed to the stance, for reasons that I lay out in Kidd (2016). I will focus my argument on the specific vice of closed-​mindedness and say more about vices that might feature in scientistic stances at the end of the chapter.

6.3 Being Closed to Epistemic Possibilities A persistent complaint about those who adopt scientistic stances is that they are closed to epistemic possibilities—​that they are guilty of being closed-​minded. Such charges of the vice of closed-​mindedness pose two questions:  First, to what ought a person be “open,” and second, what does “being open” involve or require of the virtuous agent? I suggest that the adoption of scientistic stances can lead to a person becoming “closed” off to at three types of epistemic possibility.

8. The idea of “immoral beliefs” is explored by Jones and Martin (2004). The claim that snobbery is aesthetically vicious is made by Kieran (2009). 9.  I  show the psychological and contextual complexity of imputations of some epistemic virtues and vices in Kidd (2014) and (2017).

513

Is Scientism Epistemically Vicious?  • 

153

First, a scientistic person may be closed off to the possibility that they are closed off to certain epistemic possibilities—​a self-​reflexive failure to recognize certain features of their own stance. Such closure might manifest in an obliviousness to prejudices or biases in their choices and appeals that are obvious to others. The chemist and science writer Peter Atkins (2011) is self-​attributively scientistic and advocates a thoroughly naturalistic picture of the world. But in a recent book he refers to only three other scholars, each of whom is deeply sympathetic to his claims about science, religion, and society (these being, aside from himself, Richard Dawkins and Daniel Dennett). No attempt is made to engage with a wider constituency of scholars from relevant disciplines, such as history of religion or philosophy of science, despite the strong claims made in the book about—​among other things—​the relationship of religion and science. So a scientistic person may be closed off to certain possibilities about their own stance and more general epistemic conduct and commitment. Second, a scientistic person may be closed to the possibility that they might engage productively with inquirers from other disciplines or traditions. This second sort of closure involves being closed to intersubjective possibilities involving other persons, including professionally and institutionally accredited peers. The mathematician Norman Levitt—​a robust participant in the “Science Wars” of the 1990s—​advocates a robustly scientistic picture of the world in his book Prometheus Bedevilled. But throughout that book he castigates science studies scholars for their “effusions,” “faddishness,” and “ideological obsessions” (1999: 29, 153, 299). Such remarks seem to reflect a closure to the possibility of engaging epistemically with those scholars. Certainly castigation is a poor preparation for collaboration. But the complexities of a topic like the place of science in modern intellectual and social culture mean that anyone who wants to engage with it in a thorough and informed way must rely on others. A stance that incorporates attitudes of scorn or hostility that militate against the possibility of collaboration with other scholars whose expertise one needs is a bad stance; and even if one did wish to dispute that need, there are better ways to do it than derogation. Third, a scientistic person may be closed to the possibility of there being forms and sources of knowledge, evidence, inquiry, or reason that are not scientific in character. The closure here is not to inquirers, but to epistemologies and so reflects the forms of epistemological scientism described by Peels in his contribution to this volume. It involves being closed to the possibility that our epistemic activities and values are not all legitimately describable as scientific

514

1 5 4   • I a n

James Kidd

and so—​as one critic puts it—​“grants science [a]‌monopoly over what counts as evidence, correct method, and knowledge” (Maffie 1999: 12). But this sort of closure is doubly objectionable. One is it involves an absurdly elastic application of the term “science” to a diverse and heterogeneous range of epistemic activities. Another is that it reflects a self-​serving evaluative policy of describing as “scientific” anything that “works,” a scientistic tendency noted by Dupré (2001: 113–​114f ). Much of our knowledge comes from testimony, memory, and reflection, but classifying these as scientific extends that term beyond all sense. A scientific stance can consist of attitudes, commitments, and beliefs that closes an agent off to epistemic possibilities in a way that impairs them epistemically. The stances in question may differ according to the sorts of possibilities they close one off to and the ways in which such closure manifests. It may be strongly expressed scorn that puts other inquirers off or it may be a passive reluctance to engage with certain inquirers or traditions. Some advocates of scientism seem to go out of their way to derogate disciplines they regard as inferior to science and make an art of their abuse. Others simply press on with large claims about the history and nature of science, but without citing or consulting historical or philosophical scholarship. Such differences in the composition of scientistic stances, no doubt, reflect the complexities of individual psychology and the contingencies of epistemic socialization.10 Two final points about differential viciousness of scientistic stances. The first is that such differences arise from the particular components of any given stance, such that certain attitudes or beliefs may be more liable to generate closedness than others. The second is that the dispositions to epistemic closure will affect the probably of a stance inducing or encouraging other sorts of vicious disposition. Vices like arrogance and dogmatism also plausibly reflect ways being “closed off ” to inquirers, ideas, and facts about oneself. A vicious character is a peculiarly closed character.11 To continue my analysis of scientistic closed-​mindedness, let me move to a positive account of its corresponding virtue—​open-​mindedness.

10. The complex and difficult topic of agential and collective responsibility for epistemic vice is discussed by Medina (2012: ch. 4) and Battaly (2016). 11.  Tiberius and Walker (1998) argue that arrogance prevents a person from learning or appreciating certain facts about themselves and their relations with others.

51

Is Scientism Epistemically Vicious?  • 

155

6.4 Baehr on Open-​Mindedness Jason Baehr (2011) offers a sophisticated account of the virtue of open-​ mindedness. Its conceptual core of that virtue is that the open-​minded person “departs or detaches from [or] moves beyond or transcends, a certain default . . . cognitive standpoint” (2011: 149, original emphasis). A cognitive standpoint is practically identical to a stance and so my discussion will use Baehr’s account of the virtue rather than others proposed.12 But what does it mean to “depart from” or “transcend” a stance? Baehr offers three accounts of what detachment from a default stance can involve. The first is that an agent can entertain or consider an alternative stance, one different to their default and perhaps which they have not considered before or been in the habit of employing. The agent is “open” to the possibility of adopting other stances. The second is that the agent is able to shift between their default stance and a number of alternatives so as to compare and contrast their merits, an activity that may provoke reconsideration of the merits of the default stance—​an openness, therefore, to the possibility of reappraising one’s stance. The third way to understand “detachment” is that an agent might restrain or resist the temptation to make rapid or reactive judgments about the alternative stances they are considering—​they “keep an open mind” by avoiding prejudgment or premature termination of the activity of appraisal. Baehr usefully observes that this third mode of detachment differs from the others due to its negative character, showing that “while open-​mindedness is often a matter of positively opening one’s mind, it is sometimes a matter of not closing it” (2011: 151).13 This account of virtuous open-​mindedness is consistent with many of our everyday ideas and examples about the epistemic conduct of the open-​ minded person. Such a person can “wrap her mind” around new ideas rather than being locked into a single stance—​a fixed and partial way of looking at things. She can “try on” unfamiliar ideas without dismissing them outright because she recognizes that a sense of strangeness may be due to uncritical reliance on her familiar stance. She may be able to detach from her stance in order to “take up” a very different way of perceiving a situation—​a willingness not to prematurely reject alternative stances. It should be clear that this detachment from a default stance and “engagement” with others is not always

12. Other accounts of open-​mindedness are offered by Hare (1985) and Riggs (2010). 13. A more detailed discussion of voluntarism and pluralism about stances can be found in Chakravartty (2011) and Ratcliffe (2011).

516

1 5 6   • I a n

James Kidd

going to be easy. Indeed, it will often be difficult and demanding, especially in cultures that do not encourage openness to alternative possibilities. But this underscore the important point that motivation is crucial to genuinely virtuous open-​mindedness. The agent’s openness may be motivated by a desire to acquire truths, or to do justice to a problem, or to give a strange-​seeming claim a fair hearing—​and so on. By contrast, a person who willingly detaches from their default stance to take up others because they are slavish or politically correct or wish to impress other people lacks the motivation constitutive of virtuous open-​mindedness (see Baehr 2011: 152, n. 14). Consider an example of open-​mindedness as detachment from one’s default stance—​specifically that of the historian of ideas and political theorist, Isaiah Berlin. A biographer writes that Berlin had “the ability to enter into beliefs, feelings, and attitudes alien and at time acutely antipathetic to his own” (Ignatieff 1998: 256). The “alien” ways of thinking in question were the totalitarian and fascist political stances that were starkly in contrast to the pluralistic liberalism defended by Berlin. Translated into Baehr’s terms, Berlin could detach from his default liberal stance on social and political organization in order to “take up” the diametrically opposed stances of totalitarian theorists. Moreover, he did this with the good epistemic motivation of wanting to maximize his critical capacities—​an “imaginative insight” into the “ways of thinking, feeling, imagining” of those who are historically, culturally, or morally distant from us, as Berlin put it (2013: 148). Contrast this with an example of the closed-​mindedness imputed to the Victorian anthropologist James Frazer, by Wittgenstein in his caustic Remarks on Frazer’s “Golden Bough.” His critical animus is directed at the “narrowness of spiritual life” evident in the accounts of indigenous cultures and how consequently “impossible” it was for Frazer to “understand a different way of life.” Wittgenstein was particularly critical of Frazer’s assumption that “savage” religious and magical practices should be understood as primitive attempts at scientific explanation (1979: 5, 6). Put in Baehr’s terms, this anthropological stance closed down alternative interpretive possibilities—​for instance, entrenching the assumption that “superstition” must be a form of “bad science” (see Burley 2012: ch. 1). With these examples and remarks in place, I can now offer a full characterization of Baehr’s account of the virtue of open-​mindedness. The open-​minded person is one who is (1) willing and (within limits) able (2)  to disengage from a default cognitive standpoint (3)  such

517

Is Scientism Epistemically Vicious?  • 

157

that they can take up or take seriously the merits of a distinct cognitive standpoint, (4)  in a way that makes a significant demand upon their agency and (5) which requires that they be willing to adjust their beliefs and confidence according to the outcomes of their engagement with those other standpoints. (see Baehr 2011: §8.2.2) Since this account is sufficient for my purposes, I will leave the job of providing further details and defense to Baehr (2011: §8.3–​8.4) and now use it to restate my original question: Can scientistic stances incorporate attitudes, commitments, or beliefs that diminish an agent’s willingness and ability to disengage from that stance, such that they can no longer take up or take seriously the merits of other stances? If they do, then such stances can legitimately be described as closed-​minded. It is quite possible that many non-​or even anti-​scientistic stances could build in such dispositions. Indeed, my sympathies lie with the rather pessimistic conviction that almost any stances could degenerate or be corrupted into a vicious form. But the focus of this chapter is upon the vices of scientistic stances and I have, anyway elsewhere, explored the possibility of other vicious stances (see Kidd 2017).

6.5 Sources of Scientistic Closedness There are at least four components of a scientistic stance that could encourage closed-​mindedness in the sense outlined in the last section. The first is any belief that entails a denial that one occupies anything like a stance on the grounds that there is only one stance—​that of “Reason,” perhaps. Such a belief will generate closed-​mindedness in two related ways. For a start, it will rule out or render unintelligible the very idea that one does or even could occupy a distinctive stance or “default cognitive standpoint”—​ a clear rejection of a core component of the virtue as Baehr described it. Unless one accepts that they are adopting a stance, they cannot even begin to engage with the possibility of describing or disengaging from it. But further, such a belief disables the idea of “detaching from” or “transcending” that stance. Such stances can therefore be described as dogmatically conservative for they rule out the possibility of their ever being changed or

518

1 5 8   • I a n

James Kidd

radically revised. Martin Heidegger described a variety of scientistic stances or “ways of revealing” and judged them to be “monstrous” precisely because they obscured any sense of their being a stance, one possibility among others (see Heidegger 1977: 27f ).14 Second, scientistic stances may incorporate a conviction or belief that there is a plurality of stances, but qualified by a further sense that only one’s own is sensible or plausible. Such stances impair open-​mindedness in two ways. One is that they militate against willingness to disengage from one’s default stance. Another is that they can deny sense to the idea that other stances might have merits to take seriously. An example of such closure can arguably be found in the stances of those who style themselves as “Brights” or “freethinkers” and regard themselves as already occupying a privileged stance, “disengagement” from which would be equivalent to willful lapse into irrationality. Atkins lionizes his scientistic stance as able to “shed light on every and any concept” and contrasts it with alternative religious stances (2011: vii). If one believes this about one’s default stance, then one is hardly likely to be willing to “take up” alternatives. The third possibility is that a scientistic stance may include a willingness to take up other stances, but then constrain the types of alternatives that one will take up. Such stances fulfill more of the conditions of Baehr’s account, insofar as they honor the basic sense of there being other potentially meritorious stances to engage with. But they still fall short of genuine open-​mindedness because they implicitly restrict the range of potential alternative stances and so undermine the principle that to be open to another ought to “make a significant demand upon one’s agency.” An example may be Gregory Dawes’s recent argument that certain epistemic conflicts between science and religion could be resolved, if religious traditions could offer “adequate theistic explanations of a range of phenomena” that would be “preferable to any proposed natural explanations” (2011: 6). Integral to this stance is a conception of explanation as (roughly) providing causal accounts of the “when-​then” type familiar to scientific inquiry. But that stance falls short of open-​mindedness in Baehr’s sense if it makes engagement with religious stances conditional on the style of explanation integral to the default stance. For theistic religious traditions typically employ styles of explanation that are genuinely alternative to the causal scientific style—​teleological or eschatological styles, for instance. Such

14. A rich account of the forms of occlusion and closedness to epistemic possibilities of the “technological” stance described by Heidegger is given in Cooper (1997).

519

Is Scientism Epistemically Vicious?  • 

159

alternative styles will include very different criteria of explanatory adequacy, and this makes it very difficult to rule them out prior to investigation of a sort that would indeed place “significant demand on one’s agency” (see Cooper 2002: 197–​200ff ). A fourth possibility is that a scientistic stance might contain components that lead an agent to wrongly judge that they have disengaged successfully from it, when in fact they have done so only in a partial or imperfect manner. A  stance might perhaps build in a confidence that an agent employing it can easily or transparently determine the success of one’s detachment from it. An example might be Owen Flanagan’s recent proposals for a “naturalized Buddhism” grounded in a “cosmopolitan style of philosophy.” Such a style is a sort of “open, non-​committal” stance toward culturally diverse philosophical traditions that is “ironic or skeptical about all the forms of life being  .  .  .  discussed, including one’s own” (Flanagan 2011:  2). Although the non-​committal irony of the cosmopolitan stance places it in close proximity to Baehr’s account of open-​mindedness, it seems absent from Flanagan’s own remarks on Buddhism. We are told that Buddhism is filled with “mind-​numbing and wishful hocus-​pocus” that ought to be “tamed” if it is to become “worthy of attention by analytic philosophers and scientific naturalists” (Flanagan 2011: xxiii, 3). Throughout his book, Flanagan retains the commitments of his default stance, which include the primacy of science, a naturalistic picture of the world, and the conviction that “scientific and analytic ways of speaking, writing, and arguing  .  .  .  are to be preferred when describing the way things are” (Flanagan 2011: 1–​5 passim). A scientistic stance can build in at least four types of components that encourages the closure to epistemic possibilities that is characteristic of closed-​mindedness. But the question of whether a stance does in fact do so must always be confirmed by sustained analysis of the stance in question and the agents who adopt it.

6.6 Cooperative Vices I have argued that scientistic stances can incorporate dispositions to closed-​ minded epistemic conduct. But an advocate of those stances might respond that the objection misses the mark because scientific inquiry—​whose methods and values are putatively integral to those stances—​is a paragon of open-​mindedness and humility. There is long-​standing debate in the philosophy of science about the extent to which science is actually revisable in the light of evidence and criticism—​for

610

1 6 0   • I a n

James Kidd

instance, given the positive role of dogmatism or the resistance to revision of its guiding Weltbild.15 But given the ongoing status of those debates, it is better to offer a different reply to the humility response, one that distinguishes its two components: the concession and the constraint. The first concession is the explicit acceptance by the agent—​the scientistic inquirer—​that they may be wrong in some of their claims. But the typically implicit constraint is that any criticism or challenge to those claims must proceed on the terms defined by the agent’s default stance. Perhaps a scientistic physicist concedes that their naturalistic picture of the world might be wrong, but would only accept physical scientific evidence to count as a legitimate challenge to that picture. Yet such evidence would be provided by epistemic practices that already presuppose a naturalistic picture of the world and so cannot enable radical criticism of its fundamentals. The constraint therefore implicitly rules out genuine critical possibilities such as engaging in metaphysical inquiry, to which many advocates of scientism are indeed hostile.16 The humility response is therefore suspect. Although the concession ostensibly allows for open-​minded detachment from a default stance, the underlying constraint that is doing implicit and illicit epistemological work actually reinforces it—​for it does not require the agent to detach from it. So, despite appearances, the humility response is not a form of open-​mindedness, and is, in fact, quite the opposite. The fact that scientistic stances cannot be defended by a specific appeal to humility points to the wider possibility that they might incorporate other vicious dispositions. At the start of the chapter I quoted various critics who impute to scientism and its advocates a number of other vices, such as arrogance and dogmatism, a tendency that finds support in the fact that many different vices could close an agent off to  epistemic possibilities. There are two ways to explore this further claim that scientistic stances could be vicious in a richer sense. One would be to take these other candidate vices and give to them the treatment that I  gave to closed-​mindedness—​sketching their structure, locating them within a scientistic stance, and so on. Another would be to consider the question, posed by Jason Baehr, of whether the activity of one vice can “initiate . . . sustain or support” others (2011: 157).17 It is this

15. These two arguments are rehearsed by Kuhn (1962) and Wittgenstein (1969). 16. A good example of an exercise in metaphysics of this sort is Lowe (2011). 17. Actually, Baehr asks this question in the context of virtues, but I see no reason not to ask the same question of vices.

16

Is Scientism Epistemically Vicious?  • 

161

latter claim that vices can initiate or support one another—​calling each other into play—​which I will consider in the remainder of the chapter: call these cooperative vices. The idea of cooperative vices is supported by the claim, familiar to philosophical virtue theory, that the activity of one virtue (or vice) often calls others into play. Since that term is informal, consider two examples that use Baehr’s terms. First, curiosity can inspire a person to begin inquiries or researches that in turn require the exercise of epistemic virtues such as discipline and fair-​mindedness: Curiosity initiates these other virtues—​or, at the least, the practices that express them. Second, a person who exercises the virtue of justice by defending an oppressed group from abuse may be required to also exercise the virtue of courage, because defending the oppressed often opens one up threats: Courage supports justice. These are examples of cooperative virtue, although that term does not refer to any logical entailment or psychological necessity. Nor does it invoke any strong ontological thesis about a “unity of the virtues.” One can imagine cases where curiosity fails to initiate diligence or where justice did not require courage or where certain virtues fail to “pull together” in the expected ways. Instead, the claim is that exercising certain virtues will often invite or require the cooperative exercise of other virtues in ways that are sensitive to contextual and psychological factors, even if if an agent fails to “take up” these invitations or to do what is required. I suggest there can be corresponding cooperative vices and that the structures of certain scientistic stances can encourage cooperation between such vices. The complex and contextual nature of our epistemic life makes it very unlikely that there is some algorithm for calculating which vices cooperate with each other—​a fact underlying Aristotle’s wise remark that ethics is an “inexact science.” But it is possible to identify certain affinities or reciprocities that obtain between different vicious dispositions—​a conceptual closeness or practical proximity which makes initiation or support of one vice by another more likely. In the next, final section of the chapter, I consider two candidate cooperative vices for closed-​mindedness.

6.7 Arrogance and Dogmatism Two vices that are invariably directed at scientism are arrogance and dogmatism. Since both involve being epistemically “closed,” there are good reasons to consider them as plausible candidate cooperative vices of closed-​mindedness.

612

1 6 2   • I a n

James Kidd

My discussion will use the account of these vices developed by Bob Roberts and W. Jay Wood (2007: ch. 9). Roberts and Wood argue that an epistemically arrogant person is disposed to illicitly infer from some excellence or achievement an entitlement to some privilege they do not merit (2007: 77). Imagine a Nobel Prize-​winning physicist who infers from their excellence in cosmology an entitlement to hold forth confidently on the current status of philosophy. But entitlement to offer magisterial overviews of a discipline is justified after long study and acquaintance and not achievement in another—​so the inference is arrogant. Yet many advocates of scientism make precisely these sorts of inferences when they infer from their training and achievement in a scientific discipline an entitlement to advance confident claims about other topics and subjects for which their particular scientific expertise does not equip them. Indeed, the arrogance of such inferences is underscored by the obvious fact that epistemic confidence is earned through disciplined training and practice—​a fact that will be resisted or denied by an agent whose stance is composed of beliefs about the vacuity of philosophy or attitudes of scorn to “nonscientific” disciplines. Steven Shapin nicely illustrates this sort of scientistic arrogance: There’s a story told about a distinguished cardiac surgeon who, about to retire, decided he’d like to take up the history of medicine. He sought out a historian friend and asked her if she had any tips for him. The historian said she’d be happy to help but first asked the surgeon a reciprocal favor: ‘As it happens, I’m about to retire too, and I’m thinking of taking up heart surgery. Do you have any tips for me?’ (Shapin 2015: para. 6) Shapin argues that the integral to the surgeon’s stance is a perception of “asymmetry” between medical and historical practice—​medical practice is highly specialized and its skills difficult to acquire, unlike historical practice, which can be mastered at will. This is closedness to the possibility that historical inquiry is an expert practice. Many of the component attitudes and beliefs integral to scientistic stances might encourage this sense of asymmetry between the sciences and the humanities. Another vice imputed to scientism is dogmatism, defined by Roberts and Wood (2007:  195) as a disposition to respond irrationally to attempts by others at epistemic engagement. A dogmatic person might react to probing questions in a debate by derogating the questioner or adopting an aggressive tone and manner—​a behavioral tendency of “dogmatic people.” Such dogmatic persons need not be hostile to all forms of epistemic engagement, for

613

Is Scientism Epistemically Vicious?  • 

163

many of them obviously relish opportunities to “take on” critics or to “show up” their rivals. But dogmatism does involve an aversion to critical epistemic engagement, roughly ones that are likely to put critical pressure on one’s epistemic commitments or conduct. In such cases the dogmatic person will either make efforts to avoid such engagements or to conduct themselves badly during them. A generally reliable sign of dogmatism is a tendency to derogate those adopting the role of critics, as Levitt does in his talk of the “quasi-​ scholarly” and “ideological” status of science studies scholars. Arrogance, dogmatism, and closed-​mindedness are plausibly likely to initiate and support one another and, if so, would class as cooperative vices. Imagine the case of a person who adopts a stance that is closed to certain types of epistemic possibilities—​a diehard champion of scientism who derogates philosophers of science, say. This person might not be able to reliably avoid encounters with those philosophers and so may at times be required to engage with them—​in public debate, say. During such encounters it may become difficult for that person to maintain their epistemic closure without relying upon illicit entitlements—​to set the terms of debate, say—​or to resort to irrational forms of response. In these situations the vice of close-​mindedness can initiate and support the vices of arrogance and dogmatism, especially if an encounter is particularly prolonged or the mood very charged. Once again, the claim is not that every instance of closed-​minded behavior will trigger arrogant and dogmatic conduct as well—​only that many cases are likely to, the best proof for which is that such vices do tend to cooperate and manifest together in the conduct of those such stances. But it is quite possible that a vicious stance can weaken and that critical engagement may be a cause of this. So it is not inevitable that a weak, localized, or incipient vicious disposition will come to “infect [one’s] whole character” (Baier 1995: 274). The best way to test claims about cooperative vices is to undertake a close study of the conduct and character of those who adopt vicious stances. The cooperativeness of certain vices is likely to depend very much on a stance’s structure and components—​a point anticipated by Ratcliffe (2011: 129) and Rowbottom and Bueno (2011: 9). It is very likely that scientistic stances will be more disposed to viciousness if their component attitudes and commitments are stronger or more rigid and inflexible.

6.8 Conclusions I argued that scientism can be usefully characterized as a stance and that this offers a way of interpreting familiar charges that scientism is epistemically

614

1 6 4   • I a n

James Kidd

vicious. A  stance is vicious if either accepting or expressing its component attitudes or commitments or beliefs will require one to be vicious. I argued that scientistic stances can build in a disposition to be closed to different sorts of epistemic possibility, in a way that makes such stances viciously closed-​ minded. The realization of such dispositions is not inevitable or necessary but will be sensitive to the contingencies of psychology and of context. But there are also affinities between certain vices insofar as the exercise of one can initiate or support others, and these are what I called cooperative vices. The study of such vicious stances would be a profitable project for virtue epistemology and offer new ways to think critically about scientism.18

References Aberdein, Andrew. 2014. “In Defence of Virtue:  The Legitimacy of Agent-​Based Argument Appraisal.” Informal Logic 34(1): 77–​93. Atkins, Peter. 2011. On Being: A Scientist’s Exploration of the Great Questions of Existence. Oxford: Oxford University Press. Baehr, Jason. 2011. The Inquiring Mind: On Intellectual Virtues and Virtue Epistemology. Oxford: Oxford University Press. Baier, Annette. 1995. Moral Prejudices:  Essays on Ethics. Cambridge, MA:  Harvard University Press. Battaly, Heather. 2010. “Attacking Character:  Ad Hominem Argument and Virtue Epistemology.” Informal Logic 30(4): 361–​390. Battaly, Heather. 2014. “Varieties of Epistemic Vice.” In The Ethics of Belief, edited by Jon Matheson and Rico Vitz, 51–​67. Oxford: Oxford University Press. Battaly, Heather. 2016. “Epistemic Virtue and Vice: Reliabilism, Responsibilism, and Personalism.” In Moral and Intellectual Virtues in Western and Chinese Philosophy, edited by Chienkuo Mi, Michael Slote, and Ernest Sosa, 99–120. New York: Routledge. Berlin, Isaiah. 2013. Against the Current: Essays in the History of Ideas. Edited by Henry Hardy, foreword by Mark Lilla. Princeton, NJ: Princeton University Press. Burley, Mikel. 2012. Contemplating Religious Forms of Life. London: A&C Black. Cassam, Quassim. 2016. “Vice Epistemology.” The Monist 99(3): 159–​180. Chakravartty, Anjan. 2011. “A Puzzle About Voluntarism About Rational Epistemic Stances.” Synthese 178: 37–​48.

18. I offer my thanks to Jason Baehr, David E. Cooper, audiences at Durham, Lancaster, and Leeds, and to the editors for very helpful comments and discussion. This research was funded by an Addison Wheeler Fellowship. Since this paper was written, there has been new work on the vice of closed-mindedness by Heather Battaly. Considering its relation to my claims here is a task for another time.

615

Is Scientism Epistemically Vicious?  • 

165

Clarke, Steve and Adrian Walsh. 2011. “Scientific Imperialism and the Proper Relations between the Sciences.” International Studies in the Philosophy of Science 23(2): 195–​207. Clarke, Steve and Adrian Walsh. 2013. “Imperialism, Progress, Developmental Teleology, and Interdisciplinarity.” International Studies in the Philosophy of Science 27(3): 341–​351. Cooper, David E. 1997. “Wittgenstein, Heidegger, and Humility.” Philosophy 72: 105–​123. Cooper, David E. 2002. The Measure of Things:  Humanism, Humility, and Mystery. Oxford: Clarendon Press. Dawes, Gregory. 2011. “In Defense of Naturalism.” International Journal for Philosophy of Religion 70: 3–​25. Dupré, John. 2001. Human Nature and the Limits of Science. Oxford: Clarendon Press. Feyerabend, Paul. 1993. Against Method, 3rd ed. London: New Left Books. Flanagan, Owen. 2011. The Bodhisattva’s Brain:  Buddhism Naturalised. Cambridge, MA: MIT Press. Haack, Susan. 2000. Manifesto of a Passionate Moderate:  Unfashionable Essays. Chicago: University of Chicago Press. Hayek, Friedrich A. 1952. The Counter-​ Revolution in Science. Glencoe, IL:  The Free Press. Heidegger, Martin. 1977. The Questioning Concerning Technology, and Other Essays. Translated by William Lovitt. New York: Harper & Row. Ignatieff, Michael. 1998. Isaiah Berlin: A Life. London: Random House. Jones, Ward E. and Thomas Martin. eds. 2004. Immoral Beliefs, Philosophical Papers 33(3). Hare, William. 1985. In Defence of Open-​mindedness. Montreal: McGill-​Queens University. Hawking, Steven and Leonard Mlodinow. 2010. The Grand Design. London: Bantham Press. Kidd, Ian James. 2011. “Three Cheers for Science and Philosophy! Reflections on Hawking’s The Grand Design.” Think 10(29): 37–​41. Kidd, Ian James. 2014. “Was Sir William Crookes Epistemically Virtuous?” Studies in History and Philosophy of Biological and Biomedical Sciences 48: 67–​74. Kidd, Ian James. 2015. “Doing Science an Injustice.” In Science and the Self: Animals, Evolution, and Ethics: Essays in Honour of Mary Midgley, edited by Ian James Kidd and Elizabeth McKinnell, 151–​167. London: Routledge. Kidd, Ian James. 2016. “Charging Others With Epistemic Vice.” The Monist 99(3): 181–​197. Kidd, Ian James. 2017. “Epistemic Vices in Public Debate: The Case of ‘New Atheism.’” In New Atheism:  Critical Perspectives and Contemporary Debates, edited by Christopher Cotter and Philip Quadrio, 51–​68. Dordrecht: Springer. Kieran, Matthew. 2009. “The Vice of Snobbery:  Aesthetic Knowledge, Justification and Virtue in Art Appreciation.” The Philosophical Quarterly 60: 243–​263.

61

1 6 6   • I a n

James Kidd

Kitcher, Philip. 2012. Science in a Democratic Society. New York: Prometheus Books. Kuhn, Thomas. 1962. The Structure of Scientific Revolutions, 1st ed. Chicago: University of Chicago Press. Ladyman, James. 2011. “The Scientistic Stance: The Empirical and Materialist Stances Reconciled.” Synthese 178: 87–​98. Levitt, Norman. 1999. Prometheus Bedevilled:  Science and the Contradictions of Contempporary Culture. New Brunswick, NJ: Rutgers University Press. Lowe, E.J. 2011. “Naturalism, Imagination, and the Scientific Worldview.” In Turning Images in Philosophy, Science, and Religion: A New Book of Nature, edited by Charles Taliaferro and Jill Evans, 91–​113. Oxford: Oxford University Press. Maffie, James. 1999. “Naturalism, Scientism, and the Independence of Epistemology.” Erkenntnis 43: 1–​27. Mäki, Uskali. 2013. “Scientific Imperialism:  Difficulties in Definition, Identification, and Assessment.” International Studies in Philosophy of Science 27(3): 325–​339. Medina, José. 2013. The Epistemology of Resistance:  Gender and Racial Oppression, Epistemic Injustice, and Resistant Imaginations. Oxford: Oxford University Press. Midgley, Mary. 2001. Science and Poetry. London: Routledge. Peels, Rik. 2018. “A Conceptual Map of Scientism.” This volume. Ratcliffe, Matthew. 2011. “Stance, Feeling and Phenomenology.” Synthese 178: 121–​130. Riggs, Wayne. 2010 “Open-​mindedness.” Metaphilosophy 41: 172–​188. Roberts, Robert C. and W. Jay Woods. 2007. Intellectual Virtues: An Essay in Regulative Epistemology. New York: Oxford University Press. Rowbottom, Darrell P. and Otávio Bueno. 2011. “How to Change It:  Modes of Engagement, Rationality, and Stance Voluntarism.” Synthese 178: 7–​17. Shapin, Steven. 2008. The Scientific Life: A Moral History of a Late Modern Vocation. Chicago: University of Chicago Press. Shapin, Steven. 2015. “Why Scientists Shouldn’t Write History.” The Wall Street Journal, February 13. Stenmark, Mikael. 2001. Scientism: Science, Ethics, and Religion. Aldershot: Ashgate. Teller, Paul. 2004. “What Is a Stance?” Philosophical Studies 121: 159–​170. Tiberius, Valerie and John D. Walker. 1998. “Arrogance.” American Philosophical Quarterly 35(4): 379–​390. Van Fraassen, Bas C. 2002. The Empirical Stance. New Haven, CT: Yale University Press. Williams, Bernard. 2006. Philosophy as a Humanistic Discipline. Edited by A.W. Moore. Princeton, NJ: Princeton University Press. Wittgenstein, Ludwig. 1969. On Certainty. Edited by G.E.M. Anscombe and G.H. von Wright, translated by Denis Paul and G.E.M. Anscombe. Oxford: Basil Blackwell. Wittgenstein, Ludwig. 1979. Remarks on Frazer’s “Golden Bough.” Translated by A. Miles. Retford: Brynmill.

617

7

AN EPISTEMOLOGICAL CRITIQUE OF SCIENTISM

René van Woudenberg

7.1 Introduction “Scientism” is normally used to pejoratively characterize the views of others than oneself. But not always. This chapter discusses two recent views that have self-​consciously been labeled by their proponents as “scientism.” One is propounded by Alex Rosenberg in his book The Atheist’s Guide to Reality,1 and the other by Don Ross, James Ladyman, and David Spurrett in the first chapter (entitled “In Defence of Scientism”) of their book Every Thing Must Go: Metaphysics Naturalized.2 While Every Thing Must Go is a contribution to academic philosophical discussions, The Atheist’s Guide to Reality is a popular book geared toward a wide audience. For two reasons, however, a discussion of Rosenberg’s scientism in a scholarly volume is called for. First, as his contribution to the current volume indicates, Rosenberg intends his scientism to be a philosophically commendable and defensible position. Second, in the formulation it receives in Rosenberg’s hands, scientism is a view that seems to be widely, if mostly unreflectively, held; it is, we may say with a reference to Francis Bacon, an epistemology of the tribe.3 As such it merits, invites, and needs critical discussion.

1. Page references to Rosenberg (2011) are preceded by AGR. 2. Page references to Ladyman and Ross (2010) are preceded by ETMG. 3. See De Ridder (2014). Rosenberg’s formulation of scientism seems widely albeit unreflectively endorsed by many. It should be noted, however, that the implications that Rosenberg draws from scientism are far more radical than those that others draw from it.

618

1 6 8   • 

René van Woudenberg

There are further differences between the two books. They have different targets they oppose. Rosenberg’s premier targets are theism and common sense.4 Ross, Ladyman, and Spurrett’s target is contemporary analytic metaphysics that seems to proceed (i) from the idea that at the most fundamental level of reality there are “little things” engaged in “microbangings”; and (ii) from the idea that metaphysical “intuitions” about the world have epistemic significance, that is, are guides to the way the world really is. Second, the scientisms proposed take quite different attitudes toward metaphysics. Whereas Rosenberg discards metaphysics in its entirety, Ross, Ladyman, and Spurrett see one big task for metaphysics, namely “to show how the separately developed and justified pieces of science . . . can be fitted together to compose a unified world-​view” (ETMG 45). Or, more normatively, metaphysics “is the enterprise of critically elucidating consilience networks across the sciences” (ETMG 28). Part A of this chapter is an exposition and critical evaluation of Rosenberg’s scientism; and part B of Ross, Ladyman, and Spurrett’s scientism.

7.2 Part A: Rosenberg’s Scientism Examined Rosenberg’s scientism is “the conviction that the methods of science are the only reliable ways to secure knowledge of anything; that science’s description of the world is correct in its fundamentals, and that when ‘complete’, what science tells us will not be surprisingly different from what it tells us today” (AGR, 6–​7). Rosenberg also says:  “being scientistic means treating science as our exclusive guide to reality, to nature—​both our own nature and everything else’s” (AGR 8). I take it that these two statements hang together in the following way: If you want to acquire knowledge of what is real (or what is so, or what is the case, or what is true), you should treat science as your exclusive guide. So scientism is a conviction that is an answer to such questions as “What should I believe if I want my beliefs to be rational?,” “What should

4. Rosenberg answers the big questions of life such as “Is there a God?”; “What is the meaning of life?”; “Why am I here?”; “Is there an immortal soul?”; “Is there free will?”; “Is there life after death?”; “Is there a moral difference between right and wrong, good and bad?”; “Is abortion, euthanasia, suicide, paying taxes, foreign aid or anything else you don’t like forbidden, permissible, or sometimes obligatory?” Rosenberg answers all these questions in the negative (see AGR 2–​3). We can thus see that one of his main targets is theism, but not exclusively. Other targets are the commonsensical views that there are moral differences between right and wrong and that, when it comes to actions, not anything goes.

619

An Epistemological Critique of Scientism  • 

169

I do if my aim is to acquire true rather than false beliefs? i.e., it is an ‘ethics of belief ’.”5 The first question I will address is whether Rosenberg’s scientistic conviction is true. So, is it true that the methods of science are the only reliable way to secure knowledge of anything? First we need some clarifications. For, what are the methods of science? Rosenberg describes them as the combination of “controlled experiment and careful observation with mainly mathematical requirements on the shape theories can take” (AGR 24). These, he says, are the only right methods for acquiring any knowledge. From this it follows that science is demarcated from non-​science by the application of methods. Finally, what does Rosenberg take knowledge to be—​what is required for knowing that p? He nowhere explicitly says, but it is charitable to ascribe to him the view that at least the following two conditions must be satisfied: (i) p is true, and (ii) p results from the application of the methods of science. With these clarifications in place I  now turn to the question whether the scientisitic conviction that “the methods of science are the only reliable way to secure knowledge of anything” is true. The answer must be that of course it is not. The counterexamples are legion. I know that I am now thinking about Rosenberg’s view of knowledge. I know what schools I went to. I know the names of my siblings, of the streets I lived on, etc. Moreover, I know that my neighbor’s house is white, that now there are no leaves on the branches of the trees in my garden, that the person who now walks past my house wears a yellow coat, etc. In addition, I know that 7 + 5 = 12, that there is only one even prime number, and that the Laws of De Morgan are true.6 I also know that what king David did to Bathseba was deeply wrong, and that promises ought to be kept. Finally, I know that birds don’t bark, and dogs don’t sing. I know all these things (that out of reverence to G. E. Moore (1925) I shall call Moorean truths), but none of them through anything that merits the title “application of the methods of science.” My knowledge of none of the Moorean truths results from the combination of “controlled experiment and careful observation with mainly mathematical requirements on the shape theories

5. See Chignell (2017). 6. Rosenberg acknowledges that mathematics is a problem for his scientism; his chapter in the current volume aims to address it.

