Lacey_1999_Is Science Value Free.pdf - [PDF Document] (2024)

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (1)

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (2)

Is Science Value Free?

‘Hugh Lacey’s book gives a careful and well-rounded treatmentof the unceasingly topical and controversial question of the roleof values in scientific inquiry. It is a particular virtue of thebook that it combines a treatment of the epistemological andmethodological debates with an engagement in moresubstantive questions about whose values may or may not besocially well placed to inform science, and about science’s rolein international development’

Miranda Fricker, Heythrop College

The view that science is value free has been challenged from a number ofdifferent sides, including post-modernists, feminists, radical ecologists,third-world advocates and religious fundamentalists.

In this book, Hugh Lacey explicates and appraises the view that scienceis value free. Lacey discusses how science and values interact, with a focuson a discussion of development and science’s place in development—particularly in third-world countries.

Is Science Value Free? not only offers us a unique perspective on theongoing debate—above all—in defining “levels,” one at which strategiesare adopted and the other at which theories are chosen, but it is also themost comprehensive book completely devoted to the theme of science asvalue free.

Is Science Value Free? gives us a refreshing and intriguing account ofhow we see and study science. Anyone interested in science and thephilosophy of science will find this book an invaluable read.

Hugh Lacey is Professor of Philosophy at Swarthmore College. He is alsothe co-author (with Barry Schwartz) of Behaviorism, Science, and HumanNature.

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (3)

Philosophical Issues in ScienceEdited by W.H.Newton-Smith

Balliol College, OxfordReal HistoryMartin Bunzl

Brute ScienceHugh LaFollette and Niall Shanks

Living in a Technological CultureMary Tiles and Hans Oberdiek

The Rational and the SocialJames Robert Brown

The Nature of the DiseaseLawrie Reznek

The Philosophical Defence of PsychiatryLawrie Reznek

Inference to the Best ExaminationPeter Lipton

Time, Space and PhilosophyChristopher Ray

Mathematics and Image of ReasonMary Tiles

Evil or Ill?Lawrie Reznek

The Ethics of Science: An IntroductionDavid B.Resnik

Philosophy of Mathematics: An Introduction to aWorld of Proofs andPicturesJames Robert Brown

Theories of Consciousness: An Introduction and AssessmentWilliam Seager

Psychological Knowledge: A Social History andPhilosophyMartin Kusch

Is Science Value Free? Values and scientific understandingHugh Lacey

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (4)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (5)

Is Science Value Free?

Values and scientific understanding

Hugh Lacey

London and New York

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (6)

First published 1999by Routledge

11 New Fetter Lane, London EC4P 4EE

Simultaneously published in the USA and Canadaby Routledge

29 West 35th Street, New York, NY 10001

Routledge is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2005.

“To purchase your own copy of this or any of Taylor & Francis or Routledge’scollection of thousands of eBooks please go to”

© 1999 Hugh Lacey

All rights reserved. No part of this book may be reprinted orreproduced or utilised in any form or by any electronic,

mechanical, or other means, now known or hereafter invented,including photocopying and recording, or in any

information storage or retrieval system, without permission inwriting from the publishers.

British Library Cataloguing in Publication DataA catalogue record for this book is available from the British


Library of Congress Cataloging in Publication DataLacey, Hugh

Is science value free?: values and scientific understandingp. cm.

Includes bibliographical references and index.1. Sciences—Philosophy. 2. Objectivity. 3. Values.

I. Title.Q175. L157



ISBN 0-203-98319-X Master e-book ISBN

ISBN 0-415-20820-3 (Print Edition)

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (7)

For Maria Ines, Andrew and Daniel

Like last summer’s tomato harvest and many goodthings, this book is a product of living in our new


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (8)


Preface viii

Acknowledgments x

1 Introduction: the idea that science is value free 1

2 Values 23

3 Cognitive values 45

4 Science as value free: provisional theses 67

5 Scientific understanding 89

6 The control of nature 111

7 Kuhn: scientific activity in different ‘worlds’ 147

8 A “grassroots empowerment” approach 177

9 A feminist approach 197

10 Science as value free: revised theses 219

11 Conclusion 249

Notes 255

Bibliography 271

Index 279

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (9)


This book aims to explicate and appraise the view that science is valuefree: making a contribution both to analytical philosophy of science and(more speculatively) to substantive moral reflection on the place of sciencein contemporary society. Regarding the latter, I have discussed how scienceand values interact, keeping an eye toward discussions of “development,”and the place of science in it, that are taking place in many “third-world”countries.

Only in passing (and through the intermediary of interlocutors whom Iidentify in the text) do I interact with other philosophical perspectives thatdiscuss the interaction of values and scientific understanding, namely:critical theory, phenomenology, post-structuralism, pragmatism and socialstudies of science. This reflects my personal biography, not a judgment thatimportant insights into the issues cannot be obtained from theseperspectives. I hope that the readers of the book will bring my argumentsinto interaction with theirs.

In order to focus on my chosen themes—scientific understanding, values,and the relations between them—I have inevitably had to short-changeothers, concerning which of mine have presuppositions and implications.Thus, for example, I have skipped over issues about the nature of scientifictheories and about how to interpret them (realism, empiricism,constructivism); and about whether scientific knowledge should beregarded as the possession of individuals or groups of individuals as socialor as belonging to an abstract domain. While my arguments are intended tobe independent of where one stands on the issues I do not discuss, I havenot been able to develop an idiom that is completely neutral with respect tothem. Throughout the book, I have used a realist idiom, and, except whenexplicitly noted, I discuss the objectives of science in broadly realist terms.My intention is not so much to endorse realist interpretations of science, asto show that, even with realist interpretations, which are usually bearers ofthe idea of science as value free, important criticisms of it can arise. I amconfident that my arguments can readily be restated within, for example,empiricist perspectives. Moreover, their force does not depend uponadopting any controversial conception of the nature of scientific theories,

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (10)

other than that generally the acceptance of theories is of specified domainsof phenomena. Thus, I believe, my arguments can be stated (criticized andappraised) without entering into current controversies about the nature oftheories and realist (and other) interpretations of science.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (11)


I have worked on the interaction of values and the sciences for many years,in the course of which I have incurred many philosophical debts. Myteacher, Michael Scriven, first brought the issues to my attention. ElizabethAnderson, John Clendinnen, Richard Eldridge, J.A.Giannotti, GeoffreyJoseph, Joseph Margolis, Braulio Muñoz, Hans Oberdiek, RichardSchuldenfrei, Barry Schwartz, Miriam Solomon, Mary Tiles and dozens ofmy students at Swarthmore College have all been helpful from time to timeover the years. As the book has been taken shape during the past threeyears, Elizabeth Anderson, Marcos Barbosa de Oliveira, Eduardo Barra,Otávio Bueno, John Clendinnen, Alberto Cupani, Luiz Henrique Dutra,Richard Eldridge, Brian Ellis, Ernan McMullin, Lynn Hankinson Nelson,Graham Nerlich and Howard Sankey have made useful comments andcriticisms. I am especially grateful to Elizabeth Anderson, Marcos Barbosade Oliveira, Lynn Hankinson Nelson, Graham Nerlich and HowardSankey for correspondence about early versions of sections of the book,and to Otávio Bueno who crafted the title from a long-winded, provisionalone. Miranda Fricker, as reader for Routledge, made a number of helpfulsuggestions.

Jackie Robinson provided assistance in all sorts of helpful ways. So, too,did the editorial staff at Routledge (Anna Gerber, Lisa Carden, CeriPrenter). It has been a pleasure to work with them all.

During 1996, several universities provided me with extended hospitalityand opportunities to test my developing ideas: the Department of Historyand Philosophy of Science at The University of Melbourne, Australia(especially Rod Home and Howard Sankey); the Philosophy Department atUniversidade de São Paulo, Brazil, (especially Pablo Mariconda); and thePhilosophy Department at Universidade Federal de Santa Catarina,Florianópolis, Brazil.

In several chapters, I have drawn upon previously published work whichhas been in most cases significantly rewritten and developed. Partsfrom Lacey and Schwartz (1996) “The formation and transformation ofvalues,” in W.O’Donohue and R.Kitchener (eds) (1996) The Philosophy ofPsychology, London: Sage, are reprinted in Ch. 2 by permission of Sage

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (12)

Publications Ltd; from Lacey (1986) “The rationality of science,” inJ.Margolis, M. Krausz and R.A.Burian (eds) (1986) Rationality, Relativismand the Human Sciences, Dordrecht: Kluwer Academic Publishers (pp. 127–50) in Chs 5 and 6 with the kind permission of Kluwer AcademicPublishers; from Lacey (1990) in Ch. 6 and Lacey (1997c) in Chs 3 and 4with permission of the editor of Journal for the Theory of Social Behavior,Lacey (1997b) in Chs 3 and 5 with permission of the editor of Principia;Lacey (1998) in Ch. 8 with permission of the production editor ofDemocracy and Nature, and Lacey (1999a, b) in Chs 5 and 6 withpermission of the editor of Science and Education.

I am also grateful for financial support from the Eugene M.LangResearch Professorship at Swarthmore College (1993–96), the SwarthmoreCollege Faculty Research Fund, FAPESP (Fundação para a Amparo aPesquisa do Estado de São Paulo, Brazil), and (for earlier developments ofthis and related work, especially as it pertains to psychology) the NationalScience Foundation (1975–6, 1979–80, 1983–4).


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (13)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (14)


The idea that science is value free

The idea that the sciences are value free has long played a key role in theself-understanding and the public image of modern science. Poincaré,writing early in this century, captured its core as follows:

Ethics and science have their own domains, which touch but do notinterpenetrate. The one shows us to what goal we should aspire, theother, given the goal, teaches us how to attain it. So they neverconflict since they never meet. There can be no more immoral sciencethan there can be scientific morals.

(Poincaré 1920/1958:12)

Science and values only touch; they do not interpenetrate. To deny this isoften perceived as to challenge that science is the pre-eminent or exemplaryrational endeavor, to demean the cognitive credentials of science and toundercut its claim to produce knowledge. Lately, however, it has beenmuch contested from an eclectic variety of viewpoints: feminism, socialconstructivism, pragmatism, deep ecology, fundamentalist religions, and anumber of third world and indigenous people’s outlooks. Exactly what is atissue does not always emerge clearly in these contestations. The rhetorictends to be at high volume, but the argument thin. Incommensurabilityseems to reign. From one viewpoint, the mounting threat of multipleirrationalities and empty voluntarism looms large; from the other, theentrenchment of ideologies.

I will attempt to sort out what is at stake in the contestation of “scienceis value free,” an idea that incorporates several distinct views about ways inwhich science and values do (ought) not interpenetrate. But those whoaffirm it have always recognized Poincaré’s distinction, and held thatscience and values touch in various ways with more or less significanteffects. Too often the critics point only to aspects of the touch, but evenwhen the focus is on alleged interpenetrations a further ambiguityarises. For “science is value free” in general hardly represents a fact.Perhaps it represents an idealization of fact.1 It also represents a value, agoal or aspiration of scientific practices and a criterion for appraising its

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (15)

products and their consequences. The fact and value components cannot beseparated. To the extent that “science is value free” represents a fact, or anidealization of a fact, that is because “science is value free” has been heldas a value; and its being held as a value is without foundation if it is notpossible for it to be increasingly manifest in fact. Thus to refute “science isvalue free” it is not enough to display cases where it is not manifested infact; rather, the cognitive resources of the practices of science must beassessed for their ability and likelihood to bring about its manifestationsincreasingly and systematically.

In this introductory chapter I will provide an overview of the varioussources of the idea that science is value free, leading to the proposal that itshould be regarded as constituted by three component views: impartiality,neutrality and autonomy. Then I will outline some of the important waysin which science and values may interact without (from the proponents’viewpoint) the idea being challenged. Finally I will preview the focus,argument and methodology of the book.