710

1 7 0   • 

René van Woudenberg

can take.” This enables me to present my argument against Rosenberg’s scientism in succinct form: P1 If scientism is true, we have no knowledge of Moorean truths. P2 We do have knowledge of Moorean truths. C: Hence scientism is false. This, I  take it, is as good as a philosophical argument can get:  simple and sound. But this means that Rosenberg’s scientistic ethics of belief is deeply problematic, as it is based on an obviously false conviction. However, couldn’t Rosenberg bite the bullet and deny (2), so deny that I, or anyone else, knows Moorean propositions? Of course he could, but he would thus throw his scientism in jeopardy—​for science is in many respects based on our ordinary perceptions, on our memories, on the testimony of others that we accept. For as Susan Haack has argued, science is the long arm of common sense.7 When we conduct scientific experiments using apparatuses, we will have to read the measures, to remember what the experiment was all about, and to accept many things that others have said. So this reply doesn’t work. But perhaps Rosenberg can respond in one of two ways. First, he could say that “methods of science” must be taken in a capacious way so that the Moorean truths are also known by their application. But this is unconvincing in that it is counterintuitive to say that I know the names of my siblings by the application of a “method of science.” Moreover, the ways in which the Moorean truths are known, are the same ways by which many propositions about morality that Rosenberg rejects, are believed by many. Second, Rosenberg could retort that what we know are only those propositions that either in fact do or in principle could (also) result from the application of scientific methods. In that way we could still be said to know many Moorean propositions even if in fact they don’t result from the application of scientific methods—​so long as they could in principle result from scientific inquiry. And at least a number of Moorean propositions could plausibly be claimed to have that property. Perhaps my knowledge of the proposition that my neighbor’s house is white in fact doesn’t result from the application of scientific methods, but it is surely conceivable that in principle it could result from the application of such methods. And so it might be with many other Moorean propositions. This may be attractive for Rosenberg, as many of the propositions that he wants to

7. Haack (2007, ch. 4).

17

An Epistemological Critique of Scientism  • 

171

claim are false—​propositions such as that there is no God, that there is no design in nature, and that free will is illusory—​are conceivably such that even in principle they could not result from the application of scientific methods. But there is a problem with this response that I can best bring out by discussing a number of claims that express if not the letter then certainly the spirit of Rosenberg’s scientism. One of those claims is that “Physics Tells us the Whole Truth About Reality,” the argument for which is this: The phenomenal accuracy of its prediction, the unimaginable power of its technological application, and the breathtaking extent and detail of its explanations are powerful reasons to believe that physics is the whole truth about reality. As for the reality above the subatomic, all we need to know is what things are physically composed of and how the parts are arranged in order to explain and predict their behaviour to equal detail and precision. That goes for people too. . . . The physical facts fix all the facts. . . .—​including the chemical, biological, psychological, social, economic, political, and other human facts. (AGR 25–​26) The conclusion here is that physics is the whole truth about reality. This is, of course, a somewhat enigmatic saying. After all, physics is a field of inquiry, so not the sort of thing that can be true. But interpreted charitably, I take it, what it is intended to say is that every truth there is, is (or perhaps: supervenes on) a truth of physics—​and a “truth of physics” is a truth that physics has, or can, or will conjure up, that is, a truth that we do, or can, or will know through the application of the methods used in physics. How does Rosenberg argue for this conclusion? The quotation just given suggests the following: P1 P2 P3 C1:

Physics is accurate in its predictions Physics has unimaginable technological applications Physics provides extensive and detailed explanations Hence, physics is the whole truth about reality (or: Every truth there is, is a truth of physics)

This argument is perhaps best thought of as an inductive argument:  The premises raise the probability of the conclusion, even if they don’t demonstrate it. Now P1, P2, and P3 are surely reasons to believe that physics tells us

712

1 7 2   • 

René van Woudenberg

truths about reality. But they aren’t reasons to believe that physics tells us the whole truth about reality, nor are they reasons to believe that only physics tells us truths about reality. So his argument is unconvincing. But the conclusion itself may, of course, still be true. But it isn’t, as I will now argue. For the conclusion implies all of the following: (a) The truth that “to procrastinate” means “to put things off ” is a truth that we do, or can, or will know through the application of the methods of physics. (b) The truth that Goldbach’s conjecture holds for all even numbers smaller than 100.000 is a truth we do, or can, or will know through the application of the methods of physics. (c) The truth that Mozart wrote a Requiem is a truth that we do, or can, or will know through the application of the methods of physics. (d) The truth that when you make a promise to someone, you thereby incur the obligation to do as promised, is a truth that we do, or can, or will know through the application of the methods of physics. However, (a)–​(d) are quite obviously false.8 And so, we have the following argument against the claim that physics is the whole truth about reality: P4 If physics is the whole truth about reality, then (a)–​(d) are true. P5 (a)–​(d) are false C2 Physics isn’t the whole truth about reality. I should next like to address a further claim that is clearly in the spirit of Rosenberg’s scientism, even though it isn’t the letter of it, namely a wieldy claim about what science entails. After Rosenberg has claimed, unsupported by argument, that the application of the methods of science is the only way to secure knowledge of anything, he goes on to argue that the application of those methods show us that:

8. Rosenberg acknowledges that conceptual and mathematical truths such as (a) and (b) are problematic for his view and need special attention—​some of which he gives in his contribution to the current volume. As to (c) and (d) a number of philosophers have claimed that the truths at hand “supervene” on physical facts, or can be “reduced” to physical facts. Rosenberg doesn’t do that. Had he said that, my argument would have to be more complicated. Here I deal only with Rosenberg views as he has stated them.

713

An Epistemological Critique of Scientism  • 

173

(i) There is no God. (AGR 17) (ii) There is no design in nature. (AGR chs. 3 and 4) (iii) Immortal souls don’t exist. (AGR ch. 7) (iv) Free will is an illusion. (AGR 236–​238) (v) There is no difference between right and wrong, good and bad; morally speaking anything goes. (AGR ch. 5) It is very hard to see, however, why the claim that the application of the methods of science have these entailments should be true. Of course, on the assumption that physics tells us the whole truth about reality, on the assumption of scientism, the application of the methods of science may very well have these entailments. However, I  have shown that this assumption is false. But if this assumption is false, then we should open our minds for the following point: From the fact that we can do physics without reference to God; or without needing the notion of intentional design; or from the fact that physics doesn’t entail that immortal souls exist; or doesn’t tell us that we have free will; or doesn’t tell us there is a difference between right and wrong, we cannot seriously conclude that physics entails the nonexistence of God, design, souls, free will, or the difference between right and wrong. The point is that physics may just be the wrong place to start thinking about these matters—​as many physicists and philosophers have argued.9 Absence of evidence from physics for the existence of any of the items mentioned, isn’t evidence from physics for the nonexistence of these items. Let me quickly illustrate this for the topic of design in nature.10 Design hypotheses, such as the hypothesis that the world is created by God, are compatible with all we know from physics. This means that for all we know, physical reality may be the product of intentional design. The proposition that physical reality is designed is not proved false by anything we know from physics. There is an analogy here with radical skeptical hypotheses. The proposition that we are brains in vats, hooked up to computers that are operated by super scientists that feed us with experiences that we think of as experiences of the “external world,” is not proved false by any of our experiences. Just as the brain in a vat hypothesis cannot be proved false, the design hypothesis cannot be

9. See Medawar (1986), Rescher (1999). 10. For extended argument on the point made in this paragraph, see Van Woudenberg and De Ridder (2017).

174

1 7 4   • 

René van Woudenberg

proved false either by anything we know from physics. Physics doesn’t have the entailment Rosenberg claims for it. A final point I should like to make about the spirit of Rosenberg’s scientism concerns his handling of the question Why is there something rather than nothing? Quantum physics, Rosenberg tells us, shows that the correct answer to this question is that there is no reason why there is something: There could have been nothing at all, our universe is just one of those random events that “sometimes” occur and at other “times” don’t. . . . The same goes for fermions and bosons. Their existence is just the way the cookie of quantum randomness crumbled. . . . Since the big bang is just . . . a quantum event, it, too, is a wholly indeterministic one. It is an event that just springs up out of the multiverse’s foam of universes without any cause at all. . . . Why is there a multiverse in which universes pop into existence for no reason at all? No reason at all! It’s just another quantum event. What science and scientism tell those who hanker for more is “Get over it!” (AGR 38–​39) This quotation displays some strange bit of reasoning. Normally when we ask a question of the sort “Why do Xs exist and why don’t they not exist?,” we answer by referring to things that existed prior to the existence of the Xs and that are, somehow, responsible for the (coming into) existence of the Xs. But when we ask “Why is there something rather than nothing?,” we cannot proceed in this way—​for since we ask why there exists anything at all, we cannot, in answering this question, refer to something that existed prior to the existence of everything and that is, somehow, responsible for the existence of everything. For there is no such thing—​that is, there is no thing that exists prior to the existence of anything at all. Still, Rosenberg does proceed in exactly that way! For he says that the fact that there exists something and not nothing, is due to quantum randomness. But this, of course, is not an answer of the right sort to the question. For quantum randomness surely is already something and not nothing.11,12 11. The same mistake is made by Krauss (2012). 12. Related to this is the following confusion: Rosenberg discusses the anthropic principle—​ that is, the principle according to which the universe is fine-​tuned for life—​as if that principle is intended to answer the question why there exists something rather than nothing (see AGR 39–​4 0). But this is wrong. The anthropic principle doesn’t address this question, but the question of whether the universe’s apparent fine-​tuning is a matter of chance or a matter of intentional design. See Leslie (1989).

715

An Epistemological Critique of Scientism  • 

175

I am bringing this up because the question “Why is there something rather than nothing?” is a question that cannot be answered by physics, that is, not by reference to anything that physics can discover. It is a question the answer to which will have to draw on sources other than physics. Unless one thinks, as Rosenberg apparently does not, that the question is somehow ill-​conceived. All of this goes to show that Rosenberg’s scientism suffers from a number of fatal problems. And in addition to the problems I have raised, there is the debilitating further point that Rosenberg’s scientism is self-​referentially incoherent. By this I mean that if one follows Rosenberg’s scientistic ethics of belief consistently, one must reject his scientism. Rosenberg’s ethics of belief, recall, is to treat science as our exclusive guide to reality. So suppose one treats science as one’s exclusive guide to reality, and then reflects on the question Q: “Should I treat science as my exclusive guide to reality?” In order to find an answer to Q, Rosenberg’s advice is to take science as one’s exclusive guide. Now does science tell us that we should answer Q affirmatively? That is, does science tell us that we should treat science as our exclusive guide to reality? The answer must be: No, it doesn’t! Neither physics, nor chemistry, nor biology, nor neuropsychology tell us that science is our exclusive guide to reality. Science tells us about the world, not about how to treat science. But this means that if one adopts Rosenberg’s scientistic advice, one has a solid reason to reject scientism. Scientism, as the saying goes, is hoist with its own petard. The overall conclusion of my discussion thus far is that the ethics of belief that Rosenberg advocates, namely to treat science as our exclusive guide to reality, is deeply problematic, and that the conviction on which it is based, namely that the methods of science are the only reliable ways to secure knowl­ edge of anything, is folly.

7.3 Part B: Ross, Ladyman, and Spurrett’s Scientism Examined Ross, Ladyman, and Spurrett’s scientism is quite a bit more complex than Rosenberg’s. It has three what I call “elements”: (1) A view as to what science is, (2) non-​positivist verificationism, and (3) the idea that scientism is a “stance” rather than a claim. I will present and examine these elements in this order. 7.3.1 First Element: What Science Is Ross, Ladyman, and Spurrett’s scientism, of course, has to do with science. But how do they understand “science,” how is it demarcated from non-​science?

716

1 7 6   • 

René van Woudenberg

They explain their understanding of science by, first, specifying what they take the aim of science to be, and second by offering a criterion by which science can be distinguished from non-​science. As to the first point, Ross, Ladyman, and Spurrett aver that the aim of science is to “discover the objective character of the world”: Science . . . is our set of institutional error filters for the job of discovering the objective character of the world—​that and no more but also that and no less—​science respects no domain restrictions and will admit no epistemological rivals (such as natural theology or speculative metaphysics). With respect to anything that is a putative fact about the world, scientific institutional processes are absolutely and exclusively authoritative. (ETMG 28) The quote leaves no doubt that Ross, Ladyman, and Spurrett intend their view to be fully general: Only science and nothing else is our guide to “the objective character of the world.” As to the second point: Ross, Ladyman, and Spurrett deny that science is demarcated from non-​science by something like the scientific method and that is supposed to be common to all of the sciences and not used by non-​ science.13 Instead, they propose that .  .  . science is demarcated from non-​science solely by institutional norms:  requirements for rigorous peer review before claims may be deposited in “serious” registers of scientific belief, requirements governing representational rigour with respect to both theoretical claims and accounts of observations and experiments, and so on. . . . We can, however, achieve significant epistemological feats by collaborating and creating strong institutional filters on errors. (ETMG: 28) Science, then, Ross, Ladyman, and Spurrett claim, is demarcated from non-​science institutionally—​by the institutions of peer review, and representational rigor. In addition, Ross, Ladyman and Spurrett urge that science is demarcated from non-​science by norms having to do with fundability:  “ ‘Bona fide institutional science’ ” consists in working on

13. Other friends of scientism do hold on to the idea that method sets science apart from non-​ science. A critical discussion of that claim is Van Woudenberg and Rothuizen (2016).

71

An Epistemological Critique of Scientism  • 

177

“hypotheses one cannot propose as the targets of investigation in a grant proposal to a “serious” foundation or funding agency with non-​ zero prospects of success” (ETMG 33). Even though Ross Ladyman and Spurrett offer an institutional demarcation of science, they also formulate a constraint on anything that qualifies as science, which they call the Primacy of Physics Constraint (PPC):  Special science hypotheses that conflict with fundamental physics, or such consensus as there is in fundamental physics, should be rejected for that reason alone. Fundamental physical hypotheses are not symmetrically held hostage to the conclusions of the special sciences. (ETMG 44) Examples that Ross, Ladyman, and Spurrett offer of breaches of this constraint are the following: substance dualism and various theories of agency and volition, such as libertarian ones (ETMG 45). I will first examine the claim that “Science is the one and only authority on the objective character of the world.” This claim is ambivalent. For what is the “objective” character of the world supposed to be? John Searle’s distinction between ontologically objective and epistemically objective features is relevant here.14 Feature F of the world is an ontologically objective feature of the world =df The world has F independently of any subject ever experiencing, believing, or asserting that the world has F. The statement that iron expands when heated, states an ontologically objective feature of the world: The world’s having that feature is independent of any subject ever experiencing, believing, or asserting that iron expands when heated. By contrast, the statement that Sam is suffering, does not state an ontologically objective feature:  The world’s having that feature is not independent of anyone experiencing pain. Other examples of statements that state ontologically subjective features of the world are the following:  “The zero meridian passes over Greenwich, England”; “John Adams was married to Abigail Smith”; and “This particular piece of paper is a $100 bank note.”

14. See Searle (1995: 7–​13).

718

1 7 8   • 

René van Woudenberg

Searle contrasts the notion of “ontologically objective” (a status that attaches to features of the world) with the notion of “epistemically objective” (a status that attaches to statements): A statement (judgment) is epistemically objective =df. The truth-​value of the statement is determined solely by facts in the world, not by anyone’s feelings, attitudes, or points of view vis-​à-​vis its truth-​value. The statement that iron expands when heated is epistemically objective, as its truth-​value is solely determined by the world, not by anyone’s feelings or attitudes. By contrast, the statement that Rembrandt is a better artist than Rubens, is epistemically subjective, as it is determined by the feelings and attitudes of persons. We must notice that we can make epistemically objective statements about things that are ontologically subjective. That Sam is suffering is an epistemically objective statement about something that is ontologically subjective. Also, we can make epistemically subjective statements about things that are ontologically objective. That Mt. Everest is more beautiful than Mt. Rushmore is an epistemically subjective statement about an ontologically objective feature. Ross, Ladyman, and Spurrett’s claim that science is our only guide to the objective character of the world, then, is potentially ambivalent between the claim that science is our only guide to the ontologically objective features of the world, and the claim that the only way to make epistemically objective statements is by doing science. Although they don’t explicitly say so, I take it that they have the latter option in mind. My reason is that Ross and his colleagues devote an entire chapter of their book to the status and ontologies of the special sciences and draw freely on examples from psychology, sociology, and economics, which they clearly count as science. But the subject matters of these sciences are, of course, at least in part, ontologically subjective features of the world: They depend on people’s beliefs, feelings, or attitudes. So they must think that the ontologically subjective features of the world belong to the jurisdiction of science just as much as its ontologically objective features. Hence, their scientism can be further specified as the claim that science is exclusively and absolutely authoritative with regard to any epistemologically objective claim about the world. The question is why we should think this is true. Consider the following statements, and ask yourself “are these statements epistemically objective—​ and if so: Is science the exclusive and absolute authority on them?”:

719

An Epistemological Critique of Scientism  • 

179

(a) I now have no headache; (b) Some birds are black; (c) A proposition and its negation cannot both be true; (d) Promises ought to be kept; (e) It is wrong to torture others just for fun; (f ) There is a God; ( g) Human persons have identity over time; (h) Properties are universals. I take it that it is plausible to say that these statements are epistemically objective: Their truth-​value is determined by the facts in the world alone. But why should we think that science is the one and only authority on them? Take (a), the statement that I now have no headache. Why shouldn’t I hold on to it, even if science is silent about what I feel in my head? Don’t I have any authority with respect to statements like (a)? It would seem I have—​but the view under consideration denies it to me. Or take (b), the statement that some birds are black. Many people know this through simple forms of perception that cannot be classified as “doing science.” Does it therefore have no authority whatsoever? That strains credulity—​also because science itself is at least partly based on simple forms of perception. Or take (c), the statement that a proposition cannot be both true and false. Is science the one and only authority on this? One problem is that in order for science to do its work, (c) must be presupposed. The authority that science has, depends, in part, on (c). But then it is unclear how science can be an authority on (c). Rather, the very fact that we do accept the elementary principles of logic indicates that there is epistemic authority outside of science!15 As to (d), the statement that promises ought to be kept and (e), the statement that it is wrong to torture others just for the fun of it:16 Why should we think that here too science is the one and only authority? Many people across times and places have firmly held on to (d) and (e), believing them to be without defeaters. Doesn’t that have any authority whatsoever? Even more radically: Why should we even think that science has any authority when it comes to moral, or normative statements more generally, statements in the first place. After all, as we all learned from

15. One account of the nonscientific authority on the principles of logic, as well as on other a priori justified propositions, is BonJour (1998). 16. As indicated in the body of the text, I take it that (d) and (e) are epistemically objective. Moral antirealists of various stripes will deny this. But even if I  were to leave out (d)  and (e) from the discussion, there is enough left for my argument to go through.

810

1 8 0   • 

René van Woudenberg

Hume, science tells us about what is the case, not about what ought or ought not to be the case. Again, take (f ), the proposition that there is a God. Why should we think that science is the one and only authority in this matter? Many theists believe (f ) for a variety of different and interlocking reasons—​ reasons having to do with mystical and other experiences, with abstract forms of reasoning, or with the testimony of instances they deem reliable. Don’t these sources have any authority whatsoever? Should we discard them simply because they aren’t sources that science works with? That would be ill-​advised, it seems to me. From the fact that science is silent about X, we shouldn’t conclude that there is no X. As to the metaphysical statements (g), that humans have identity over time, and (h) that properties are universals: Philosophers on both sides of these issues have worked out numerous arguments for their favored views. Do none of these arguments have any authority whatsoever, simply because they are not based on empirical science? Again this strains credulity. For it seems that at least some of (a)–​(h) are at least to some extent rationally acceptable—​for there is at least some authority that speaks for them. These considerations inform the following argument: P1 If science is the exclusive and absolute authority on epistemically objective statements, then such epistemically objective statements as (a)–​(h) cannot be rationally accepted. P2 Statements (a)–​(h) can be rationally acceptable. C: Hence, science is not the exclusive and absolute authority on epistemically objective statements. This argument rebuts one main tenet of Ross, Ladyman, and Spurrett’s scientism. I have indicated that another part of their understanding of science is that science is institutionally demarcated, and that they offer an institutional criterion for bona fide research. However, talk about bona fide science and “serious” funding bodies raises a problem for Ross, Ladyman, and Spurrett’s view on demarcation. For a straightforward application of the institutional criterion leads to the result that, in addition to the natural and the social sciences, philosophy, analytic metaphysics included, will also qualify as science. After all, analytic metaphysics, and philosophy in general, clearly satisfy various institutional norms that Ross, Ladyman, and Spurrett have laid down for science. Analytic metaphysicians have peer-​reviewed journals, academic conferences, standards of rigor in presentation; they have a proper place in universities, they receive external funding from both public and private funding bodies,

18

An Epistemological Critique of Scientism  • 

181

and so on. So on this way of demarcating science from non-​science, analytic metaphysics, and philosophy in general, are all perfectly legitimate sciences. So they are legitimate ways of finding out about the objective character of the world. But this is a result that Ross and company definitely want to reject. It is likely that Ross, Ladyman, and Spurrett may want to look for a stricter interpretation of the institutional criterion of science in order to prevent the concept of science from getting applied too broadly. It is hard to see, however, how they could do this without developing more substantive epistemological norms after all. For instance, they might attempt to spell out in more detail what constitutes a “serious” source of funding for science. One option might be to say that all public funding bodies are serious whereas private ones are not. This is both too liberal and too restrictive, however. Too liberal, because public funding bodies in some places and times tend to fund epistemically bad science in order to promote ideological agendas. Too restrictive, because some private funding sources do fund excellent science, as is amply witnessed by the fact that many of the best American universities are in fact privately funded. It won’t help either to stipulate that only privately funded universities are serious and other private funds not. That is still too restrictive. Some private funding bodies do a great job of funding excellent and unbiased science, while others are rather obviously ideology driven. But distinguishing the latter from the former will require giving some non-​institutional epistemological criteria. Anything short of epistemological criteria is likely to result in a somewhat arbitrary demarcation that includes both things that Ross and company would want excluded and excludes things they want included. Similar lines of argument can be construed for other institutional norms, such as the use of peer-​ reviewed journals, standards of rigor, etc. Trying to resolve this by tweaking the institutional norms so that only those things will come out as science that already satisfy Ross, Ladyman, and Spurrett’s more intuitive judgments about what counts as good science would of course be unacceptable, because that would make the project circular. The point of giving institutional norms was to draw a distinction between science and non-​science. To then formulate these norms on the basis of prior intuitions about what is and is not bona fide science renders the result circular and the institutional norms obsolete. If we already have a firm enough, intuitive grip on the distinction between science and non-​science, there’s no point in giving a further demarcation criterion. Relying solely on institutional norms to demarcate science from non-​ science thus spells trouble. Either too much is included and Ross and company cannot exclude the kind of philosophy they want to reject; or they are

812

1 8 2   • 

René van Woudenberg

driven to giving epistemological norms after all, on pains of drawing an arbitrary line or giving a circular demarcation criterion. Contrary to Ross, Ladyman, and Spurrett, then, I have argued, first, that science is not the exclusive and absolute authority on epistemically objective statements, and second, that the institutional demarcation criterion they propose is problematic and doesn’t do what they want it to do. 7.3.2 Second Element: Verificationism The second element of Ross, Ladyman, and Spurrett’s scientism is their non-​ positivist, pragmatist version of verificationism, which itself consists of two claims. First, that “no hypothesis that the approximately consensual current scientific picture declares to be beyond our capacity to investigate should be taken seriously” (ETMG 29), and second a claim they call the Principle of Naturalistic Closure (PNC): Any new metaphysical claim that is to be taken seriously at time t should be motivated by, and only by, the service it would perform, if true, in showing how two or more specific hypotheses, at least one of which is drawn from fundamental physics, jointly explain more than the sum of what is explained by the two hypotheses taken separately. (ETMG 37) This principle tells us, roughly, that a serious metaphysical claim is a claim that enhances, that is, broadens the scope of scientific explanation. As to the first verificationist claim, that is, the claim that no hypotheses that current science declares to be beyond our capacity should be taken seriously, it must be said that some hypotheses are beyond our capacity to investigate because it would be too expensive, or unethical, to test them. The claim at hand is not about such practical incapacities however. It is about incapacities that science itself informs us about. As an example of this our authors offer the thought that since the Big Bang is a singular boundary across which no information can be recovered from the other side, the claim that the Big Bang was caused by God is beyond our capacity to investigate. Such a claim, they hold, is therefore “uninteresting because it is silly and unmotivated” (ETMG 29), it is “pointless” (ETMG 30), meaning that it can make no contribution to objective inquiry. I shall now argue that there is little reason to accept this first claim. The hypotheses that the claim refers to are deemed “uninteresting” and

813

An Epistemological Critique of Scientism  • 

183

“pointless.” These terms are the substitutes for the old verificationist’s notion of “cognitive meaninglessness.” But there is a big difference. A  hypothesis was supposed to be cognitively meaningless provided there is no scientific method by which its truth-​value could be verified. According to Ross and his friends, however, a hypothesis that science declares beyond our capacities is “uninteresting.” This latter verdict is of course no more than an expression of personal attitudes, but this cannot prevent others from deeming them highly interesting. The qualification “pointless” has more bite: A hypothesis is pointless if it can make no contribution to objective inquiry. However, as I have suggested above, science is not the only authority on epistemically objective statements. Hence those nonscientific authorities can very well contribute to objective inquiry. And hence the statements they issue aren’t necessarily pointless either. A further point in this connection is that Ross and company suggest that science itself may tell us that certain parts of reality are beyond our capacity to investigate, because information about them cannot “be extracted for receipt in our region of spacetime or in regions of spacetime in which our instruments could in principle go.” Underlying this line of thought is the idea that the only way we can extract information from the world is through instruments. But this idea is problematic for a number of reasons. If we take “instruments” broadly here, so as to include not only telescopes, satellites, and space vessels, but also our “natural” instruments like eyes, ears, nose, tongue, and skin, we seem to be capable to extract information from the world by means other than artificial or natural instruments, for example, by means of reasoning and argument. We need artificial and natural instruments in order to extract from the world the information that this piece of iron expands when heated, and also that that piece of iron expands when heated. But in order to extract from the world the information that all iron expands when heated we need inductive reasoning. Also, we are capable to extract information from the world without the help of any of the instruments mentioned—​these instruments being entirely unfit to extract the sort of information I  have in mind, that is, information about the intrinsic modal properties of objects. Suppose we see that object X has property p, then we can ask whether X has p necessarily, or whether X could have lacked p. Very often we have information enough to answer these questions—​but the information doesn’t come to us by application of any of the instruments. For example, we have information enough to answer the question whether Obama has the property of being the president of the United States necessarily. But not through the instruments

814

1 8 4   • 

René van Woudenberg

mentioned.17 Other information that we seem capable to extract from the world without instruments is that rape is a hateful wrongdoing, and that promises ought to be kept. This remains true even if moral properties supervene on physical properties. All of this goes to show that there are other ways to extract information from the world than by applying natural and artificial instruments. And this information, and the hypotheses it gives rise to, don’t seem to be uninteresting or pointless. Of course, friends of scientism may bite the bullet and simply deny that we can extract the information from the world in the ways that I have claimed we can. However, why should one want to bite it? For what reason? After all, what is so bad about acknowledging modes of extracting information from the world other than the application of the artificial and natural instruments mentioned?18 Let us now move on to the second part of Ross, Ladyman, and Spurrett’s verificationism, the Principle of Naturalistic Closure (PNC), which, recall, says that metaphysical claims should be exclusively motivated by the service they would perform, if true, in showing how at least two scientific hypotheses, at least one of which is taken from fundamental physics, jointly explain more19 than the sum of what is explained by the two hypotheses taken separately. This is not an easy-​to-​grasp principle. It helps a bit when our authors imply that at least the following metaphysical theories fail to satisfy PNC:  Ned Markosian’s definition of physical objects as objects having spatial location, David Lewis’s doctrine of Humean supervenience, and Peter van Inwagen’s view that all material things decompose into mereological atoms. A  metaphysical theory that they argue does satisfy PNC is Ontic Structural Realism (OSR for short), which is the view that “even the identity and individuality of objects depends on the relational structure of the world.” In fewer words it is the view “that there are no things, structure is all there is” (ETMG 130). So, OSR denies that there are fundamental individuals or substances upon

17.  For extended argument see Rea (2002:  chs. 4–​5). Rea convincingly argues that the use of the instruments mentioned neither give us other sorts of information about the world that we seem to be capable of—​for example, information about whether or not something is functioning properly. 18. We can also extract information from the world through introspection, memory, proprioception, which don’t belong to the set of artificial and natural instruments either. 19.  Ross, Ladyman, and Spurrett explain what they mean by “explain more” by reference to Kitcher’s notion of “argument pattern” (ETMG 31). The basic idea is that unification in science consists in maximizing the ratio of kinds of phenomena we can explain to the number of causal processes we cite in the explanations.

815

An Epistemological Critique of Scientism  • 

185

which structures supervene. Ross and friends present OSR as an eliminative view in that it says that there are objects—​“ but they have been purged of their intrinsic natures, identity, and individuality, and they are not metaphysically fundamental” (ETMG 131). Ross and colleagues claim that OSR “is motivated by the contents of our best physical theories, namely quantum theory and general relativity” (ETMG 129) and argue furthermore that, supplemented with concepts from information theory, OSR “makes possible a plausible account of laws, causation, and explanation, and the relationships among the special sciences, and between them and fundamental physics” (ETMG 129). However, a difficulty must be noted right here. I said that OSR is presented by our authors as a claim that conforms to PNC. But it is not so clear that it does. For what they in fact say is that OSR is motivated by quantum mechanics and general relativity and that it makes possible an account of laws, causation, and explanation. But this state of affairs doesn’t seem to render OSR PNC-​conform. For in order for OSR to conform to PNC, it would have to be shown that OSR performs the following function: It shows how quantum mechanics and general relativity jointly explain more than the sum of these theories separately. And this, I think, has not been done. And this means that we lack an example of a metaphysical hypothesis that conforms to the letter of PNC. I should now like to make some critical observations regarding PNC. First, it is unclear why PNC formulates constraints only on metaphysics. Why shouldn’t the constraints hold for all claims concerning the objective character of the world, also those that are made in aesthetics, for example, or ethics, or epistemology or the philosophy of science, or the humanities? The verification criterion of the logical positivists was supposed to be applicable to each and any statement. Our authors have abandoned that criterion. But why should the conditions of PNC be restricted to metaphysical claims? I could find no such reason. Second, it is not quite clear exactly how to apply PNC. One construal is that what PNC tells us is that we should only accept a metaphysical claim when that claim, if true, in conjunction with two scientific hypotheses, explains more than just the conjunction of the two scientific theories. If we suppose this to be the spirit of PNC, a problem arises. For PNC is explicitly designed to rule out natural theology and speculative metaphysics. But PNC, on this construal, may not do so much. Suppose there is hypothesis H1 taken from fundamental physics that explains all the phenomena in set S1. Suppose furthermore there is another hypothesis, H2, also taken from fundamental

816

1 8 6   • 

René van Woudenberg

physics, that explains all the phenomena in set S2. Suppose furthermore that both of these hypotheses involve laws that have to be taken as fundamental (so as laws that cannot be reduced to more fundamental laws) and are, so far, unexplained. Suppose, finally, that there is good argument for the conclusion that there is a Prima Causa that is the cause of the so far unexplained laws. If we name the conclusion of the argument T, we can think of T as a hypothesis that is made probable, to some degree, by the argument. Then intuitively the following is true: T in conjunction with H1 and H2 explains more than the sum of what H1 and H2 explain.20 Third, PNC is a claim about the acceptability of claims. The claims about which PNC issues a claim are metaphysical claims. About metaphysical claims, PNC claims that they “should be exclusively motivated by the service they would perform, if true, in showing how at least two scientific hypotheses, at least one of which is taken from fundamental physics, jointly explain more than the sum of what is explained by the two hypotheses taken separately.” Let us abbreviate this mouthful by saying that PNC claims that metaphysical claims should only be accepted provided they have “surplus explanatory power.” Now it could be argued that PNC itself is a metaphysical claim. It certainly isn’t a scientific claim. After all, it’s not a statement of the sort that scientists make. Nor is it a claim that could make for a successful grant proposal with a reputable granting agency. And if we adopt the idea that a statement is either a scientific or a metaphysical statement, then PNC must be deemed a metaphysical claim. But then why should we accept it? Well, what is it that PNC says about the acceptability of metaphysical claims? This: Only accept them when they have surplus explanatory power. But it seems rather clear that PNC doesn’t have surplus explanatory power. Hence, if one accepts PNC, one has a very good reason to reject PNC. PNC is self-​referentially incoherent. Ross, Ladyman, and Spurrett, however, can and do try to evade this ploy. This brings me to the third element of their scientism. 7.3.3 Third Element: Scientism Is a Stance Ross, Ladyman, and Spurrett say their scientism is not a doctrine that one can believe in, but a “stance”—​something Bas van Fraassen has described as follows:

20. See Swinburne (1992), Van Holten (2003).

817

An Epistemological Critique of Scientism  • 

187

A philosophical position can consist in something other than a belief in what the world is like. We can, for example, take the empiricist’s attitude toward science rather than his or her beliefs about it as the more crucial characteristic . . . A philosophical position can consist in a stance (attitude, commitment, approach, a cluster of such—​possibly including some propositional attitudes such as beliefs as well). Such a stance can of course be expressed and may involve or presuppose some beliefs as well, but cannot be simply equated with having beliefs or making assertions about what there is. (Van Fraassen 2002: 47–​48) Ross, Ladyman, and Spurrett say: “Our fundamental principles are not propositional doctrines: Both the PNC (=Principle of Naturalistic Closure) and the PPC (=Primacy of Physics Constraint) are explicitly norms” (ETMG 64).21 The idea, then, is that if something is a norm it is not a propositional doctrine. Propositional doctrines have truth values, but norms don’t. This means that Ross, Ladyman, and Spurrett’s scientism isn’t the sort of thing that has truth-​value—​and it presumably also means that one doesn’t adopt it on the basis of truth-​relevant considerations—​one doesn’t adopt it on the basis of empirical evidence at all. Still, as Ross, Ladyman, and Spurrett say, adopting the scientistic stance (the norms embodied in PNC and PPC) is not arbitrary. Scientism, and the norms of PNC and PPC, they say, are motivated: “We motivate them as descriptively manifest in scientific practice” (ETMG 64). Ross, Ladyman, and Spurrett, then, hold that they can escape the charge that scientism is self-​referentially incoherent, by saying that it is not a claim at all, but a stance. But it is doubtful that there is this escape. For we now must think of the scientistic stance as involving the adoption of PNC as a norm. Norms are claims that tell us what we ought to do, or what ought to be done. The norm that is PNC is a claim that tells us what sort of claims we ought to accept. Does the PNC norm license the acceptance of PNC? It doesn’t seem so. After all, PNC doesn’t have that surplus explanatory power. So the problem of self-​referential incoherence hasn’t gone away by presenting PNC as a stance. PNC is not a principle that wears its application conditions on its sleeve. But perhaps there is a way around this problem. If so, the question arises why, for what reasons, on what grounds, we should adopt the scientistic stance. Now Bas van Fraassen has said that to adopt a stance “is similar or analogous to conversion to a cause, a religion, an ideology, to capitalism or to socialism,

21. See also Ladyman (2011).

81

1 8 8   • 

René van Woudenberg

or to a worldview” (quoted in ETMG 61). But if that is the way stances are adopted, stance adoption is underdetermined by evidence. That doesn’t make the adoption of a stance irrational, but it does indicate that someone who does not adopt the scientistic stance, cannot be accused of irrationality either. The point I am making lies in the neighborhood of points that Paul Helm and Mike Rea have made about belief-​policies and research programs respectively. A belief-​policy, says Helm, “is a strategy or project or program for accepting, rejecting or suspending judgment as to the truth of propositions in accordance with a set of evidential norm” (Helm 1994: 58), and he argues that we adopt some belief-​policies as the result of an overt choice. Scientism, I  aver, is an example of this. Helm then argues that “the acceptance of a belief-​policy . . . cannot be a matter of evidence (or its absence) alone and therefore may be as voluntary and free as human action” (ibid.: 58). Michael Rea defines a research program as “a set of methodological dispositions”—​ “dispositions to trust at least some of our cognitive faculties as sources of evidence and to take certain kinds of experience and arguments to be evidence” (Rea 2002: 2). Some research programs are deliberately acquired. Scientism, I aver, is an example of this. Rea then argues that it is impossible to adopt research programs on the basis of evidence. The point I should like to take away from this is that there is no rational basis for claiming that scientism must be adopted as a belief-​policy, or as a research program, or as a stance, nor for claiming that scientism is the only belief-​policy or research program, or stance, which is available for people who take science seriously. Scientism is an option. An option that I have argued suffers from so many problems, that it is utterly unattractive.22

References BonJour, Laurence. 1998. In Defense of Pure Reason. Cambridge:  Cambridge University Press. Chignell, Andrew. 2017. “The Ethics of Belief.” In The Stanford Encyclopedia of Philosophy (Spring 2017 Edition). Edited by Edward N. Zalta. https://​plato.stanford.edu/​archives/​spr2017/​entries/​ethics-​belief/​.

22. For comments and discussion I thank Jeroen de Ridder, Rik Peels, Naomi Kloosterboer, Lieke Asma, Hans van Eyghen, Gijsbert van den Brink; and the audience at the European Epistemology Network in Paris July 2016, especially Eric Olson and Klemens Kappel. This publication was made possible through the support of a grant from Templeton World Charity Foundation. The opinions expressed in this publication are those of the author(s) and do not necessarily reflect the views of Templeton World Charity Foundation.

819

An Epistemological Critique of Scientism  • 

189

De Ridder, Jeroen. 2014. “Science and Scientism in Popular Science Writing.” Social Epistemology Review and Reply Collective 3(12): 23–​39. Haack, Susan. 2007. Defending Science—​Reason. Between Scientism and Cynicism. New York: Prometheus Books. Helm, Paul. 1994. Belief Policies. Cambridge: Cambridge University Press. Kraus, Lawrence M. 2012. A Universe from Nothing: Why There Is Something Rather Than Nothing. New York: Free Press. Ladyman, James 2011. “The Scientistic Stance: The Empirical and Materialist Stances Reconciled.” Synthese 178: 87–​98. Ladyman, James and Don Ross. 2010. Every Thing Must Go. Metaphysics Naturalized. Oxford: Oxford University Press. Leslie, John. 1989. Universes. New York: Routledge. Medawar, Peter. 1986. The Limits of Science. Oxford: Oxford University Press. Moore, G.E. 1925. “A Defence of Common Sense.” In Selected Writings. Edited by Thomas Baldwin, 106–​133. London: Routledge. Rea, Michael C. 2002. World without Design. The Ontological Consequences of Naturalism. Oxford: Oxford University Press. Rescher, Nicholas. 1999. The Limits of Science. Revised Edition. Pittsburgh, PA: University of Pittsburgh Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality: Enjoying Life Without Illusions. New York: W. W. Norton. Searle, John. 1995. The Construction of Social Reality. New York: Free Press. Swinburne, Richard. 1992. The Existence of God. Revised Edition. Oxford: Clarendon. Van Fraassen, Bas C. 2002. The Empirical Stance. New Haven, CT: Yale University Press. Van Holten, Wilko. 2003. Explanation Within the Bounds of Religion. Frankfurt: Peter Lang. Van Woudenberg, René and Joëlle Rothuizen–​van der Steen. 2016. “Science and the Ethics of Belief:  An Examination of Philipse’s ‘Rule R.’” Journal for General Philosophy of Science 47(2): 349–​362. Van Woudenberg, René and Jeroen de Ridder 2017. “Design Hypotheses Behave Like Skeptical Hypotheses. Or: Why We Can’t Know the Falsity of Design Hypotheses.” International Journal for the Study of Skepticism 7(1): 23–​45.

910

8

KINDS OF KNOWLEDGE, LIMITS OF SCIENCE

Jeroen de Ridder

8.1 Introduction Knowledge comes in different kinds. Bertrand Russell famously drew a threefold distinction between propositional knowledge, knowledge how, and knowledge by acquaintance (Russell 1910, 1912). Some philosophers discuss the difference between knowledge-​that (i.e., propositional knowledge) and knowledge-​wh (i.e., knowledge what, where, who, when; see Parent 2014 for an overview). Additionally, several philosophers think first-​personal knowledge is importantly different from third-​personal knowledge (Zagzebski 2012). Eleonore Stump has drawn attention to the unique characteristics of second-​ personal knowledge (Stump 2010; cf. also Pinsent 2012). In this chapter, I want to use the insight that knowledge comes in kinds to explore two thoughts: (1) Scientific knowledge is a special kind of knowledge with distinctive characteristics, which set it apart from what we might call “everyday knowledge.” (2) That scientific knowledge is a particular kind of knowledge has implications for what we can know scientifically. More specifically:  It implies that scientific knowledge has limits. I  will proceed by laying out two plausible construals of the distinctive nature of scientific knowledge—​different, but not incompatible—​and exploring what their respective implications are for whether scientific knowledge has limits.1

1.  One caveat about the project of this chapter. Popular discussions about the limits of science often mention knowledge of normative facts—​moral knowledge in particular—​and knowledge of supernatural facts as intrinsic limitations to scientific knowledge (Hughes 2012; Wieseltier 2013; cf. also Rescher 1999: 243–​248;

19

Kinds of Knowledge, Limits of Science  • 

191

If the argument of this chapter is correct, it spells trouble for various versions of scientism. Some versions of scientism hold that science has no limits, or at least none that can be established a priori. James Ladyman (this volume) defends such a claim and Rik Peels’s chapter contains references to other examples (cf. also various contributors to Boudry and Pigliucci 2018). If the nature of scientific knowledge imposes constraints on what can be known scientifically, then we have reason to reject such no-​limits scientism. Epistemological scientism maintains that only the natural sciences provide genuine knowledge (Stenmark 2001; Peels this volume). If I  am right that scientific knowledge is inherently limited in certain ways, this constitutes at least part of an argument against epistemological scientism, since it shows that there might be truths outside the reach of scientific knowledge. To turn this into a robust objection against epistemological scientism, we would need a further argument to the effect that some of those truths can be known by other means. While my argument will hint at some possibilities, I cannot provide a full positive defense of them here. The plan for the rest of the chapter is as follows. In the next section, I will say more about what is behind the distinctions between different kinds of knowledge in order to see how we might go about distinguishing scientific knowledge from other kinds of knowledge. In section 8.3, I will introduce the first way of thinking about the nature of scientific knowledge: scientific knowledge as high-​grade knowledge. In section 8.4, I explore what limits scientific knowledge has on this proposal. Section 8.5 introduces a second way of thinking about scientific knowledge:  scientific knowledge as objectifying knowledge. Section 8.6 then discusses what this proposal means for the limits of scientific knowledge. Section 8.7 takes stock and concludes the chapter.