“Science is value free” has several sources. Its kernel is present already inthe works of Galileo and Bacon. Galileo (1623/1957:270) refers to “thefacts of Nature, which remains deaf and inexorable to our wishes”; andBacon affirms, warning us to be alert to the “Idols of the mind,” thesources of error to which we are prone: “The human understanding is nodry light, but receives an infusion from the will and affections; whenceproceed sciences which may be called ‘sciences as one would’” (Bacon1620/1960: Aphorism 49).


The Galilean input to the idea of science as value free is metaphysical. Itleads to:”…the discarding by scientific thought of all considerations basedupon value-concepts, such as perfection, harmony, meaning and aim, andfinally the utter devalorization of being, the divorce of the world of valuefrom the world of facts” (Koyré 1957:4).

Let me summarize it in contemporary dress. The world, “the facts ofnature,” the spatio-temporal totality, is fully characterizable and explicable

in terms of “its underlying order”—its underlying structures, processesand laws. All objects belonging to the underlying order can be fullycharacterized in quantitative terms; all interactions are lawful; and the laws(not necessarily deterministic) are expressible in mathematical equations.Such objects are not construed as objects of value. Qua objects of theunderlying order, they are part of no meaningful order, they have nonatural ends, no developmental potentials, and no essential relatedness to


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (16)

human life and practices. Values—and objects, qua objects of value—arenot represented as emergent from the underlying order of the world.

An object may come to acquire value through its relationship to humanexperience, practice, or social organization, but any role it plays there isplayed in virtue of its causal powers, of what it is qua part of theunderlying order of the world, so that for explanatory purposes that it mayhave acquired value is irrelevant. Since human beings are part of the world,some of the historically contingent states of affairs in the world will be aconsequence of human causal agency. But, the view maintains, thestructures, processes and laws that make up the underlying order of theworld are ontologically independent of human inquiry, perception andaction; they do not vary with the theoretical commitments, outlooks,interests or values of investigators. On this view, it is a “fact” that valuederives from an object’s relationship to human experience, practice orsocial organization (that human agents generate value), and so this “fact”is explicable, in principle, in terms of underlying structure, process andlaw. But it does not follow from a theory that explains this “fact” thathuman agents themselves are objects of value.

The underlying order of the world, and its constituent entities, are simplythere to be discovered—the world of pure “fact” stripped of any link withvalue (MacIntyre 1981:80–1). The aim of science is to represent this worldof pure “fact,” the underlying order of the world, independently of anyrelationship it might bear contingently to human practices and experiences.Such representations are posited in theories which, in order to be faithful tothe object of inquiry, must deploy only categories devoid of evaluativecontent or implications. Thus, they must not use categories that can beapplied to things only in virtue of their being related to human experiencesor practices. Concretely, simplifying a little, this means using in theoriesonly quantitative concepts, or more generally, materialist concepts (thosethat designate properties of material objects qua material objects, not quarelated to human experience) and, in any case, no teleological, intentionalor sensory concepts.

Thus we arrive at one dimension of the idea of the “neutrality” ofscience: scientific theories have no value judgments among their logicalimplications. They cannot, it is said, for they contain no value categories.A second dimension is often taken to follow: that accepting a theory has nocognitive consequences at all concerning the values one holds. A thirddimension is suggested too: that scientific theories are available to beapplied so as to further projects linked with any values. After all, theyrepresent “fact” about the world, which can—so far as science is concerned—be related to, or come to serve the interests of any values whatever. If infact they do not serve to inform the projects motivated by particularvalues, that is an entirely contingent matter. Notice that this last claim restsuneasily with another that has been heralded in the modern scientific


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (17)

tradition, that science serves especially well the projects of materialprogress; and it clashes strongly with those world-views (that “progress”intends to supplant) that consider the world to be infused with meaning orvalue.

Epistemological and methodological/Baconian

In contrast to the Galilean, the Baconian source is epistemological andmethodological. Again I summarize a contemporary version. It is throughexperience that we gain access to the world, which can be considered acomplex repository of possibilities, of which the ones that are realized maybe (increasingly) connected with our practices and our plannedinterventions. But the world is not generally what we would have it be.Not everything that we desire or imagine to be possible is among the world’srepository of possibilities. Considerations derived from values cannotdetermine what is possible. We find out what is possible only in the courseof engagement with the world, through successful practices, including mostimportantly experimental ones. A scientific theory aims to encapsulatewhatever it can of this repository of possibilities of the domains ofphenomena within its compass; hence, the centrality in methodology ofexperiment.

Sound scientific knowledge, that which we can count on for practicaladoption, is rooted in replicability and agreement. Only what is observed,especially in experimental settings, and certified by replication andagreement—independently of our desires, value perspectives, cultural andinstitutional norms and presuppositions, expedient alliances and theirinterests—can properly serve as evidence for scientific posits and forchoosing among scientific theories. As Hempel puts it: “The grounds onwhich scientific hypotheses are accepted or rejected are provided byempirical evidence, which may include observational findings as well aspreviously established laws and theories, but surely no value judgments”(Hempel 1965:91). This is one of the sources of the idea ofthe “impartiality” of science, an idea concerning the proper grounds foraccepting scientific posits or making scientific judgments.

The Baconian source of impartiality is often complemented by a viewabout the nature of scientific inference, or about how empirical data arerelated to theories so that they can serve as evidence for acceptingtheoretical posits, or choosing which theories to accept. The view is thatscientific inference can be reconstructed in terms of accordance with certainformal rules (Chapter 3). The rules mediate between empirical data andtheories in such a way that following them leads to unambiguous choicesabout which theories to accept, reject or deem as requiring furtherinvestigation, or at least to unambiguous assignments of degrees ofconfirmation to theoretical hypotheses. They provide, as it were, the means


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (18)

to transfer intersubjective acceptance from the available data to the theory.While there has been at times widespread agreement that scientificinference, and any rational inference, can be explicated in terms of formalrules, there has never been anything approaching unanimity about what therules are, or even about whether they are deductive, abductive, statistical,inductive, or some combination of these. Bacon himself is usuallyinterpreted as holding that they are inductive.

The general view (though not any particular account of what the rulesare) became reinforced with the logical empiricists’ and critical rationalists’distinction between the contexts of discovery and justification, and thencewith holding the rule-governed account of scientific inference to apply onlyin the context of justification. Oversimplifying: a theory is properlyaccepted (or justified) if and only if it is related to the data in accordance withthe rules. Values (and, for example, metaphysics) may play a role in theprocess of discovery in the course of generating and exploring the merits ofthe theory, but they can have nothing to do with assessments of its properacceptance. As Carnap (1928/1967: xvii) put it, after conceding a role to“emotions, drives, dispositions and general living conditions” in theprocess of discovery:”…for the justification of a thesis the physicist doesnot cite irrational factors [my emphasis], but gives a purely empirical-rational justification.”

The success of modern science

Both neutrality and impartiality concern the content of what is posited inscientific theories: neutrality, its implications and consequences;impartiality, the grounds for accepting it. One derives from “objectivity,”representing faithfully the object of inquiry; the other from“intersubjectivity” as a condition on empirical inquiry. In practice, the twoideas tend to fuse.2 In order that there be any scientific knowledge, theGalilean idea needs to be complemented with a methodology (orprocedures that can give it empirical content); and, methodologically, sinceobjectivity cannot be had directly, intersubjectivity seems to be the bestavailable substitute. Conversely, Baconian methodology is deployedcharacteristically in testing theories that meet requirements derived fromthe Galilean idea, although the Baconian idea itself encompasses anyinquiry that is systematic and empirical (Chapters 5 and 8).

The fusion of the Galilean and Baconian ideas underlies the manifestsuccess of modern science. Bacon promised that utility would follow fromdeploying his methodology. That is not what I have directly in mind.Rather it is the manifest success of modern science in increasing “the stockof knowledge.” One may identify this success primarily in terms of thediscovery of objects (for example atoms, electromagnetic radiation, viruses,genes) or of the definitive entrenchment of some relatively circ*mscribed


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (19)

theories (for example the heliocentric theory of planetary motion, theoriesof molecular chemistry, theories of the bacterial and viral causation ofdiseases, theories that explain the workings of instruments). Suchknowledge, of course, has been widely applied in practice: in technology, inmedicine, in interpreting various phenomena of the world of dailyexperience; and, successful application is powerful confirming evidence insupport of the knowledge. Items in the stock of knowledge have beenaccepted in accordance with impartiality and so their cognitive claims arecompelling regardless of what values one holds. The sustained success ofmodern science, as it were, speaks unambiguously to the strength (but notto the certainty or unrevisability) of its cognitive claims.

A claim that is accepted in accordance with impartiality is bindingregardless of the values that are held—so that the presuppositions of allpractices, and the beliefs that inform all actions, should (rationally) all bemade consistent with it.3 This “binding equally” should not be confusedwith what I have earlier called “neutrality” with its three dimensions:“consistent with all value judgments,” “no (cognitive) consequences in therealm of values,” “evenhandedly applicable regardless of values held”; norwith the stronger view (Chapter 3) that all practices and actions, regardlessof the values they are intended to further, should be informed by scientificknowledge to the extent possible. Neutrality presupposes impartiality; and,when the Galilean and Baconian ideas are fused—especially if themetaethical and logical views described in the next two subsections areendorsed—it may appear that impartiality implies neutrality. But, I willargue (Chapter 4) “binding equally” is not consistent with endorsing allthree strands of the idea of neutrality; and subsequent attempts to reviseneutrality into a coherent thesis confront numerous difficulties.


Components of both neutrality and impartiality have been held to gainfurther credibility from a widespread metaethical view: that valuesrepresent subjective phenomena, preferences or utilities so that “valuejudgments” are considered to be only articulations of personal preferencesnot open to rational appraisal. As such, value judgments lack truth value:“…they do not express assertions” (Hempel 1965:86). A person’s makingthem is open to scientific investigation and explanation, but notfundamentally to critical evaluation. On this view, they cannot be among atheory’s logical implications, not just on the ground that theories lackvalue categories, but because (lacking truth value) no proposition at all canhave them among its entailments. Similarly, a value judgment, in principle,cannot cognitively affect either empirical data or scientific inferences.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (20)


Closely connected with the metaethical source is a logical view: statementsof fact do not entail statements of value (Hume 1739/1968); andstatements of value do not entail statements of fact (Bacon’s Aphorism 49,quoted on p. 2). The metaethical view is often thought to explain thelogical; but the latter may be entertained in combination with other viewsabout the nature of values. The Galilean idea may be seen as a particularinstance of the general Humean schema: “Fact does not imply value,” butthe argument sketched for it there does not depend on affirming thegeneral schema. The metaphysical source is independent of the logicalsource, and arguments (Bhaskar 1979; Margolis 1995; Murdoch 1992;Midgley 1979; Putnam 1978, 1981, 1987, 1990; Scriven 1974) against thelogical view may leave the metaphysical idea untouched. On the otherhand, the Baconian schema: “Value does not imply fact,” seems to me tobe correct and not dependent on accepting the above metaethical view. It is,however, consistent with values having implications about the interest orrelevance of facts, and the adopting of values having factualpresuppositions.

Both the Humean and Baconian schemata, however, draw attentionaway from some other logical relations involving fact and value. From fact(especially as it is represented in scientific theory) one can infer certainmatters about what is possible and impossible. And judgments of value(Chapter 2) have presuppositions about what is possible and impossible.Here, at least, is an avenue through which fact and value may logicallyinteract with important implications for working out the idea of neutralityin detail.

Practical and institutional

The fundamental sources of the idea of science as value free are those frommetaphysics (ontological primacy of underlying structure, process andlaw), epistemology (intersubjectivity of data, rule-bound scientificinference) and methodology (centrality of experiment), and success inproducing knowledge. The currency of the metaethical and logical viewshas provided reinforcement, rhetorically of service, but inessential. So far Ihave considered the idea as being about the content and consequences ofscientific theoretical products, and about the character of scientificassessment and knowledge claims.