8.2 Kinds of Knowledge As I observed above, many philosophers hold that there are different kinds of knowledge. The short list I  gave can easily be extended with further proposals for distinguishing between different kinds of knowledge. Several virtue epistemologists distinguish between low-​grade and high-​grade knowl­ edge (cf. Baehr 2011; Battaly 2008; Greco 2010; Sosa 2007). On one interpretation of contextualism, it entails that knowledge has many different and René van Woudenberg’s contribution to this volume). While I  have considerable sympathy for these ideas, I will leave them to one side here, because the existence of both robust normative facts and supernatural facts is controversial.

912

1 9 2   • 

Jeroen de Ridder

senses, varying with the strength of your epistemic position (DeRose 2009; Van Woudenberg 2005). In theorizing about justification, many have found it useful to distinguish basic from non-​basic knowledge or non-​inferential from inferential knowledge (this distinction is key for any foundationalist theory about justification or knowledge; see, e.g., DePaul 2011). Perhaps there is such a thing as tacit knowledge, as opposed to explicit (or explicable) knowl­edge (Polanyi 1958). Maybe self-​knowledge is interestingly different from knowl­ edge of the rest of the world (cf. Cassam 2014; Fernandez 2013; Moran 2001). Recent epistemology has seen a number of proposals for making sense of group knowledge or social knowledge (Bird 2010a; Brady and Fricker 2016; De Ridder 2014; Gilbert 1989; Lackey 2014; List and Pettit 2011; Miller 2015; Sirtes, Schmid, and Weber 2011; Tuomela 1992). New work at the intersection of philosophy of mind and epistemology is exploring the notion of extended knowledge, that is, knowledge that is extended beyond the mind into the world (Carter et al. 2014; Carter et al. 2018). I take it that none of these distinctions are set in stone. For instance, Stanley and Williamson (2001) have argued that knowledge how is reducible to propositional knowledge and Matthew Benton (2017) has recently proposed that knowledge of persons is different from both propositional knowledge and knowledge by acquaintance. For present purposes, it is enough if this list makes it plausible that the word “knowledge” in fact masks a diversity of different kinds and sub-​kinds of knowledge. On the basis of what do philosophers distinguish among these different kinds of knowledge? It seems to me that there are a number of differences behind the distinctions. Without pretending to be exhaustive, I want to survey a few of them, in order to draw on them later to elucidate a feasible distinction between scientific knowledge and other kinds of knowledge. The point of this survey is neither to establish the reality of these kinds conclusively, nor to present the distinctions between them in rigorous detail. Obviously, a lot more would be needed to do that.2 A. Object:  One way to distinguish kinds of knowledge is by looking at the objects knowledge takes: differences in the kinds of things that are known. Propositional knowledge, of course, takes propositions as its object, but knowledge-​wh has a number of different objects, at least at first 2. For the sake of brevity, I will not discuss those kinds of knowledge that wear their differences on their sleeves, such as non-​inferential vs. inferential knowledge, individual vs. group knowl­ edge, and extended vs. non-​extended knowledge.

913

Kinds of Knowledge, Limits of Science  • 

193

sight:  persons (knowledge-​who), locations (knowledge-​where), times (knowledge-​when), objects (knowledge-​which), and whether-​clauses (knowledge-​whether).3 Different sorts of objects also seem to be what distinguishes knowledge-​that (and knowledge-​wh) from knowledge-​ how. The latter has something like skills or ways to do something as its objects and these are arguably different from propositions and wh-​ objects. Knowledge by acquaintance in turn differs from these other kinds of knowledge because it presumably has real things or persons, rather than propositions about these things or persons, as its objects. This is not to suggest that any difference in the objects of knowledge gives rise to a different kind of knowledge. Clearly, there is no interesting difference in kind between knowledge of stuff in my office and outside it. B. Mode of relating to the object: Differences between kinds of knowledge can also have be based in how a subject relates to what is known. Some examples can clarify what I have in mind. i. When you have propositional knowledge, you can be related to the proposition known in a purely theoretical way: knowing that p need not have any practical import. Knowledge-​how, however, is practical by definition, in the following sense: You cannot possess knowledge-​ how without having certain skills and being able to employ those skills in some suitable range of circumstances.4 You typically cannot know how to ride a bicycle if you have never ridden one. Nor can you know how to do long divisions if you have never been instructed in doing them or practiced doing them. ii. Some forms of knowledge necessarily involve (the possibility of ) mental access to what is known. When you know something, you can bring the object of your knowledge before your mind’s eye, so to speak. Propositional knowledge, knowledge-​wh, and knowledge by acquaintance are like this. Other kinds of knowledge can remain wholly implicit (although they need not necessarily be so). You can know how to do something without ever mentally representing your skill. Similarly, it is the very definition of tacit knowledge that it isn’t explicit; it manifests itself in cognitive and practical behavior, but 3. But perhaps all such wh-​clauses are ultimately reducible to one or more propositional clauses, so that knowledge-​wh is a subclass of propositional knowledge (cf. Parent 2014 for discussion). 4. Or perhaps having been able to employ your skills; someone who’s recently lost a hand might still know how to play the piano even if she cannot do so any longer.

914

1 9 4   • 

Jeroen de Ridder

isn’t accessed and represented consciously. The relation between yourself and the object of your knowledge isn’t mediated by mental awareness. iii. Another difference is whether your access to what is known is mediated by some representational structure. Propositional knowl­ edge, at least as traditionally conceived, is conceptually structured.5 Relevant facts are captured in the form of propositions that relate various concepts to one another and apply them to the situation at hand. To know that there is a cat on the mat involves some sort of easily accessible internal representation that has as its content a cat on the mat. Knowledge by acquaintance, in contrast, is not supposed to involve such an intermediate vehicle.6 It is a form of direct awareness of an object or person, unmediated by descriptions, propositions, concepts, or similar representational structures. To know your headache by acquaintance, for instance, is not to know that you have a headache or to apply some concept to it, but to be (or have been) directly aware of it. Something similar goes for knowledge-how. To know how to do something does not require having an internal representation of your doing it (although it of course it doesn’t exclude it either). This is why it is often claimed that knowledge-how can be tacit and need not be explicable in words or imagery. C. Strength of epistemic position: A further basis for distinguishing kinds of knowledge lies in the strength of your epistemic position vis-​à-​vis the object of your knowledge. There are at least two ways to make sense of this. i. The first is in a purely quantitative manner. A straightforward way to think of this is in terms of how strong your justification (or warrant) is. If it is strong enough, you have knowledge. If it’s even higher, you know for sure, or you’re certain. Suppose you see a horse

5. Although recent work in enactivism and embodied cognition conceives of belief without any internal representations whatsoever (cf. Hutto and Myin 2013). 6. Cf. Bertrand Russell’s characterization of it: “I say that I am acquainted with an object when I have a direct cognitive relation to that object, i.e., when I am directly aware of the object itself. When I speak of a cognitive relation here, I do not mean the sort of relation which constitutes judgment, but the sort which constitutes presentation. In fact, I think the relation of subject and object which I call acquaintance is simply the converse of the relation of object and subject which constitutes presentation. That is, to say that S has acquaintance with O is essentially the same thing as to say that O is presented to S” (Russell 1910: 108 (although Russell didn’t think we have knowledge by acquaintance of persons); cf. Gertler 2012 for recent accounts).

915

Kinds of Knowledge, Limits of Science  • 

195

approaching about 2000 feet away. If your eyesight is perfect and you’re in clear daylight, you’ll be certain that there is a horse there. However, when there’s a fog, you can’t be very sure, even if you do discern a horse-​shaped figure with enough clarity to know that there’s a horse approaching. Strength of justification varies with how much justification you have from one source (eyesight under more or less suitable circumstances) and with the number of different sources of justification (perception, testimony, memory, reasoning, etc.). On such a purely quantitative approach, however, there are no non-​arbitrary sharp boundaries we can latch onto to distinguish between different kinds of knowledge. At best, we could draw up rough categories with context-​sensitive and vague boundaries. ii. The second way to think about the strength of epistemic positions is in a qualitative way. Suppose you’re an externalist about knowl­ edge and think that reliably produced belief without defeaters suffices for knowledge. Then, your belief that you see a horse in the above example will constitute knowledge. But now suppose you also reflect on your belief and form the meta-​belief that it was formed in appropriate circumstances, through your reliable eyesight, on the basis of clear visual appearances. Or suppose that you have considered that your friend, who testified to you that a horse would be coming, is one of the most honest people you know and that she has never lied to you. Arguably, your epistemic position—​while continuing to constitute knowledge—​is now better. Suppose next that you happen to be an optometrist and know a lot about the physics and neurology of visual perception. Presumably, your epistemic position would be even better still. The general thought is this: There are different kinds of epistemic standing that might constitute your having knowledge and these standings give rise to different kinds of knowledge. Purely externalist standing might suffice for knowledge, but if you also have one or more characteristically internalist standings vis-​à-​vis a proposition, your knowledge is of a different, possibly superior, kind (cf. Sosa 2007: 129, 2009: 193).7 Such internalist standings include reflective meta-​beliefs about your first-​order

7. But cf. Grimm (2016) who argues that the value of such reflective states is not epistemic but moral.

916

1 9 6   • 

Jeroen de Ridder

belief: beliefs about what the grounds for your belief are, about the reliability of your belief-​forming mechanisms, about your belief being properly based on your grounds, about how your belief coheres with other things you know, or about your intellectual motivations in forming the belief. Differences like these are behind Sosa’s distinction between animal and reflective knowl­ edge (Sosa 2007, 2009), Greco’s distinction between knowl­edge and understanding (Greco 2010:  7–​ 10), and behind virtue epistemologists’ distinction between low-​grade and high-​grade knowledge (Baehr 2011:  40ff.; Battaly 2008:  652–​659; Roberts and Wood 2007: 109; Zagzebski 1996: 273–​280). D. Mode of acquisition: A final basis for distinguishing different kinds of knowledge involves the way in which knowledge is acquired. Some kinds of knowledge can only be acquired in specific ways. A  priori knowledge, if it exists, is only available through something like rational intuition or insight. First-​personal knowledge (such as self-​knowledge) is only available through methods that are exclusively available from the first-​personal perspective through faculties like introspection and proprioception. To the extent that there is first-​personal knowledge, it can only be acquired through these methods; there is no other route to it. Something similar goes for second-​personal knowledge. Some philosophers have argued that there are truths, particularly about other people, we can only come to know by interacting directly with them. No amount of third-​personal descriptions or propositional knowl­ edge about those people can give you such knowledge (and of course it also cannot be obtained from the first-​personal perspective). Says Eleonore Stump: Second-​person experiences cannot be reduced to first-​person or third-​ person experiences without remainder, and so they cannot be captured by first-​person or third-​person accounts either.  .  .  .  Knowledge of persons accessible in second-​person experiences is not reducible to knowledge that. (Stump 2010: 78)

The kind of knowledge that is most familiar in epistemology, third-​ personal knowledge, is available through various sorts of familiar methods (perception, testimony, reasoning, etc.) that do not depend crucially on the internal perspective of particular people or on interactions between people.

917

Kinds of Knowledge, Limits of Science  • 

197

This concludes my overview of possible ways to account for different kinds of knowledge. I will draw on this list as a toolkit to get clearer about what is specific about scientific knowledge, as opposed to everyday knowledge.

8.3 Scientific Knowledge: First Pass One might expect to find building blocks for a general characterization of the nature of scientific knowledge in the debate about a demarcation criterion between science and non-​science or in contemporary epistemological reflections on scientific knowledge. It turns out, however, that neither of these places have much to offer on the topic. Reflecting on the demarcation debate in 1983, Larry Laudan concluded that it had come to naught: It is probably fair to say that there is no demarcation line between science and non-​science, or between science and pseudo-​science, which would win assent from a majority of philosophers. Nor is there one which should win acceptance from philosophers or anyone else. (Laudan 1983: 112) This remains the majority opinion and the demarcation question has largely lost the interest of philosophers of science.8 Equally remarkably, neither recent mainstream epistemology nor philosophy of science say much about the general nature of scientific knowledge. The former remains mostly focused on more mundane instances of knowledge, while the latter appears to have become wary of general claims about science, focusing instead on narrower philosophical questions pertaining to the special sciences.9 This doesn’t mean that we’re completely at a loss for ideas about the distinctive nature of scientific knowledge. For starters, we can revisit the old philosophical ideal of scientific knowledge:  scientia or episteme. As used by philosophers such as Aristotle and Descartes, this notion of knowledge 8. However, there are a few recent dissenters, who think a feasible demarcation criterion can be provided after all. See Ladyman et al. (2012) and several contributors to Pigliucci and Boudry (2013). 9. Some notable exceptions are Alexander Bird’s work on the epistemology of science (Bird 2010b), John Hardwig’s (1985, 1991)  papers on trust and epistemic dependence in science, Elizabeth Fricker’s (2002) paper on testimony in science, and Philip Kitcher’s (2002) epistemology-​oriented overview of prominent debates in the philosophy of science. Even though these authors make insightful points about the goal of science, scientific progress, scientific inference, and the social character of scientific knowledge, none of them attempt to characterize what’s special about scientific knowledge when compared to nonscientific knowledge.

918

1 9 8   • 

Jeroen de Ridder

entailed a particularly elevated epistemic status, which still seems like a fitting aim for science. In science, we’re not content to “just know,” we want to understand the reasons why things are as they are and to grasp how everything hangs together. I’ll say more about this in the next section. Next, one of the few contemporary philosophers of science whose work does contain suggestions about the nature of scientific knowledge is Bas van Fraassen. According to him, it is “objectifying knowledge.” It has a special mode of representing its objects, which we don’t find in other forms of rational engagement with the world. I’ll elaborate on his thoughts in section 8.6. These two proposals set the agenda for the rest of the chapter: I’ll develop them in more detail and explore what they imply for the limits of scientific knowledge.

8.4 Scientific Knowledge as High-​Grade Knowledge If anything, scientific knowledge is supposed to be high-​grade knowledge. This has been part and parcel of philosophical reflection on knowledge from the very beginning. Plato already contrasted episteme, “genuine” knowl­ edge, with mere doxa or opinion. In a similar vein, Descartes distinguished scientia from cognitio. In contemporary epistemology, this distinction is typically rendered as the contrast between knowledge and mere (true) belief. While there is something to be said for this rendering, it is clear that earlier philosophers had a more demanding epistemic state in mind than what many contemporary epistemologists—​especially those of an externalist bend—​ call knowledge. For Plato and Descartes, as for many others (BonJour 1985; Fumerton 1995), merely having a reliably produced true belief or a belief that is true because of an exercise of one’s intellectual virtues (understood as reliable faculties) wouldn’t have sufficed for “genuine” knowledge. Arguably, then, a more fitting contemporary rendering of episteme and scientia would be high-​grade knowledge, scientific knowledge, or understanding rather than simply knowledge.10 But regardless of these historical-​interpretive questions, there is a distinction between low-​grade and high-​grade knowledge and it makes sense to think of scientific knowledge as high-​grade. What I said above in section 8.3 (under C) about strength of epistemic position provides two options to think about high-​grade knowledge. First, we can construe it as differing quantitatively from low-​grade knowledge. High-​ grade knowledge is knowledge that has a lot of positive epistemic status. This certainly clarifies part of what is epistemically good about scientific knowledge. 10. See Grimm (2001) and Greco (2014) for arguments to this effect.

91

Kinds of Knowledge, Limits of Science  • 

199

Scientists observe phenomena very carefully and systematically; use the best available observation techniques; repeat their observations where they can with independent methods; and engage in a lot of double-​checking, both of their own observations and inferences as well as those of their peers. As a result, at least the low-​level observational claims on which scientific models and theories are built typically have a great deal of justification. The results of science, of course, form a complex edifice of claims, models, and techniques with different levels of justification. Even if low-​level observational claims are typically highly justified, we cannot assume that such high-​grade justification transfers losslessly all the way up to higher-​level theoretical claims, since the latter can be quite loosely connected to the former via more or less risky interpretational and ampliative inferential steps. But in so far as high-​level claims are based on lower-​level ones through valid interpretations and inferences, and have withstood independent critical scrutiny, they, too, can be highly justified. Do such quantitative differences amount to a difference in kind between scientific knowledge and more mundane types of knowledge? Does low-​grade knowledge transform into a different kind of knowledge once the amount of justification or warrant reaches a certain level? As I said above, that seems implausible. It is unclear and arbitrary both where the threshold between low-​grade and high-​grade knowledge should be and why a continuum of levels of justification would divide into just two kinds of knowledge, rather than three, four, or an infinity. There may be significant quantitative differences between the amount of justification for ordinary everyday knowledge claims and low-​level scientific knowledge claims, but this doesn’t constitute a difference in kind. The second—​and more plausible—​way of thinking about high-​grade knowl­ edge construes it as qualitatively different from low-​grade knowledge. We can take a cue from Ernest Sosa’s virtue epistemology (Sosa 2007, 2009, 2011). He distinguishes between animal and reflective knowledge. Animal knowledge is true belief that is creditable to an agent’s exercise of intellectual virtue. As Sosa thinks of it, an intellectual virtue is a reliable belief-​forming faculty. A subject thus has animal knowledge when her belief is true because of the reliability of the faculty used in forming it. It is important to note that animal knowledge does not require the agent to have any kind of access to, or beliefs about, the reliability of her cognitive faculties, or the fact that her belief is true because of the exercise of reliable faculties. It is an entirely externalistic account of knowledge.11 11. There are different construals of the internalism/​externalism distinction in epistemology. Here I am working with one default understanding of it in terms of whether or not knowl­ edge requires cognitive access to the factors that confer the knowledge status on one’s belief (BonJour 2010).

02

2 0 0   • 

Jeroen de Ridder

Reflective knowledge demands more. It requires “that the knower have an epistemic perspective on his belief, a perspective from which he endorses the source of that belief, from which he can see that source as reliably truth conducive” (Sosa 2009: 135). Hence, to know reflectively, you not only need animal knowledge but also a “perspective” on it, that is, a true meta-​belief to the effect that your first-​order true belief was indeed formed through reliable faculties. Moreover, this meta-​belief, too, should be formed by a reliable cognitive faculty and it should be true because it was formed this way. For you to have reflective knowledge that there is a cat on the mat, you should not only form a true belief through your use of reliable vision, but also reliably form a further true belief that your vision is indeed reliable (in the current circumstances) and that this accounts for why you have a true belief about the cat on the mat. Reflective knowledge thus requires internalist epistemic standing:  access to some of the factors that account for the strength of one’s epistemic position. Abstracting away from the particulars of Sosa’s position, we can think of high-​grade knowledge more generally as knowledge that requires internalist standing. There are several options for such standing (cf. Alston 2005 for an overview): access to (a) the grounds for your belief, (b) the quality of the grounds for your belief (or the reliability of the cognitive process by which your belief has been produced), (c) the fact that your belief has been properly based on good grounds (or non-​deviantly produced by a reliable belief-​forming mechanism), (d) the coherence of your belief with your further beliefs. These standings can also be combined. The most high-​grade (and demanding) type of knowledge would involve all of (a)  through (d). In order to have high-​ grade knowledge, you not only need to find yourself with a reliably (or virtuously) formed true belief, you also need to know something about how or why you find yourself in this epistemic position. On this proposal, the qualitative difference between low-​grade and high-​grade knowledge is that the latter requires not just more of the same status that low-​grade knowledge also has, but another sort of epistemic standing added to it, namely internalist awareness. This construal of high-​grade knowledge fits well with scientific knowledge. What matters in science is not just reliable (virtuous) observation and reasoning (i.e., animal knowledge), but also insight into why your observations and inferences are reliable (i.e., reflective knowledge).12 This is why scientists 12.  To be clear:  The proposal is to think of internalist standing as a necessary condition on scientific knowledge and not a sufficient one. There may be nonscientific knowledge with internalist standing too.

210

Kinds of Knowledge, Limits of Science  • 

201

care so much about methodology, research design, proper experimental setup and execution, and correct use of statistics and other inferential methods. Scientific knowledge thus requires internalist epistemic standing:  access to the grounds for one’s conclusions and to the epistemic goodness of those grounds.13

8.5 The Limits of High-​Grade Knowledge Let’s now consider what follows from the above proposals for the limits of scientific knowledge, thought of as high-​grade knowledge. Since the first proposal—​high-​grade knowledge as quantitatively highly justified knowledge—​seemed unpromising, I’ll be relatively brief about it. Are there limits to how much justification we can acquire for certain propositions? Presumably there are, and they have to do mostly with contextual factors, such as what is known, what methods are available, technological developments, and our spatiotemporal position in the universe. To illustrate, consider Galileo. Given what was known and what could be observed and measured in his time with a reasonable degree of precision, it was very difficult, if not outright impossible, for him to acquire a lot of justification for heliocentrism. Even though later research confirmed that he was right, it is probably false that, at the time, Galileo really knew that heliocentrism was true and certainly false that he knew it with a high degree of justification. And so it goes: As our knowledge, methods, and technology develop and become more reliable, we can acquire more justification for extant beliefs and acquire new highly justified beliefs. There are no in-​principle limits to scientific knowledge in sight here, with two important but uncontroversial exceptions. First, given the laws of nature and our position in space-​time, parts of the universe must remain unobservable to us, as they are located outside our light cone. It is physically impossible for information from these parts of the universe to reach us (barring sci-​fi scenarios). We might still be able to know various things about these parts of the universe by inferring them from our knowledge about the observable part of the universe, but verifying this knowledge through direct observation and measurement is impossible. Hence, there is at least a limit on the kinds of justification we can possess here. Whether this translates into a limit on the 13. Elsewhere, I have argued that this feature makes much scientific knowledge collective knowl­ edge (De Ridder 2014). Often, no single individual is cognitively able to bear the burden of having such internalistic epistemic standing.

20

2 0 2   • 

Jeroen de Ridder

amount of justification we can have is an open question, the answer to which depends on how reliable our inferential methods are. Second, our natural human cognitive equipment might pose limits to what we can know with high degrees of justification.14 Given the sorts of brains and perceptual capacities that we have, there must be limits to what we can grasp and know. Of course, our perceptual capacities have been greatly enhanced and extended by all manner of equipment and this process will no doubt continue. But even so, everything we learn must come to us in sensory modalities that we can take in. Our brains, too, might pose limits to what we can know. In the philosophy of mind, so-​called mysterians have argued that we will never be able to know the full explanation of consciousness because that is impossible given the abilities of brains like ours (McGinn 1991). Perhaps a similar case could be made for other highly complex phenomena. Less radically, and in line with the present discussion, perhaps there are propositions for which we cannot acquire a very high degree of justification because their complexity outstrips our cognitive abilities. If that is indeed the case, it would follow that we cannot know such propositions scientifically. Let’s consider the second way of thinking about high-​grade knowl­ edge: scientific knowledge as knowledge with internalist epistemic standing. As noted above, there are various sorts of internalist standing. We can get a handle on one class of beliefs for which it is impossible to acquire internalist standing of kind (a)—​access to the grounds for your belief—​ by revisiting popular foundationalist views about epistemic justification. Foundationalists distinguish between basic and non-​basic beliefs (DePaul 2011; Huemer 2010). The latter are beliefs that we accept on an inferential basis, on the basis of other beliefs; and the former are accepted in a direct way, not on the basis of other beliefs. Ultimately, the foundational structure of all our non-​basic beliefs must bottom out in basic beliefs. If things go well epistemically, in the sense that our beliefs are justified, this foundational structure also lends justification to our beliefs. At the bottom, justified basic beliefs are justified, not because of their inferential connections with other beliefs, but in virtue of some other epistemically good-​making feature (e.g., their indubitability; self-​evidence; or their being produced by well-​functioning, basic belief-​forming mechanisms). Justified non-​basic beliefs, in turn, get

14. I’m disregarding the possibility of radical changes in our cognitive make-​up as a result of evolutionary or (bio)technological developments.

230

Kinds of Knowledge, Limits of Science  • 

203

their epistemic status from other (justified) non-​basic beliefs and ultimately from justified basic beliefs. From this, we can see that justified basic beliefs, by definition, lack one type of ground: other (justified) beliefs. But some of them will have other grounds that are accessible upon reflection. Consider a lowly perceptual belief: I see that there is a tree in my garden. Assume this belief of mine is justified (there was nothing wrong with my eyes, conditions for tree-​perception were suitably normal). Presumably, the ground for this belief is something like my perceptual experience, my being appeared to tree-​ly, or it seeming to me that I see a tree. For such beliefs, internalist standing (a)—​access to the grounds for belief—​is possible. But some basic beliefs appear to be groundless. What would constitute the grounds for beliefs that basic logical, mathematical, or conceptual truths hold, or for simple memory beliefs? Such beliefs simply strike us as true, but they aren’t based on perceptual or other experiences. At best, they are accompanied by non-​experiential seemings, but such seemings are not their grounds. The relation between them and the relevant beliefs is quite different from the paradigmatic grounds-​for-​belief relation as it holds between experiential beliefs and the experiences on which they are based, or between inferential beliefs and the beliefs on which they are based. For one thing, the seemings that accompany these beliefs are much less specific and conceptually structured than the typical grounds for beliefs. Seeing a tree is an experience with rich content, which can only serve as the ground for fairly specific beliefs about that tree. The seeming that accompanies belief that modus ponens is valid is just a general sense that it must be so, which is hardly distinguishable from the seeming that accompanies, say, belief that 2 + 2 equals 4. We don’t have different seemings for various basic conceptual, logical, and mathematical truths, whereas our experiences for different experiential beliefs are clearly different (no one would confuse seeing a tree with hearing a train approaching). Similarly for memory beliefs: When I remember that I lived in Rotterdam in 1999 this is just how it seems to me. A different memory would elicit roughly the same seeming (although there might be differences in how compelling and strong the seemings feel, but it’s not as if each memory belief has its own unique seeming). While this is not a rigorous argument that there are groundless basic beliefs, it at least suggests that some categories of basic beliefs lack grounds. If this is right and there are indeed groundless but nonetheless epistemically upstanding beliefs, then they constitute a limit for scientific knowl­ edge. Perhaps they can be known (that will depend on your exact theory of

024

2 0 4   • 

Jeroen de Ridder

knowledge), but they cannot be known scientifically, since it is impossible for these beliefs to have the required internalist standing. Internalist standing of type (b) requires access to the epistemic goodness of the grounds for your belief (or of your belief-​forming process). This gives rise to an in-​principle limit for scientific knowledge. To know—​or believe with justification—​that your grounds are truth-​conducive or your belief-​ forming processes reliable, requires you to form an epistemically justified perspective on the reliability of one or more of your belief-​forming practices. There are reasons to doubt that this can be done. As Alston (1993) has argued, arguments for the reliability of our basic belief-​forming practices are inevitably epistemically circular: Epistemically justified belief in their premises presupposes the truth of their conclusion. When you want to argue for the reliability of sense perception, you inevitably rely on claims that are only justified if sense perception is in fact reliable. For instance, to know that my perceiving a tree is an epistemically good ground for my belief that there is a tree, I rely on the claim that my visual perception is reliable (in suitable circumstances). If I were to try to justify that claim, I will end up appealing to past cases of visual perception that turned out to be successful. But to know that these past cases were successful, I must rely on either my own visual perception, that of others, or perhaps on other modes of sensory experience. As Alston argues at length, there is no escaping this circle: To believe the premises of arguments for the reliability of sense perception, you must presuppose their conclusion. And the same goes for other basic practices of belief formation: memory, induction, testimony, rational intuition, etc. Opinions differ about what general lessons we should draw from this. Whether epistemic circularity really prevents us from knowing that our basic belief-​forming practices are reliable, depends on further views about the nature of knowledge and justification.15 This doesn’t matter for my purposes. The conclusion I’m drawing is that there is at least a kind of limit to our access to the quality of our grounds for belief—​and thus another limit to what we can know scientifically. Epistemic circularity precludes one thing that seems particularly relevant for high-​grade knowledge: to wit the possibility of giving an unproblematic (i.e., non-​circular) argument for the conclusion that the grounds for some of your beliefs are epistemically good. Since epistemic circularity affects all of our basic belief-​forming practices,

15.  See Stroud (1989, 1994, 2004)  and Sosa (2004:  315–​317, 2009:  166–​177, 204–​120) for discussion.

205

Kinds of Knowledge, Limits of Science  • 

205

this will be true for the grounds of our perceptual beliefs, memory beliefs, testimonial beliefs, inductive beliefs, etc. That is, pretty much all of our beliefs.16 Thinking about internalist standing of type (b)—​access the quality of the grounds for our beliefs—​yields a further limit for scientific beliefs in particular. This one is not an in-​principle limit, but an empirically based and developing one. Start with the platitude that science, like all of our cognitive endeavors, is fallible. We should not think that the mere possibility of error throws the quality of our grounds for belief into doubt. That’s a quick route to radical skepticism. We can be more specific about how fallibility creates problems for scientific knowledge. First of all, Laudan’s (1981) famous pessimistic meta-​induction suggests that we should not be too confident about the epistemic status of our current best scientific theories. The history of science shows a steady and sure turnover of theories (explanations, models): Older theories are rejected as mistaken in favor of newer and superior ones. Most past scientific theories have turned out to be false. A simple inductive argument should lead us to conclude that, most likely, our current scientific theories will one day be rejected as false. Granted, the pessimistic meta-​induction is not uncontested. Several philosophers have argued that, once we restrict the inductive basis to developed, mature scientific theories or to the core tenets of scientific theories, the turnover rate becomes much less impressive. Instead, we witness significant amounts of continuity and accumulation in the development of science.17 Such observations, however, work with the benefit of hindsight. In retrospect, we can see which elements and aspects of scientific theories are preserved in successor theories, but scientists who were working on those earlier

16. Perhaps some readers are tempted to take this argument as showing that high-​grade knowl­ edge of anything is impossible. Since (a) epistemic circularity besets all our basic belief-​forming practices and (b) all our beliefs ultimately spring from one or more of these practices, it might seem that we lack the right sort of access to the quality of the grounds for all of our beliefs. I don’t think this follows, though. The thing to see is that there are other considerations bearing on the quality of the grounds for a belief, besides a non-​circular argument for the reliability of the relevant belief-​forming practice. If I see someone familiar approaching nearby, I can notice that my observation takes place in broad daylight, that I often see this person around here, that I’m not under the influence of any substances, etc. All these things tell me something about the quality of the grounds for my belief, even if I don’t have an unproblematic argument for the reliability of visual perception in general. Hence, the argument doesn’t automatically generalize into a skeptical one. 17. See Leplin (1981) and Chakravartty (2007: Ch. 2) for discussion.

026

2 0 6   • 

Jeroen de Ridder

theories would have had at best a partial and uncertain view of which parts of their theories would stand the test of time. The same will be true of current scientists. While these qualifications give us reason to be more confident about the quality of the grounds for long-​established central scientific claims and theories, once we approach the frontiers of science, judgments about the epistemic quality of the grounds for belief ought to become less and less certain.18 This conclusion is reinforced by recent work in scientometrics. John Ioannidis (2005) has shown that it is highly likely that most published findings are simply false! He draws this conclusion from a simple but realistic probabilistic simulation model that takes into account several parameters of possible research designs and the current socio-​economic organization of scientific fields. His startling conclusion is borne out by empirical scientometric investigations. Samuel Arbesman (2012: Ch. 3) reports that collections of what he—​misleadingly—​calls scientific facts have a “half-​ life.” For any given body of published scientific literature (journal articles, books), there is a more or less predictable time period within which half of it will have been overturned. For instance, after 45 years, biomedical experts reject half of the published findings on cirrhosis and hepatitis as false or superseded (ibid.:  28–​29). In another study, researchers found that after about 10 years, half of the papers in the prominent Physical Review journals were never again cited.19 While this doesn’t quite show that these papers are flat-​out rejected, it does indicate that they are no longer considered relevant. Surely, for some cases, this will be because they are considered false. The exact numbers for half-​lives vary among fields and publication types, but the general picture is clear:  For any novel published scientific claim, there is a good (i.e., higher than 50 percent on average) chance that it will be discarded sooner or later. Compare this with a situation in which you assess your own reasoning powers. When you argue, you try to be as careful as you can, and you thus have a pretty high initial confidence the conclusions you reach. You’re aware of the grounds for your conclusions and you think they are pretty solid. But

18.  My impression is that most practicing scientists are keenly aware of this, which is why, when pressed, they are usually careful in making explicit knowledge claims. Most will talk of plausibility and probability when characterizing the confidence they have in the latest findings. 19. That is, as far as the researchers could tell at the time. It remains a (distant) possibility that some papers will be resuscitated at some point in the future. This is unlikely to happen to a great many papers, however, so the general conclusion stands.

027

Kinds of Knowledge, Limits of Science  • 

207

then a trustworthy expert tells you that you only make sound inferences, say, 50 percent of the time. (Or, if you will, you discover this yourself by revisiting conclusions you drew from your past reasoning.) Surely, this undermines your having high-​grade knowledge of the conclusions of your reasoning. Perhaps you can maintain some high-​grade knowledge by comparing your reasoning with that of others and by finding out what sort of conclusions you have arrived at time and again through multiple independent strands of reasoning. But even so, you should definitely grant that the amount of high-​ grade knowl­edge you have is significantly less than you initially thought. The above empirical facts thus support the claim that there is a limit to what we can know scientifically. Relatively novel scientific claims cannot be known with robust internalist standing, because we do not have reliable access to the quality of the grounds on which they are based. These grounds may look epistemically superb to us at the time, but sober facts about the development of science demonstrate that there is a big chance that they are really not. At best, we may possess genuine scientific (high-​grade) knowledge of mature theories and claims. The upshot of this section is that construing scientific knowledge as high-​ grade knowledge gives us four in-​principle limits and one developing practical limit to what we can know scientifically: (1) facts and events beyond our light cone, (2) things that are beyond our cognitive capacities, (3) groundless basic propositions, (4)  the reliability of our basic belief-​forming processes, and (5) the great majority of recent published claims.

8.6 Scientific Knowledge as Objectifying Knowledge I’ll now consider the second proposal for thinking about the distinctive nature of scientific knowledge. This proposal focuses on the object of scientific knowledge or, more precisely, on the specific mode by which scientific knowl­edge represents or construes its object. It thus takes its inspiration from item B on the list in section 8.2. My primary source here will be the philosopher of science Bas van Fraassen, but the idea that there is something unique about the way in which science engages its objects of knowledge is not unique to him. Different versions of it can be found in Neo-​Kantianism (e.g., Rickert 1926)  and in the work of the Dutch philosopher Herman Dooyeweerd (1997 [1953]). To warm up to the idea that there is something distinctive about the way in which scientific knowledge represents its objects, I want to revisit the distinction between observing something and observing that something is the case

208

2 0 8   • 

Jeroen de Ridder

(cf. Van Fraassen 1980: 15).20 You see something fly by your window and notice that it’s a sparrow. Of course, you might have failed to notice that it was a sparrow, or even that it was a bird. But you would nonetheless have seen the sparrow flying by your window—​except you wouldn’t have recognized it as such. That is, you wouldn’t have observed that a sparrow flew by your window. A  natural way to understand the difference is that, in the former case, you have represented the object of your experience in a specific way or conceptually interpreted it, whereas, in the latter, you just have your “raw” experience without (much) conceptual interpretation and representation.21 This illustrates the thought that a cognitive relation—​in this case, seeing—​ can take its object in different ways. The object stays the same, but our way of engaging it, representing it, or cognizing about it is different. Something similar occurs in science, according to Van Fraassen. Scientific inquiry, he contends, is objectifying inquiry (Van Fraassen 2002:  156ff ). Science doesn’t just observe the world willy-​nilly, it objectifies the world and the things in it. Van Fraassen never gives a crisp and clear definition of what objectification amounts to, but he does discuss five defining features (ibid.: 160–​164). In general terms, the goal of objectifying inquiry is to take ourselves out of the picture: our contingent location in space and time; the contingencies of the senses we happen to have; and our subjective limitations, biases, preconceptions, etc. By doing so, science aims to study the world as it is independently of us and our observations and thinking.22 The first thing that is required for scientific inquiry—​as opposed to other forms of cognitive engagement—​is that a domain of study is delimited. Scientists specify in advance what sort of things they will study. This doesn’t mean that they say that physics is about physical stuff and biology about living beings; the idea is that scientists specify which specific properties and relations they will study by stipulating the quantities and parameters that are allowed to figure in their descriptions of the phenomena. It’s not that every scientist does this on her own every time she begins a new project. In everyday science, the quantities and parameters are given by the context of broader lines of inquiry, (sub)disciplines, and scientific fields.

20. Or, alternatively, between object perception and fact perception (Dretske 1969). 21. I don’t mean to take a stand on whether experience is always and necessarily conceptually interpreted. If that’s the case, then the difference between the former and the latter case should be cashed out in terms of interpretation with more and less specific concepts. 22.  In characterizing science as objectifying inquiry, Van Fraassen is laying out an ideal of science. He is not committed to the claim that actual science always fits this ideal perfectly.

029

Kinds of Knowledge, Limits of Science  • 

209

Second, not just any specification of parameters will do. Parameters must be independent of the scientists carrying out the research. Specifications in terms of how things look or feel to us will not do. This constraint was also what inspired the old distinction between primary and secondary qualities. Parameters must capture how the phenomena are in themselves and not how they appear to us.23 Independence is often interpreted as requiring quantitative measurement. Although quantification can indeed be a good way of making sure that parameters do not depend on us, it does not by itself guarantee the right sort of independence. We could quantify an object’s size as a multiple or fraction of our own length with perfect precision, but such a parameter would still improperly depend on us. Similarly, if we could devise reliable ways to quantify our subjective taste, that wouldn’t suddenly make it an appropriate parameter in, say, chemistry. If we think of ourselves as a measuring apparatus, the thought behind independence is that acceptable scientific parameters cannot depend on the contingencies of the sort of apparatus that we happen to be. Third, the set of properly independent parameters and quantities must be restricted to those that are relevant: Scientific questions and answers must be expressible in terms of the relevant parameters. Unfortunately, Van Fraassen doesn’t tell us in general terms what makes parameters relevant or irrelevant, but he does mention Galileo’s list of primary qualities for physics (shape, position, motion, contact, and number) as an example. The point of this feature of objectifying inquiry is to avoid introducing empty terminology: meaningless or uninformative notions that lack genuine empirical content. Using a substance’s “dormative virtues” to explain why people get sleepy when they use it exemplifies such empty verbiage. Even if we had a way of measuring such virtues, they would be unsuitable as a parameter in good science as they don’t latch onto a relevant or informative quantity. What counts as relevant can develop over time. Before Newton, a parameter like force, describing mysterious action at a distance, would not have been acceptable, but it is a central element of Newtonian mechanics. We can think of the previous three features as creating a language for scientific inquiry.24 The fourth feature involves experimentation and observation:  “nature put to the question” (ibid.:  163, my italics). Scientists observe 23. Of course, when humans are the object of study, parameters will depend on us in the sense that they will describe qualities of human beings. But they must not do so in a way that depends on how these qualities look, feel, or seem to us. 24.  Note that the metaphor of language is mine. Van Fraassen might not approve of it. It should certainly not be associated with the positivists’ ideal of an observational and theoretical language.