But science should not be identified with its theories. We do not graspenough of the character of scientific theories if we abstract them from theprocesses in which they are generated, tested, assessed, reproduced,transformed, interlinked with other theories, adopted in practice,transmitted and surpassed. Scientific theories are both products of and of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (21)

instrumental importance to scientific practices, and our cognitive attitudestoward theories are shaped within these practices. Members of thescientific community engage in these practices, which are made intelligiblein the light of a long tradition; and they are conducted within various kindsof scientific institutions. These institutions, in turn, depend on otherinstitutions in society at large for the provision of their necessary materialand social conditions.

We can look at scientific theories from various points of view: theappropriate cognitive attitudes to hold toward them in virtue of theirrelations with empirical data; as products of a practice; as produced bypractitioners who have certain characteristics (including qualifications andperhaps moral qualities); and as produced within certain types ofinstitutions which express particular values, perhaps linked with those ofthe institutions that provide the material and social conditions needed forresearch or whose interests are best served by practical applications ofscientific results. Since scientific practice must be conducted withininstitutions, the possibility of there being constraints on its conduct andoutcomes, derived from the institution’s interests and values, cannot besummarily excluded. The potential for tension with “science is value free”is obvious; not an idle potential, for sometimes it turns into outrightconflict. I indicated an avenue on p. 7 whereby fact and value may interactthrough the intermediary of what is possible and impossible. A possibilitypresupposed by a valuative outlook (endorsed widely in society, ininstitutions that materially support science or by a significant politicalmovement) may be confirmed to be impossible in a scientific theory Then,in the name of the values, there can be a strong motive to overrulethe scientific claim. Or, more subtly, where such a conflict is incipient, thescientific community (consciously or not) may simply withhold frominvestigating the inconvenient possibility. My point is that it is quiteintelligible that values intrude on the scientific claims that are heldwhetheror not this intrusion is considered rationally admissible.

The idea that science is value free regards all such intrusion as distortion,and thus to be kept out of scientific practices:

One of the strongest, if still unwritten, rules of scientific life is theprohibition of appeals to heads of state or to the populace at large inscientific matters. Recognition of a uniquely competent professionalgroup and acceptance of its role as the exclusive arbiter ofprofessional achievement has further implications. The group’smembers, as individuals and by virtue of their shared training andexperience, must be seen as the sole possessors of the rules of the gameor of some equivalent basis for unequivocal judgments.

(Kuhn 1970:168)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (22)

Because there is an intelligible mechanism through which such intrusioncan readily occur, counter-mechanisms need to be operative withinscientific practice. Thus arises the further idea of the “autonomy” of thepractices and institutions in which scientific theories are generated,entertained, tested and evaluated. In practice, according to this idea,autonomy is a condition for gaining impartiality of theoretical appraisaland neutrality of theoretical claims.

Autonomy tends to be a rather slippery idea—reflecting diverse and evencontradictory currents—and one that is often and easily trivialized. At onelevel it is a political proposal: leave science to the scientists, but alsoprovide them with the resources to conduct their inquiries with no stringsattached. Appeals to neutrality, and to success in gaining knowledge andinforming practical applications, are often made to support this proposal.It also presumes that the growth of scientific knowledge (and of the bodyof accepted theories that manifest impartiality and neutrality) will takeplace most effectively within practices that involve and are under thecontrol of practitioners of the scientific community. A certain reading ofthe history of science might support this presumption. Autonomy alsodraws on the idea that science has its own internal dynamic, that sciencedefines its own problems, asks its own questions, identifies its own researchpriorities, seeking to gain ever more accurate, more unified, moreencompassing representations of the underlying order. The internaldynamic, it is said, responds only to the evidence and to the appropriatecriteria of cognitive value. According to this view, in the long run thehistory of science is the unfolding of this internal dynamic, punctuated bymoments of intrusion from outside values and interests which alwaysretard the process.

The proposal for autonomy normally grants sole authority to thescientific community with regard not only to defining problems andappraising theories, but also to determining the qualifications required formembership in the scientific community, and deciding the content ofscience education. This draws on the sociological posits that members of thescientific community conduct their scientific practice motivated by theobjectives of impartiality and neutrality or, more likely, that their activitiesare so structured that in the long run accord with impartiality andneutrality is virtually assured; and that scientific education adequatelyattunes them to accept as knowledge only that which accords withimpartiality.

These posits are bolstered by the claim, common in the public imagepresented by the tradition of modern science, that the scientific communityhas successfully cultivated among its members in their conduct as scientiststhe “scientific ethos” (Merton 1957)4, the practice of such virtues ashonesty, disinterestedness, forthrightness in recognizing the contributions(and opening one’s own contribution publicly to the rigorous scrutiny) of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (23)

others, humility and courage to follow the evidence where it leads. Clearlythis is the stuff of which myths are made.

Autonomy is not easy to render in a precise thesis, and its historical andsociological presuppositions are open to further empirical investigation. Itis better regarded, I think, as a reaction affirming a value in the face ofunhappiness occasioned by its perceived rejection or subordination—areaction that provides a ready and unthreatening explanation of whysometimes science has gone astray. The meaning of autonomy is shaped inopposition to troubling events, symbolized by the trial of Galileo, thehorrors of Lysenkoism, the bemusing stubbornness of the creationists and,among some, also by the ready willingness of scientists to engage inclassified research when called to do so for the sake of national securityand to keep their results secret or legally limited in their use for the sake ofcorporate profits.

While the idea of autonomy arises as a reaction to certain kinds of“outside interferences,” hinted at in symbols rather than specified sharply,there is one kind of “outside influence” that generally is tolerated, evenovertly welcomed—when the institutions which fund and support scienceare granted an important role in determining research agendas, theproblems to be investigated and the domains of phenomena to be studied.Where this happens (and it happens commonly enough) research prioritiesare generally not set according to the posited internal dynamic of science,but by negotiation with the bearers of non-scientific values and interests—typically for a practical reason. This need not undermineimpartiality, though it may (Lewontin 1993), for the role of the values andinterests may be restricted to the choice of research domain and need notextend to having impact on which specific theory comes to be accepted ofthat domain. We will see much later on p. 251 (Chapter 10) what impact itmay have on neutrality. Impartiality, however, might be threatened if therewas in fact an identity of (personal and social) interest among the scientificcommunity and the agencies of support. The myth of the scientific ethosfunctions to deny that there are such identities of interest. Others counterthat a greater diversity (of personal and social values and interests) amongthe practitioners of science would make a more convincing argument; butpublic pressure to bring about such greater diversity tends to be opposed inthe name of autonomy.

It is a compromise of the idea of autonomy of scientific practice to granta role to non-scientific values and interests in choosing a domain ofinvestigation. I do not criticize such compromises per se, for scientificpractice may be impossible without some of them (and less than completemanifestation of a value does not mean that it is not a seriously adoptedvalue). However, in particular cases, I do criticize the choice from theperspective of other values. In its most compelling form, autonomy isclaimed so that the responsibility of scientists—concerning impartiality and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (24)

neutrality—can be exercised. In a trivial form, which has become morecommon in recent years, it amounts to little more than the special plea tobe free to enter into compromises with whatever agencies one sees fit,without regard to the broader social interests that may be affected by thechoices.

Scientific method

The ideas about empirical data and scientific inference, often functioning inconcert with the Galilean idea, may be put together under the idea thatmodern science has a method. The accepted theories of modern science arethe product of following a method in which intersubjectivity and oftenconstraints grounded in the Galilean metaphysical idea are the definingelements. The method matters; not who is following the method. Imentioned on p. 10 the related idea that the practitioners of science,insofar as they engage in scientific practice, are the bearers of the virtues ofthe scientific ethos. Qua bearers of these virtues they are interchangeable,reinforcing that who is following the method does not matter, subject tothe condition that the practitioner has the relevant competencies(observational, experimental, mathematical, inferential, conceptual,theoretical) necessary to follow the method. “Method,” as used here,pertains principally to how theories come to be properly accepted orappraised, not (except as a constraint) to how they come to be put forwardand entertained in the first place; it is held to pertain to the context ofjustification, not that of discovery.

According to common views, the other side of method is free creativity,for that is what supposedly enables a theory to be put forward forconsideration. (A theory is created; then it is appraised following the normsof the method.) In the “context of discovery,” individuality is celebratedand no potential (conscious or unconscious) stimulus to creativity (whichflourishes on analogies) including values, is ruled out a priori. Perhaps valuescan slip in here and play unnoticed roles. Theory appraisal is comparative:it involves choice among competing theories, but the competitors have firstto be “created.” Values may be hidden because a competing theory thatwould enable the values to become manifest may not have been “created.”Intersubjective agreement, obtained through following the method, maynot be enough to overcome this, especially if the agreement is amongpractitioners selected in a context where competence is the only explicitlyrecognized necessary requirement, and the assumption prevails that theyembody the scientific ethos. For they, the “creative innovator,” the researchinstitution and its funders may share identical interests that, in the absenceof tension derived from competing interests, may simply fall into theunproblematic and unrecognized background. Thus, it is possible thatvalues are in play and not “noticed” because the intersubjective agreement


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (25)

extends to include agreements about them. Here is a hint that who thepractitioners are may matter.

Perhaps impartiality can be regularly achieved only if there is a diversity(with respect to values and interests) of practitioners in critical interactionand some diffusion of cognitive authority. “Method” may require clashingvalue perspectives rather than the activities of practitioners who actindividually out of the scientific ethos.5 Scientific appraisal may becommunal or social: the product of interaction rather than the sum ofindividual acts of following the method (Longino 1990; Solomon 1992;1994).


The ideas of impartiality, neutrality and autonomy sum up what I think isthe core of the idea that science is value free. Endorsing them is compatiblewith values playing many roles in connection with science, mostimportantly: values may play decisive roles in connection with the stancesadopted toward theories prior to their acceptance; cognitive values helpto articulate the idea of impartiality; and the three ideas themselvesfunction as values that may not always be well reflected in actual scientificpractice.

Theories: acceptance, application, significance

Earlier, I have used expressions like “accepting” and “choosing” a theory.The idea that science is value free concerns characteristics of the theoriesthat we accept and ought to accept, their consequences and the practices inwhich they are considered and come to be accepted. I will now introduce insome detail the key notion of “accepting a theory” (which is deployedfrequently throughout the text) and distinguish it from several otherstances that may be taken toward theories. Values may play a variety ofroles in connection with the other stances.

Accepting theories

I will stipulate a usage6 of “to accept a theory,” and distinguish it fromsome other important stances that may be taken toward theories(hypotheses, proposals, posits, or conjectures): provisionally entertainingthem, adhering to them in research practices, endorsing their greaterevidential support (compared to rival theories) and applying them inpractical life. A theory (T) is accepted of a domain (D) or domains ofphenomena; one “accepts T of D.”


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (26)

To accept T of D is to judge that, in the light of the available evidence, Tof D is sufficiently well supported that it need not be submitted to furtherinvestigation—where, for example, it is judged that further investigationcan be expected only to replicate what has already been replicated manytimes over and to bring minor refinements of accuracy and sharperidentification of the bounds of D. It is to consider it among the items. ofrationally consolidated beliefs (Chapter 3) or to include it in the stock ofknowledge so that ceteris paribus it is sufficiently well established to beapplied to inform practical projects (pertaining to the phenomena of D).Acceptance is a strong stance to take toward a theory. It also alwaysremains, in principle, open to revision that might be occasioned by newdevelopments concerning either empirical data or theory. To reject T of DD is to accept T′ of D, where T and T′ are held to be inconsistent.

Accepting T (of D) is a stronger stance than endorsing that, on balancethe available evidence points more toward T than toward rival theoriesthat have been entertained, for then one anticipates that the balance maywell be disrupted by further research including that which mayprovisionally entertain novel rival theories. Acceptance is a stance adoptedwhen relevant research has become considered as effectively completed,like (ideally) in the cases mentioned (in the preceding section) as successesof modern science. In these cases it seems reasonable to maintain, and theconsensus of the scientific community confirms, that further research willnot could!—not lead to a change of judgment about T, except at the levelsof refinement and meeting standards of accuracy. It involves (ideally)judging that the degree of evidential support is sufficiently high accordingto the highest available standards for estimating it, so that consistency withan approximation of T (of D) becomes a constraint upon any theory thathas more encompassing scope than T (Joseph 1980). Accepting T (of D) isaccompanied by a sort of (pragmatic) certitude, but that should not beconfused with (epistemic) certainty.