120

2 1 0   • 

Jeroen de Ridder

the phenomena carefully, often under meticulously prepared (experimental) conditions, and record their observations in the previously accepted language of science, that is, in terms of the relevant parameters and quantities. The result is a systematic set of data, or a “data model,” which can subsequently be studied to find patterns. The fifth feature characterizes scientific theorizing. Hypotheses, models, and theories are introduced to account for the data. They, too, must respect the language of science and must therefore be formulated in terms of the relevant parameters and quantities. But sometimes it will be necessary to introduce novel quantities at the level of theory construction. What quantities are permissible? Van Fraassen says that it is impossible to give a general and informative answer to this, but it depends on the preceding theoretical development of a scientific (sub)discipline, on the relation of the proposed novel quantities to older models and theories, and, for the special sciences, on the parameters and quantities already in use in more fundamental sciences. Theoretical quantities can only be observed indirectly, by observing the quantities with which they are connected in the model or theory. The quality of hypotheses, models, and theories is ultimately decided by how well they fit the data.25 Now we can see more clearly what it might mean to say that scientific knowledge involves a special mode of representing of its object. Objectifying inquiry requires the deliberate construction of a system of representing the world—​a scientific language of sorts—​consisting of measurable parameters and quantities that are properly independent of us and that delimit the possible domain of inquiry. This language constrains what can be expressed and observed, and what can make it into scientific theories, because every observation must be put in terms of the relevant parameters and quantities, with possible additions of theoretical quantities. Let me guard against two potential misunderstandings. First, to draw attention to the objectifying character of scientific knowledge is not in any way to suggest that science distorts reality. On the contrary, the fit between  a scientific account of a phenomenon and that phenomenon itself might be excellent. But even in cases of optimal alignment, science homes in on its objects in a specific way, by conceiving of them in pre-​specified 25.  Van Fraassen is optimistic about how widely his notion of objectifying inquiry applies: “With some minor adjustments the pattern of objectifying inquiry is certainly gen­ eral enough to include the sciences and in fact all disciplines pursued with academic discipline” (ibid.: 165, my italics). This strikes me as a bold claim. I do not find it easy to see how various forms of humanistic inquiry could fit the mold of objectifying inquiry. But it is sufficient for my purposes if objectifying inquiry at least captures what is special about scientific inquiry.

21

Kinds of Knowledge, Limits of Science  • 

211

relevant terms that we ourselves have authored. The actual phenomena always outstrip their scientific descriptions, in the sense that they have more properties—​perhaps even radically different properties—​than any particular scientific description captures. Second, none of this should be read as implying that scientific knowledge is a social construction, or that the acceptability of hypotheses and theories is largely determined by social processes. According to Van Fraassen, the ultimate arbiter for the success of scientific theories—​and thus for the parameters and quantities in which they are formulated—​is their fit with the data. While he recognizes that observation is theory-​laden and that the interplay between theory and observation is often subtle and complex, he remains firmly committed to empiricism. In the long run, empirical observations determine what goes in science.

8.7 The Limits of Objectifying Knowledge Let’s now consider whether this conception of scientific knowledge poses limits on what we can know scientifically. The first three features of objectifying inquiry discussed above impose one limit on science. Something that cannot be described in the language of science—​that is, in terms of items on the list of relevant parameters and quantities that define the domain of inquiry and that are allowed to figure in the descriptions of measurements and in theories—​cannot figure in science. This is not to say that the sciences all have their own sets of parameters and quantities that are fixed once and for all, so that some phenomena can never become objects of scientific study. This would make radical innovation and novelty—​genuine scientific development and progress—​impossible. Without the introduction of new vocabulary to describe, say, electromagnetic forces, physics would have remained stuck in the first half of the 19th century. So limits imposed by the set of accepted parameters and quantities are relative to time and the state of scientific development. They can be overcome when inexplicable phenomena, puzzling experiments, or theoretical difficulties force scientists to introduce novel parameters or to develop new theories. Giving specific examples of this requires in-​depth knowledge of cutting-​edge scientific developments, but perhaps the well-​known incompatibility of general relativity and quantum mechanics in physics can serve as a toy illustration. Both theories are extremely well established, but also known to be mutually incompatible. Hence, they cannot both be correct as they stand. Recasting this in terms of Van Fraassen’s notions, we might say that the currently accepted

21

2 1 2   • 

Jeroen de Ridder

language of physics precludes us from finding a solution to the recognized inconsistency. To unify both theories, novel parameters and quantities might be needed.26 At any given time, the currently accepted language of science puts limits to what science can study. This isn’t a limit that is fixed once and for all, as science may progress by revising and extending its vocabulary, but it is a limit nonetheless. Another limit of scientific knowledge derives from the fact that parameters used in science must be independent of observers. This requirement excludes the use of parameters that depend on how things seem to us or on what it is like for us to experience something. In other words, it precludes what are often called phenomenal qualities from playing any role in scientific knowledge. This is reasonable enough for most parts of the study of the natural world,27 but significantly less so for scientific and humanistic disciplines that involve individual and collective human experience in its many varieties: psychology; anthropology; sociology; economics; the full range of the humanities; and many interdisciplinary fields like communication science, business administration, political science, etc. Of course, not every inquiry in these disciplines must involve parameters tied to the subjective characteristics of human experience, but many of them in fact do. That is because people’s subjective takes on the world will often be an important factor in explanations of behavior. This holds for the study of history, for studying works of art or other cultural achievements, and for many other social-​scientific and humanistic inquiries. It would thus seem that Van Fraassen’s conception of science poses severe limitations on what can be known scientifically. So much so, in fact, that Van Fraassen’s suggestion that the objectifying style of inquiry is characteristic for all intellectual pursuits carried out with academic rigor loses much of its plausibility. Perhaps, however, these observations bring out an ambiguity in Van Fraassen’s characterization of the independence requirement. It might be interpreted (a) as banning parameters and quantities that are essentially tied to the subjective character of human experiences from science altogether; or (b) as banning only those parameters and quantities that are essentially tied

26. I’ll readily admit that what I say here is speculative and vague. I nonetheless hope that it conveys the thrust of how we could think about one kind of limit of science with the help of Van Fraassen’s picture of science. 27. Although it would seem that scientific subdisciplines that are essentially concerned with things that affect human experience—​such as nutrition science, pharmaceutical science, medicine, and many engineering disciplines—​cannot avoid taking onboard such parameters.

213

Kinds of Knowledge, Limits of Science  • 

213

to the subjective character of human experience that are not intersubjectively accessible or verifiable, that is, that are not independent of the particularities of the individuals conducting the inquiry. The worries I  raised in the previous paragraph primarily affect the first interpretation. On this interpretation, much of what goes in psychology, the social sciences, and the humanities would fall short of proper objectifying inquiry. This is probably not what Van Fraassen had in mind. On the second interpretation, there is no problem with using subjective human experience in scientific inquiry as such, but it must be made to satisfy the demands of objectifying inquiry. Experiences must be codified in terms of parameters and qualities that are independent of any particular inquirers. And this is precisely what the methods of the human sciences and the humanities are for. Variables involving subjective experiences are operationalized so that they can be measured on a suitably independent scale, for instance, by means of surveys. Qualitative information obtained in interviews has to be coded in terms of intersubjectively verifiable variables. Historians have to account for their interpretations of the sources in such a way that others can follow and evaluate their reasoning. And so on. Since Van Fraassen intended his notion of objectifying inquiry to capture every sort of investigation pursued with academic discipline, adopting this second interpretation is the charitable thing to do. The central question about the limits of scientific knowledge thus transforms into the question whether there are limits to what can be captured about subjective experience in intersubjective terms. There is good reason to answer affirmatively. First, there is a large intuitive difference between experiencing, say, disgust from the first-​personal perspective and filling out a survey about what sort of things you find disgusting or scoring the intensity of your disgust on a five-​point scale. Similarly, how satisfied you are at your job, when experienced from within your own subjective point of view is very different from the list of answers you might give in a work satisfaction survey or in a face-​to-​face interview, even if you’re completely forthcoming. The same holds, a fortiori, for more complex experiences such as appreciating Edward Hopper’s Nighthawks, being engrossed in Marilynne Robinson’s Lila, playing volleyball, or maintaining a romantic relationship with your significant other. These and many other subjective experiences are so richly textured, multifaceted, and holistic that they resist description by standardized quantities. This is not to deny that parts and aspects of such experiences can be described by means of intersubjectively accessible variables, but much in them eludes such description. To the extent that first-​personal perspectives can be conveyed to others, novels and other

214

2 1 4   • 

Jeroen de Ridder

works of art are arguably more suitable to the task. This is precisely because they don’t adhere to the rules of objectifying inquiry. In addition to these intuitive considerations, further support comes from well-​known arguments for the inherent subjectivity of conscious mental states (Nagel 1974, 1986) and the irreducibility of phenomenal consciousness (Chalmers 1996). If Nagel and Chalmers are right that consciousness cannot be accounted for in purely physical terms, then its contents are not fully explicable through objectifying inquiry either. This is because phenomenal consciousness and the subjective, first-​personal point of view are essentially tied to experiencing “what things are like” for me as an individual. Phenomenal consciousness is defined in terms of how things look and feel from the perspective of particular individuals. Hence, attempts to describe it will violate Van Fraassen’s independence requirement by definition, since that requirement explicitly excludes the use of variables and quantities that are tied to a subjective point of view. Some aspects of phenomenal consciousness might be shared by many individuals so that they are susceptible to intersubjective access, but not everything is. What it is like for me to experience things has an essentially private quality. This, then, constitutes another in-​principle limit for scientific knowledge.28 This section has shown that, when scientific knowledge is construed as objectifying knowledge, we find two more limits to what can be known scientifically. (6) One constituted by what Van Fraassen calls the “relevant parameters and quantities” for a field of scientific inquiry. Whatever cannot be described in terms of them, cannot be studied scientifically (or at least not until science has progressed to revise its vocabulary). (7)  Subjective, first-​personal experiences:  “what it is like” to experience various kinds of things.

8.8 Conclusion Thinking about the nature of scientific knowledge throws cold water on scientistic assertions to the effect that scientific knowledge knows no limits. I  hope to have shown that there are at least seven different kinds of limits to scientific knowledge. We cannot obtain scientific knowledge of: (1) facts

28. If there indeed is distinctly second-​personal knowledge (cf. Pinsent 2012; Stump 2010), that is, knowledge that is only available through interpersonal interactions with another person, then a parallel argument could be constructed that this constitutes a similar limit for scientific knowledge.

251

Kinds of Knowledge, Limits of Science  • 

215

and events beyond our light cone; (2) things that are beyond our cognitive capacities; (3) groundless basic propositions; (4) the reliability of our basic belief-​forming processes; (5) the great majority of recently published scientific claims; (6) things that cannot be described in terms of currently accepted “relevant parameters and qualities” for various fields of inquiry; (7) subjective, first-​personal experiences: “what it is like” to experience various kinds of things. As I  said at the beginning of this chapter, establishing that scientific knowledge has limits, isn’t a full-fledged objection to epistemological scientism, for it might be that the limits of scientific knowledge simply coincide with those of knowledge in general:  What cannot be known scientifically cannot be known at all. Defending the claim that there is knowledge to be had outside science falls beyond the scope of this chapter, although I  do think that for at least a few of the items on the above list, for instance 3, 4, 6, and 7, it is at least intuitively plausible that we can know them by other means than scientific inquiry. I want to close with two final remarks to put the project of this chapter in perspective. First, the above list of limits isn’t radically novel or different than what other authors have proposed (e.g., Rescher 1999). While this may seem disappointing initially, I actually think it’s a good thing. It would be rather surprising if careful thinking about the nature of scientific knowledge would suddenly entail that the limits of science are very different from what has been proposed on other grounds. Where I nonetheless see my project making progress, is in proposing and exploring a different and promising approach to thinking about the limits of science. Rather than relying on more or less intuitive judgments about what science can and cannot do or what sorts of questions it can and cannot address, my approach is more systematic. Hopefully, thinking about the nature of scientific knowledge isn’t as divisive as thinking about the limits of science directly. If I’m right about this, my approach gives both proponents and detractors of the idea that scientific knowledge has its limits common ground for further debate.29

29. Thanks to Stephen Grimm, Rik Peels, Scott Robbins, and René van Woudenberg for very helpful comments on a previous version of this chapter. This publication was made possible through the support of a grant from Templeton World Charity Foundation. The opinions expressed in this publication are those of the author and do not necessarily reflect the views of Templeton World Charity Foundation.

126

2 1 6   • 

Jeroen de Ridder

References Alston, William P. 1993. The Reliability of Sense Perception. Ithaca, NY:  Cornell University Press. Alston, William P. 2005. Beyond “Justification”:  Dimensions of Epistemic Evaluation. Ithaca, NY: Cornell University Press. Arbesman, Samuel. 2012. The Half-​Life of Facts:  Why Everything We Know Has an Expiration Date. New York: Current. Baehr, Jason. 2011. The Inquiring Mind. New York: Oxford University Press. Battaly, Heather. 2008. “Virtue Epistemology.” Philosophy Compass 3(4): 639–​663. Benton, Matthew. 2017. “Epistemology Personalized.” Philosophical Quarterly 67(269): 813–​834. Bird, Alexander. 2010a. “Social Knowing: The Social Sense of ‘Scientific Knowledge.’” Philosophical Perspectives 24: 23–​56. Bird, Alexander. 2010b. “The Epistemology of Science—​A Bird’s-​Eye View.” Synthese 175(S1): 5–​16. BonJour, Laurence. 2010. “Externalism/​Internalism.” In A Companion to Epistemology, edited by Jonathan Dancy, Ernest Sosa, and Matthias Steup, 364–​ 368. Oxford: Wiley-​Blackwell. BonJour, Lawrence. 1985. The Structure of Empirical Knowledge. Cambridge, MA: Harvard University Press. Boudry, Maarten and Massimo Pigliucci. eds. 2018. Science Unlimited? The Challenges of Scientism. Chicago: Chicago University Press. Brady, Michael and Miranda Fricker. eds. 2016. The Epistemic Life of Groups: Essays in the Epistemology of Collectives. Oxford: Oxford University Press. Carter, J. Adam, Andy Clark, Jesper Kallestrup, Orestis Palermos, and Duncan Pritchard. eds. 2018. Extended Epistemology. Oxford: Oxford University Press. Carter, J. Adam, Jesper Kallestrup, Orestis Palermos, and Duncan Pritchard. 2014. “Varieties of Externalism.” Philosophical Issues 24(1): 63–​109. Cassam, Quassim. 2014. Self-​Knowledge for Humans. Oxford: Oxford University Press. Chakravartty, Anjan. 2007. A Metaphysics for Scientific Realism:  Knowing the Unobservable. Cambridge: Cambridge University Press. Chalmers, David. 1996. The Conscious Mind. New York: Oxford University Press. De Ridder, Jeroen. 2014. “Epistemic Dependence and Collective Scientific Knowledge.” Synthese 191(1): 37–​53. DePaul, Michael. 2011. “Foundationalism.” In The Routledge Companion to Epistemology, edited by Sven Bernecker and Duncan Pritchard, 235–​244. London: Routledge. DeRose, Keith. 2009. The Case for Contextualism: Knowledge, Skepticism, and Context, Vol. 1. Oxford: Clarendon Press. Dooyeweerd, Herman. 1997 (1953–​1958). A New Critique of Theoretical Thought, 4 Vols. Lewiston, NY: Edwin Mellen Press. Dretske, Fred. 1969. Seeing and Knowing. Chicago: University of Chicago Press.

217

Kinds of Knowledge, Limits of Science  • 

217

Fernández, Jordi. 2013. Transparent Minds: A Study of Self-​Knowledge. Oxford: Oxford University Press. Fricker, Elizabeth. 2002. “Trusting Others in the Sciences:  A Priori or Empirical Warrant?” Studies in History and Philosophy of Science 33(2): 373–​383. Fumerton, Richard. 1995. Metaepistemology and Skepticism. Lanham, MD: Rowman and Littlefield. Gertler, Brie. 2012. “Renewed Acquaintance.” In Introspection and Consciousness, edited by Declan Smithies and Daniel Stoljar, 93–​128. Oxford: Oxford University Press. Gilbert, Margaret. 1989. On Social Facts. London: Routledge. Greco, John. 2010. Achieving Knowledge. Cambridge: Cambridge University Press. Greco, John. 2014. “Episteme:  Knowledge and Understanding.” In Virtues and their Vices, edited by Kevin Timpe and Craig Boyd, 285–​301. New  York:  Oxford University Press. Grimm, Stephen. 2001. “Ernest Sosa, Knowledge, and Understanding.” Philosophical Studies 106(3): 171–​191. Grimm, Stephen. 2016. “The Value of Reflection.” In Performance Epistemology, edited by Miguel Ángel Fernández Vargas, 183–​195. New York: Oxford University Press. Hardwig, John. 1985. “Epistemic Dependence.” Journal of Philosophy 82(7): 335–​349. Hardwig, John. 1991. “The Role of Trust in Knowledge.” Journal of Philosophy 88(12): 693–​708. Huemer, Michael. 2010. “Foundations and Coherence.” In A Companion to Epistemology, 2nd ed., edited by Jonathan Dancy, Ernest Sosa, and Matthias Steup, 22–​33. Chichester: Wiley-​Blackwell. Hughes, Austin L. 2012. “The Folly of Scientism.” The New Atlantis 37: 32–​50 Hutto, Daniel and Eric Myin. 2013. Radicalizing Enactivism:  Basic Minds Without Content. Cambridge, MA: MIT Press. Ioannidis, John. 2005. “Why Most Published Research Findings Are False.” PLoS Medicine 2(8): e124. Kitcher, Philip. 2002. “Scientific Knowledge.” In Oxford Handbook of Epistemology, edited by Paul K. Moser, 385–​4 07. New York: Oxford University Press. Lackey, Jennifer. ed. 2014. Essays in Collective Epistemology. New  York:  Oxford University Press. Ladyman, James, Don Ross, with David Spurrett and John Collier. 2012. Every Thing Must Go. Oxford: Oxford University Press. Ladyman, James. 2018. “Scientism with a Humane Face.” This volume. Laudan, Larry. 1981. “A Confutation of Convergent Realism.” Philosophy of Science 48(1): 19–​49. Laudan, Larry. 1983. “The Demise of the Demarcation Problem.” In Physics, Philosophy and Psychoanalysis, edited by Robert S. Cohen and Larry Laudan, 111–​ 127. Dordrecht: Reidel. Leplin, Jarrett. 1981. “Truth and Scientific Progress.” Studies in History and Philosophy of Science 12(4): 269–​292.

218

2 1 8   • 

Jeroen de Ridder

List, Christian and Philip Pettit. 2011. Group Agency. New York: Oxford University Press. McGinn, Colin. 1991. The Problem of Consciousness. Oxford: Blackwell. Miller, Boaz. 2015. “Why (Some) Knowledge Is the Property of a Community and Possibly None of Its Members.” The Philosophical Quarterly 65(260): 417–​4 41. Moran, Richard. 2001. Authority and Estrangement:  An Essay on Self-​Knowledge. Princeton, NJ: Princeton University Press. Nagel, Thomas. 1974. “What Is It Like to Be a Bat?” The Philosophical Review 83(4): 435–​450. Nagel, Thomas. 1986. The View From Nowhere. New York: Oxford University Press. Parent, Ted. 2014. “Knowing-​Wh and Embedded Questions.” Philosophy Compass 9(2): 81–​95. Peels, Rik. 2018. “A Conceptual Map of Scientism.” This volume. Pigliucci, Massimo and Maarten Boudry. eds. 2013. Philosophy of Pseudoscience. Chicago: University of Chicago Press. Pinsent, Andrew. 2012. The Second-​Person Perspective in Aquinas’s Ethics: Virtues and Gifts. London: Routledge. Polanyi, Michael. 1958. Personal Knowledge. London: Routledge & Kegan Paul. Rescher, Nicholas. 1999. The Limits of Science. Pittsburgh, PA:  University of Pittsburgh Press. Rickert, Heinrich. 1926. Kulturwissenschaft und Naturwissenschaft, 6th and 7th expanded editions. Tübingen: Mohr Siebeck. Roberts, Robert C. and Jay Wood. 2007. Intellectual Virtues:  An Essay in Regulative Epistemology. Oxford: Oxford University Press. Russell, Bertrand. 1910. “Knowledge by Acquaintance and Knowledge by Description.” Proceedings of the Aristotelian Society 11: 108–​128. Russell, Bertrand. 1912. The Problems of Philosophy. Oxford: Oxford University Press. Sirtes, Daniel, Hans Bernard Schmid, and Marcel Weber. eds. 2011. Collective Epistemology. Frankfurt: Ontos Verlag. Sosa, Ernest. 2004. “Replies.” In Sosa and His Critics, edited by John Greco, 275–​325. Oxford: Blackwell. Sosa, Ernest. 2007. A Virtue Epistemology: Apt Belief and Reflective Knowledge, Vol. I. New York: Oxford University Press. Sosa, Ernest. 2009. Reflective Knowledge: Apt Belief and Reflective Knowledge, Vol. II. New York: Oxford University Press. Sosa, Ernest. 2011. Knowing Full Well. Princeton, NJ: Princeton University Press. Stanley, Jason and Timothy Williamson. 2001. “Knowing How.” Journal of Philosophy 98(8): 411–​4 44. Stenmark, Mikael. 2001. Scientism: Science, Ethics and Religion. Aldershot: Ashgate. Stroud, Barry. 1989. “Understanding Human Knowledge in General.” In Knowledge and Skepticism, edited by Marjorie Clay and Keith Lehrer, 31–​ 50. Boulder, CO: Westview.

219

Kinds of Knowledge, Limits of Science  • 

219

Stroud, Barry. 1994. “Scepticism, ‘Externalism’, and the Goal of Epistemology.” Proceedings of the Aristotelian Society Supplementary Volume 68: 291–​307. Stroud, Barry. 2004. “Perceptual Knowledge and Epistemological Satisfaction.” In Sosa and His Critics, edited by John Greco, 165–​173. Oxford: Blackwell. Stump, Eleonore. 2010. Wandering in Darkness. New York: Oxford University Press. Tuomela, Raimo. 1992. “Group Beliefs.” Synthese 91(3): 285–​318. Van Fraassen, Bas. 1980. The Scientific Image. Oxford: Clarendon Press. Van Fraassen, Bas. 2002. The Empirical Stance. New Haven, CT: Yale University Press. Van Woudenberg, René. 2005. “Contextualism and the Many Senses of Knowledge.” Grazer Philosophische Studien 69(1): 147–​164. Van Woudenberg, René. 2018. “An Epistemological Critique of Scientism.” This volume. Wieseltier, Simon. 2013. “Crimes Against Humanities.” The New Republic, online at https://​newrepublic.com/​article/​114548/​leon-​wieseltier-​responds-​steven-p​ inkers​scientism. Zagzebski, Linda. 1996. Virtues of the Mind: An Inquiry into the Nature of Virtue and the Ethical Foundations of Knowledge. Cambridge: Cambridge University Press. Zagzebski, Linda. 2012. Epistemic Authority. New York: Oxford University Press.

20

9

SCIENTISM

Who Needs It? Alvin Plantinga

I’ve been asked to write about scientism—​presumably about its success (or lack thereof ), its prospects, and about how we should think of it. But there is a prior question: What exactly, or even approximately, is scientism? Well, whatever it is, scientism involves a certain attitude toward science, an attitude of respect, or reverence, or allegiance. And of course this raises still another prior question: the nature of science.

9.1 What Is Science? We might think this question unnecessary: Don’t we all know what science is? Maybe not. This question isn’t as easy as you might think. For example, at one time theology, so it was said, was the queen of the sciences; at present this suggestion would not be met with widespread enthusiasm. Why not? Partly because the term “science” has both a broad and narrower interpretation. In the broad interpretation, “science” would be like the German “Wissenschaft”; the term would denote nearly any systematic intellectual activity or investigation. So understood, science would include mathematics and logic as well as theology, and if you really liked theology, you might sensibly call it the queen of the sciences. In the more narrow interpretation, any science would have to have a substantial empirical connection; it would have to have some substantial connection with observation and experiment. It’s not easy to say just how far this goes: Physics, for example, is a paradigm science, but of course there is much in physics that seems at

21

Scientism: Who Needs It?  • 

221

best distantly connected with observation and experiment. Consider superstring theory, according to which the fundamental constituents of reality are very small strings. This is clearly a scientific theory, and clearly part of physics, but at present there is apparently no evidence for this theory, and no way of empirically testing it. Still, taking “science” in this way, that is, as involving empirical content, theology—​systematic theology, at any rate—​could hardly be thought of as a science. Even here, however, it’s easy to exaggerate the contrast between theology and paradigm examples of science. Theology can and does involve empirical claims. It’s a part of Christian theology, for example, to endorse the reality of miracles; according to the gospels, Jesus performed many miracles, including raising Lazarus from the dead, multiplying the loaves and fishes, and much else. And surely if Jesus multiplied the loaves and fishes, it was empirically evident that he did so. Still further his own resurrection from the dead, an essential element in Christian theology, was or at least involved empirically in evident events and states of affairs. Nevertheless, those who endorse scientism are unlikely to include theology among the disciplines they so enthusiastically endorse. And this is true, even though it has proven exceedingly difficult to give a serious definition of “science.” Much energy has gone into the attempt to give a definition of science, thus distinguishing it from astrology and pseudoscience, as well as from other more legitimate but nonscientific activities. This is the so-​called Demarcation Problem. And although much effort has been expended on it, this project has come to grief; no sensible set of necessary and sufficient conditions for something’s being science, or a science, has been forthcoming (cf. Laudan 1983). For present purposes, however, perhaps we could think of science, a bit informally, as a disciplined and careful attempt to learn about ourselves and our world, ordinarily but not necessarily involving a substantial empirical element.

9.2 Scientism Characterized Given this account of science, we can turn to scientism:  What is scientism? But this isn’t quite the right question. That is because this question presupposes that there is a particular phenomenon, scientism, which is denoted by the English word “scientism”; the question then is about the nature of that phenomenon. But this presupposition is false: There is no one particular phenomenon denoted by the term “scientism.” The word “scientism,” like such words as “naturalism,” “democracy,” “materialism,” and “faith”

2

2 2 2   • A l v i n

Plantinga

doesn’t have a single established English meaning; it means different things to different people and is used in different ways by different people. In this way it is unlike, say, the word “triangle.” That word has an established meaning: As any dictionary will tell you, a triangle is a three-​sided polygon. Of course it’s a free country; you can use “triangle” however you wish; but if you (taking a leaf from Humpty Dumpty) use it to mean, for example, “a nice knock-​down argument” you won’t be using it in accord with its established English meaning. Things stand differently with “scientism.” Some people use it to denote one phenomenon, others use it denote a different phenomenon, still others use it to denote still a third phenomenon, and none of these different uses is contrary to its established meaning. So what are some of these ways of using the term? According to Steven Pinker (2013), “[scientism] is distinguished by an explicit commitment to two ideals, and it is these that scientism seeks to export to the rest of intellectual life.” On this account it seems that scientism is a kind of program, an attempt to export to the rest of intellectual life two characteristics of science: “The first of these is the assumption that the world is intelligible,” says Pinker, and a second like unto it is that “the acquisition of knowledge is hard” (ibid., author’s italics). Given Pinker’s account of scientism, there would be wide agreement that scientism is a very good thing: No doubt it is good, in pursuing the intellectual life, to assume that our world is intelligible; and it certainly does seem that certain kinds of knowledge, at any rate, are not easy to come by. (Other kinds of knowledge—​e.g., knowledge that at the moment I’m suffering from a slight pain in my left knee—​are not very hard to come by, but perhaps that’s not the kind of knowledge Pinker has in mind.) Scientism, thus defined, is a fine thing and we should all endorse it. But of course this is an idiosyncratic definition of the term. In much more common uses, “scientism” is at least a bit pejorative: Scientism is undue reliance upon or exaltation of science. So taken, there would be widespread agreement that scientism is a bad thing; after all, if something is undue, then shouldn’t it be discouraged? So how shall we understand scientism? Here are some other definitions of the term. (1) Scientific knowledge is the only kind of knowledge; all knowl­ edge is scientific knowledge. (2)  The only correct method of fixing belief, or coming to believe some proposition, is by way of scientific investigation. (3)  The methods of science are universally applicable; any question can be investigated by way of scientific methods. (4) When there is conflict between science and any other alleged source of belief, the results of science are always to be preferred. Here is still another, due to Alexander Rosenberg, and

23

Scientism: Who Needs It?  • 

223

encapsulating (1)—​(4) above: “Scientism is the conviction that the methods of science are the only reliable ways to secure knowledge of anything. . . . Being scientistic just means treating science as our exclusive guide to reality, to nature—​both our own nature and everything else’s” (Rosenberg 2011: 6–​7). So how shall we understand the term? Let’s provisionally go with Rosenberg and take scientism to be the idea that science is our only reliable means of coming to know about ourselves and our world. Rosenberg endorses scientism thus thought of, and he is not alone. For example, according to Maarten Boudry (2015: 3.2): . . . belief in a supernatural Creator amounts to a scientific hypothesis in the sense that it has testable empirical consequences, is amenable to scientific investigation, and would make a genuine difference for science if true . . . That is precisely the heart of the conflict thesis: because it makes factual claims about reality, religion encroaches on the domain of science and is vying for the same explanatory domain, even if does not remotely look like science. Here the suggestion seems to be that any factual claim about reality that is not part of science is encroaching on the domain of science. Apparently Boudry believes that science alone properly makes factual claims about reality. And here the usual self-​referential problem rears its ugly head:  “Science alone properly makes factual claims about reality” seems to be a factual claim, but it is certainly not part of physics, chemistry, biology, or any other science. If this claim is true, then it is itself false: Therefore it is false. The claim is also clearly false on less dialectical grounds. I remember what I had for breakfast this morning: coffee and oatmeal. Clearly enough, this is a claim about reality; equally clearly, it is not part of any current science; but even the most enthusiastic devotee of scientism would hesitate to insist that I am encroaching on the domain of science in so claiming. I look out of the window, see a squirrel, and claim that I see a squirrel in my yard. That I see a squirrel is no part of science: But surely I am not encroaching on science in making that claim.

9.3 Origin of Scientism Now what sorts of belief underlie scientism? Where does it come from? For the most part it would seem to originate in versions of empiricism. Empiricism comes in several forms, but fundamentally it is the idea that experience is our

24

2 2 4   • A l v i n

Plantinga

only source of knowledge. Empiricism thus precludes the idea that we can attain knowledge by way of reason as well as by way of experience. Empiricists reject the idea that there is a priori knowledge, that is, knowledge that is independent of or precedes experience. Empiricism started cautiously with John Locke (1632–​1704) and picked up steam with David Hume (1711–​1776), who is perhaps the paradigm empiricist. Hume isn’t entirely easy to understand. Under one common interpretation, however, he claimed that all knowledge is really empirical knowledge, knowledge that is derived from experience. But then what about mathematics, for example, where we seem to have knowledge, but knowledge that isn’t empirical? Hume’s suggestion is that mathematical knowledge, like all allegedly a priori knowledge, is really just knowledge of “the relations of ideas,” as he put it. Knowledge of the relations of ideas, as he thought of it, seems to be something like knowledge of definitions, knowl­ edge of the meanings of terms, or perhaps knowledge of how certain words are used. This account has several unhappy results, including the result that knowledge of mathematical truths such as, for example, Gödel’s Theorem, is really knowledge of contingent propositions about the meanings of terms or knowledge about how language is used. That’s bad enough, but there also seems (once more) to be a self-​referential problem here: (a) “All knowledge is either empirical or a matter of relations of ideas” does not itself seem to be empirical; hence if (a) is true, then knowledge of (a) is a matter of knowing something about the relations of ideas, that is, knowledge of how certain words are used. If so, (a) is not a substantive thesis about knowledge, but a comment about how the word “knowledge” is ordinarily used. More recently, the logical positivists have taken up the cudgels on behalf of radical empiricism. The most widely popular exposition of logical positivism was Alfred J. Ayer’s Language, Truth and Logic, first published in 1936 and republished with a new introduction in 1946. This form of positivism took the form of a criterion of meaningfulness. A candidate for meaningfulness, for the status of being meaningful, so Ayer said originally, is in fact meaningful if and only if it is empirically verifiable. It was soon pointed out that this criterion too, sadly enough, has self-​referential difficulties: It isn’t empirically verifiable, so if it is correct, it is not itself meaningful. Consequently, in the preface to the later 1946 edition, Ayer moved on to the following suggestion: A candidate for being meaningful is in fact meaningful, if and only if some observation statement can be deduced from it in conjunction with other premises without being deducible from those other premises alone (Ayer 1946: 13). This didn’t work out well either. Let O be just any observation statement and let P be just any candidate for meaningfulness:  For example, Martin

25

Scientism: Who Needs It?  • 

225

Heidegger’s somewhat oracular suggestion that “the not nothings itself.” P and if P then O entails O; since O isn’t deducible from if P then O alone, it follows that (according to Ayer’s criterion) P is meaningful. So this criterion, obviously, confers meaning far too liberally, for it implies that “the not nothings itself ” is meaningful, as are any number of equally Delphic propositions. After all, the observation statement O can be deduced from it in conjunction with the premise if P then O without its being deducible from if P then O alone. Given this criterion, any statement whatever can be shown to be meaningful. Subsequent attempts to state the verifiability criterion encountered similar difficulties, and people eventually gave up the quest as misguided. The conclusion, I think, is that if scientism is grounded in these sorts of empiricism, it is poorly grounded indeed.

9.4 Varieties of Scientism Scientism, clearly enough, comes in several varieties. I can’t canvass them all, but one important variety would be the thought that empirical science is our best or even our only real source of knowledge. (Remember Boudry’s idea that any factual claim about reality that is not part of science is encroaching on the domain of science.) Now it is difficult in excelsis to maintain that empirical science is our only source of knowledge. I remember that I had oatmeal for breakfast; isn’t this a case of knowledge, even though it has nothing to do with empirical science? And aren’t there many other kinds of knowledge that in addition to whatever knowledge we have by way of empirical science? What about ethics, for example: Don’t we know that it’s wrong to hurt someone just for the fun of it? But this is certainly not something we could discover by way of scientific investigation. By means of the latter we could discover some of the results of hurting people just for the fun of it—​results that we find unpalatable. But how could we discover by empirical investigation that such behavior is morally wrong? And what about mathematics and logic and the whole realm of the a priori? The very point of describing a bit of knowledge as a priori is to claim that it is not achieved by empirical investigation. We may need empirical science to determine that the number of stars is even (ignoring for the moment irrelevant complications involving the fact that stars are always coming into being, that it’s not entirely clear how to distinguish stars from other kinds of heavenly bodies, etc.), and we may need it to determine that the number of stars is odd; we don’t need it to determine that

26

2 2 6   • A l v i n

Plantinga

that number is either even or odd. It is therefore pretty clear that science isn’t our only source of knowledge. What about that other suggestion, the idea that empirical science is our best source of knowledge? Again, the suggestion certainly needs more work. Science is a fine way of acquiring knowledge, but not in the area of ethics or mathematics. So perhaps we should say that science is our best source of empirical knowledge. But even that is far from obvious. I know that at the moment it is snowing, and that there is a good deal of snow on the ground. I suppose this is empirical knowledge; but I don’t need science to know this—​all I have to do is take a look. I know that I had oatmeal for breakfast, but not by way of scientific investigation—​I just remember it. Science, clearly enough, is not our only way of coming to empirical knowl­ edge. Is it our best way of coming to such knowledge? Perhaps in some areas:  but clearly not in others. At the moment I  am looking out of the window: It seems to me that I see grass and trees. I know that it seems to me that I see grass and trees, and I know this just by way of introspection. Further, for this sort of knowledge, science isn’t of much use at all. So the thought that science is our best source of knowledge needs serious qualification:  In some areas it is our best source of knowledge (for example, knowledge of the microscopic world), but in others it is not.

9.5 Moderate Scientism We can’t sensibly say, therefore, that science is our only or even our best way of coming to empirical knowledge; scientism so characterized is clearly mistaken. So how can we characterize a reasonably defensible scientism? Let’s leave that question for homework. Let’s instead consider a moderate version of scientism:  the idea that when any nonscientific belief comes into conflict with a scientific belief, it is always the scientific belief that should prevail. In any conflict between a scientific belief and a nonscientific belief, it is the latter that should give way. Again, however, this seems mistaken. Suppose neuroscientists, after careful scientific examination of my brain, declare that (in Roderick Chisholm’s terminology), I  am “being appeared to redly”; they declare, that is, that I am in the condition of being such that it seems to me that I see something red. But now suppose it seems to me that I am not being appeared to redly; either I’m not being appeared to at all or I am being appeared to, but not redly. Then I should think the right thing for me to believe is that I am not being appeared to redly, no matter what those

27

Scientism: Who Needs It?  • 

227

neuroscientists think or claim. So scientific claims don’t automatically trump nonscientific claims. Once again the claim is much too broad.

9.6 Science and Religion But even if scientific claims don’t automatically trump nonscientific claims, the devotee of scientism will certainly claim that when a scientific claim is incompatible with a religious claim, a claim based on religious faith, it is the religious claim that should give way. If science comes up with the denial of some belief I hold on faith, I should give up that religious belief. All forms of scientism seem to unite in the thought that when science and religious faith clash, it is always faith that has to give way. In any conflict between a deliverance of faith and a deliverance of science, we should accept the deliverance of science. In any clash between some established scientific theory or result, on the one hand, and any religious belief on the other, it is always the latter that is obliged to give way. (“When faith and science clash, ‘tis faith must go to smash.”) Again, however, why should that be true? I believe that God has created the world—​that is, that at some time in the past, God created our universe. Now at the moment, scientific thinking would agree, at least with the assertion that the universe came into existence at some time in the past: Our world, so they tell us, came into being some 13.8 billion years ago or so. Still, not long ago, the scientific view was that our universe never came into existence but has always existed. Perhaps in the near future that belief will again come to be scientific orthodoxy. If that happens, should I give up my view that the universe had a beginning some finite number of years ago? Suppose science comes to deny that anyone has ever died and then come back to life; should I give up my belief in the death and resurrection of Jesus? Why should we think so? As far as I can see, the argument would have to go along the following lines. My belief in the resurrection, clearly enough, has as its source whatever are the sources of religious belief; scientific belief, on the other hand, has its source in the faculties or processes responsible for scientific belief. And (so the claim would go) the former are less, perhaps much less, reliable than the latter. Endorsing this line of thought, Philip Kitcher (2007) argues as follows: There are a host of religions, with a host of conflicting claims and views. So the source, whatever it is, of religious belief is unreliable; it produces a cacophony of beliefs, many of which contradict each other. Of course to

28

2 2 8   • A l v i n

Plantinga

complete Kitcher’s argument, we’d have to claim that the same does not go for scientific beliefs:  Their source, whatever it is, doesn’t display this distressing promiscuity. (Perhaps that is true, although at the moment science endorses contradictory beliefs, so that the source of scientific belief must also be deemed less than totally reliable.) But why think all religious beliefs have the same source? If Christian belief, at least in certain versions, is true, the source of Christian belief would be something like the internal testimony of the Holy Spirit; the same would not be true for other religions, at least not for any item of another religion that was incompatible with Christian belief (see my 2000 book for more on this). In other versions, the origin of Christian belief is not in the internal testimony of the Holy Spirit, but perhaps in something like tradition and testimony, perhaps Spirit-​led tradition and testimony. If something like this is correct, then here too Kitcher’s account of the origin of Christian belief would be false. Indeed, Kitcher’s account of the source and origin of Christian belief appears to be true only if Christian belief is false. What is the source of Christian belief ? The answer depends on whether Christian belief is true: If it is true, the source would be something like internal testimony of the Holy Spirit, or Spirit-​led tradition and testimony, or some other reliable source; if it is false, maybe something more like what Kitcher suggests. So if Christian belief is true, its source is not unreliable in the way Kitcher suggests. Accordingly, if Christian belief is true, there would be no reason to think that when a scientific belief conflicts with a religious belief, it’s the religious belief that has to yield. The universe in which we live seems to be fine-​tuned for life, and indeed for intelligent life; this fine-​tuning requires that the values of many physical parameters have to fall within very narrow ranges. A scientific response on the part of some who reject theistic belief has been to suggest that there are a whole host of universes, with those parameters taking different values in different universes. If that’s true, however, then it is very likely that at least one universe displays parameters that support life. As it happens ours is such a universe—​but no recourse to a Designer or Creator is needed. With these parameters assuming different values in different worlds, it is just by happenstance or chance that in our universe they take values permitting life and intelligent life. Now suppose this suggestion rises to scientific orthodoxy. Would I then be obliged to believe or assume that it is just by happenstance that our universe is fine-​tuned for life? I can’t see why. I can perfectly properly continue to believe that our universe is life-​friendly because God designed it that way; he intended that it be such as to support intelligent life, the life of creatures made in his image.