Accepting a theory comes at the end of a process of research in which ithas been developed from predecessors, which have been provisionallyentertained and adhered to by committed investigators, and separated outfrom both their predecessors and other rival theories by way of numerousjudgments of comparative evidential support or rational acceptability—acceptance of a theory follows (properly) after having made numeroustheory choices. A theory (an early version of one that may eventually beaccepted) may be provisionally entertained for the sake of exploring itsimplications, its potential to generate and solve problems, and itsrelationship with empirical data and with other theories. Generally thisinvolves endorsing the plausibility of T: To hold that T providesconceptual and hypothetical resources sufficient to shape (reshape,contribute toward) a research agenda, and that the agenda is sufficientlypromising to warrant material, financial and institutional support. In order


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (27)

to be developed and reshaped into an acceptable form a theory must alsobe adhered to, that is, a research agenda framed by it must be participatedin and commitment must be made to its furtherance.

Applying theories

Finally, accepting T (of D) is not the same thing as applying it.“Application” concerns the role of T in the realm of daily life and practicalactivity. T may not be applicable because D does not include significantphenomena of daily life and experience (Chapter 7) or because it cannot bedeployed significantly in practices which express one’s adopted valuecomplex. To apply T, I stipulate, is to apply T to significant phenomena ofdaily life and experience and/or to apply it in practical activity. T is appliedto phenomena when it is used (by way of providing representations of themwith its categories and principles) to provide understanding of them—sothat when the relevant phenomena of daily life are included in D,acceptance of T (of D) suffices to ensure its applicability to them. Tis applied in practical activity when it is used to inform practical (oftentechnological) activities related to the phenomena to which it appliesconcerning such matters as the workings of things, means to ends, theattainability of ends and the consequences of realizing the possible.

My usage of “apply” is more general than the one commonly associatedwith the phrase “applied science”, which limits “apply” to the secondcomponent, that is, essentially to technological applications, when scientificknowledge is deployed as an instrument that informs effectively practicalinnovations in daily life, and particularly the development, introduction,operation and maintenance of technological devices and practices, wherethe outcomes of scientific inquiry become causal factors in transformingthe social “world.” On this common usage, the making and exploding ofan atomic bomb is referred to as an application of physical theory, butexplaining how the sun is a source of light and heat in terms of itsthermonuclear activity is not. As a widespread and socially significantphenomenon technological application is relatively recent, dating from onlyabout 150 years ago (White 1968), though applications of theories toexplain the workings of technological objects, and theoretical reflection ontechnological objects as a source of scientific ideas, date at least from thetime of Galileo, as do theoretical applications to numerous otherphenomena of daily life and experience.

Significance of a theory

Clearly (moral and social) values must play a role when a theory is applied.No matter how strongly a theory is taken to be accepted it is applied onlyif applying it accords with one’s values. One may apply a theory to


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (28)

significant phenomena of daily life, but not in one’s central practicalactivities. In the case of technological applications, a condition of(legitimately) applying T is the moral propriety of the intendedconsequences and the anticipated side effects. The highest degree ofcognitive value is never sufficient to legitimate practical application, so thatany move from acceptance to application in practical activity shouldalways explicitly involve considerations of moral and social values.Accepting a theory does not imply the desirability or legitimacy of applyingit practically, but only that there is no cognitive barrier to doing so. Thelegitimate applicability of a theory in practical activity requires supportfrom one’s adopted value complex.

I will say that T (of D) is significant for specified values if T is applicableto important phenomena of daily life and experience and/or is applicable inpractical activities in ways that further (and do not undermine) theinterests shaped from adherence to the values. A theory is more or lesssignificant for given values; and it may be highly significant, for example,concerning applications to phenomena but not concerning applications inpractical activities. Significance is a matter of degree, multifaceted andsubject to historical variation, and it does not follow from acceptance (cf.Anderson 1995b).

The role of cognitive values

The idea of impartiality denies that value judgments are among thegrounds for accepting and rejecting theories. But to accept or reject atheory is itself to make a judgment of cognitive value (worth, merit)(Scriven 1991). One interpretation of impartiality is that judgments of“non-cognitive” (personal, moral, social, aesthetic, etc.) value play no rolein choosing theories. Another, drawing heavily from the metaethical andlogical sources, wants to keep out all value judgments. It proposes to do soeffectively by reducing theory appraisal to the recognition of the outcomesof rule-governed operations involving formal relations between theoriesand empirical data. I favor the first interpretation (Chapter 3), and I willdevelop it in detail (Chapters 4 and 10). Regardless of interpretation,however, to choose a theory is to grant it cognitive value, to affirm (atleast) that it is a better theory than some other competitor. Suchaffirmations are intended to be “objective”; there is a fact of the matterabout which theory has greater cognitive value, causing if not outrightparadox at least perplexing tension with both the metaethical and logicalviews that partially ground the second interpretation (Scriven 1974, 1991;Putnam 1981).

It fits with the interpretation of impartiality that I favor that judgmentsof cognitive value can be construed as the outcomes of estimates of howwell theories fare when appraised in the light of certain criteria (for


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (29)

example, empirical adequacy, explanatory and predictive power), that is,estimates of how well theories manifest certain cognitive values(Chapter 3). “Science is value free,” thus, should be considered compatiblewith the view that cognitive value judgments play essential roles in theaccepting and rejecting of theories; it thus presupposes that cognitivevalues can be clearly distinguished from other kinds of values.

Throughout the book I will follow the terminological convention thatthe word “values,” used without qualification, will mean “personal, moral,social and other values, but not cognitive values.” I beg no questions ofsubstance in doing so; should it be established that cognitive cannot bedistinguished from other kinds of values the convention will have to bedropped—so too will impartiality. The role granted to cognitive values ofcourse penetrates to the very heart of scientific practices. So long as no non-cognitive values penetrate in similar ways, there is nothing that theproponents of “science is value free” need regard as threatening.