29

Scientism: Who Needs It?  • 

229

But could it ever be reasonable to reject a deliverance of reason or to accept a belief that goes contrary to the deliverances of reason? Suppose reason, in the guise of science, tells us that there is bound to be one universe with parameters enabling it to support intelligent life, so that there need be no particular reason why our universe has those parameters. Could it be reasonable to reject this deliverance of reason? How could reason even tell us to reject a deliverance of reason? Reason sometimes tells us to do exactly that. There is a very famous example:  Russell-​type paradoxes. Reason seems to tell us that every property either exemplifies itself or does not exemplify itself; fair enough. Reason also seems to tell us that there is such a property as non-​self-​exemplification: the property a property has just if that property does not exemplify itself. (Thus, for example, the property of being a tree does not exemplify itself, since that property is not a tree.) But now consider this property of non-​ self-​exemplification. If it exemplifies itself, then it exemplifies non-​self-​ exemplification, and thus does not exemplify itself. That means that it does not exemplify itself. But if it does not exemplify itself, then it does not exemplify non-​self-​exemplification, in which case it exemplifies self-​exemplification—​ and thus exemplifies itself. This property, therefore, both does and does not exemplify itself. Reason seems to tell us this; but reason also tells us that it is false that non-​self-​exemplification both does and does not exemplify itself. So here reason tells us to reject a proposition that seems to be among the deliverances of reason. It goes on to tell us that at least one of the premises involved in this derivation is false—​despite the fact that each of the premises seems to be a deliverance of reason. This is a famous example; a much simpler example is given by the scenario envisaged a few paragraphs back, where scientists tell me that I am being appeared to redly when in fact I am not being so appeared to. The scientific case for my being appeared to redly might be ever so strong; but if I am not being appeared to redly, the right thing for me to believe is that I  am not being thus appeared to and that the scientific case is mistaken. So it can certainly happen that reason can tell us that in certain circumstances, some other source of belief is to be preferred, in those circumstances, to the deliverances of reason. Now many Christians believe that there is a source of knowledge or justified belief that is independent of reason and common sense and thus independent of science:  the internal testimony of the Holy Spirit (hereafter “ITHS”). They also believe that this source of knowledge or justified belief is the source of belief in the main lines of the Christian faith. Science, of course,

320

2 3 0   • A l v i n

Plantinga

tells us that it is extremely improbable that someone has risen from the dead; the ITHS, on the other hand, may tell us that Jesus Christ arose from the dead. And according to scientism, we should believe what science says here, rejecting the belief that Jesus rose from the dead. It’s worth noting that science itself doesn’t say that when there is a conflict between a scientifically arrived at belief and one with some other source—​for example, the ITHS—​then scientific belief should be preferred. That belief is an extra-​scientific belief. The deliverances of science do not include the proposition that the deliverances of science are to be preferred, in any conflict between science and some nonscientific view or proposition. Neither physics, biology, chemistry, nor any other science tell us that when there is conflict between any of these sciences and some other belief, it is that other belief that should give way. That belief itself is not part of any science. Scientism, therefore, obviously goes beyond any science. Its justification, if it has a justification, would have to be epistemological, or in any event philosophical, not scientific. So what might its philosophical justification be? Presumably something like the following: The faculties or sources of belief that underlie and produce science are somehow more trustworthy or more reliable than any source of belief that could come in conflict with science. But now suppose Christian belief is true. Suppose there is such a person as God and that his divine son, Jesus Christ, died for our sins and then rose from the dead. And suppose further, as Christians believe, that faith is a gift from God. Faith, of course, carries a doxastic component. It involves an affective component, certainly (something like trust or confidence in the Lord), but it also involves a doxastic component: belief in the main lines of the Christian gospel. This belief, therefore, is a gift from God. According to Calvinistic ways of thinking, faith is the result of the internal testimony of the Holy Spirit; but in other Christian traditions, there will be something similar. And now the point: Surely the deliverances of science are not to be preferred to the deliverances of the internal testimony of the Holy Spirit. Surely the deliverances of science are not more reliable that those of the ITHS. If science tell us that no one has ever died and then risen from the dead, but the ITHS tell us that Jesus Christ, the second person of the divine trinity did die and then rise from the dead, it is the deliverance of science, not the ITHS, that must give way. Scientism, as I’ve characterized it, tells us that in any conflict between a deliverance of science and some nonscientific belief, the scientific belief is to be preferred; if Christian belief is true, this is false; therefore scientism, as I’ve characterized it, is incompatible with Christian belief.

213

Scientism: Who Needs It?  • 

231

The devotee of scientism, therefore, takes on a very considerable load of baggage; he must also accept the idea that Christian belief is false. Any reason he has for accepting scientism must also be a reason for supposing that Christian belief is false. Scientism begins in a relatively innocent-​looking epistemological declaration: Science is our best means of coming to know about ourselves and our world, and hence in any conflict between a scientific belief and a belief of some other sort, it is the scientific belief that should be accepted. But that innocent epistemological declaration has weighty metaphysical or ontological consequence: It implies that Christian belief is false.

9.7 Conclusion By way of conclusion: There are many different ways of defining “scientism,” or perhaps many different varieties of scientism. Under many of these, scientism is clearly false. Science is a wonderful institution and a wonderful source of knowledge or warranted belief; but it is not our only source of knowledge about ourselves and our world. Indeed, with respect to some facets of ourselves and our world—​facets about which we have knowledge, such as our own interior mental lives—​science tells us nothing at all. And even where science does give us knowledge, it may not be our best source of knowledge. When we turn to a more moderate version of scientism—​the thought that in any conflict between science and a nonscientific belief, we should always go with the scientific belief—​we find that scientism, so characterized, is inconsistent with Christian belief. It is therefore difficult to see how anyone could reasonably accept both Christianity and scientism, and any successful argument for scientism will also be a successful argument against Christian belief.

References Ayer, Alfred J. 1946. Language, Truth and Logic, 2nd ed. London: Victor Gollancz. Boudry, Maarten. 2015. “The Relentless Retreat:  Kelly James Clark’s Religion and the Sciences of Origins.” Reports of the National Center for Science Education 35(4): 3.1–​3.6. Kitcher, Philip. 2007. Living with Darwin: Evolution, Design, and the Future of Faith. New York: Oxford University Press. Laudan, Larry. 1983. “The Demise of the Demarcation Problem.” In Physics, Philosophy, and Psychoanalysis, edited by Robert S. Cohen and Larry Laudan, 111–​ 127. Dordrecht: Springer.

32

2 3 2   • A l v i n

Plantinga

Pinker, Steven. 2013. “Science Is Not Your Enemy.” New Republic, available online at: https://​newrepublic.com/​article/​114127/​science-​not-​enemy-​humanities. Plantinga, Alvin. 2000. Warranted Christian Belief. New York: Oxford University Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality: Enjoying Life Without Illusions. New York: W.W. Norton.

23

10

COGNITIVE SCIENCE AND MORAL PHILOSOPHY

Challenging Scientistic Overreach William FitzPatrick

10.1 Introduction Empirical work in cognitive science and moral psychology can obviously be morally significant. As Joshua Greene points out, for example, empirical information about implicit racial bias and its influence on juries is morally significant in the following sense:  When combined with the moral assumption that legal decisions shouldn’t be influenced by racial considerations, it implies that there is often a problem with jury deliberations and that measures should be taken to correct for this (Greene 2014:  711). Those are broadly moral conclusions. But this form of moral significance for empirical information is nothing new or surprising (though the details sometimes are):  Moral judgments always have to take account of relevant empirical facts, and these include psychological facts. What is controversial, and of broader philosophical interest, is whether the empirical research has a deeper bearing on matters of general theoretical relevance to moral philosophy. A central issue here concerns the nature and significance of ethical intuition, and one prominent theme emphasized by those who do take the empirical work to have deep philosophical relevance is the idea that the empirical research—​which teaches us “how the mind actually works,” as Greene (ibid.: 726) puts it—​can be enlisted to do broad debunking work in moral philosophy. I will focus here on two recent examples where research in neuroscience, cognitive science, and empirical moral psychology is claimed to pose powerful

324

2 3 4   • 

William FitzPatr ick

challenges to mainstream philosophical views through undermining targeted ethical or meta-​ethical intuitions. The first is Greene’s updated attack on deontology in “Beyond Point-​and-​ Shoot Morality: Why Cognitive (Neuro)Science Matters for Ethics.” Greene here claims that the neuroscientific work he discusses “reveals the hidden inner workings of our moral judgments,” indicating that deontological ethical intuitions and theories are the products of non-​rational causal influences operating by way of evolved “automatic settings” within human moral psychology. This in turn implies that once we become aware of these debunking facts “we should distrust [these] automatic settings” and so reject the deontological beliefs and theories that flow from them. We should, he thinks, instead embrace an allegedly more rationally grounded morality, such as act-​ consequentialism (ibid.:  695, 717). The second is Shaun Nichols’s “Process Debunking and Ethics” (2014) and his claim is that the empirical work he cites provides, or at least with further research will likely provide, materials for similarly debunking the meta-​ethical belief in moral objectivity (much in the way that others have appealed to scientific results to debunk religious beliefs, for example). The idea is that the data can show this belief to result merely from epistemically defective processes, in which case those of us who hold such a belief should give it up. My concern is with such debunking projects, which attempt to use empirical work to debunk traditional normative ethical or meta-​ethical views. Such projects, I  shall argue, tend to embody a problematic scientism, privileging scientific causal explanation of targeted ethical or meta-​ethical beliefs while ignoring or downplaying important philosophical alternatives, thereby overreaching significantly from the scientific results to draw unwarranted conclusions about debunking the beliefs in question. More specifically, I will take up two central tasks in sections 10.2–​10.4. The first is a critique of these two recent projects. I’ve chosen this work because it represents some of the most prominent arguments currently on offer in this area, covering both normative ethics and meta-​ethics, and it also exhibits features of debunking arguments I want to highlight for more general critique, bringing out difficulties that are likely to plague other work with similar debunking ambitions.1 I will argue that the cases made by Greene and Nichols for this

1. As the subtitle of Greene’s article indicates, he is purporting to tell moral philosophers “why cognitive (neuro)science matters for ethics,” suggesting that the data and arguments marshaled here represent what Greene takes to be the best case for this sort of impact of scientific work on ethics; indeed, it is presented as a more philosophically refined version of his main earlier

235

Cognitive Science and Moral Philosophy  • 

235

sort of impact of experimental work on moral philosophy fail. The empirical results they cite are interesting and do have some role to play in the philosophical debates, but the reasoning from those results to the philosophical conclusions they seek is deeply problematic. In the end, the threats presented to their philosophical targets are much weaker than they are often taken to be. My second task is to go on to draw some more general lessons from these case studies. Although the two critiques I offer are naturally limited, they illustrate broader issues that arise for many such projects, as I have explored elsewhere, for example, in connection with recent evolutionary debunking arguments in meta-​ethics (see my 2014, 2015, 2017). A central point that will emerge in connection with both Greene and Nichols is that there are alternative philosophical accounts of the workings of at least some ethical intuition and of the etiology of at least some of our ethical or meta-​ethical beliefs, and these accounts are equally compatible with all the empirical data cited. These alternatives tend to be ignored or downplayed by those working on ethics from an empirical perspective, and this isn’t surprising. The alternative accounts do not lend themselves as fully to empirical investigation as the debunking forms of causal explanation do, making them less tractable from an empirically oriented perspective and less attractive to those with scientistic leanings. Nonetheless, they remain live and plausible philosophical possibilities. This matters for the dialectical situation because if these alternative accounts turn out to be correct, then they obviously undercut the purported empirical debunking. Everything therefore turns on how each of us should weigh, from where we currently stand, the advantages and disadvantages of the two very different stories on offer—​the debunking accounts vs. alternative accounts amenable to the traditional views under attack. And my central point is that this is not in fact settled for everyone by the empirical data but remains a philosophical issue as controversial and contestable as the original philosophical debates in question (i.e., between deontology and consequentialism, or objectivity and nonobjectivity). I will argue that proper attention to philosophical alternatives to the debunking stories shows that while the empirical work may indeed be philosophically significant, it cannot do the strong debunking work it is often imagined to do. It can lend modest support to explanatory models that, if true for our relevant beliefs, would indeed

work on the topic, and so is a suitable target for focused critical examination in its own right, as is Nichols’s article.

326

2 3 6   • 

William FitzPatr ick

cause problems for them. That is a significant contribution to the philosophical debate, but it is a far cry from using scientific leverage to debunk our beliefs by showing us that our beliefs are in fact “best explained” in ways that show them to be ill-​founded, thus defeating our justification for them. That, I will argue, is what the empirical work does not, and probably cannot, do, at least not in any very general way. Finally, I will end by offering a dilemma and challenge to experimentalists with strong debunking ambitions similar to those of Greene and Nichols, and I will urge a recasting of the philosophical role for such work.

10.2 Greene on the Dual Process Model and “Automatic Settings” At the heart of Greene’s case, stemming largely from his well-​known fMRI studies, is the “dual process theory” of moral judgment. On this view, there are two very different processes by which we arrive at moral beliefs or judgments—​ one involving quick, emotionally laden responses and emphasizing efficiency over flexibility, the other involving controlled conscious reasoning that brings greater flexibility at the cost of less efficiency (Greene 2014:  698). Greene presents extensive scientific evidence for this psychological claim, and for present purposes I will simply grant it. I am happy to do so because this psychological claim is in itself entirely innocuous from the point of view of moral theory. What is not innocuous is a further interpretation of the quick, intuitive, emotionally laden responses. This is what Greene introduces from the start with a metaphor involving “automatic settings” on a camera that allow for efficient though inflexible “point-​and-​shoot” operation. These settings are contrasted with the “manual mode” of operation, which corresponds instead to controlled, conscious reasoning. This metaphor is so pervasive in the discussion that intuitive moral responses are all soon referred to simply as “automatic settings,” encouraging the impression that the moral dispositions in question are, by analogy with the preprogrammed auto settings designed into a camera, innate features of our Darwinian evolutionary design. This in turn encourages the thought that the moral dispositions associated with intuitive moral responses are nothing more than evolved emotional dispositions—​elements of our “inner chimp” with little claim to rational respectability or reliability when it comes to the kinds of moral problems we contemplate in normative theorizing, which are not the kinds of problems such evolved responses were designed to solve. And if that is how things are, it is easy to be drawn into Greene’s

237

Cognitive Science and Moral Philosophy  • 

237

dismissive take on deontology once he links “characteristically deontological judgments” (involving concepts of rights and duties) to such “automatic emotional responses” and adds that the judgments are “driven by” the emotional responses.2 Ultimately, he doesn’t move quite that fast, since he recognizes that dispositions for automatic responses can also be acquired through cultural and personal experience. So he has a separate argument (which I’ll take up below) for why these too should be thought unreliable for moral theorizing. But in both cases the fundamental problem with Greene’s argument lies in the unsupported slide—​lubricated by the “automatic settings” metaphor—​ from the empirically measurable features of intuitive moral responses, such as their being quick and emotionally laden, to their unflattering characterization as unreliable or rationally barren, mere relics of our evolutionary past or upshots of deficient experience. Consider the subset of negative, characteristically deontological judgments of particular interest to Greene, which pertain to bringing about bodily harms through “personal force,” where “the agent directly impacts the victim with the force of his/​her muscles,” as by pushing a large person off a footbridge in front of a trolley in order to stop it before it hits five others (Greene 2014: 709). (It is worth noting here that there are plenty of common negative deontological judgments that don’t fit this model. Many cases involve no direct or personal bodily harm, e.g., giving the order to have someone shoved off the footbridge. Others involve no bodily harm at all and even bodily benefits, e.g., embezzling money in order to donate more to charities that provide bodily aid. Still others involve no harm of any kind, e.g., paternalistic deception or manipulation for someone’s own good. So Greene’s focus is artificially narrow and that is important to the debate. But I will set that aside and focus on Greene’s trolleyology.) Let’s grant the (entirely unsurprising) claim that such judgments, as in the footbridge case, are made quickly and are associated with emotional arousal. And let’s grant also the further claim that these judgments actually depend on the healthy functioning of brain regions associated with emotions, as indicated by the data on diminished deontological judgment patterns in subjects with damage to these regions (ibid.:  701–​703). This is interesting, though of course if these brain regions have additional functions (beyond 2. See Greene (2014: 699, 713). Characteristically deontological judgments are judgments that are “naturally justified in deontological terms (in terms of rights, duties, etc.) and that are more difficult to justify in consequentialist terms.”

328

2 3 8   • 

William FitzPatr ick

emotional ones) that also influence moral judgment, then we could not conclude straightaway that deontological judgments depend specifically on proper emotional functioning: Perhaps judgment is affected by the impairment of these other functions in the damage cases.3 But again, set that complication aside and let’s just grant that the data establish that when emotional functioning is impaired, this tends to inhibit at least some deontological judgment. What follows? This might indeed rule out an extreme rationalist view according to which all deontological judgments flow from exercises of pure reason that take place independently of healthy emotional functioning. But it needn’t bother more moderate deontologists in the least. Philosophers ranging from sensibility theorists back through Ewing and Broad and all the way back to Aristotle have embraced the importance of proper emotional development and functioning to sound ethical deliberation, judgment, and behavior (Dancy 2014). It is hardly a blow, then, to be told that people with life-​impairing forms of brain damage and dysfunction (such as psychopaths, those suffering from dementia, or VMPFC damaged patients) tend to answer moral questions on surveys in ways that are incompatible with the moral theory we endorse. What would cause problems for the deontologist would be a demonstration that emotional involvement in deontological judgment is merely distorting or deeply unreliable across the board. This would be the case, for example, if the emotional involvement associated with characteristically deontological judgments were simply a matter of “automatic settings” built into our evolutionary design for Darwinian reasons having nothing to do with any deontological moral facts (even if such facts exist), being geared instead simply toward maximizing a hunter-​g atherer’s inclusive biological fitness. In that case, there would be good reason for deontologists to be worried: It would require a “cognitive miracle” for such “automatic settings” to lead us reliably to have accurate moral beliefs. Greene himself speaks not in terms of reliability with respect to tracking moral facts (as is typical in current debates over evolutionary debunking), but rather in terms of the automatic settings “functioning well” or not, leaving open the meta-​ethical understanding of this. Still, his worry is similar, with the addition of the premise that, barring miracles, “automatic settings” will tend to function well only when either: 3. For example, Mikhail (2014: 777) points to research from Young and Saxe indicating that the VMPFC, which Greene associates with emotions, is also implicated in “the task of ascribing intent to harm.”

239

Cognitive Science and Moral Philosophy  • 

239

(i) they are adaptations (resulting from evolutionary “trial and error”), being employed in a context similar to that in which they evolved (as with dispositions for perceptual judgment), or (ii) they are dispositions cultivated through social and individual “trial and error experience.” (Greene 2014: 714) The problem, he thinks, is that neither condition is met with respect to the dispositions operative in our characteristically deontological judgments. This is especially so for the judgments involved in much of our normative theorizing, which appeal to cases that are often “unfamiliar.” By this he means:  cases outside of actual personal experience or involving types of situations that have come on the scene long after our evolutionary conditioning—​such as opportunities to help distant strangers, or to affect the distant future by mitigating global warming, or challenges presented by global terrorism, or complications arising from biomedical technologies (ibid.: 716). In the absence of cognitive miracles, deontological judgments would thus seem highly suspect if they stemmed from “automatic settings” that are either nothing but (i)  vestiges of ancient adaptations being employed in novel contexts, or equally badly, (ii) cultivated dispositions employed in thinking about cases that are so unfamiliar that it would take a miracle for these dispositions to function well in Greene’s sense. The question, then, is whether Greene shows that the dispositions that yield the quick, intuitive moral judgments used to support deontology are nothing but outmoded adaptations or cultivated dispositions that are insufficiently rationally informed to allow for sound application in normative theorizing. In fact, we find almost nothing in Greene’s discussion to provide support for this claim or to undermine the natural alternative available to deontologists. It is one thing to demonstrate empirically that many deontological judgments occur quickly, without deliberation, and engage emotions, so that they stem from automatic moral belief-​forming dispositions; it is quite another thing to speculate about what exactly did or did not go into the formation of those particular dispositions. No doubt some dispositions involving quick, non-​ deliberative responses are the direct result of evolutionary shaping, “designed” for meeting the adaptive challenges of Pleistocene hunter-​gatherers. Such dispositions can be expected to be reliable only with respect to such biological aims in similar environments—​and not in novel environments or with respect to the aim of truth of a sort that is not neatly aligned with the biological aims. But, as Greene himself acknowledges (ibid.: 714), other psychological dispositions equally involving quick, non-​deliberative responses might

420

2 4 0   • 

William FitzPatr ick

be acquired through experience and training, and in a way that is rationally informed (as described below). These are obviously the relevant focus for the deontologist, so the question is whether Greene has said anything to preclude understanding at least many deontological judgments on this model. I shall argue that he has not.

10.3 Problems for Greene’s Argument, and an Alternative Model To begin with, the system involved with automatic processing is far more flexible than theorists such as Greene suggest, and its dispositions can be highly intelligent and responsive to environmental stimuli (Railton 2014). As Peter Railton has pointed out, there is plenty of evidence that this system, far from being limited to rigid response types, was “designed to inform thought and action in flexible, experience-​based, statistically sophisticated and representationally complex ways—​grounding us in, and attuning us to, reality,” thus allowing for “spontaneous yet apt” responses (ibid.:  846–​ 847). But even apart from the wide-​ranging evidence Railton discusses in connection with a variety of organisms, it should already be clear from examples of human skill that the system involved with quick, automatic responses can be developed through experience and training to yield intelligent, rationally informed, flexible responses—​or as Julia Annas helpfully puts it, “educated responses” (Annas 2011:  28–​29). And often these responses will be emotionally engaged as well, where emotional dispositions are developed through the same experience and training, becoming equally refined and rationally informed, functioning in ways that needn’t be oriented specifically toward evolutionarily given aims.4 Consider the development of artistic skill and sensitivity. Once someone has mastered jazz piano, as Annas points out, “the result is a speed and directness of response comparable to that of mere habit, but unlike it in that the lessons learned have informed it and rendered it flexible and innovative” (ibid.:  29). If the pianist’s cognitive, emotional, and physical dispositions

4. These developed emotional dispositions might, of course, still reflect some degree of evolutionary shaping, and where they are playing an epistemic role it is possible that this influence will be distorting. But as I  have argued elsewhere (see earlier references to work on evolutionary debunking arguments), debunkers have not offered any compelling case for general distorting influence here, once the points in the text concerning culturally developed and refined dispositions are properly taken into account.

214

Cognitive Science and Moral Philosophy  • 

241

constitute “automatic settings,” they are in any case very different from the inbuilt, rigid settings of a camera in “point-​and-​shoot” mode. They may be quick, non-​deliberative, and emotionally laden, belonging to the system associated with automatic responses rather than conscious, effortful reasoning. But they are no less informed by all the conscious, reason-​involving effort, experience, and refinement that went into the years of intelligent training that crafted those dispositions within artistic practices. Far from being evolutionarily given settings keyed to Darwinian goals, they are intelligently formed dispositions oriented toward goals reflecting standards internal to autonomously developed cultural practices, such as jazz piano. Similarly with games, sports, and crafts. Indeed, Greene should be happy to acknowledge this, since again he allows for the development of dispositions within the automatic processing system through “trial and error experience.” Presumably he would recognize that this captures such things as the mastery of chess or the skills of the pianist, painter, or diagnostician or surgeon, and not just simple cases of “trial and error” such as learning not to eat certain poisonous plants. The point, however, is that in addition to capturing a wide and familiar range of human skills and dispositions, the above model can very plausibly be employed to capture dispositions involved in a great deal of ethical judgment as well, including deontological judgment. Just as in the case of artistic, intellectual, and practical skills and pursuits, individual human beings can, through a decent ethical upbringing, intelligently cultivate ethical dispositions oriented toward certain goals and reflecting certain standards internal to our culturally inherited traditions of ethical inquiry—​standards that are no more micromanaged by specific evolutionary influences than standards of jazz piano or contemporary scientific practice are, reflecting instead cultural developments and refinements that are significantly independent of specific evolutionary shaping. Greene never seriously considers this model in connection with ethical judgment, but if artists, chess masters, and physicians can make quick, intuitive, and often emotionally laden judgments in ways that nonetheless reflect the intelligence of the slower, effortful processes that went into training and refining the dispositions for such judgment, then why should this possibility be closed off for ethical thought? When we make a quick, emotionally laden, intuitive judgment that it is wrong to push innocent people in front of trains even to do good for others, this may well be a rationally informed expression of a moral disposition formed and refined through moral experience, reflection, and training. It needn’t be merely a response caused by irrational, alarm-​like emotion. Our developed emotionally

42

2 4 2   • 

William FitzPatr ick

laden intuitions, or what Dancy calls “practical seemings,” may at least sometimes constitute an intelligent appreciation of non-​consequentialist values and their normative implications (Dancy 2014:  795–​801). And if they do, then Greene’s case against deontological judgments does not go through. So, as he recognizes, he needs an argument to rule out this construal of (at least many) intuitive moral judgments, at least if he wishes to do more than simply to flag an alternative possibility that, if true for deontological judgments across the board, would present a genuine problem for deontologists. He does, of course, purport to do more than that, so let us consider the further argumentation he provides to that end. The primary consideration he cites has to do with the “unfamiliar” nature of many of the cases discussed by deontologists, which is supposed to cast doubt on the possibility of our competently addressing them using our “automatic settings” (Greene 2014: 714–​715). Part of what this is supposed to mean is that these cases involve cultural developments very unlike the circumstances of our Pleistocene evolutionary background—​trolleys, biomedical technologies, or the possibility of actions with global impact. But again, since we are not here appealing to evolutionarily given dispositions, this is irrelevant. The question is whether there is a problem applying our culturally and individually developed dispositions for quick moral responses to such cases, on the grounds that we lack the relevant “cultural or personal experience” to underwrite such applications (ibid.: 714). Greene thinks there is a problem here, and likens this to trying to apply the skills it takes to ride a bike to the unfamiliar task of driving a car. It is hopeless to try to apply our automatic bike riding skills to driving a car, which will instead require switching to “manual mode” until the car driving skill is developed; similarly, he thinks, it is hopeless (or would require a “cognitive miracle” not to be hopeless) to apply our dispositions for quick, automatic moral responses to the cases at issue in debates in normative ethics. But why should we think that applying our culturally developed ethical intuitions to trolley cases, or to cases involving bioethics or global terrorism or climate change, is anything like trying to drive a car using only bike riding skills (ibid.: 716)? Greene’s claim here should not be confused with the plausible worry that some science-​fiction cases contrived by moral philosophers (and metaphysicians) are so outlandish that intuitive responses to them cannot count for much. That may be true. But Greene provides no argument for thinking that this worry threatens the deontologist’s support for basic non-​consequentialist claims. For one thing, although deontologists often do appeal to artificial cases, they can in fact make the same points using

243

Cognitive Science and Moral Philosophy  • 

243

more “familiar” cases, as they often do in case studies drawn from medicine, business, or war. And even apart from that, most of the artificial cases deontologists discuss are hardly so far out as to preclude our ability to recognize and assess their morally salient features. It is true that few (if any) of us have actually encountered a runaway trolley while standing near a switch that could redirect it—​and no one will ever find herself on a bridge next to a person both large enough to stop one and yet somehow still light enough that one could handily push him off ! But it is hardly obvious why this should pose a problem for quickly recognizing the morally salient features of these cases already familiar from everyday life. We have driven cars if not trolleys, swerved to minimize damage, and faced many situations in which an innocent person could be metaphorically “thrown under the bus” to spare trouble for others. People have plausibly developed, in everyday contexts, an emotionally laden disposition to consider the deliberate harmful use of one person (even to do good for others) as a wrong-​making factor, especially where the harm is egregious. We have similarly developed a disposition to recognize harm brought about through mere damage control, as a side effect of diverting a public threat, as being a very different matter. And the development of these dispositions involves not just the sort of “trial and error” Greene considers, based on actually encountered situations, but also consideration of fictional or hypothetical cases. The critical and imaginative discussion of such cases is a standard part of moral training and development. Consider, for example, the social role of fiction and film in shaping our moral sensibilities, and similarly with conversation about historical events, current affairs, and merely possible cases, which help us to clarify our values through ongoing discourse. The moral philosopher’s discussions about cases are just more systematic extensions of this, not some brand new activity as ill-​served by our developed sensibilities as driving a car would be by bike riding skills. When we consider pushing someone in front of a train to help others, our response may be quick and emotionally laden, but it may nonetheless be the exercise of a rationally informed moral disposition to recognize the morally salient features and to see them as wrong-​making. This is what Railton (2014: 832) calls a “manifestation of an underlying moral competency,” built up through conscious experience, training, and refinement. And, for all Greene has shown, this may well be a matter of having developed a reliable sensitivity to deontological moral properties and facts, allowing for the intuitive recognition of good reasons for such responses (no cognitive miracles required). Nothing in Greene’s argument discredits this possibility.

42

2 4 4   • 

William FitzPatr ick

Indeed, Greene himself presumably thinks that in the even more “unfamiliar” case of having to push a button that will kill one innocent person in order to prevent the destruction of most of the planet, the equally quick intuitive judgment that this is permissible—​which on his schema counts as a “characteristically consequentialist judgment”—​is a matter of recognizing good reasons for thinking it permissible. But if we can have reasons-​responsive intuitions in this (extremely) “unfamiliar” case, why can’t we equally have them in the much more “familiar” kinds of cases employed by deontologists to support deontology? Greene’s view relies on an indefensible double standard here. It is true, of course, that some automatic responses are caused by emotional factors independently of recognition of good reasons. And such blindly emotional responses may tend to yield more characteristically deontological judgments than consequentialist ones. But it doesn’t follow from this that (going in the opposite direction) most deontological judgments are the result of blindly emotional responses. Even if all blindly emotional responses yielded deontological judgments, this might account for only a subset of deontological judgments, with many other quick and intuitive deontological judgments instead resulting from rationally informed, emotionally laden dispositions on the reasons-​responsive model. Empirical evidence that blind emotions lean toward the deontological is not empirical evidence that deontological judgments are generally driven by blind emotions. Greene (2014: 708) asks: “What, for example, is it about pushing the man off the footbridge that makes us feel that it is wrong? Experiments are answering this question, among others.” But experiments are “answering” such questions only insofar as it is just assumed, scientistically, that the alternative, reasons-​responsive model I have sketched is false. If that model instead turns out to be correct, and what it is about that case that makes (some of ) us feel it is wrong is our recognition of good reasons for thinking it is wrong through a recognition of the wrong-​making factors in play, then the experiments are not in fact correctly answering the question why we believe what we do: The explanatory appeals to extraneous causal factors will just be misguided here. Of course, experiments can show that some judgments made by some people are unreliable, by revealing ways in which certain subjects are influenced by factors that all would agree are morally irrelevant and so should not be influencing judgment (ibid.:  709–​710). For example, suppose some subjects are found to judge it to be permissible to knock the large man off the footbridge by using a switch that activates a lever that pushes him off, while also judging it impermissible to shove him off directly using their muscles.

245

Cognitive Science and Moral Philosophy  • 

245

In that case, it is indeed plausible that their judgment is being influenced by morally irrelevant factors along the lines Greene identifies, since there is plainly no morally relevant difference between the two cases, as all agree. Deontologists should grant that such influence occurs, and they owe a debt to experimentalists for exposing such potential sources of error for some deontological judgments, which might help with weeding out certain mistakes. But this does not add up to anything like the sort of powerful case against deontology that Greene imagines, because it does not support the general, deflationary interpretation of intuitive deontological judgments he relies upon in his argument. In fact, given the lack of actual support for his conclusions by the scientific data, it is tempting to turn around a favorite move deployed by Greene himself and hypothesize that much of his argument concerning the ethical significance of the data can best be explained as an elaborate post hoc rationalization of a consequentialist philosophical view held for unrelated reasons—​rather than being a scientifically driven undermining of deontology.5 But however that may be, my more modest claim is just that his interpretations are not supported by the science, so that his claim to be drawing strong conclusions based on scientific results is a matter of scientistic overreach. Indeed, it is far from clear what kind of experimental work could possibly show us that our deontological judgments across the board are mere “automatic settings” in Greene’s loaded sense, as opposed to spontaneous expressions of rationally informed, intelligently developed moral dispositions.

10.4 Nichols on Process Debunking and Belief in Moral Objectivity Nichols focuses his primary debunking efforts on the meta-​ethical sphere, targeting the common belief in the objectivity of morality, that is, the notion that the truth conditions for moral claims “are independent of the attitudes and feelings people have toward the claim” (Nichols 2014: 734). His interest is in what he calls “process debunking arguments” wherein “one attempts to undercut the justificatory status of a person’s belief by showing that the belief was formed by an epistemically defective psychological process” (ibid.: 727). A paradigm of such a debunking argument would be one purporting to show that “the sole source of [a person’s] theistic belief ” lies in wishful thinking.

5. For Greene’s own move of this sort against deontologists, see Greene (2014: 718).

426

2 4 6   • 

William FitzPatr ick

Since that is clearly an epistemically defective process, this revelation of the etiology of the theistic belief should undermine any confidence she might have had in its truth (ibid.: 733). Nichols’s concern, then, is with debunking arguments seeking to establish that the sole source of our belief in moral objectivity lies in equally epistemically defective processes. If this could be shown, then we should likewise lose any confidence we might have had in the objectivity of morality, and so abandon moral objectivism. Now the first thing to notice here is what a tall order this turns out to be. It is not enough for a debunking argument just to establish the existence of epistemically defective causal influences that have some effect on some people’s belief in moral objectivity. Such a claim is entirely unsurprising and does nothing so far to undermine a given agent’s belief in moral objectivity. If I am told that some morally irrelevant causal factor C helps to incline some people to believe in moral objectivity, this should perhaps inspire self-​scrutiny to see whether C might be operative in myself as well. But it does not by itself give me reason to conclude that it is. And even if I do come to that conclusion, this is so far compatible with other, more epistemically sound processes also being at work in leading me to the belief in question. If these other, more respectable processes are sufficient on their own to lead me to this belief, and C merely provides further inclination to believe what I would believe anyway for good reasons, then the mere presence and operation of C fails to be undermining. This is true even if it were somehow established that C would by itself also have been sufficient to lead to the belief. Suppose a close friend is accused of a grisly murder and I have excellent evidence that he is innocent, which I cite in defense of my belief that he is. Someone might object that I would believe this even in the absence of such evidence, simply due to my attachment to him and desire to protect him from prosecution. Even if this is true, it does nothing to impugn the evidence I have and does not undermine my justification for the belief I base on that evidence, though it may prompt increased scrutiny to be sure the evidence really is as good as it seems. My justification would be threatened only if there were independent grounds for doubting the evidence in question. Similarly, the mere presence and additional influence of C needn’t deflate my confidence in my belief in the objectivity of morality, at least in the absence of independent grounds (presented to me in a non-​question-​begging manner) to doubt the quality of the reasons I cite as good reasons for that belief. In order for me to lose confidence in my belief in moral objectivity, I will have to be given good reason to think not merely that some people’s belief in moral objectivity is influenced to some extent by C, but that my own belief in moral objectivity

247

Cognitive Science and Moral Philosophy  • 

247

is to be chalked up solely to C and/​or other epistemically defective processes.6 But does the experimental work really show those of us not already inclined on other grounds to reject moral objectivity that our own belief in moral objectivity is deeply suspect due to its having originated solely from epistemically defective processes? Nichols, to his credit, doesn’t claim that it does. Instead, he limits himself to much weaker theses that do seem to be supported by the empirical research. The problem, however, is that claims weak enough to be credibly supported by the science are also too weak to do any strong, general debunking work. Nichols claims, for example, that many moral judgments are accompanied by emotions and that “there is some reason to think that emotion facilitates judgments of objectivism,” that is, the belief that the moral claims in question are objectively true (ibid.: 738). But this in itself is entirely innocuous. First, an emotion’s facilitating judgments of objectivism is compatible with the agent’s also appreciating good reasons for belief in objectivism, as Nichols recognizes. Second, it is also compatible with the emotion’s playing an epistemically valuable role in facilitating such judgments, so that it is actually part of the agent’s coming to appreciate good reasons for the judgment. For example, indignation over someone’s cruel mistreatment in a case of human trafficking, say, may be an expression of a developed moral-​emotional disposition that is partly constitutive of grasping the values that make such treatment wrong. Our emotionally laden intuitive sense that this is wrong and should be stopped may be an insightful “practical seeming” in Dancy’s sense, attuning us to the values in play and to their normative implications. If we are then asked whether the wrongness here is objective, we might reflect on the values to which our emotions have attuned us—​such as the dignity of the person being mistreated and the badness of her misery. And we might then recognize that these values ground the wrongness of such cruel treatment in a way that is independent of subjective or merely conventional factors. This will then lead us rationally to conclude that this wrongness is objective. In this way, emotions may facilitate our belief in moral objectivity by helping us to grasp values that we understand upon reflection to have direct normative implications apart from conventional or subjective factors, thus justifying our belief in objectivity. This is exactly as it should be. And it would also nicely explain such empirical facts as that “children with psychopathic tendencies, 6. In speaking of the belief as being chalked up solely to such influences, I am just following Nichols’s language in formulating his paradigm example involving debunking theistic belief, quoted above.

248

2 4 8   • 

William FitzPatr ick

[who therefore] have diminished sensitivity to distress in others, . . . are more likely to judge that moral transgressions (e.g., hitting another child) are authority dependent,” that is, not objective (ibid.:  736). Their emotional impairment is an epistemic impairment, akin perhaps to emotional impairments in the realm of aesthetics, posing obstacles to aesthetic experience and understanding. Consider next Nichols’s claim that “there is a tiny bit of evidence suggesting that emotional processes inflate judgments of objectivity in epistemically defective ways” (ibid.: 739). Here he discusses evidence that induced disgust can contribute to a tendency to view a judgment as objective, and that induced anger over an unrelated matter can do the same in some subjects, though the results are not very robust.7 This is interesting, if again unsurprising: People worked up from thinking about a conflict that made them angry are likely to be in a more combative, stubborn mood, and more ready to dig in their heels in an argument; so they might be expected also to be slightly more likely to say that someone who disagrees with them about something is clearly mistaken. There are a number of worries one could raise about whether such experiments are adequately measuring genuine belief in objectivity, rather than picking up on other things, but let us set such worries to one side and grant that there can be such emotional distortions in judgments of objectivity.8

7. Nichols (2014: 736–​738). He notes that in the induced anger study the effect was found only for female subjects, and it consisted merely in a shift from 2.92 to 3.96 on a 1–​6 point scale in terms of agreement with the proposition on the survey question meant to capture objectivity for the judgment in question. 8.  To take just one example, Nichols mentions a study showing that “as compared to their responses on whether robbing a bank is wrong, participants were more likely to say that whether abortion or euthanasia is wrong is a matter of opinion or an issue about which there is no correct answer” (2014: 742). But a natural explanation for this effect is just that while robbing a bank is a paradigm case of clear wrongdoing in nearly everyone’s judgment, abortion and euthanasia are classic hard cases falling into notoriously complex gray areas. This means that abortion and euthanasia are both matters about which reasonable people disagree and matters that most people see as typically depending for their evaluation on the complexities of the circumstances, entirely unlike bank robbery. When people say, then, that it is “a matter of opinion” whether abortion or euthanasia are wrong, they likely mean nothing more than that the latter are controversial and that there is “no correct answer” in the sense of a single agreed upon answer to moral questions about them, or in the sense that there is in any case no simple one-​size-​fits-​all answer to moral questions about these things (again, unlike with bank robbery), since answers will vary with the details. Even staunch objectivists will agree with the above thoughts, and people not trained to make distinctions among the many issues in play here can be expected to answer survey questions in ways that might appear to suggest subjectivism, relativism, or selective denial of objectivity, but are in fact just reflections of these other rather obvious thoughts about notoriously hard cases. All of this complicates arguments relying on such survey data to draw conclusions about effects on objectivity judgments.

249

Cognitive Science and Moral Philosophy  • 

249

Another proposal Nichols discusses is that belief in moral objectivity can be influenced by the desire to punish (ibid.: 740). On the face of it, of course, this is far too limited to support any general or strong debunking argument about belief in moral objectivity. It cannot naturally explain belief in the objective wrongness of historical acts (unless we form strong desires to punish people we know are long dead, such as Nero, which seems unlikely), or of acts that we judge to be objectively wrong but also to be such that the agent is excused for them, so that we think punishment inappropriate. So even if there is some effect in some cases, it could at most play a very limited role in a debunking argument that stitched together a variety of psychological factors to cover all the relevant types of cases.9 Still, let’s grant for the sake of argument that there are some real, measured effects on belief in moral objectivity for some cases. The real question is:  What follows? Certainly nothing that should make reflective people feel that their belief in moral objectivity is in imminent danger of being debunked. The experimental results serve only as warnings to be on the lookout for pitfalls they identify: If you are wrestling with meta-​ethical questions about objectivity, don’t base your views on inclinations you feel when disgusted or hopping mad about something, or when itching to throttle someone, since we know those emotional factors may be distorting your judgment; instead, consider the question of moral objectivity in cooler moments. This is good advice, but the points raised don’t take us very far down any path to debunking. Again, Nichols is aware of these limitations, so his thesis is stated very modestly: “Insofar as we believe in moral objectivity because the associated feelings of anger and disgust trigger greater certainty in one’s assessment, the belief in objectivity is not justified,” or again, “to the extent that people believe in objectivity

9. Again, there are questions to raise about the genuineness of the effect. It is interesting that people tend to give somewhat lower ratings for objectivity when considering an act by someone who has already been severely punished, but this might be less a measure of effects on belief in objectivity concerning the wrongness of the crime than a reflection of people’s tendency to conflate different issues when answering survey questions. Are people who give slightly lower objectivity ratings when considering a case in which the wrongdoer was severely punished really thinking that what he did was less objectively bad, the badness somehow being more dependent on conventions or subjective factors than they would otherwise have thought? Or are their answers just confused reflections of their feeling that as objectively bad as the crime was, the wrongdoer has paid for it and therefore a less harsh response to the crime is called for at this point (this being poorly expressed through a weaker response to questions meant to measure “objectivity”)? Given how hard it is to keep students from repeatedly conflating different issues in ethics classes, we should not have much confidence that similar conflations are not occurring in the minds of subjects filling out questionnaires in many of these studies.

520

2 5 0   • 

William FitzPatr ick

because of their desire to punish, their belief is prima facie unjustified” (ibid.: 739–​740, my italics). In other words, any boost we might get from a mere desire to punish or from anger-​based or disgust-​based confidence (where this is epistemically irrelevant anger or disgust) is epistemically worthless. This conclusion is entirely reasonable, but it tells us very little because it remains a wide open question how much of our belief in moral objectivity is attributable to such effects and how much is attributable to more epistemically respectable processes of the sort mentioned earlier. Nichols concedes, then, that “this is just the barest sketch of a debunking argument” (ibid.: 740). But in fact it is not even that. What it provides are only some ingredients that could in principle be used in a debunking argument if somehow one could fill in the rest of the argument to show that the etiology of our belief in moral objectivity is to be cashed out exclusively in terms of such extraneous causal factors. All of the empirical work goes toward supporting the existence and operation of these ingredients to some extent in some cases, but none of it shows us that such ingredients exhaust the explanation of “our” belief in moral objectivity. That is where the crucial work of a real debunking argument would have to lie, and in order to be effective against those of us who do currently believe in moral objectivity this work would have to be done in a non-​question-​ begging way, giving us good reasons to think our belief in moral objectivity can be chalked up entirely to such epistemically defective processes. But the sort of empirical research Nichols discusses does not provide that. I  therefore reject his claim that “whether the argument can support a strong conclusion that the lay belief in moral objectivity is unjustified will depend on what science reveals about why we have the belief ” (ibid.: 741, my italics). My point has been that this is precisely what science cannot do, at least without just scientistically begging the question against those of us who take ourselves to have recognized good reasons for the belief—​for example, good reasons for thinking it to be objectively true that things like rape, slavery, and depriving girls of education are morally wrong. The scientific work can reveal causal factors that, if operative in a given case of belief formation, would be distorting, but it does not thereby show that there are not in fact good reasons for holding the belief in question or that we are not (at least in part) responding intelligently to those reasons in forming those beliefs. Indeed, it is hard to see how any experiment could accomplish such a thing, especially with respect to “our” moral or meta-​ethical beliefs and not just in certain particular cases.

215

Cognitive Science and Moral Philosophy  • 

251

What we are left with, then, are just the much weaker and plausible theses from Nichols described earlier, which do not constitute an actual debunking argument against belief in moral objectivity. Nichols claims that there is at least a burden on objectivists “to defuse the debunking argument” (ibid.: 741). But I deny even this. There is, in fact, no looming debunking argument that requires defusing, at least if we stick to the modest, scientifically supported claims Nichols makes. One could, of course, construct an actual debunking argument in this area, employing as premises much stronger explanatory claims about our belief in moral objectivity, but those claims (as with Greene’s strong claims) would then go well beyond the limited and cautious conclusions actually supported by the scientific work, again bringing on the charge of scientism. To be sure, objectivists do owe an epistemic account of how we might justifiably believe in moral objectivity, perhaps along the lines I have sketched. But that demand is nothing new in the philosophical debate.

10.5 Debunking Arguments, the Philosophical Dialectic, and a Plausible Role for Empirical Work in Ethics Having raised problems for the sorts of debunking projects offered by Greene and Nichols, it is necessary now to step back and consider more carefully the nature of the philosophical dialectic between debunkers and those whose views they target, so that we can see more clearly what exactly each side is trying to accomplish and which moves are or are not fair ones in this context. Some will object, for example, that the responses I  have given to the challenges from Greene and Nichols are unfair because they set the bar for debunking arguments unreasonably high. We need to be able to assess such worries in order to assess whether any real progress has been made. The answer to the above objection is that what an argument needs to accomplish in order to count as a successful debunking argument depends on how we understand the idea of debunking, and debunkers themselves tend to understand this quite strongly. A common understanding, familiar from recent discussions of evolutionary debunking arguments against ethical realism, goes roughly like this: A debunking argument aims to show that those who hold certain beliefs should give them up, by showing that their justification for those beliefs is defeated by explanations for those beliefs that reveal them to have been formed in epistemically defective ways; in particular, the explanations show the beliefs in question to have been formed through causal processes of a sort such that it would be a mere lucky coincidence if those beliefs turned out to be true. This is clearly Nichols’s understanding, for example: He

25

2 5 2   • 

William FitzPatr ick

construes the sort of debunking he has in mind as attempting “to undercut the justificatory status of a person’s belief by showing that the belief was formed by an epistemically defective psychological process,” as by showing that “the sole source of [a person’s] theistic belief ” lies in wishful thinking (Nichols 2014: 727, 732–​733, my italics). Importantly, such arguments are supposed to be directed not only at third parties who can then appreciate from a safe distance that someone else’s beliefs have been debunked, but also at the believer herself, so that the argument has rational debunking force for her: She is supposed to be shown by the argument that her relevant beliefs are merely the results of epistemically defective processes, whereupon she should lose confidence in them. This is clear in Richard Joyce’s paradigm involving an agent’s learning that her beliefs about Napoleon are simply the result of a pill that causes such belief formation utterly independently of any historical facts about Napoleon ( Joyce 2006: 179): Her becoming aware of this truth about the etiology of her beliefs should undermine her confidence in their truth, since it defeats her justification for those beliefs. The idea is that debunking arguments provide us with similar revelations about the etiology of the beliefs being targeted, appealing to extraneous evolutionary or psychological causal processes rather than magic pills, with similar results for the believer. The bar here is indeed set high, but it is debunkers themselves who set that bar, and this is no accident since nothing weaker than this will succeed in demonstrating to us that our beliefs are mere bunk. Merely showing that there are some epistemically defective processes commonly involved in beliefs of this type and that it is possible that our beliefs are fully attributable to such sources alone is not the same thing as showing that our beliefs are in fact properly and exhaustively explained in this way. We have seen that alternative explanations remain open possibilities and will seem far more plausible to those of us who do hold the beliefs in question and take them (and our arguments for them) seriously. So in order for debunking arguments to have their intended force against those of us in this position, they would have to present us with compelling reasons to take the proposed debunking explanations more seriously than the alternative explanations favorable to our beliefs. And many debunkers seem to think they have done just that, by presenting their strong explanatory claims as underwritten by science. They take themselves to be showing that the scientific results ought to lead us to accept that these (debunking) explanations of our beliefs are indeed compelling and the best explanations of our beliefs—​especially in light of the fact that they are more parsimonious than the alternative explanations that appeal to

253

Cognitive Science and Moral Philosophy  • 

253

our coming to appreciate good reasons for believing the contents of certain moral or meta-​ethical propositions to be true. Many debunkers thus embrace the high bar set for successful debunking but are not bothered by this because they are confident they can meet it.10 It is far from clear, however, why those of us who take the relevant beliefs and supporting arguments seriously should feel compelled to grant the superiority of the debunking explanations over the alternatives that strike us as more compelling despite being less parsimonious. While parsimony is one significant methodological desideratum, it does not automatically trump all other considerations in all contexts. And for many of us the evident plausibility of the vindicating explanations of certain moral or meta-​ethical beliefs is far more compelling than the parsimony being sold by accounts that explain away these beliefs and their apparent justifications. We would gladly pay the cost of less parsimony in order to preserve the structure of belief and justification that we continue to find, upon reflection, to be overwhelmingly plausible—​for example, the idea that I believe that slavery or rape are objectively wrong because they are objectively wrong and I’ve come to recognize that fact by recognizing the good reasons for thinking it to be so.11 We might, of course, be wrong about that in the end, in the uninteresting sense that we might always be wrong about anything worth arguing about. But we cannot be rationally forced out of our position simply by appeals to science and parsimony. It is in this sense that the debunking arguments are guilty of scientistic overreach, whether they aim to show us that out of respect for cognitive science (and neuroscience, evolutionary biology, and so on) we deontologists must give up deontology, or that we realists must give up realism, as science has shown such things to be “bunk.”12 10. Both Joyce (2006) and Sharon Street (2006) think that their debunking explanations of our moral beliefs are the best explanations for those beliefs, which should be accepted by all, including those who begin with realist sympathies, insofar as we are appropriately respectful of evolutionary biology and related sciences, and of what they believe such scientific work shows. They do not shy away from such strong claims, which is why they take themselves to be in a position to provide strongly debunking arguments that ought to be effective against their targets (rather than merely preaching to the choir of skeptics or subjectivists). 11. I have developed this line of response in connection with evolutionary debunking arguments in the articles cited earlier. 12. I am not here denying that if science had actually succeeded in giving us compelling reason to doubt the reliability of our moral faculties, then there would be problems for the justification of our moral beliefs and for meta-​ethical beliefs partly based on them. My point is that the scientific results have not actually given us compelling reasons for such general doubt: If they seem to do so, that is only because of supplemental scientistic assumptions of the sort I have discussed, such as the assumption that the distorting causal influences identified are the

524

2 5 4   • 

William FitzPatr ick

That said, however, there is a legitimate point underlying the complaint that the bar for debunking arguments is being set too high, and it is important to be equally clear about this. While it is true that the high bar is appropriate for the strong debunking ambitions I have discussed, which tend to characterize these debates, there are more modest ambitions for which the bar is appropriately set lower. The empirical results cited by experimentalists do indeed provide some interesting support for the pictures offered by those who challenge the traditional views in question. In Nichols’s case, for example, the empirical results enable the skeptic about moral objectivity to move beyond mere speculation that our belief in moral objectivity is caused by extraneous psychological factors, to the more interesting position of being able to point to specific factors and provide evidence of their operation in some cases. While this remains far from debunking the beliefs in question, as I have emphasized, and it cannot be expected to dislodge those of us who continue to find our reasons for believing in moral objectivity compelling, it does at least add to the credibility of the skeptical position. And it also provides some further reason to scrutinize the belief in question, making more salient the possibility that it might be poorly grounded. That is significant. Even if it does not have the broad debunking power commonly imagined, and does not generally defeat justification for the targeted beliefs, this sort of work might still prove important for some participants in these philosophical debates who are close enough to the fence that this added credibility for the skeptical position and extra potential doubt for the traditional positions might tip the balance.13 Someone who is only weakly attracted to deontology in the first place, and only tentatively moved by the rationales offered for deontological claims, might well find the sorts of explanation offered by Greene, given the empirical support he cites, to be on balance more compelling than the sorts of explanation that would vindicate deontological beliefs. Someone who is only weakly tempted by moral objectivism and is especially concerned with parsimonious explanation might well be most attracted on balance to Nichols’s more parsimonious explanation of what is really behind her temptation to believe in moral objectivity, and so might then renounce that belief. What this shows is that the sorts of undermining explanations of certain moral or meta-​ethical beliefs that many offer with the support of empirical

exhaustive explanation of beliefs generally (within the targeted domain), and that the alternative “good reasons” model I have sketched is not significantly in play. 13. Matthew Braddock has emphasized this point to me.

25

Cognitive Science and Moral Philosophy  • 

255

results can indeed have an important role to play for some people in the philosophical debates, shifting them away from certain traditional philosophical positions. And they can also play a role even for those who are not so moved by them. The kinds of studies Nichols cites, for example, might at least make us question some of our objectivity judgments in certain contexts. And they might provide a plausible theory of error for cases where even objectivists might agree someone has gone wrong in her objectivity judgments (for example, because she made them while angry). Similarly, some of the considerations Greene cites might provide a plausible account of some judgments that even deontologists will agree are erroneous (for example, a case where mere squeamishness about hands-​on violence leads to certain morally implausible deontological judgments). Even those of us who hold the traditional philosophical views targeted by experimental philosophers can therefore grant that this empirical work is philosophically significant and interesting. What I deny is that the empirical work in the service of potential undermining explanations of certain beliefs succeeds in providing us with anything strong enough to be worth calling a “debunking argument,” as this is typically understood in current debates.

10.6 Conclusion: A Dilemma and a Challenge Let me conclude by posing a dilemma for experimentalists with strong debunking ambitions. On the one hand, they may make strong, sweeping claims about the etiology of the beliefs they are targeting that, if true, would debunk them; but then it is highly dubious whether the scientific data they cite really support such strong etiological claims, at least in a way that doesn’t just beg central questions against opponents in the ways we have seen. On the other hand, they may instead make more cautious, modest claims that are indeed plausibly supported by the scientific research; but then these claims are too weak to debunk the beliefs in question, at least in the manner hoped for. I have argued that Greene falls prey to the first horn: His debunking claims extend far beyond the very modest conclusions actually supported by the empirical research he cites. By contrast, Nichols falls prey to the second horn: He is mostly careful to state his conclusions from the empirical research very modestly, so that they are plausibly supported by that work; but these reasonable claims—​and any similarly reasonable claims we can expect from further research in the same vein—​are far too weak to debunk our belief in moral objectivity in the sense described above. Nichols’s claims could of course be strengthened so that they would, if true, genuinely support strong debunking conclusions, but

526

2 5 6   • 

William FitzPatr ick

then this leads right back to the first horn, as noted earlier. Indeed, it is hard to see how that first horn will ever be avoided by arguments employing premises making the very strong and sweeping explanatory claims that would need to be made in order to support strong debunking conclusions: Scientific work may well support weaker and more limited explanatory claims about our beliefs, but it is far from clear how it could go further without devolving into scientism, just ignoring or discounting plausible alternative philosophical explanations that, if true, would block the proposed debunking. Experimentalists with strong debunking ambitions thus face a challenge going forward. They need to show that they can avoid this dilemma by doing two things:  (i) basing their debunking conclusions on explanatory claims that are genuinely strong enough to support those conclusions (as Greene does but Nichols does not), while at the same time (ii) making a compelling, non-​question-​begging case that the explanatory claims they rely upon are genuinely supported by the empirical research they cite (as Nichols does but Greene does not). It is this combination that is required for a successful experimentalist debunking argument. But I have argued that it is plainly missing in the prominent work discussed here. Perhaps there are better ways of trying to meet these two conditions. But this is the challenge, in any case, that should be explicitly addressed and met by those who purport to be offering actual debunking arguments in ethics or meta-​ethics, based on empirical research. Alternatively, and far preferably, those interested in experimentalist lines of critique of traditional views could simply drop the strong debunking ambitions and settle instead for the more modest and plausible aims I have suggested. This would involve bringing empirical research to bear on the philosophical debates in a way that lends some support to already familiar challenges to the views being targeted, while avoiding overreaching in the philosophical conclusions drawn from the scientific work. This would still be important, empirically informed work that would have a significant impact on the philosophical debates. It might even provide a nudge past a tipping point for some participants in the debate. But it would avoid misrepresenting or exaggerating the implications of scientific work for normative ethics or meta-​ethics in the way that much current talk of “debunking” has tended to do. The interdisciplinary debates over these topics would be much improved as a result.14

14. I am grateful to Rik Peels, Jeroen de Ridder, and René van Woudenberg for very helpful comments on an earlier draft of this chapter.

257

Cognitive Science and Moral Philosophy  • 

257

References Annas, Julia. 2011. Intelligent Virtue. Oxford: Oxford University Press. Dancy, Jonathan. 2014. “Intuition and Emotion.” Ethics 124(4): 787–​812. FitzPatrick, William J. 2014. “Why There Is No Darwinian Dilemma for Ethical Realism.” In Challenges to Moral and Religious Belief: Disagreement and Evolution, edited by Michael Bergmann and Patrick Kain, 237–​ 255. Oxford:  Oxford University Press. FitzPatrick, William J. 2015. “Debunking Evolutionary Debunking of Ethical Realism.” Philosophical Studies 172(4): 883–​904. FitzPatrick, William J. 2017. “Why Darwinism Does Not Debunk Objective Morality.” In The Cambridge Handbook of Evolutionary Ethics, edited by Michael Ruse and Robert Richards, 188–​201. Cambridge: Cambridge University Press. Greene, Joshua D. 2014. “Beyond Point-​and-​Shoot Morality: Why Cognitive (Neuro) Science Matters for Ethics.” Ethics 124(4): 695–​726. Joyce, Richard. 2006. The Evolution of Morality. Cambridge MA: MIT Press. Mikhail, John. 2014. “Any Animal Whatever? Harmful Battery and Its Elements as Building Blocks of Moral Cognition.” Ethics 124(4): 750–​786. Nichols, Shaun. 2014. “Process Debunking and Ethics.” Ethics 124(4): 727–​749. Railton, Peter. 2014. “The Affective Dog and its Rational Tale:  Intuition and Attunement.” Ethics 124(4): 813–​859. Street, Sharon. 2006. “A Darwinian Dilemma for Realist Theories of Value.” Philosophical Studies 127: 109–​166.

528

11

PHYSICALISM, NOT SCIENTISM

Alyssa Ney

11.1 Introduction Physicalism, a position that has for some time been the received view in philosophy (certainly in contemporary metaphysics), is usually formulated roughly as the proposition that the world is simply the way physics says it is. As appealing an idea as this is to many of those impressed by the history of successes of physical science, demystifying a range of phenomena from ordinary chemical reactions to the motions of stars and planets to biological processes in plants and animals, others view it as a form of scientism, and thus problematic, as scientism is a controversial and much maligned position. Although rough formulations of physicalism can make it look as if physicalism is simply a reductionist form of scientism, as I’ll argue here, attention to more nuanced and plausible characterizations of the position will show this is not so. As is well demonstrated by the contributions in this volume, scientism, like physicalism, may be understood in a variety of subtly different ways. The core idea I  mean to be distinguishing from physicalism in the present chapter is what Susan Haack has called that “kind of over-​enthusiastic and uncritically deferential attitude towards science, an inability to see or an unwillingness to acknowledge its fallibility” (2012), and not just its fallibility but also its limitations. As Rik Peels argues in his “A Conceptual Map of Scientism,” versions of scientism vary from one philosopher or scientist to another, but those seen to advocate scientism are typically those who believe that our scientific theories, or perhaps more narrowly, the physical sciences, should be expanded to cover more or perhaps all domains of inquiry, often because it is thought that science or scientific methods are best positioned to bring us

529

Physicalism, Not Scientism  • 

259

reliable knowledge about the world. As a corollary, those who adopt the position commonly believe that when it comes to conflicts between science and other modes of discovery—​common sense, naïve perception, religion, philosophy—​it is never our scientific theories that are in the end wrong.1 I wish to show here that physicalism simply fails to entail these strong claims either about the imperialistic reach of physics or that in cases of conflict, that physics may never be corrected; it is a more moderate position than may appear from rough formulations.

11.2 Fundamentality What is physicalism? A typical formulation is: Physicalism: the view that the world is the way physics says it is. That is to say, it is within physics and physics alone that one can find a true account of reality. But what does this entail? If physics doesn’t ever mention spaghetti, does this mean the physicalist must deny its existence? Or what about the more interesting cases of phenomena that appear unaddressed by physics: free will, intentionality, consciousness, God? The answer is that no, physicalists do not typically deny the existence of spaghetti, nor free will, nor intentionality, nor consciousness (e.g., Dennett 1991; Jackson 1998; Kim 2005; Montero 1999); although God is a separate matter as I  will explain shortly. And this is so even despite the fact that mainstream physics discusses none of these phenomena. Some physicalists, Alex Rosenberg (2011, this volume) is a good example, adopt a strong scientistic attitude, and accordingly reject what is not mentioned by physics. But here these physicalists go beyond what is required by physicalism. The reason why physicalism per se does not entail eliminativism about those entities or phenomena not mentioned by physics is made transparent by noting that more carefully formulated, physicalism is not a position merely about what exists, but rather about how the world is fundamentally:

1. This is strictly speaking weaker than a claim of outright infallibility that may be read into Haack’s interpretation of scientism above, but this is what I take her to be suggesting. It is not that our scientific theories cannot be wrong. Of course they can. Science evolves and theories change. Rather, the idea is that in cases of conflict between a scientific theory and the result of some other method of inquiry, it is always that other method, not the scientific theory, that must be corrected or reinterpreted.

620

2 6 0   • A l y s s a   N e y

Physicalism: the view that the world is fundamentally the way physics says it is. And of course one may accept this claim, accept that physics does not mention spaghetti, and nonetheless accept its delicious existence. The consequence will simply be that spaghetti does not exist fundamentally. The same applies for free will, intentionality, and consciousness. According to the physicalist, these are not fundamental phenomena. And so their existence is compatible with physicalism given this more accurate formulation. The case of God must be treated differently for the simple reason that theists typically endorse not merely the claim that God exists, but that God is a fundamental being. And so, given that physics does not mention the existence of God, at least this typical kind of theism will be in tension with physicalism.2 All of this raises the question of what it means to say that something is fundamental. What is the difference after all between existing and fundamentally existing? This is a vexed topic for philosophers, but there are a couple of ways to understand it that have been developed in metaphysics. On one way of thinking about fundamentality, our fundamental theories provide a catalog of the entities in terms of which all others are built or made up.3 The non-​fundamental entities are those that are built up out of the fundamental entities. The character of this building relation will vary depending on the types of entities one finds in the fundamental domain and the types of entities one wants to see as constituted out of them.4 It could be that the fundamental physical entities are the parts out of which everything else is built up mereologically (Oppenheim and Putnam 1958) or that the fundamental physical entities realize the non-​fundamental in being the things that play the causal roles constitutive of being strands of spaghetti or cells or organisms (Melnyk 2003; Putnam 1975). Many argue that physical entities should be viewed as constituting a supervenience basis for all other facts so that once all of the facts about the arrangement of physical objects and features are fixed, the existence of everything else follows necessarily ( Jackson 1998; Lewis 1983; Stoljar 2016). There is a contentious debate among physicalists about

2. How might one argue that God exists as a non-​fundamental being, so that theism is not in tension with physicalism? One might claim that God is simply the totality of the cosmos as described by physics, or more contentiously, that God is a fictional character, one whose being is constituted by other non-​fundamental physical objects like religious texts and people’s beliefs (see, e.g., Schaffer 2009). 3. See the various formulations in Melnyk (2008). 4. See Bennett (2011) for a survey of the various kinds of building relations.

216

Physicalism, Not Scientism  • 

261

whether mere supervenience of all facts on the physical facts is sufficient for physicalism. Some insist that unless we can find some entities in the physical domain to identify with our target phenomena physicalists should deny their existence (Kim 2008; Ney 2007; Polger 2004), even if the relevant supervenience relations obtain. But the point remains, many physicalists are happy to accept the existence of entities not explicitly mentioned by physics as long as they may be viewed as built out of, realized by, identified with, or supervenient upon physical entities. There is another way of understanding the physicalist’s claim that the world is fundamentally the way physics says it is. The difference here is subtle, but it is worth mentioning as this latter interpretation of fundamentality has become influential in recent years. Start by thinking of the fundamental not as a collection of objects, but rather as a collection of facts or theories. The fundamental domain then will be something like a comprehensive theory of the way the universe is so that all facts ought to have a complete explanation or grounding in these facts the theory describes. The explanation of non-​fundamental facts in terms of the fundamental theory may be complicated and indirect, but if there are to be any non-​fundamental facts, they must have an ultimate explanation in terms of this theory. This is a perspective (really a class of perspectives) on fundamentality developed by Gideon Rosen (2010), Ted Sider (2011), and Kit Fine (2012). It is a useful alternative way to think about fundamentality because sometimes when we are trying to see the connections between physics and our pre-​or nonscientific beliefs, we are not interested so much in accommodating the existence of a particular entity or class of entities in a physicalist ontology, but rather trying to see how a certain set of facts may obtain given what we know physics says. To give an example, no physical theory mentions anything about morality, what would be ethical or unethical to do in a certain situation, how to be a good person and flourish in the world. And yet one might reasonably think there are moral facts, such as that the arbitrary killing of innocents is wrong. Since the physicalist believes the world is fundamentally the way physics says it is, if there are to be such moral facts according to the physicalist, there must be some way of understanding how these facts may possess an ultimate explanation in physical terms. The explanation of how morality fits into a fundamentally physical world will likely be complicated and need to proceed in stages. A physicalist might, for example, explain the moral facts in terms of facts about human desires and plans or suffering, then explain these psychological facts in turn using facts about certain neurobiological processes, and then finally provide a physical explanation of these. The mark of success for

26

2 6 2   • A l y s s a   N e y

such an explanation is, after all is said, that there be no leftover questions remaining about whether if our physical theory is true, the arbitrary killing of innocents is wrong. We can see in this example how the second understanding of what it means for physics to be fundamental is more appropriate than the first. What we are looking for when we attempt to accommodate morality in a physical world isn’t a way of seeing some distinctively moral entities as built up out of physical entities, but instead a way of finding explanations for moral facts in the terms of physical theory. Properly understood, we may now see that physicalists are not forced to be skeptics about spaghetti or morality or in general what is not mentioned in our physical theories. So long as these phenomena may be seen as constituted out of or explained by the more fundamental physical theory, they may be accepted, even embraced by the physicalist, even if they are never mentioned by physics.5 Establishing this suffices to show that physicalists need not think that the sciences or physical science should be expanded to cover more or all domains. For the physicalist is committed only to physics’s successful description of the fundamental.6 Other sciences, ordinary perception, introspection, philosophy, the examination of literature and art, and perhaps even common sense all may provide legitimate (even essential) means of discovering various realms of non-​fundamental facts. The task of connecting the non-​fundamental to the fundamental through constitutive or grounding explanations falls some of the time to the physicist (for example, in quantum mechanical explanations of atomic and molecular states and chemical reactions or electrochemical explanations of biological processes). But just as often such inter-​level explanations fall to scientists in other disciplines or even philosophers (as in the sort of explanations of morality mentioned earlier). And so to know all of the facts, one must move beyond physics. Physics doesn’t have the resources to discover everything or all of the facts (only those that are fundamental). Still, there might be a concern that there are limits to what these other methods of inquiry might show us. Sure, the physicalist might allow that we may make legitimate discoveries using naïve perception or common sense,

5. See also Poland (1994: 98) who emphasizes that the physicalist is only committed to physics’s providing “realization, determination, and explanation of all that there is in nature.” 6. She need not deny the obvious fact that physics in fact covers more than the fundamental. There are of course solid state physicists, astrophysicists, etc., whose job is at least partially to study non-​fundamental physical phenomena. But the physicalist’s central position only concerns physics’s ability to discover the fundamental.

236

Physicalism, Not Scientism  • 

263

discover the existence of genuine entities. But there is still a concern that for these methods to allow us to state facts or discover more of what there is, there remains a constraint, namely that these discoveries and postulates be compatible with and explainable in terms of what is provided by our physical theories, and thus that physics is the final arbiter of what is true and real. Is this not still scientism?

11.3 Hempel’s Dilemma The key will be to recognize that the way physicalism regulates one’s beliefs about what sorts of things fundamentally exist, what is real and true, is dynamic and evolving. Physics and more broadly the physical sciences change and so the physicalist’s view of the fundamental (if she is informed) will change as well. From what we have just noted, it is clear that even if the physicalist were committed to everything’s possessing an explanation in terms of one static physical theory, this would still not be scientism in the sense that that position was characterized in the opening pages of this chapter, namely as the view that science should expand to cover more or perhaps all domains of inquiry, since there may be better means of grasping the non-​fundamental.7 But we just saw there is a worry that because according to the interpretation of physicalism we have been considering, all nonscientific facts must be explainable by physical theories, all entities discovered by nonscientific means must be constituted out of physical entities, that where this is not so, the physicalist must deem the putative non-​fundamental facts non-​facts; the entities discovered by nonphysical means non-​entities. And so for the physicalist, it looks as if the successes of nonscientific or non-​fundamentally scientific methods of inquiry are contingent on whether they may be fit into the picture of the world we recover from physics. If a nonscientific or non-​fundamentally scientific claim is one for which it is difficult to see how it could be fit into the picture of the world we get from physics, then the physicalist must reject it. And then it looks as if physicalists are committed to what I above referred to as a corollary of the scientistic position, that in cases of conflict between science and other modes of inquiry, it is never the scientific theory that must be revised. The physicalist doesn’t avoid scientism after all. We will see why the physicalist is not committed to what I am calling this corollary of physicalism by way of considering a puzzle the physicalist faces,

7. For defense of this as a reasonable characterization of scientism, see Peels (this volume).

624

2 6 4   • A l y s s a   N e y

a puzzle that arises out of the dynamic and changing character of physics and physical theories. It is by recognizing and addressing this puzzle that we may better understand why the physicalist is not committed to rejecting verdicts of other methods of inquiry even when they appear difficult or even impossible to reconcile with physical theory. So let us put scientism to one side momentarily and see the issue. Following the interpretation of the previous section, physicalists believe that the world is fundamentally the way physics says it is. But what are we talking about when we say the world is fundamentally the way physics say it is? The most obvious thing this might mean is that the world is fundamentally the way our best current physical theories say it is. This would then suggest that physicalism is the claim (depending on how we understand fundamentality talk) that everything there is is built out of the entities mentioned by our current physical theories, or that every fact has an explanation that bottoms out in the claims of our current physical theories. But if this is what physicalism comes to, then physicalism is straightforwardly false. For as successful as physics has been, the end is not yet in sight. Physics includes claims that will most definitely be corrected by future theories, and there remain many phenomena within its own domain that physics does not yet have the resources to explain. To emphasize how far we are from having in our hands such a true and complete physical theory, consider how many results taken for granted by our best confirmed and most resilient theories, such as the Standard Model of particle physics, are continually being revised by experiment. For example, until the late 1990s and the Nobel Prize-​winning work of Kajita and McDonald, it was assumed that the Standard Model was correct in its claim that neutrinos lack mass. Experimental work now continues trying to give us a more accurate picture of the nature of these particles, including a precise value for their mass. Or, for a case in which we know there is even more we have to learn, consider the discoveries of dark matter and dark energy. These strange new phenomena are estimated to make up 27 percent and 68 percent of our universe respectively and we have nearly no idea what they are. Even when it comes to what we do seem to know, what we’ve learned about matter from the extremely successful quantum field theories making up the Standard Model and about space-​time from general relativity, these too reveal the limitations of current physics. For these two pillars of fundamental physics have yet to be reconciled into a single agreed upon theory of quantum gravity. Doing so, it is widely recognized, is a far from trivial task and will require significant revolutions in physics. So we have good reason to believe that the world isn’t, as a matter of fact, fundamentally the way current physics says it is. However maybe this isn’t the

265

Physicalism, Not Scientism  • 

265

only possible way to interpret the “physics” in physicalism. Maybe instead of understanding physicalism as the view that the world is fundamentally the way current physics says it is, we should take it as the view that the world is fundamentally the way the future, true, final, completed physical theory says it is. This interpretative strategy relieves physicalism of the problem of being false and so has been embraced by many philosophers (Dowell 2006; Lewis 1983; Loewer 2001). However, as many others have noted, there are problems with this construal of physicalism as well. I’ll discuss three. First, many view physicalism as committing its adherents to at least some substantive ontological claims about the nature of our world. Physicalists believe in fermions and bosons, electromagnetic fields, quantum entanglement, gravitation, and curved space-​time. But since no one now knows what future physics will look like, no one knows whether any of these things will be preserved by the time (if there is a time) in which physics is completed. Physics could continue for millennia and just as the things physics postulates today look very different from what was postulated a millennium ago, we can’t be sure what future physical theories will contain. And so if physicalism is supposed to be the view that the world is the way this future, completed, true, but unknown theory says it is, then we don’t really know what physicalists are committed to (Hempel 1980). A second concern some have raised (Chomsky 2003; Montero 1999) is that understanding physicalism as the view that the world is fundamentally the way the future, completed physical theory says it is threatens to make the view trivial. This is because one might think that for physics to be completed, it will have to expand to have the resources to explain all of the phenomena that exist in our world. Physics differs from other scientific theories like biology, psychology, and economics precisely in having within its purview the fundamental structure of the universe and all objects in it, not a subset of those objects. Physics doesn’t just study living systems or thinking systems or financial systems. It studies all systems. But if this is so, then it is a trivial matter that the world is the way that a completed physical theory says it is, since that by definition would be a complete theory with the resources to explain everything. One illustration of how this way of understanding physicalism threatens to trivialize the position concerns the case of phenomenal consciousness. Most of those who reject physicalism today, for example dualists (Chalmers 1996; Nagel 1974), do so because they think that physics is in principle incapable of capturing the existence of phenomenal consciousness, that our experiences have certain qualities constituting the way they look, feel, taste,

62

2 6 6   • A l y s s a   N e y

and so on. The thought is that no amount of information about the distribution of matter or charges or gravitation will be able to explain the way pains feel or strawberries taste. But suppose the dualist is right. Then for physics to be completed, for it to have an account encompassing everything, it can’t rest at giving a complete account of the distribution of matter and charges and the like. It must also add descriptions of all of the various kinds of conscious experiences one might have. So suppose at some time in the future, to have a complete description of everything, physics expands and supplements its theories with the postulation of additional phenomenal features:  tastes, smells, bodily sensations, and so on. This would be a scenario in which the world was fundamentally the way the true final completed physics says it is. But intuitively it would be a situation in which physicalism turned out to be false—​the central debate between physicalists and dualists after all turns on this very issue, whether we need to postulate qualitative features of experience as additional, fundamental features of the world. And so it seems that we cannot understand physicalism as the view that the world is the way the completed future physics says it is. This would be to make physicalism trivial, consistent with the existence of just about anything. So our first two problems are that the understanding of physicalism as the view that the world is fundamentally the way a final, completed physics says it is makes the view contentless (unable to say anything positive about what the world contains) and trivial. Finally, the third problem with this formulation of physicalism concerns an assumption that this way of seeing physicalism commits the physicalist to, an assumption that is contentious and should really be orthogonal to the issue of whether physicalism is correct. This is the assumption that there at some point will be a final, completed physical theory. To see why it is contentious whether there will ever be such a theory, consider the many obstacles standing in the way: computational limitations, bounds on human ingenuity, restrictions in financial or other resources, the possibility of apocalyptic scenarios like catastrophic climate change, and so on. But whether a completed physical theory is ever actually achieved is not relevant to the question of physicalism as it matters to many of us today. Here we may take the opportunity to say a bit about what it is exactly that the physicalist is interested in when adopting this position so we can see why it does not require the existence of a final, completed theory. There are three main components to physicalism or what it is to be a physicalist. The first is that by adopting physicalism, one makes some substantive positive claims about what the world is like. The physicalist is one who takes the lessons of physics seriously, so (as mentioned already) she believes

267

Physicalism, Not Scientism  • 

267

in what physics has shown us to exist so far: electrons and electromagnetic fields, curved space-​time, quantum entanglement, superconductivity, and on and on. Someone who believed in nothing but spaghetti or rocks or the power of love would not be a physicalist.8 Second, being a physicalist also involves having some negative commitments at a given time regarding what the scientific community has good reason to believe is incompatible with or not capable of explanation by current physical theory. So a physicalist, qua physicalist, will reject the existence of things like entelechies, Cartesian souls, irreducible vital or mental forces, and so on. Third, so long as physics has not yet reached completion, physicalists have the expectation that (near) future physical theories will continue to improve in their ability to accurately describe our world and provide the basis for explanations of the non-​fundamental. These central components of the physicalist’s position together distinguish her from the idealist or dualist. But none commits her to the view that at some point in the future, physics will be completed, that it will ever have an account of everything. The physicalist, knowing that physics is not now finished and has a lot left to explain, will believe that future physical theories will improve on current ones. She may also be committed to the fact that physics, rather than other theories, is in the best position going forward to eventually reach a complete theory. But this is not the same as saying it will reach this complete theory. Agnosticism on the issue of a final, completed physics does not undermine physicalism and the positive and negative attitudes just described. And so the position should not be defined as one that can only be correct if there is such a thing as the final, completed physical theory. So we have seen two alternatives for how to interpret physicalism: one that interprets it as the claim that the world is fundamentally the way current physics says it is and the other that has it as the claim that the world is fundamentally the way some future, completed physics says it is. The first interpretation makes physicalism false; the second makes it unsubstantive, trivial, and potentially meaningless (if there is never any completed physics). So both options are problematic. This problem for interpreting physicalism

8. How exactly to understand the positive commitments of the physicalist will depend on how the physicalist addresses the issue of scientific realism. A physicalist may be an object-​oriented scientific realist, a structural realist (one who commits only to the structure of physical theories), or even a constructive empiricist (being committed only to the empirically accessible consequences of our physical theories).

268

2 6 8   • A l y s s a   N e y

has come to be known as Hempel’s Dilemma, after Carl Hempel who perhaps first noted this difficulty in defining physicalism (Hempel 1980). Note (as Hempel himself did) it will not help to define physicalism instead in terms of some future, moderately improved theory striking a middle position between the two horns of the dilemma, for example, in the way David Lewis did: [Physicalism] is the thesis that physics—​something not too different from present-​day physics, though presumably somewhat improved—​ is a comprehensive theory of the world, complete as well as correct. (Lewis 1983: 33–​34) For even though this will do better at making the content of physicalism clear to us and we can be sure there will still be slightly better physical theories in the future, and it isn’t trivial that such a theory will be correct, we have only slightly less reason to say that this theory will be false. As has already been noted, as of right now, we have good reason to think any near descendent of current physics will be missing out on a lot and need to be replaced with a more accurate and more complete theory. To get close to completion, physics will have to undergo substantial revolutions. Some look at this problem for defining physicalism and conclude that it is impossible to make sense of the position and so it is hopeless and we should all set physicalism aside (Crane and Mellor 1990; Stoljar 2010). However, as I  have argued elsewhere (Ney 2008), there is a solution to this problem and that is to reject the assumption that physicalism is a descriptive proposition or claim that the world is the way this or that physical theory says it is. Instead, following Bas van Fraassen (2002), it is best to conceive of physicalism as a complex psychological attitude. And so it is a position in a broader sense, a stance one takes toward the world. The complex attitude involved in being a physicalist primarily involves the three components mentioned above:  the two positive and the other negative. One is a physicalist insofar as one has the disposition to form ontological and more broadly metaphysical commitments to the kinds of things and propositions our best, orthodox, current physical theories describe and one does not form commitment to those things we have good reason to believe are incompatible with or cannot be explained by current physical theories. Physicalism is also constituted by the expectation (so long as physics has not reached completion) that future physical theories will continue to improve in their ability to describe the world as it is. A reasonable physicalist will also

269

Physicalism, Not Scientism  • 

269

recognize, as she should be informed about current physics, although this is not an essential part of the position, that physics does not have anything close to the final theory.9 Note that physicalism is not partly constituted by the belief that the world is the way current physics says it is. This is something any reasonable physicalist should reject. The physicalist, qua physicalist, will have beliefs that the phenomena described by our (best, orthodox) current physical theories exist. She will believe in the existence of electrons, electromagnetic fields, curved space-​time, and so on, and not believe in irreducible qualia or entelechies. (Note these specific beliefs are not required by physicalism per se. Rather one who holds the physicalist attitude will be disposed to believe in whatever she believes current physics says exist.) But this is strictly speaking different from believing that the world is simply the way this theory says it is. Some object to this understanding of physicalism, which ties the physicalist’s metaphysical commitments to current physics, that it does not improve sufficiently on the first interpretation of physicalism we discussed. Adopting the physicalist attitude I describe still commits the physicalist to what are probably lots of false beliefs since we know physics will have to be revised significantly on its way to completion. And so physicalism, even if it is an attitude, can seem to amount to a bad attitude. This is an interesting feature of the position, but I do not think it makes the interpretation of physicalism wrong. Rather, what bringing out this objection shows is that the physicalist is one who, to a certain extent, sticks her neck out in the sense that she adopts commitments to lots of things, some of which she acknowledges might turn out not to be real. The situation of the physicalist is similar to one facing the familiar paradox of the preface.10 The physicalist has excellent reason for each of her beliefs: in the existence of electrons, in quantum entanglement, in curved space-​time, and so on. But she also has excellent reason to believe at least some of these beliefs are false. She doesn’t know which ones since there are always very many opposing views about the correct way to extend

9. In the future, as physics develops and hopefully gets closer to completion, the physicalist will believe that physics is closer to having the complete theory. 10. The preface paradox (Makinson 1965) is that facing the writer of the work of nonfiction who after carefully researching all of the very many claims made in her book recognizes that given human fallibility, it is likely she made at least one mistake somewhere. So to be honest, she acknowledges this fact in her book’s preface. In this (common) situation, the author has excellent reason for making each claim in the main text of the book. But she also has excellent reason to believe at least one of these claims is false. Thus the paradox.

270

2 7 0   • A l y s s a   N e y

current physics. Now one might think that given the fact that she has good reason to believe physics will undergo substantial revolution that all of these beliefs are false. But this probably isn’t so. For typically after theory change, it isn’t that all earlier postulates of a theory are rejected. Of course some postulates are rejected (as most believe was the case with Cartesian souls, phlogiston, and the ether), but what often happens is that what was thought to be some kind of fundamental phenomenon is revealed to be explainable in terms of something more fundamental underlying it and so the commitments are thereby preserved. So the physicalist, precisely because each of her commitments has been so well justified, has good reason to believe many of her individual commitments won’t be shown to be false. There will likely be some future theory to explain most of them. And so she can still be confident in each of her commitments and the picture they provide of fundamental reality.

11.4 Physicalism and Scientism We can now bring the discussion back to the question of scientism. The way things were left at the end of Section 11.2, we saw that the physicalist has no problem allowing that there are many facts that are not at all mentioned by physics and so may be discovered by other scientific as well as by nonscientific methods of inquiry. Indeed she might believe other methods are best placed to discover non-​fundamental truths. This is unquestioningly accepted by the physicalist since her claim is not that the only facts there are are those described by physics, but instead that physics provides a fundamental set of claims that can provide explanatory grounds for all others. There was still a concern, however, that this limited the role of other domains since for the physicalist, other disciplines or methods look to still be ultimately required to agree with what is established by physics, since to be legitimate, their discoveries must all have an explanation in terms of physics. Moreover, if there is ever a conflict between physics and the results of another method of inquiry, the physicalist must always side with the verdict of physics. And this looks still rather scientistic. If physicalism is simply the thesis that the world is fundamentally the way current or a final, completed physics says it is, then this, I think, is the conclusion we are left with. Physicalism is committed to a form of scientism, at least by commitment to what I earlier called a corollary of scientism. Other disciplines can make discoveries, but then to be legitimized, these must be explained by our best current or final physical theories. And

217

Physicalism, Not Scientism  • 

271

so if there is a conflict between some verdict of another means of inquiry and physics, then it is the former that must be tossed out. But as I have just argued, this is not the correct way to understand physicalism. Physicalism isn’t the proposition that the world is the way some specific physical theory says it is. Instead it’s an attitude that involves believing in the kinds of things that current physics says there is and not believing in what one believes is incompatible with or cannot be explained in terms of current physics, with the expectation that future physical theories will be able to explain more. But this leaves a lot of room open for what we might learn about from other disciplines. The physicalist indeed may have a fourfold classification of putative entities and facts: 1. What is described by current physics; 2. What possesses an explanation in terms of current physics or may be seen as constituted by the entities of current physics; 3. What it is reasonable to believe will never have an explanation in terms of current or near future physics or ultimately be seen as constituted by the entities of current or near future physics; 4. What it is not reasonable to believe cannot ultimately be explainable in terms of, or seen as constituted out of entities of current or near future physics. The last three categories correspond to the results of nonscientific or non-​ fundamental scientific methods. We have already discussed those falling under the second category and how belief in them is compatible with physicalism. But we may now see that the physicalist attitude is also compatible with belief in facts or entities in the fourth category. Nothing in the physicalist attitude commits the physicalist to rejecting what falls in this category. She must only have the expectation (and as we saw, this is one component of the physicalist attitude) that those facts that now lack explanations in terms of current physics may be explained in the future by the resources of current or near future physics, and that those that do not presently have an account of their constitution in terms of the entities of current physics may be reasonably seen in the future as constituted out of the entities of current or near future physics. To see this more concretely, consider again the case of phenomenal consciousness. Most philosophers believe that there are such things as phenomenal features of experiences, qualia. We have very good reason to think that pains hurt (they feel a certain way), that there is something it is like to taste a

27

2 7 2   • A l y s s a   N e y

strawberry,11 and so on. This is so even though at this stage, physical science has not progressed to the point of providing an agreed upon explanation of how these phenomenal features arise from physical matter. Some have argued that there is good reason to think that qualia are irreducible in a strong sense, that they could never be explained in terms of a physical science nor do they supervene on physical phenomena (Chalmers 1996; Nagel 1974). If the physicalist accepts these arguments and their conclusion that qualia are indeed irreducible, then the physicalist may reject their existence. This would be to view them as falling under the third category above—​what is irreducible can never be explained physically. However, another option is to think that qualia may be unexplained but at the same time physical, and that a future, more complete understanding of physical theory will allow for their understanding. This is a way of viewing them as falling under the fourth category above. This is to reject only the current reducibility of phenomenal features, not their existence. With respect to phenomena that we have good (perhaps nonscientific) reason to believe exist but do not yet have a physical explanation, so long as there is no compelling reason to think their existence cannot be explained by current or near future physics, the physicalist may believe in them or remain neutral. The physicalist need not think that everything there is has an explanation in terms of her current physical theory for the simple reason that physicalism does not commit her to saying that science is complete. Moreover, it is not incompatible with physicalism that what we learn using nonscientific methods (like introspection or naïve perception or even common sense) can motivate changes to or reinterpretations of current physical theories. For example, one of the main philosophical challenges for quantum mechanics is what is known as the measurement problem. This is the problem that the laws of the theory entail that systems may evolve into states in which it is indeterminate what features they have, where they are located, or how fast they are moving. Erwin Schrödinger famously used this consequence of quantum mechanics to show that given a simple experimental setup, the theory would predict that a cat could be put into a state in which it was indeterminate whether it was alive or dead. We don’t ever observe states like this. Instead what we find upon measurement is that physical systems are always in determinate states of position or momentum, whether they are living or dead. This is the measurement problem:  how

11. Perhaps this is different for different people.

273

Physicalism, Not Scientism  • 

273

to reconcile (i)  the fact that quantum theories permit systems to evolve into states that are indeterminate with (ii) what we learn from naïve perception,12 even common sense.13 It is possible to take a strong scientistic approach to this problem, arguing that scientific theories cannot reasonably be challenged this way. One may dismiss the problem as merely philosophical and not something scientists need concern themselves with. Another response to the problem, which strikes me as similarly scientistic,14 is to embrace the consequences of quantum mechanics and argue that they don’t actually conflict with our experience. If our scientific theory says that systems must evolve into states that are indeterminate, then despite appearances, this must be so. This is a strategy that was proposed by the physicist Hugh Everett in the 1950s; it is now referred to as the many worlds theory. The core idea is to argue that when a quantum system is in a (pure) state that is indeterminate between A  and B,15 this means there is a system that is genuinely in both state A and state B. To reconcile this with the fact that a human observer will always see only one state A or B in a given instance, Everett suggested that just as both states are present, so two observers are present. So there are never indeterminate experiences. One observer will always see A and the other will see B (Everett 1957). The result is a theory very much in tension with common sense and background beliefs about how many objects and observers may exist in any given location. However, if one wants to maintain the original scientific theory, sticking to it even in the face of challenges from perception and common sense, then Everett provides a way to do this. On the other hand, one might argue it is reasonable to pursue a less scientistic approach, that naïve perception and common sense do give us reason to revise quantum mechanics, to avoid these puzzling consequences. Although no such revision to quantum mechanics is widely accepted, many have been proposed.16 What I  wish to point out here is that recognizing the flaws in

12. By “naïve perception,” I simply mean to refer to the ordinary use of perceptual (visual, auditory, etc.) faculties, rather than the more sophisticated perceptual techniques used in scientific research. 13. A now classic introduction to this problem may be found in Albert (1992). 14. Though more reasonable since rather than ignoring perceptual facts, it tries to reconcile them with the scientific theory. 15. By saying “pure state,” I mean to indicate that the indeterminacy is not due to our ignorance of the full details of the situation, but rather is a feature of the system itself. 16. See Albert (1992) again for a useful overview.

247

2 7 4   • A l y s s a   N e y

quantum mechanics, that there is a measurement problem, is not at all in tension with the physicalist attitude. Not knowing which part of quantum mechanics exactly will have to be revised in order to solve the measurement problem, the physicalist will adopt the ontological commitments entailed by current formulations, while expecting that future formulations will do better at reconciling the theory with perception and common sense. To conclude this section, we may consider more generally what happens in a situation in which physical science and the results of other methods of inquiry appear to conflict and compare how this looks from the perspectives of physicalism and scientism. Those with a scientistic attitude face a choice about what to do with the recalcitrant information. The choice is between eliminativism and compatibilism. Scientism forces one to reject the claim that is in tension with scientific theory or find a way to reinterpret what seems to conflict with the scientific theory so that it no longer conflicts. In some cases, we have seen, this will lead to a tension with common sense. However, the physicalist will recognize that current physics is incomplete, and that in order to improve on current formulations, the scientist may need to acquire information using other nonscientific or non-​fundamentally scientific methods. Though the physicalist will as part of her commitment have an expectation that future theories will be better at explaining recalcitrant phenomena, she need not believe there will ever be such a theory. It may in the end be impossible to provide explanations of these phenomena in terms of physical theory.17 And so physicalism permits a third way (neither eliminativist nor compatibilist) of tolerating the tension, in contrast to the imperialism of the scientistic attitude.

11.5 The Domain of Physicalism So far, in discussing the places where physics may seem to come into conflict with the results of other methods of inquiry, I have assumed the recalcitrant information involves a particular subject matter. All of the examples I  have used concern factual information about actual and contingently existing objects known about through a posteriori means. However, when we think about all of the means by which we may acquire information and all of the facts we might learn by scientific or nonscientific means, some concern not how things actually are, but what is possible. Some concern not what may contingently exist, but rather what is necessary. Some concern the nature

17. A predecessor here is McGinn (1989).

275

Physicalism, Not Scientism  • 

275

of things, basic matters of metaphysics. Some concern mathematical truth. Some concern not objective factual information at all, but rather information that is subjective in some respect, such as that concerning matters of taste, or normativity, concerning what one ought to do or believe. The key difference between these topics and the examples that were discussed above is that in each case there is reason to think that these facts cannot even in principle find explanations in terms of current or even any future physics. And this is because there may be reason to think that in each case the subject matter is not about how our world is as a matter of fact. This is why these facts cannot be derived from facts about how physically our world is. What to say about each of these cases is an issue on which physicalists will differ, but many recognize that although physics aims at comprehensiveness, it is not a comprehensiveness on all matters. For example, Jeffrey Poland claims that: We should understand that [physicalist] theses are concerned only with the natural order, with the universe in which we live and interact. Physicalism applies to all claims that are objectively true or false in nature. This means that the theses do not apply to abstract realms, if there are any: mathematical structures, universals, propositions, and the like. (1994: 227) Interestingly, Poland does believe that the aesthetic and the moral lie in the domain of physicalism, since he claims, these do concern the natural order. Andrew Melnyk restricts his physicalism to concern only those things that exist contingently or play some sort of role in the causal order (2003: 10).18 So, if mathematical entities or universals are necessary existents and don’t causally interact with anything, then their existence would also fall outside of the realm of concern for a physicalist of Poland’s or Melnyk’s stripe. A physicalism restricted in this way would thus open the door for legitimate nonscientific or nonphysical methods of investigation into some topics lying in the third category above. What about those domains just mentioned for which it is more plausible that the relevant topics do concern the natural order, what there is in

18. Frank Jackson stays neutral on this issue stating physicalism “claims that a complete account of what our world is like, its nature (or, on some versions, a complete account of everything contingent about our world), can in principle be told in terms of a relatively small set of favoured particulars, properties, and relations, the ‘physical’ ones” (1998: 6).

267

2 7 6   • A l y s s a   N e y

our world and how it must be? Here, we must distinguish different cases. Although modal facts about what is necessary and possible may seem not to depend on the results of physics, some argue that there is nothing beyond the kind of necessity and possibility we find in the laws of physics (e.g. Shoemaker 1998). Objectively normative, for example, moral, aesthetic, and epistemic, facts may also perhaps find an explanatory basis in physics. Discovering such bases has been a long and difficult research program involving many philosophers.19 If such projects are successful, then normative facts may again fall under the second category of facts and entities described in Section 11.4: facts that we discover using nonscientific or non-​fundamentally scientific methods, and those it is consistent for one with the physicalist attitude to endorse since they possess explanations using the resources of current or near future physics. So, there may be reason to hope that even if the realms of modal and normative facts may be within the purview of physicalism, commitment to them may be compatible with physicalism, even if these facts do not yet possess physical explanations. However, one might think there are a further class of facts that are about how our world actually is and yet these are not even in principle capable of physical explanations. In particular, some have argued that the questions of metaphysics concern the nature of our world as it actually is and yet are not explainable in terms of physics because they are more fundamental than the physical facts (Paul 2012) or they are much broader and more abstract (Sider 2008). For example, metaphysicians may discuss deep questions about physical objects (whether they are mere bundles of properties or properties inhering in substrata, how or even if they may persist through change over time, whether time is in fact real, and what essential features if any objects possess). Physicalism is compatible with a range of views about the status of metaphysical questions. One may reject them wholesale. One may argue that there is no sharp way to make the distinction between questions of physics and questions of metaphysics. One may view metaphysics and its methods as distinct from those of physics, but view it as a way of getting a deeper understanding of the same facts that physics investigates. Physics, for example, may claim that there are fermions and bosons, where metaphysics may fill in what we are talking about when we talk about fermions and bosons. In this way, metaphysics may be viewed as a way of interpreting the claims of

19.  For approaches to explaining epistemic normativity, see Hazlett (2013). For one among very many approaches to explaining morality, see Jackson (1998).

72

Physicalism, Not Scientism  • 

277

physics. Alternatively, a physicalist might regard metaphysics as an enterprise that is not trying so much to state facts about the nature of the world, but rather to offer alternative pictures to supplement the physical story and try to get a better imaginative grasp on the theories physics provides (Ney 2012). Holding the physicalist attitude certainly puts constraints on what one may reasonably believe one can learn from metaphysics: For example, it rules out a view according to which metaphysics provides access to a realm of fundamental truths altogether distinct from the truths of current physics. But as we have just seen, we can see that physicalism is compatible with a range of ways of accommodating a fertile discipline of metaphysics.

11.6 Conclusion So we have seen that although superficially, physicalism may seem to amount to something like a reductionistic version of scientism, by paying attention to more subtle and careful formulations of the position, we can see that the physicalist need not be scientistic. Indeed, when the physicalist properly acknowledges the limitations of current physical theory, she can acknowledge with modesty the possibility of gaining valuable information from other domains of inquiry, not just for matters not discussed by physics but to help guide physics in the agenda it sets for itself as well.

References Albert, David Z. 1992. Quantum Mechanics and Experience. Cambridge, MA: Harvard University Press. Bennett, Karen. 2011. “Construction Area (No Hard Hat Required).” Philosophical Studies 154(1): 79–​104. Chalmers, David. 1996. The Conscious Mind. Oxford: Oxford University Press. Chomsky, Noam. 2003. “Replies.” In Chomsky and His Critics, edited by Louise Antony and Norbert Hornstein, 255–​328. Oxford: Blackwell. Crane, Tim and D.H. Mellor. 1990. “There Is No Question of Physicalism.” Mind 99 (394): 185–​206. Dennett, Daniel. 1991. Consciousness Explained. Boston: Little Brown. Dowell, Janice. 2006. “Physicalism: Empirical, Not Metaphysical.” Philosophical Studies 131 (1): 25–​60. Everett, Hugh. 1957. “Relative State Formulation of Quantum Mechanics.” Review of Modern Physics 29: 454–​4 62. Fine, Kit. 2012. “Guide to Ground.” In Metaphysical Grounding, edited by F. Correia and B. Schnieder, 37–​80. Cambridge: Cambridge University Press.

287

2 7 8   • A l y s s a   N e y

Haack, Susan. 2012. “Six Signs of Scientism.” Logos and Episteme 3(1): 75–​95. Hazlett, Allan. 2013. A Luxury of the Understanding. Oxford: Oxford University Press. Hempel, Carl. 1980. “Comments on Goodman’s Ways of Worldmaking.” Synthese 45(2): 193–​199. Jackson, Frank. 1998. From Metaphysics to Ethics. Oxford: Oxford University Press. Kim, Jaegwon. 2005. Physicalism, or Something Near Enough. Princeton, NJ: Princeton University Press. Kim, Jaegwon. 2008. “Reduction and Reductive Explanation: Is One Possible Without the Other?” In Being Reduced: New Essays on Reduction, Explanation, and Causation, edited by Jakob Hohwy and Jesper Kallestrup, 93–​114. Oxford: Oxford University Press. Lewis, David. 1983. “New Work for a Theory of Universals.” Australasian Journal of Philosophy 61: 343–​377. Loewer, Barry. 2001. “From Physics to Physicalism.” In Physicalism and its Discontents, edited by Carl Gillett and Barry Loewer, 37–​ 56. Cambridge:  Cambridge University Press. Makinson, D.C. 1965. “The Paradox of the Preface.” Analysis 25: 205–​207. McGinn, Colin. 1989. “Can We Solve the Mind-​Body Problem?” Mind 98: 349–​366. Melnyk, Andrew. 2003. A Physicalist Manifesto:  Thoroughly Modern Materialism. Cambridge: Cambridge University Press. Melnyk, Andrew. 2008. “Can Physicalism Be Non-​Reductive?” Philosophy Compass 3(6): 1281–​1296. Montero, Barbara. 1999. “The Body Problem.” Noûs 33(2): 183–​200. Nagel, Thomas. 1974. “What Is It Like to be a Bat?” Philosophical Review 83: 435–​450. Ney, Alyssa. 2007. “Can an Appeal to Constitution Solve the Exclusion Problem?” Pacific Philosophical Quarterly 88(4): 486–​506. Ney, Alyssa. 2008. “Physicalism as an Attitude.” Philosophical Studies 138(1): 1–​15. Ney, Alyssa. 2012. “Neo-​Positivist Metaphysics.” Philosophical Studies 160(1): 53–​78. Oppenheim, Paul and Hilary Putnam. 1958. “The Unity of Science as a Working Hypothesis.” In Minnesota Studies in Philosophy of Science 2, 3–​36. Minneapolis: University of Minnesota Press. Paul, L. A. 2012. “Metaphysics as Modeling:  The Handmaiden’s Tale.” Philosophical Studies 160: 1–​29. Peels, Rik. 2017. “A Conceptual Map of Scientism.” This volume. Poland, Jeffrey. 1994. Physicalism:  The Empirical Foundations. Oxford:  Oxford University Press. Polger, Thomas. 2004. Natural Minds. Cambridge, MA: MIT Press. Putnam, Hilary. 1975. “Philosophy and Our Mental Life.” In Mind Language, and Reality:  Philosophical Papers, Volume 2, 291–​303. Cambridge:  Cambridge University Press.

729

Physicalism, Not Scientism  • 

279

Rosen, Gideon. 2010. “Metaphysical Dependence:  Grounding and Reduction.” In Modality:  Metaphysics, Logic, and Epistemology, edited by Bob Hale and Aviv Hoffmann, 109–​136. Oxford: Oxford University Press. Rosenberg, Alex. 2011. The Atheist’s Guide to Reality: Enjoying Life Without Illusions. New York: W. W. Norton. Rosenberg, Alex. 2018. “Philosophical Challenges for Scientism (and How to Meet Them?).” This volume. Schaffer, Jonathan. 2009. “On What Grounds What.” In Metametaphysics, edited by David Chalmers, David Manley, and Ryan Wasserman, 347–​383. Oxford: Oxford University Press. Shoemaker, Sydney. 1998. “Causal and Metaphysical Necessity.” Pacific Philosophical Quarterly 79: 59–​77. Sider, Theodore. 2008. “Introduction.” In Contemporary Debates in Metaphysics, edited by Theodore Sider, John Hawthorne, and Dean Zimmerman, 1–​ 7. Oxford: Blackwell. Sider, Theodore. 2011. Writing the Book of the World. Oxford: Oxford University Press. Stoljar, Daniel. 2010. Physicalism. London: Routledge. Stoljar, Daniel. 2016. “Physicalism.” In The Stanford Encyclopedia of Philosophy (Spring 2016 Edition), edited by Edward N. Zalta, http://​plato.stanford.edu/​archives/​ spr2016/​entries/​physicalism/​. Van Fraassen, Bas. 2002. The Empirical Stance. New Haven, CT: Yale University Press.

280

12

M O D E R AT E S C I E N T I S M I N   P H I L O S O P H Y

Wesley Buckwalter and John Turri

12.1 Introduction Scientism comes in many varieties (Stenmark 1997; Peels, this volume). One radical version of scientism is the view that science is the only way to acquire knowledge about reality (cf. Rosenberg, this volume; Trigg 1993: 90). Although radical scientism is a coherent view, it is either clearly false or trivial. On the one hand, many organisms gain knowledge about reality but are incapable of practicing science, in any recognizable sense of that phrase, in which case radical scientism is false. The list includes human infants, dolphins, snakes, frogs, sharks, octopi, spiders, and many others. On the other hand, suppose that “(practicing) science” is understood so loosely as to count these organisms as scientific practitioners. In that case, any way of knowing counts, by stipulation, as “science,” and radical scientism becomes trivial and uninteresting. A less radical version of scientism is the view that science is a good way of answering any evaluable question (see also Atkins 1995; Radnitzky 1978: 1008). The plausibility of this view depends mainly on what counts as practicing science and what makes for a good way to answer a question. For example, suppose your friend asks whether you heard what he just said. This is an evaluable—​ and in some contexts important—​question. You immediately say, “Yes,” because it is fresh in your memory. Given the relative efficiency of simply relying on memory, it would be bad to instead respond by practicing science. To take another example, suppose your partner asks whether you still love them. It would be bad to begin evaluating the hypothesis “I still love you” via the scientific method. Of course, if simply relying on memory or attending to

218

Moderate Scientism in Philosophy  • 

281

one’s feelings counts as practicing science, then this version of scientism also becomes trivial and uninteresting. In this chapter, we will defend a more moderate claim concerning science:  moderate scientism. Moderate scientism is the view that science can help answer questions in disciplines typically thought to fall outside of science. (This is very similar to what Stenmark (2004) calls “scientific expansionism”), according to which “the boundaries of science can and should be expanded in such a way that something that has not been understood as science can now become a part of science” (Stenmark 2004: xi–​xii). Stenmark also calls this view “scientism” (Stenmark 2004: xii; see also Stenmark 2000). As a proof of concept for moderate scientism, we will examine the role that empirical science has played in a discipline often perceived as far removed from empirical science: philosophy. When practicing their trade, philosophers often appeal to ordinary usage of words and patterns of judgment or behavior. The basic assumption behind this approach is that patterns in ordinary thought and talk—​at least about categories central to social cognition, such as knowledge, morality, belief, assertion, or freedom—​can be used as evidence for philosophical theories of important categories. This approach is common throughout the history of philosophy. Aristotle, for instance, defended this approach when he wrote that one way to gather evidence in philosophy was to find a balance between different views about a philosophical topic “in the light not only of our conclusion and our premises, but also of what is commonly said about it” (Aristotle 1941 (350 BCE):  1098b, 9–​11). Thomas Reid thought, “Philosophy has no other root but the principles of Common Sense,” and that “severed from this root, its honours wither, its sap is dried up, it dies and rots” (Reid 1764/​ 1997:  19). The approach remains popular today. For example, J.  L. Austin advised that “ordinary language” should get “the first word” in philosophical theorizing (Austin 1956: 11). Wilfrid Sellars argued that identifying the defining features of ordinary thought—​“the manifest image”—​is “a task of the first importance” for philosophers (Sellars 1963: ch. 1). And David Lewis warned, “When common sense delivers a firm and uncontroversial answer about a not-​too-​far-​fetched case, theory had better agree” (Lewis 1986: 194). To characterize ordinary thought and talk, philosophers often draw on their own experiences, social observation, and reflections about what we would say about certain situations (Ducasse 1941:  ch. 10; Fodor 1964; Jackson 1998: ch. 2). Seminal philosophical work has relied on this kind of introspection and social observation, which is a natural place to start (e.g., Locke 1975 (1690): bk. 4.11.3–​8; MacIver 1938; Wittgenstein 1969). Indeed,

28

2 8 2   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

we find it such a natural place to start that we ourselves have written papers contributing such observations to the literature (e.g., Buckwalter and Turri 2014). However, this method of gathering evidence is limited and in some cases has mischaracterized ordinary thought and talk (for recent reviews, see Blouw, Buckwalter, and Turri 2017; Turri 2016a, 2016b). In the remainder of this chapter, we review several ways that empirical science has helped philosophers accurately represent ordinary thought and talk. Experimental, observational, and statistical techniques have significantly contributed to research in epistemology, action theory, ethics, philosophy of language, and philosophy of mind. We also review several ways that empirical science has contributed to our understanding of the judgments and decisions of professional philosophers themselves. Based on this research, we conclude that the case for moderate philosophical scientism is very strong: Science has promoted significant progress in philosophy and its further development should be welcomed and encouraged.

12.2 Epistemology: Direct and Indirect Effects A major debate in contemporary epistemology involves the influence that stakes have on knowledge attribution (Buckwalter 2010; Buckwalter and Schaffer 2015; Cohen 1999, 2013; DeRose 1992, 1995, 2009; Fantl and McGrath 2002, 2007; Hawthorne 2004; May et  al. 2010; Pinillos 2012; Sripada and Stanley 2012; Stanley 2005). Many philosophers claim that ordinarily whether we attribute knowledge depends on how much is at stake or the consequences of error. In support of this claim, philosophers ask us to consider our intuitive reactions to pairs of cases that vary the stakes while stipulating that (something like) a “justified true belief ” is held fixed. If knowl­edge attribution seems to vary along with stakes, then this is interpreted as important data to be accounted for by epistemological theory. For instance, some philosophers interpret this as “evidence of the very best type” that “knows” is a semantically context-​sensitive expression (DeRose 2009: 81), while others interpret this as evidence that how much is at stake is part of what “makes true belief into knowledge” (Stanley 2005: 2). One limitation of this research is that it often proceeds by explicitly stipulating some crucial details of the scenarios. But this does not correspond to the way people ordinarily make such attributions. Ordinarily people must decide these things for themselves in the context of a knowledge judgment. This raises questions about the ecological validity of judgments about the cases, and whether findings from them generalize to situations in which

238

Moderate Scientism in Philosophy  • 

283

knowledge judgments naturally occur. Moreover, researchers stipulate crucial details without taking sufficient precautions to prevent other details of the case from interfering with the stipulations. This is potentially important because recent research shows that many social evaluations, including the attribution of mental states like belief and knowledge, can occur implicitly and automatically (Bargh et al. 2012). Another limitation of this research is that it features cases containing too many variables for introspection or social observation to reasonably track. For instance, in one influential exposition, readers are asked to sequentially consider five permutations on a single case, varying between approximately 100–​150 words each (Stanley 2005: 3–​5). An optimal approach would not only keep track of all the variables but also estimate their interactions with each other and the cumulative impact that this has on knowledge attribution. Social and cognitive scientists have developed tools that can accomplish this. Combined with a randomized assignment in a properly controlled experiment, statistical techniques ranging from basic regression to causal path analysis can be used to discern complex and unexpected relationships among variables (Chickering 2003; Hayes 2013; Iacobucci, Saldanha, and Deng 2007). For instance, such techniques can be used to evaluate whether varying stakes affects knowledge attribution and, if so, whether the affect is direct or indirect (i.e., mediated by other variables). These tools can also evaluate the relative strength of these effects and the overall contribution that they make in predicting knowledge attribution. For precisely these reasons, researchers recently used these tools to study knowledge attributions (Turri and Buckwalter 2017; Turri, Buckwalter, and Rose 2016). They randomly assigned participants to consider one of two minimally matched scenarios that varied the stakes. The construction of these scenarios was guided by prior empirical work from the judgment and decision-​making literature, which identified several factors that influence the perception of stakes (Beach and Mitchell 1978; McAllister, Mitchell, and Beach 1979). And instead of stipulating matters typically considered obvious or crucial for knowledge, researchers had the participants judge those matters for themselves, alongside judging whether the agent had knowledge. Here is one pair of cases involving a protagonist who must make a decision in a low-​stakes context (submitting a provisional report about whether an individual is “jogging” and on “a low-​carb diet”) or a high-​stakes (submitting a final report about whether the individual is a “threat” and “selling arms to terrorists”):

284

2 8 4   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

Jennifer is an intelligence analyst developing a file on Ivan, an elusive foreign operative. Jennifer has a source who tells her that Ivan stopped [his low-​carb diet/​selling arms to terrorists] and is no longer [jogging regularly/​a threat]. Jennifer must submit a [provisional/​final] report on Ivan to her supervisor within the hour. She will [definitely/​definitely not] have a chance to revise her [provisional/​final] report, and she [will not/​will] be held accountable for decisions based on her [provisional/​final] report. After seeing either a low-​stakes or a high-​stakes case, participants judged the important details of the case for themselves. More specifically, they were asked to rate their agreement with these attributions: 1. Jennifer thinks that Ivan no longer [jogs regularly/​is a threat]. 2. It’s true that Ivan no longer [jogs regularly/​is a threat]. 3. Jennifer has good evidence that Ivan no longer [jogs regularly/​is a threat]. 4. Jennifer should write in the report that Ivan no longer [jogs regularly/​is a threat]. 5. Jennifer knows that Ivan no longer [jogs regularly/​is a threat]. When researchers tested these cases on hundreds of participants, they found that stakes affected knowledge attributions (Turri, Buckwalter, and Rose 2015:  Experiment 1). People were more likely to attribute knowledge in a low-​stakes case than in a high-​stakes case. In order to ascertain how stakes affected knowledge attribution, researchers used causal path analysis to model the relationships among stakes (the independent variable in this experimental design) and the five dependent variables (the five judgments that participants made about Jennifer’s situation). On the best fitting causal model, stakes directly affected judgments about how Jennifer should act, and these judgments in turn directly caused judgments of what she knew. These results provide evidence that powerfully vindicates certain theoretical hypotheses about the connection between knowledge and action (e.g., Fantl and McGrath 2009; James 1948 (1879)), with a level of detail and precision unattainable without the tools of empirical science.

12.3 Ethics: Ought Implies Can A long-​standing assumption in moral philosophy is that obligations entail the ability to fulfill them, typically glossed with the slogan “ought implies

258

Moderate Scientism in Philosophy  • 

285

can.” According to this principle, an agent is not morally obligated to act if she is unable to do so. Many contemporary philosophers endorse this principle (Copp 2008; Dahl 1974; Feldman 1986; Flanagan 1991; Hare 1963; Howard-​Snyder 2006; Littlejohn 2012; Moore 1922; Streumer 2002; Van Fraassen 1973; Vranas 2007; Zimmerman 1993). Agreement on ought-​ implies-​can is so widespread that direct arguments for it are rarely articulated (as observed by Stocker 1971:  303). When arguments do appear for the principle, they often appeal to judgments about particular thought experiments, intuitions about ordinary usage, and claims to the effect that “ought implies can” is a core principle of common-​sense moral cognition (Moore 1922: 317). But not all philosophers endorse ought implies can without reservation. Some claim that the principle is “only partially correct” (Stocker 1971: 303) or that “ought” only pragmatically conversationally implicates “can” (Sinnott-​ Armstrong 1984). A handful of philosophers have even rejected the principle outright (Graham 2011; Ryan 2003). Critics frequently rely on intuitions about thought experiments. For example, they often ask us to imagine agents with various “psychological compulsions” such as kleptomania or addiction. If agents in these cases have an obligation to stop stealing or smoking but are unable to do so, then ought implies can is false (Blum 2000; Ryan 2003). Still others have constructed complex and imaginative thought experiments (Fischer 2003; Frankfurt 1969) or cases hinging on other decisive moral principles (Graham 2011) in an attempt to prime intuitions and arrive at a compelling counterexample. In response, champions of ought implies can have claimed that is “easy to deny” that these scenarios are genuine counterexamples (Graham 2011: 342). Given that there seems to be fundamental disagreement in the field, it is natural to wonder whether ought-​implies-​can actually is a core principle of our moral cognition. Researchers recently set out to investigate this question experimentally (Buckwalter and Turri 2015). Instead of using complicated or fanciful thought experiments, they administered straightforward cases featuring ordinary agents with uncontroversial moral obligations frequently encountered in everyday life. When several simple and straightforward cases were tested on hundreds of participants, the overwhelming majority of people attributed moral obligations but denied the ability to fulfill them. In other words, the result was that common-​sense moral cognition utterly rejects ought-​implies-​can. In one experiment, for instance, researchers presented participants with this basic scenario about an innocent bystander named Michael:

286

2 8 6   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

Michael is relaxing in the park when he sees a small girl fall into a nearby pond. She is drowning and definitely will die unless someone quickly pulls her out. This part of the park is secluded and Michael is the only person around. After seeing this scenario, participants saw one of two endings to the story that manipulated the protagonist’s ability. In the unable condition, the protagonist was physically unable to act: But Michael is stricken with a sudden paralysis in his legs and cannot swim to save the girl. As a result, Michael is not physically able to save the girl. While in the able condition, the protagonist was physically able to act: And Michael is a normal adult male and can swim fast enough to save the girl. As a result, Michael is physically able to save the girl. After seeing one of these cases, participants were asked to select the best option that applies from the list of options below concerning moral obligation and ability: 1. Michael is morally obligated to save the girl, and Michael is physically able to do so. 2. Michael is morally obligated to save the girl, but Michael is not physically able to do so. 3. Michael is not morally obligated to save the girl, but Michael is physically able to do so. 4. Michael is not morally obligated to save the girl, and Michael is not physically able to do so. If ought implies can is a central principle of moral cognition, then when people consider which option best applies to Michael, we would expect them to answer very differently depending on whether Michael is able or unable to act. Specifically, participants in the unable condition should strongly disprefer option 2, which says that Michael has an obligation he is unable to fulfill. However, researchers instead found that Michael’s inability made no difference at all to people’s judgments about moral obligation. In both conditions, the overwhelming majority of participants judged that Michael

287

Moderate Scientism in Philosophy  • 

287

was morally obligated despite being able or unable to act. This same basic pattern of response is robust across a wide variety of narrative contexts, types of inability, and manner of probing for moral obligation.

12.4 Philosophy of Mind: Mechanisms and Concepts of Belief A core question in philosophy of mind concerns whether beliefs can be voluntary. According to doxastic voluntarism, it is possible to have the same kind of willful control over our beliefs as we do over of our actions. Doxastic involuntarists deny the possibility of controlling our beliefs this way. Philosophers have traditionally been split on this issue. Descartes, Augustine, Aquinas, Pascal, James, and others have favored doxastic voluntarism. But the dominant view in contemporary philosophy of mind is that beliefs are involuntary (Alston 1988; Bennett 1990; Scott-​Kakures 1994; Pojman 1999; Williams 1973). Each side has claimed that their preferred view is obviously correct. For example, Descartes thought that our ability to voluntarily control belief was “so obvious” that “it must be regarded as one of the first and most common notions that are innate in us” (Descartes 1985 (1644): §39). Doxastic involuntarists respond that the notion of voluntary belief-​formation is absurd, “chokingly unswallowable” (Bennett 1990: 90), or obviously impossible “as a conceptual matter” (Scott-​Kakures 1994: 96). Science is relevant to this debate in two important ways. First, cognitive science is best positioned to investigate the human mind’s powers, including whether it is capable of forming beliefs at will. If belief-​formation was found to be, say, insensitive to experimental interventions on willpower or volitions, then it would support involuntarism. By contrast, if belief-​formation was found to be directly sensitive to such interventions, then it would support voluntarism. Recent evidence suggests that basic physiological processes, such as heart rate, are subject to voluntary control (Lehrer, Sasaki, and Saito 1999), but we are unaware of any related research on belief-​formation. Nevertheless, if something as basic as cardiac rhythm is open to some degree of voluntary control, then it would not be surprising if belief-​formation was too (compare Naylor 1985) Second, science can help evaluate whether the ordinary concept of belief rules out voluntarism. Initially, this would involve a simple experiment randomly assigning people to assess one of two closely matched scenarios. Minimally matched scenarios differ from one in another only with respect

28

2 8 8   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

to the variable of interest their comparison is intended to test. One group reads a scenario in which an agent professes to choose to form a certain belief, and the other group reads a scenario in which the agent does not do this. Then participants rate whether the agent believes the relevant proposition. If voluntary belief-​formation is conceptually impossible, then we would expect no difference in belief-​attribution across the two conditions. As it turns out, this experiment has already been done (Turri, Rose, and Buckwalter 2017). The agent’s professed choice to believe (or disbelieve) the proposition had an extremely large effect on whether participants attributed belief to him. When the agent professed to choose to believe the proposition, participants attributed the belief to him. When the agent professed to refuse to believe the proposition, participants did not attribute the belief to him. Follow-​up studies revealed that this same basic pattern persists across different narrative contexts and ways of probing for belief-​attribution, that belief-​attribution was affected by interventions on the agent’s perceived willpower, and that the agent’s professed volitions can be a much stronger cue to belief-​attribution than even the agent’s evidence is. In one follow-​up study, for instance, participants were divided into six groups. They each read a story about Malcolm receiving a weather report that indicated there was either a 5 percent, 50 percent, or 90 percent chance that it would rain the next day. Then participants were told either that Malcolm is optimistic and says, “I refuse to believe it will rain,” or that he is pessimistic and says, “I choose to believe it will rain.” Participants overwhelmingly denied belief in the optimistic condition and attributed belief in the pessimistic condition. By contrast, the strength of Malcolm’s probabilistic evidence (5 percent, 50 percent, or 90 percent) had only a small effect on belief-​attributions. These results suggest that the ordinary concept of belief fully countenances the possibility of voluntary belief. Indeed, ordinary belief-​attributions can be more sensitive to the agent’s volitions than his evidence.

12.5 Action Theory: Representation of Thought Experiments Philosophers frequently use thought experiments to generate evidence for or against theories. Thought experiments often feature highly fanciful situations involving complex events, underspecified details, and unrealistic outcomes. For example, the following thought experiment was designed to test whether people intuitively judge that acting freely is compatible with a deterministic scenario in which perfect prediction of human choices is possible:

289

Moderate Scientism in Philosophy  • 

289

Recent brain scanning studies have shown that specific patterns of brain activity can be used to predict simple decisions several seconds before people are consciously aware of those decisions. Imagine that in the future brain scanning technology becomes much more advanced. Neuroscientists can use brain scanners to detect all the activity in a person’s brain and use that information to predict with 100% accuracy every single decision a person will make before the person is consciously aware of their decision. The neuroscientists cannot, however, do anything to change brain activity and hence they cannot directly influence thoughts and actions. Suppose that in the future a woman named Jill agrees, as part of a neuroscience experiment, to wear this brain scanner for a month (it is a lightweight cap). The neuroscientists are able to use real-​time information about her brain activity to predict everything that Jill will think or decide, even before she is aware of these thoughts or decisions. However, they cannot alter her brain activity to change what she thinks and does. On election day, Jill is considering how she will vote for President and for Governor. Before she is aware of making any decisions, the neuroscientists can see, based on her brain activity, that she is about to decide to vote for Smith for President and Green for Governor. Just as the neuroscientists predicted, Jill votes for Smith for President and Green for Governor. As with her decisions to vote for Smith for President and Green for Governor, the neuroscientists are able to predict every decision Jill ends up making with 100% accuracy while she is wearing the scanner. Occasionally, Jill tries to trick the neuroscientists by changing her mind at the last second or by stopping herself from doing something that she just decided to do, but the neuroscientists predict these events as well. Indeed, these experiments confirm that all human mental activity is entirely based on brain activity such that everything that any human thinks or does could be predicted ahead of time based on their earlier brain activity. (Nahmias, Shepard, and Reuter 2014: 514) Some researchers found that people overwhelmingly agreed that Jill voted of her own free will (Nahmias et al. 2014, Experiment 1). Since it was stipulated in the thought experiment that “neuroscientists are able to predict every decision Jill ends up making with 100% accuracy,” they concluded that our concept of free will is not threatened by and is compatible with the future possibility of perfect neuro-​predictability of human behavior. However, one potential worry is that people’s judgments might be due to misunderstood,

920

2 9 0   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

ignored, or unspecified details that they systematically “fill in” by relying on background assumptions. This might be especially likely to happen when participants are asked to consider highly contrived and unfamiliar scenarios. A properly conducted experiment can address this worry and thereby help philosophers discover how we would ordinarily think about or categorize Jill’s behavior in this scenario. For example, another group of researchers recently tested whether key details of the thought experiment just described were being represented when people agreed that Jill acted freely (Rose, Buckwalter, and Nichols 2015). They administered the same thought experiment verbatim but also included a series of follow-​up questions. Crucially, one question measured whether participants accepted that Jill’s action was determined. It turns out that they did not. Instead, they judged that Jill could have voted for a different candidate, despite the occurrence of a perfectly predictive pattern of brain activity to the contrary. In fact, mediation analysis suggested that ascribing free will actually caused people to interpret the case in a way that is inconsistent with determinism. These results cast doubt on the claim that such cases demonstrate that people are “intuitive compatibilists.” For although people attributed free will, they did not view the agent’s actions as determined. This result also serves as another demonstration of how scientific tools can be used to aid philosophical inquiry. When philosophers engage in thought experiments, these tools can help to estimate whether the key variable of interest is accompanied and informed by other philosophically relevant variables.

12.6 Philosophy of Language: Assertion Assertion is the main way in which human beings communicate information to each other. An important question at the intersection of philosophy of language and epistemology involves the standards for assertion. What rules or norms govern when an assertion should be made? Some philosophers defend factive accounts of the norm. A factive account entails that only true assertions should be made. The most popular factive account is the knowledge account, which says that you should assert a proposition only if you know that it is true (Hawthorne 2004; MacIver 1938; Moore 1962; Schaffer 2008; Turri 2011, 2013a; Unger 1975; Williamson 2000; for a review, see Benton 2014). Many critics reject factive accounts on the charge that they do not reflect ordinary thought and talk associated with the social practice of assertion (e.g., Douven 2006; Hill and Schechter 2007; Lackey 2007; Kvanvig 2009). In place of the knowledge norm, critics have defended a range of non-​factive accounts that base the standard for assertion on belief (Bach 2008; Bach and

219

Moderate Scientism in Philosophy  • 

291

Harnish 1979) or justified belief (Douven 2006; Lackey 2007), which do not demand that an assertion be true. These philosophers argue that non-​factive accounts better “explain” or “accommodate” “our intuitions” about assertions (Douven 2006; Lackey 2007). There is a straightforward way to test these claims about what is intuitive or accommodating: Study how ordinary language users evaluate assertions. When researchers studied this in controlled experiments, they found that truth had a profound effect on whether people thought an assertion should be made (Turri 2013a). In one study, for example, participants read a simple story about a watch collector, Maria, who maintains a detailed inventory of the thousands of watches she owns. Someone asks her whether she owns a 1990 Rolex Submariner, so she consults the inventory, which she knows is imperfect but extremely accurate. One group of participants saw the story continue in a way that made Maria’s belief about the watch true: And this is just another case where the inventory is exactly right: She does have one. Another group of participants saw the story continue in a way that made Maria’s belief false: But this is one of those rare cases where the inventory is wrong: She does not have one. When Maria’s belief was true, nearly all participants (97 percent) judged that she should tell her guests that she has a 1990 Rolex Submariner. But when Maria’s belief was false, the vast majority of participants (80 percent) judged that Maria should not make the assertion. In other words, holding fixed the assertion’s evidential basis and changing only its truth-​value caused an enormous shift in how people evaluated the assertion. This same basic pattern persisted across different narrative contexts and ways of probing for evaluations across several experiments. The results demonstrate that most people do not find factive accounts of assertion counterintuitive. Science revealed that non-​ factive accounts do not cohere well with our ordinary social practice.

12.7 Methodology: Philosophical Judgments, Decision-​Making, and Behavior Science has recently begun shedding light on how professional philosophers make decisions about important questions. One group of studies pertains to

92

2 9 2   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

whether philosophers are susceptible to well-​known cognitive biases ( Jones and Nisbett 1971; Tversky and Kahneman 1974) when reasoning about philosophical matters. The main finding here is that professional philosophers are equally susceptible to these biases as other people. For example, professional philosophers are susceptible to order effects when making philosophical judgments about cases (Liao, Wiegmann, Alexander, and Vong 2012; Schwitzgebel and Cushman 2012, 2015; Swain, Alexander, and Weinberg 2009). Professional philosophers also exhibit framing effects, such as whether a scenario is considered from the perspective of an actor or an observer (Tobia, Buckwalter, and Stich 2013; Tobia, Chapman, and Stich 2013), or whether salient features are described in terms of a gain or a loss (Schwitzgebel and Cushman 2015). Another group of studies pertains to how individual differences affect professional philosophers’ judgments. For example, researchers have found that heritable personality traits such as extraversion or emotional stability correlate with judgments about free will and moral responsibility in lay populations (Feltz 2013; Feltz and Cokely 2009), and that this may also be true to some extent of professionals (Schultz, Cokely, and Feltz 2011). Researchers in experimental epistemology have argued that intuitions among professional philosophers about knowledge may be influenced by their native language (Vaesen, Peterson, and Van Bezooijen 2013). Research has also suggested that intuitions about the reference of proper names vary according to academic areas of specialization within philosophy and linguistics (Machery 2012) as well as across Western and East Asian cultures (Machery, Olivola, and Blanc 2009; Sytsma et al. 2015). At the same time, researchers have also found evidence that some philosophical intuitions may be shared by individuals across many different cultures, which some have argued is essential for understanding their evidential significance (Kim and Yuan 2015; Machery et al. 2015; Turri 2013b). A related but distinct line of research investigates how specialization in philosophy impacts the behaviors of professional philosophers in their daily lives. This work has mainly examined whether training in ethics encourages moral and other pro-​social behavior such as answering student emails, voting, staying in touch with one’s mother, or being considerate to others in various ways at professional conferences (Rust and Schwitzgebel 2013; Schwitzgebel 2009; Schwitzgebel and Rust 2009, 2010, 2013; Schwitzgebel, Rust, Huang, Moore, and Coates 2011). Researchers have found little if any evidence that professional training in ethics promotes better behavior. The results of one recent meta-​analysis found no difference between ethicists and non-​ethicists,

293

Moderate Scientism in Philosophy  • 

293

despite including over 18,000 individual observations across 18 different measures of behavior (Schwitzgebel and Rust 2016).

12.8 Conclusion Moderate scientism is the view that science can help answer questions in disciplines typically thought to fall outside of science. In good scientific spirit, we have approached moderate scientism as a hypothesis to be evaluated in light of existing evidence. As a case study, we reviewed a range of recent scientific work focused specifically on questions of long-​standing philosophical interest, and on the judgments and behavior of professional philosophers themselves. In each case, scientific research deepened our understanding of the underlying issues and advanced the debate. For example, when it comes to assessing the relationship between stakes and knowledge attributions, advanced statistical techniques produced a clear and testable model of a complex set of variables, which even the most astute introspection or observation is helpless to evaluate when unaided by the tools of cognitive and social science. Behavioral experiments showed that common-​sense morality implicitly rejects the “ought implies can” principle, that the ordinary concept of belief allows for the possibility of voluntary belief-​formation, that subtle misinterpretations can lead us to misunderstand the true meaning of our intuitive reactions to philosophical thought experiments, and that objections to certain theories about assertion are deeply mistaken. A series of recent empirical studies has also revealed the extent to which professional philosophers, just like ordinary people, exhibit a suite of cognitive biases affecting judgments about philosophically important categories. In light of all this, we conclude that, at least with respect to philosophy, moderate scientism is true. Scientific research has promoted significant progress in philosophy, and its further development within the field should be welcomed and encouraged. Philosophers often appeal to ordinary patterns of judgment and behavior. The methods and techniques of science have helped philosophers to accurately represent how these things work. In some cases, these techniques track complex relationships beyond anything that introspection or social observation could otherwise reasonably track. (The best example of this reviewed here is the work on knowledge attributions and stakes, which used multiple linear regression and causal modeling.) This suggests that sometimes scientific tools are a practical necessity when addressing certain philosophical issues.

924

2 9 4   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

There is at least one noteworthy limitation to our conclusion. Nearly all of the research we reviewed focused on concepts central to social cognition and evaluation. Broadly speaking, it focused on judgments pertaining to knowledge, belief, moral obligations, abilities, and actions. But some areas of philosophy are not primarily concerned with these categories. For example, some logicians focus on the relationship of logical consequence and some metaphysicians focus on the ultimate basis of attribute agreement. It might be claimed that common-​sense categories and even the entirety of empirical science are completely irrelevant to philosophical inquiry in these areas. We acknowledge that nothing we have said here casts doubt on this claim. Ultimately, such things are best evaluated on a case-​by-​case basis (for examples of such evaluations, see Paul 2016; Ripley 2016). We suspect that our conclusion about moderate scientism in philosophy generalizes to neighboring fields in the humanities. Although details will vary across different areas of inquiry, all responsible researchers seek accurate and detailed evidence for their conclusions. As the name “humanities” suggests, humanities research essentially involves human perceptions, reactions, and evaluations of worldly affairs. We deny that science is essential to all knowl­ edge and evidence (see the Introduction), and we do not believe that all phases of humanistic scholarship must be modeled after science. Nevertheless, social and cognitive science provide excellent tools for collecting, analyzing, and interpreting human perceptions, reactions, and evaluations. Thus, we expect that the benefits of philosophical science will be reflected in other areas of the humanities.1

References Alston, William P. 1988. “The Deontological Conception of Epistemic Justification.” Philosophical Perspectives 2: 257–​299. Aristotle. 1941 (350 BCE). Nichomachean Ethics. Translated by W.D. Ross. In The Basic Works of Aristotle, edited by R. McKeon. New York: Random House. Atkins, Peter 1995. “Science as Truth.” History of the Human Sciences 8(2): 97–​102. Austin, J.L. 1956. “A Plea for Excuses.” Proceedings of the Aristotelian Society 57: 1–​30. Bach, Kent. 2008. “Applying Pragmatics to Epistemology.” Philosophical Issues 18(1): 68–​88.

1. For helpful feedback, we thank Adam Mantha, Jonathan Simard, and Angelo Turri. This research was supported by the Social Sciences and Humanities Research Council of Canada and the Ontario Ministry of Economic Development and Innovation.

925

Moderate Scientism in Philosophy  • 

295

Bach, Kent and Robert Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press. Bargh, John A., Kay Schwader, Sarah Hailey, Rebecca Dyer, and Erica Boothby. 2012. “Automaticity in Social-​Cognitive Processes.” Trends in Cognitive Sciences 16(12): 593–​605. Beach, Lee Roy and Terence R. Mitchell. 1978. “A Contingency Model for the Selection of Decision Strategies.” Academy of Management Review 3(3): 439–​4 49. Bennett, Jonathan. 1990. “Why Is Belief Involuntary?” Analysis 50(2): 87–​107. Benton, Matthew. 2014. “Knowledge Norms.” Internet Encyclopedia of Philosophy, http://​www.iep.utm.edu/​kn-​norms/​. Blouw, Peter, Wesley Buckwalter, and John Turri. 2017. “Gettier Cases: A Taxonomy.” In Explaining Knowledge: New Essays on the Gettier Problem, edited by Rodrigo Borges, Claudio de Almeida, and Peter Klein, 242–​252. Oxford: Oxford University Press. Blum, Alex. 2000. “The Kantian versus Frankfurt.” Analysis 60(3): 287–​288. Buckwalter, Wesley. 2010. “Knowledge Isn’t Closed on Saturday: A Study in Ordinary Language.” Review of Philosophy and Psychology 1(3): 395–​4 06. Buckwalter, Wesley and John Turri. 2014. “Telling, Showing, and Knowing: A Unified Theory of Pedagogical Norms.” Analysis 74(1): 16–​20. Buckwalter, Wesley and Jonathan Schaffer. 2015. “Knowledge, Stakes, and Mistakes.” Noûs 49(2): 201–​234. Buckwalter, Wesley and John Turri. 2015. “Inability and Obligation in Moral Judgment.” PLoS ONE 10(8): e0136589. doi:10.1371/​journal.pone.0136589 Chickering, David Maxwell. 2003. “Optimal Structure Identification with Greedy Search.” Journal of Machine Learning Research 3: 507–​554. Cohen, Stewart. 1999. “Contextualism, Skepticism, and the Structure of Reasons.” Philosophical Perspectives 13: 57–​89. Cohen, Stewart. 2013. “Contextualism Defended.” In Contemporary Debates in Epistemology, edited by Matthias Steup, John Turri, and Ernest Sosa, 69–​75. Malden, MA: Wiley-​Blackwell. Copp, David. 2008. “‘Ought’ Implies ‘Can’ and the Derivation of the Principle of Alternate Possibilities.” Analysis 68(297): 67–​75. Dahl, Norman O. 1974. “Ought Implies Can and Deontic Logic.” Philosophia 4(4): 485–​511. DeRose, Keith. 1992. “Contextualism and Knowledge Attributions.” Philosophy and Phenomenological Research 52(4): 913–​29. DeRose, Keith. 1995. “Solving the Skeptical Problem.” The Philosophical Review 104(1): 1–​52. DeRose, Keith. 2009. The Case for Contextualism. Oxford: Oxford University Press. Descartes, René. 1985 (1644). The Principles of Philosophy. In The Philosophical Writings of Descartes, Vol. 1. Cambridge: Cambridge University Press. Douven, Igor 2006. “Assertion, Knowledge, and Rational Credibility.” Philosophical Review 115(4): 449–​485.

926

2 9 6   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

Ducasse, C.J. 1941. Philosophy as a Science:  Its Matter and Its Method. New York: Oskar Piest. Fantl, Jeremy and Matthew McGrath. 2002. “Evidence, Pragmatics, and Justification.” The Philosophical Review 111(1): 67–​94. Fantl, Jeremy and Matthew McGrath. 2007. “On Pragmatic Encroachment in Epistemology.” Philosophy and Phenomenological Research 75(3): 558–​589. Fantl, Jeremy and Matthew McGrath. 2009. Knowledge in an Uncertain World. Oxford: Oxford University Press. Feldman, Fred. 1986. Doing the Best We Can:  An Essay in Informal Deontic Logic. Dordrecht: Reidel. Feltz, Adam. 2013. “Pereboom and Premises:  Asking the Right Questions in the Experimental Philosophy of Free Will.” Consciousness and Cognition 22(1): 53–​63. Feltz, Adam and Edward T. Cokely. 2009. “Do Judgments about Freedom and Responsibility Depend on Who You Are? Personality Differences in Intuitions about Compatibilism and Incompatibilism.” Consciousness and Cognition 18(1): 342–​350. Fischer, John Martin. 2003. “‘Ought-​Implies-​Can’, Causal Determinism, and Moral Responsibility.” Analysis 63(279): 244–​250. Flanagan, Owen J. 1991. Varieties of Moral Personality: Ethics and Psychological Realism. Cambridge, MA: Harvard University Press. Fodor, Jerry. 1964. “On Knowing What We Would Say.” The Philosophical Review 73(2): 198–​212. Frankfurt, Harry G. 1969. “Alternate Possibilities and Moral Responsibility.” Journal of Philosophy 66(3): 829–​839. Graham, Peter. 2011. “‘Ought’ and Ability.” Philosophical Review 120(3): 337–​382. Hare, R.M. 1963. Freedom and Reason. Oxford: Clarendon Press. Hawthorne, John. 2004. Knowledge and Lotteries. Oxford: Oxford University Press. Hayes, Andrew F. 2013. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-​Based Approach. New York: Guilford Press. Hill, Christopher S. and Joshua Schechter. 2007. “Hawthorne’s Lottery Puzzle and the Nature of Belief.” Philosophical Issues 17(1): 102–​122. Howard-​Snyder, Frances. 2006. “‘Cannot’ Implies ‘Not Ought.’” Philosophical Studies 130(2): 233–​246. Iacobucci, Dawn, Neela Saldanha, and Xiaoyan Deng. 2007. “A Meditation on Mediation:  Evidence That Structural Equations Models Perform Better Than Regressions.” Journal of Consumer Psychology: 17(2): 139–​153. Jackson, Frank. 1998. From Metaphysics to Ethics. Oxford: Oxford University Press. James, William. 1948 (1879). “The Sentiment of Rationality.” In Essays in Pragmatism, edited by A. Castell, 3–​36. New York: Hafner Press. Jones, Edward Ellsworth and Richard E. Nisbett. 1971. The Actor and the Observer: Divergent Perceptions of the Causes of Behavior. Morristown, NJ: General Learning Press.

927

Moderate Scientism in Philosophy  • 

297

Kim, Minsun and Yuan Yuan. 2015. “No Cross-​Cultural Differences in Gettier Car Case Intuition: A Replication Study of Weinberg et al. 2001.” Episteme 12(3): 355–​361. Kvanvig, Jonathan. 2009. “Assertion, Knowledge, and Lotteries.” Williamson on Knowledge, edited by Patrick Greenough and Duncan Pritchard, 140–​ 160. Oxford: Oxford University Press. Lackey, Jennifer. 2007. “Norms of Assertion.” Noûs 41(4): 594–​626. Lehrer, Paul, Yuji Sasaki, and Yoshihiro Saito. 1999. “Zazen and Cardiac Variability.” Psychosomatic Medicine 61(6): 812–​821. Lewis, David. 1986. “Postscripts to ‘Causation.’” Philosophical Papers Vol. II, 172–​213. Oxford: Oxford University Press. Liao, S. Matthew, Alex Wiegmann, Joshua Alexander, and Gerard Vong. 2012. “Putting the Trolley in Order: Experimental Philosophy and the Loop Case.” Philosophical Psychology 25(5): 661–​671. Littlejohn, Clayton. 2012. “Does ‘Ought’ Still Imply ‘Can’?” Philosophia 40(4): 821–​828. Locke, John. 1975 (1690). An Essay Concerning Human Understanding. Edited by P.H. Nidditch. Oxford: Clarendon Press. Machery, Edouard. 2012. “Expertise and Intuitions about Reference.” Theoria 27(1): 37–​54. Machery, Edouard, Christopher Y. Olivola, and Molly de Blanc. 2009. “Linguistic and Metalinguistic Intuitions in the Philosophy of Language.” Analysis 69(4): 689–​694. Machery, Edouard, Stephen Stich, David Rose, Amita Chatterjee, Kaori Karasawa, Noel Struchiner, Smita Sirker, Naoki Usui, and Takaaki Hashimoto. 2015. “Gettier Across Cultures.” Noûs Early View. doi:10.1111/​nous.12110. MacIver, A.M. 1938. “Some Questions about ‘Know’ and ‘Think.’” Analysis 5(3): 43–​50. May, Joshua, Walter Sinnott-​Armstrong, Jay G. Hull, and Aaron Zimmerman. 2010. “Practical Interests, Relevant Alternatives, and Knowledge Attributions:  An Empirical Study.” Review of Philosophy and Psychology 1(2): 265–​273. McAllister, Daniel W., Terence R. Mitchell, and Lee Roy Beach. 1979. “The Contingency Model for the Selection of Decision Strategies:  An Empirical Test of the Effects of Significance, Accountability, and Reversibility.” Organizational Behavior and Human Performance 24(2): 228–​244. Moore, G.E. 1922. “The Nature of Moral Philosophy.” Philosophical Papers. London: Routledge and Kegan Paul. Moore, G.E. 1962. Commonplace Book: 1919–​1953. London: George Allen & Unwin. Nahmias, Eddy, Jason Shepard, and Shane Reuter. 2014. “It’s OK if ‘My Brain Made Me Do It’:  People’s Intuitions about Free Will and Neuroscientific Prediction.” Cognition 133(2): 502–​516. Naylor, M.B. 1985. “Voluntary Belief.” Philosophy and Phenomenological Research 45(3): 427–​436. Paul, L.A. 2016. “Experience, Metaphysics, and Cognitive Science.” In Companion to Experimental Philosophy, edited by Justin Sytsma and Wesley Buckwalter, 419–​433. Wiley-​Blackwell.

928

2 9 8   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

Peels, Rik. 2018. “A Conceptual Map of Scientism.” This volume. Pinillos, Ángel. 2012. “Knowledge, Experiments and Practical Interests.” In Knowledge Ascriptions, edited by Jessica Brown and Mikkel Gerken, 192–​219. Oxford: Oxford University Press. Pojman, Louis P. 1999. “Believing, Willing, and the Ethics of Belief.” The Theory of Knowledge: Classical and Contemporary Readings, edited by Louis P. Pojman, 525–​ 543. Belmont, CA: Wadsworth. Radnitzky, G. 1978. “The Boundaries of Science and Technology.” The Search for Absolute Values in a Changing World. Proceedings of the 6th International Conference on the Unity of Sciences, 1007–​1036. New York: International Cultural Foundation Press. Reid, Thomas. 1764/​1997. An Inquiry into the Human Mind on the Principles of Common Sense. Edited by D.R. Brookes. University Park:  Pennsylvania State University Press. Ripley, David. 2016. “Experimental Philosophical Logic.” In Companion to Experimental Philosophy, edited by Justin Sytsma and Wesley Buckwalter, 523–​534. Chichester: Wiley-​Blackwell. Rose, David, Wesley Buckwalter, and Shaun Nichols. 2015. “Neuroscientific Prediction and the Intrusion of Intuitive Metaphysics.” Cognitive Science Early View. doi: 10.1111/​cogs.12310. Rosenberg, Alex. 2018. “Philosophical Challenges for Scientism (and How to Meet Them?).” This volume. Rust, Joshua and Schwitzgebel, Eric. 2013. “Ethicists’ and Nonethicists’ Responsiveness to Student E-​mails:  Relationships Among Expressed Normative Attitude, Self-​ Described Behavior, and Empirically Observed Behavior.” Metaphilosophy 44(3): 350–​371. Ryan, Sharon. 2003. “Doxastic Compatibilism and the Ethics of Belief.” Philosophical Studies 114(1–​2): 47–​79. Schaffer, Jonathan. 2008. “Knowledge in the Image of Assertion.” Philosophical Issues 18(1): 1–​19. Schultz, Eric, Edward Cokely, and Adam Feltz. 2011. “Persistent Bias in Expert Judgments about Free Will and Moral Responsibility:  A Test of the Expertise Defense.” Consciousness and Cognition 20(4): 1722–​1731. Schwitzgebel, Eric. 2009. “Do Ethicists Steal More Books?” Philosophical Psychology 22(6): 711–​725. Schwitzgebel, Eric and Joshua Rust. 2009. “The Moral Behaviour of Ethicists:  Peer Opinion.” Mind 118(472): 1043–​1059. Schwitzgebel, Eric and Joshua Rust. 2010. “Do Ethicists and Political Philosophers Vote More Often Than Other Professors?” Review of Philosophy and Psychology 1(2): 189–​199. Schwitzgebel, Eric and Fiery Cushman. 2012. “Expertise in Moral Reasoning? Order Effects on Moral Judgment in Professional Philosophers and Non-​Philosophers.” Mind & Language 27(2): 135–​153.

92

Moderate Scientism in Philosophy  • 

299

Schwitzgebel, Eric and Joshua Rust. 2013. “The Moral Behavior of Ethics Professors:  Relationships Among Self-​Reported Behavior, Expressed Normative Attitude, and Directly Observed Behavior.” Philosophical Psychology 3: 1–​35. Schwitzgebel, Eric and Joshua Rust. 2016. “The Behavior of Ethicists.” In Companion to Experimental Philosophy, edited by Justin Sytsma and Wesley Buckwalter, 225–​233. Oxford: Blackwell. Schwitzgebel, Eric and Fiery Cushman. 2015. “Philosophers’ Biased Judgments Persist Despite Training, Expertise and Reflection.” Cognition 141: 127–​137. Schwitzgebel, Eric, Joshua Rust, Linus Ta-​Lun Huang, Alan T. Moore, and Justin Coates. 2011. “Ethicists’ Courtesy at Philosophy Conferences.” Philosophical Psychology 25(3): 331–​340. Scott-​Kakures, Dion. 1994. “On Belief and Captivity of the Will.” Philosophy and Phenomenological Research 54(1): 77–​103. Sellars, Wilfrid. 1963. Science, Perception, and Reality. Atascadero, CA: Ridgeview. Sinnott-​ Armstrong, Walter. 1984. “‘Ought’ Conversationally Implies ‘Can.’” Philosophical Review 93(2): 249–​261. Sripada, Chandra Sekhar and Jason Stanley. 2012. “‘Empirical Tests of Interest-​Relative Invariantism.” Episteme 9(1): 3–​26. Stanley, Jason. 2005. Knowledge and Practical Interests. New  York:  Oxford University Press. Stenmark, M. 1997. “What Is Scientism?” Religious Studies 33(1): 15–​32. Stenmark, M. 2000. Scientism: Science, Ethics, and Religion. Aldershot: Ashgate. Stenmark, M. 2004. How to Relate Science and Religion: A Multidimensional Model. Grand Rapids, MI: Eerdmans. Stocker, Michael. 1971. “‘Ought’ and ‘Can.’.\” Australasian Journal of Philosophy 49(3): 303–​316. Streumer, Bart. 2003. “Does ‘Ought’ Conversationally Implicate ‘Can’?” European Journal of Philosophy 11(2): 219–​228. Swain, Stacey, Joshua Alexander, and Jonathan Weinberg. 2008. “The Instability of Philosophical Intuitions:  Running Hot and Cold on Truetemp.” Philosophy and Phenomenological Research 76(1): 138–​155. Sytsma, Justin, Jonathan Livengood, Ryoji Sato, and Mineki Oguchi. 2015. “Reference in the Land of the Rising Sun: A Cross-​cultural Study on the Reference of Proper Names.” Review of Philosophy and Psychology 6(2): 213–​230. Tobia, Kevin, Wesley Buckwalter, and Stephen Stich. 2013. “Moral Intuitions:  Are Philosophers Experts?” Philosophical Psychology 26(5): 629–​638. Tobia, Kevin, Gretchen Chapman, and Stephen Stich. 2013. “Cleanliness Is Next to Morality, Even for Philosophers.” Journal of Consciousness Studies 20: 11–​12. Trigg, Roger E. 1993. Rationality and Science:  Can Science Explain Everything? Oxford: Blackwell. Turri, John 2011. “The Express Knowledge Account of Assertion.” Australasian Journal of Philosophy 89(1): 37–​45.

03

3 0 0   • 

W e s l e y B u c k w a l t e r a n d J o h n   T u rr i

Turri, John. 2013a. “The Test of Truth: An Experimental Investigation of the Norm of Assertion.” Cognition 129(2): 279–​291. Turri, John. 2013b. “A Conspicuous Art:  Putting Gettier to the Test.” Philosophers’ Imprint 13(10): 1–​16. Turri, John. 2016a. “How to Do Better:  Toward Normalizing Experimentation in Epistemology.” In Advances in Experimental Philosophy and Philosophical Methodology, edited by Jennifer Nado, 35–​51. London: Bloomsbury. Turri, John. 2016b. “Knowledge Judgments in ‘Gettier’ Cases.” In A Companion to Experimental Philosophy, edited by Justin Sytsma and Wesley Buckwalter, 337–​348. Chichester: Wiley-​Blackwell. Turri, John and Wesley Buckwalter. 2017. “Descartes’s Schism, Locke’s Reunion: Completing the Pragmatic Turn in Epistemology.” American Philosophical Quarterly 54(1): 25–​45. Turri, John, David Rose, and Wesley Buckwalter. 2017. “Choosing and Refusing: Doxastic Voluntarism and Folk Psychology.” Philosophical Studies, https://doi.org/10.1007/ s11098-017-0970-x. Turri, John, Wesley Buckwalter, and David Rose. 2016. “Actionability Judgments Cause Knowledge Judgments: Evidence From Two Experiments.” Thought 5(3): 212–​222. Tversky, Amos and Kahneman, Daniel. 1974. “Judgment Under Uncertainty: Heuristics and Biases.” Science 185(4157): 1–​9. Unger, Peter. 1975. Ignorance: A Case for Skepticism. Oxford: Oxford University Press. Vaesen, Krist, Martin Peterson, and Bart Van Bezooijen. 2013. “The Reliability of Armchair Intuitions.” Metaphilosophy 44(5): 559–​578. Van Fraassen, Bas C. 1973. “Values and the Heart’s Command.” Journal of Philosophy 70(1): 5–​19. Vranas, Peter B.M. 2007. “I Ought, Therefore I  Can.” Philosophical Studies 136(2): 167–​216. Williams, Bernard. 1973. “Deciding to Believe.” Problems of the Self, 136–​151. Cambridge: Cambridge University Press. Williamson, Timothy. 2000. Knowledge and Its Limits. Oxford: Oxford University Press. Wittgenstein, Ludwig. 1969. On Certainty. Editors G.E.M. Anscombe and G.H. von Wright. Oxford: Blackwell. Zimmerman, Michael J. 1993. “Obligation, Responsibility and Alternate Possibilities.” Analysis 53(1): 51–​53.

310

INDEX

arrogance 150, 154, 160, 161–​3 assertion  290–​1 atheism, 62–​3,  72–​3 Atkins, Peter 34, 53, 63, 153, 158, 280 authority of science 10, 150, 177–​80, 182, 183 Baehr, Jason 155–​61 belief passim basic vs. non-​basic 202–​5 Christian 23, 228–​31 moral 16, 37, 236–​45 rational 1, 3–​4, 11, 16, 19, 44, 48, 63–​8,  79 religious 16, 39, 46–​7, 64, 66, 67, 79, 227–​31, 234 Berlin, Isaiah 156 Byerly, Henry 1, 26, 30, 35, 55 causal path analysis 283–​4 closed-​mindedness  152–​63 cognitive biases 292–​3 cognitive science, 128, 133–​9, 233, 253, 287, 294 common sense 4, 14–​16, 29, 36, 48, 50, 52–​3, 168, 170, 229, 259, 262, 272–4, 285, 293–​4 Comte, August 6–​8, 25

debunking arguments 16, 47, 233–​6, 238,  245–​56 decision-​making 24, 121, 124, 128, 282–​ 4, 289, 291 demarcation criterion, 12, 14, 116, 181–​2, 197, 221 deontology  234–​54 Descartes, René 12, 87, 144, 197–​8, 287, 295 determinism 84, 288–​90 dogmatism 115, 125, 150, 157, 160–​3 doxastic deliberation 127–​8, 130–​2 doxastic (in)voluntarism 287–​8 dual-​process theory, 133–​4,  236–​45 Dupré, John 149–​50, 154, 165 emotion 132, 236–​49 empiricism 9, 93–​6, 110–​14, 151, 211, 223–​5 ethics 13, 50–​3, 61, 101, 119–​20, 123, 161, 185, 225–​6, 234–​5, 242, 256, 284–​7,  292 of belief 170, 175 meta-​ 84–​5,  256 evolution 11, 16, 37–​9, 60, 74, 85, 116–​18, 235–​42,  251–​3 expansionism, scientific 29, 46, 58–​9,  107

302

3 0 2   • 

Index

experimental philosophy 280–​300 experiments 90, 118, 137–​8, 145–​6, 170, 209–​11, 244–​5, 250, 264 fictionalism 95–​6, 103 Field, Hartry 69, 81, 96, 104 Fodor, Jerry 99–​100, 104, 133, 147, 281, 296 folk psychology 16, 92 foundationalism 192, 202 framing effects 292 free will 4, 16, 28, 29, 36, 69, 83–​5, 104, 131, 139–​4 0, 171, 173, 259–​60, 289 Gödel, Kurt 93–​5, 224 Greene, Joshua 233–​45 Haack, Susan 2, 6, 25, 29, 32, 54, 107, 109, 116, 123–​4, 149–​50, 170,  258–​9 Heidegger, Martin 158, 165, 225 Helm, Paul 188, 189 Hempel, Carl 9, 25, 278 Hempel’s dilemma, 116, 263–​70 humanism 73–​5, 108, 114, 120 humanities 3, 11, 16, 32–​3, 40, 49, 58, 60–​3, 67–​8, 71, 73, 75, 92, 104, 107–​9, 114, 124–​5, 162, 185, 210, 212–​13,  294 Hume, David 6–​8, 25, 84–​5, 93–​4, 104, 180, 184, 224 inference 13, 87, 91, 95, 117, 136, 200 intentionality 70–​1, 84, 92, 97–​8, 100–​4,  259–​60 internalism /​externalism 199 introspection 3, 11, 18, 91, 131, 137, 184, 196, 226, 272, 281, 283, 293 intuition 37, 48, 109–​10, 114, 168, 181, 196, 204, 233–​4, 242, 244, 285,  291–​2 Ioannidis, John 206, 217

Kandel, Eric 98, 100, 105 Kant, Immanuel 13, 87 Kitcher, Philip 5, 65–​6, 81, 95, 105, 124, 126, 150, 165, 197, 217, 227–​8, 231 knowledge passim animal vs. reflective knowledge 196, 199–​200 first-​personal 146, 196 kinds of, 191–​6 mathematical, 84–​7, 93–​5, 224 a priori 94, 111–​14, 196, 224–​5 scientific, 197–​201,  207–​11 second-​personal 196, 214 sources of 11, 16–​17, 72, 113, 119, 123, 144, 153, 180–​1, 195, 227, 230 Korsgaard, Christine 128–​9, 139–​4 0 Ladyman, James 2, 10, 13, 18, 29–​30, 106–​26, 151, 167–​8, 175–​8, 180–​2, 186–​7, 191, 197 Laudan, Larry 12, 15, 197, 205, 217, 221 Levitt, Norman, 153, 163 Lewis, David 96, 184, 268, 281 limits of science 59, 64, 190–​219 logical positivism, see verificationism manifest image 84, 104, 127–​8, 132–​3, 138–​9, 142–​4,  281 materialism 68, 84, 110–​14, 118 mathematics 84–​7, 92–​6, 117–​18, 220,  224–​6 metaphysics 7–​8, 13, 32, 89, 94, 95, 120, 151, 168, 180–​2, 185–​6, 275–​7 methodology, philosophical 145–​6 methodology, scientific 90, 117 methods of science 62, 89, 107–​8, 116–​ 18, 124, 168–​75, 222–​3, 263, 270–​6 Midgley, Mary 19, 149 Mill, John Stuart 11, 26, 95 Millikan, Ruth 99, 105 Moorean truth 169–​70 moral epistemology 233–​57

30

Index  • 

moral norms 37, 85–​6, 116, 119 moral objectivity 234, 245–​51, 254–​5 moral psychology 233–​4 naturalism 68, 84, 70–​8, 80, 84, 87–​9,  93–​6 different meanings of 59, 221 liberal 57, 70–​2, 80 philosophical,  70–​1 religious 57 scientific 68 neuroscience 33, 91, 92, 97–​8, 101–​3, 107, 233, 253, 289 Nichols, Shaun 234–​6, 245–​56 objectifying inquiry 208–​14 objectivity  247–​9 epistemic vs. ontological 177–​82 of facts 53, 76, 275–​6 moral 52–​3,  245–​51 ontic structural realism 184–​5 open-​mindedness  154–​60 ordinary language, 281, 291 ought implies can 284–​7 pessimistic meta-​induction 205 Philipse, Herman 10, 26 physicalism 84, 110, 112, 116, 118, 258–​77 physics 10, 29–​33, 49, 57, 61, 63, 76, 80, 86, 117–​18, 171–​7, 182, 184–​6, 221–​2,  230 Plato 94, 198 Positivism, see Comte, August; logical positivism pragmatism 89, 102, 182 prediction 11, 14, 87, 90–​2, 118, 120, 123, 171, 283, 288–​9 Principle of Naturalistic Closure 113,  182–​8 quantum mechanics 88, 90, 185, 211, 262,  272–​4

303

rationalism 94, 111–​12, 238 Rea, Michael 184, 188, 189 reasons 16, 58 for belief 66, 131–​2, 135–​6, 138, 244, 246–​7, 250, 253 vs. causes 139, 141–​4 reductionism 60, 70–​1, 107, 109, 118, 124, 258, 277 reflection, 91, 127–​48, 154, 203, 241, 247 Rorty, Richard 76, 81, 89, 105 Rosenberg, Alexander 2, 5, 10, 13, 16, 26, 23, 30, 33, 34, 44, 59, 62–​3, 67, 72, 83–​105, 110, 168–​75, 222–​3, 259, 280 Ross, Don 10, 13, 26, 29, 30, 109–​10, 112–​13, 117,  175–​88 Russell, Bertrand 34, 35, 55, 62, 81, 190, 194, 218, 229 science passim scientific image 127–​47 scientism passim arguments for 14–​16, 83–​148, 280–​300 arguments against 17–​20, 149–​257 forms of 31–​49, 60–​70, 221–​3, 225–​7,  280–​1 moderate 226, 281 self-​referential incoherence 10, 17–​18, 175, 186–​7,  223–​4 Sellars, Wilfrid 35, 55, 62, 82, 113, 127, 132, 148, 281, 299 Shapin, Steven 151, 162, 166 social constructionism  75–​7 Sosa, Ernest 131, 144, 148, 195–​6, 199–​ 200, 204, 217–​19 Spurrett, David, see Ladyman, James; Ross, Don stance 29, 110–​12, 268 intentional 100 scientism as a 18, 113–​25, 151–​6 4,  186–​8

034

3 0 4   • 

Index

Stenmark, Mikael 3, 18, 26, 30, 31, 33, 34, 37–​4 0, 46–​7, 51–​3, 57–​82, 149, 280, 281, 299 Stevenson, Leslie, see Byerly, Henry Street, Sharon 16, 26, 253 system 1 and 2, see dual-process theory teleosemantics  98–​100 theism 77–​80, 168, 260 theology 7–​8, 10–​11, 13, 31, 35, 46, 50–​1, 176, 185, 220–​1

value-​free 11, 122, 123, 248–​9 van Fraassen, Bas 27, 110–​13, 122, 151, 186–​7, 207–​14, 268, 285 verificationism 9–​10, 182–​6,  224–​5 vice epistemology 149–​66 virtue epistemology 149–​50, 154–​7, 161, 191, 196, 198, 199 Williams, Bernard 73, 82, 149, 287, 300 Wittgenstein, Ludwig 24, 156, 160, 165–6, 281, 300

305

306

307

308