Where science and values “touch”: a miscellany

To use Poincaré’s evocative words there are many places where science andvalues may touch but not interpenetrate. I will list some of what have beenconsidered the more important places of “touch,”7 without furthercomment, simply for the sake of clarifying what is, and what is not, atstake (for its proponents) in the idea that science is value free:

• Science itself is a value (not necessarily an unsubordinated one). Thisaffirmation comes in many versions: knowledge (truth) is a value;science informs practices that produce value; its own practice requiresthe exercise of rationality, a universal value (Nagel 1961), or moregenerally, it cultivates in its practitioners characteristics that areconducive to human flourishing or well-being (Putnam 1981, 1990); itcreates beauty (Poincaré 1920/1958).

• The making of value judgments, and relations among value judgments,can be informed (and criticized) by scientific knowledge of means to endsand the attainability of ends.

• There can be scientific (psychological, sociological, historical andperhaps biological) studies of values: Of their being held, manifested andembodied in persons, institutions and cultures, and of how particularvalues come to be held and transformed (Lacey and Schwartz 1996).

• There can be ethical evaluation of, and restrictions on, scientific practiceand applications. There are, for example, ethical issues that arise inconnection with the choice of research goals, the staffing of researchactivities, the selection of research methods (and experimental subjects),the specification of standards of proof, the dissemination of researchfindings, the control of scientific information, and the credit for research


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (30)

accomplishments (Rescher 1965:274). Deploying a soundly acceptedtheory in practical application, its specific manner of application, andjudgments about its “significance” reflect ethical evaluation. That atheory is sufficiently well supported to warrant its practical application,in view both of the “side effects” of applications (Scriven 1974) and ofthe risks of its application should it turn out to be false, involves ethicaljudgments (Rudner 1953).

• Values may play numerous roles (either positive or negative) in the“context of discovery,” concerning judgments made in connection withthe various stances that precede acceptance of a theory; in sensitizingresearchers to the importance of certain facts; in motivating researchefforts (Rescher 1965); and in assessing “scientific performances,” suchas carrying out experiments or writing papers (Scriven 1974).

• Values may play a role in connection with the compromises reachedinvolving autonomy (discussed in the preceding section), for exampleconcerning questions raised, research supported and problems selected;and in making judgments about whether a certain line of researchshould be carried out in view of probable applications that would follow.

• Commitment to certain values may motivate scrutiny of commonscientific practices for “biases,” focus on particular problems andpolicies regarding membership of the scientific community; there may bevalue-based criticism of scientific practices and institutions.

• The practices of science may require that their practitioners manifestcertain personal and moral values (the “scientific ethos”) and reinforcethe valuing of certain personal traits (for example, creativity,mathematical and experimental capabilities). Since it has social andmaterial conditions, it may progress, or its rate of progress may beaffected at a given time, where particular social and personal values aredominant, and what these values are may vary with the historicalmoment (Hull 1988:76). It may also, for the sake of the fullermanifestation of impartiality, require that a variety of (social and moral)values be held among its many practitioners.

• The practitioners of science may incur special moral and socialresponsibilities in the light of their activities and discoveries.

Science as value free: fact, idealization, or value?

That science is value free, I repeat, does not mean that there is no interplaybetween science and values; only that what interplay there is leaves thethree component views untouched. Thus, matters of values may illuminateall sorts of aspects of the practice, sociology, institutionalization andhistory of science. It is not enough to impugn that science is value free todisplay ways in which science and values “touch” each other.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (31)

Furthermore it is not enough to impugn “science is value free” that oneor other of the components is actually not highly reflected in some aspectsof scientific practice. Neutrality, for example, may not be highly manifestedin actual fact since available scientific knowledge (given current conditions)may be significantly applicable only in support of certain values; yet it mayremain open to fuller manifestation in principle.

Among its proponents, each of the components of “science is value free”is itself a value, to be expressed in scientific practices and embodied inscientific institutions, a value embedded in the objectives of science itself.There is nothing paradoxical about this. As values they are manifested tovarying degrees in scientific practices and in the acts of accepting orrejecting theories. Its proponents suggest that “science is value free” is alsoactually reflected—at least as an idealization—in many fields of science, sothat in the modern scientific tradition the value, “science is value free,” hasbecome well manifested; as the slogan goes: “How else can one explainthat we got to the moon?” In some fields of science, for example,psychology, it is much more difficult to back up the suggestion that it isactually manifested even as idealization. Nevertheless, as long as it isdeemed possible of fuller manifestation (in a field of inquiry), it can serveas a guiding value or regulative ideal. It is not impugned unless it is shownthat the trajectory of science is not or ought not be in the direction of itsfuller manifestation in the making of theory choices, that it cannot orought not serve as a regulative ideal—perhaps by showing that cognitiveand other kinds of value cannot be distinguished or by arguing that there isno way to institutionalize scientific practice that can ensure that theoriesare chosen only in view of considerations of cognitive value.

I am not putting “science is value free” beyond the scope of criticalappraisal, but pointing to the genuine complexity and multifacetedness ofthe interplay of science and values. The interplay is not all at the levels oflogic, method or even metaphysics. How fully “science is value free” isexpressed in scientific practices can depend in part on the social conditionsin which research is conducted. The question, “What conditions of researchare conducive to the fuller expression of ‘science is value free’?” is nottrivial. It is a question that needs to be informed by empirical inquiry,historical as well as sociological and psychological, that may also point tohow, for example factors, other than the data and the cognitive norms, couldhave (unobtrusively) played a role in certain theories becoming accepted inviolation of impartiality. (Thus, heuristically, the sociology of inquiry isrelevant to the “context of justification.”) I take the components of“science is value free” to be theses about the actual (past, present andfuture) acceptance of theories, not logical theses about an idealized science.It is idle to affirm that they represent values of scientific practice if thesocial conditions for their progressive fuller realization are neither presentnor plausibly available to be implemented.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (32)


My principal goal is to explicate and to appraise the three ideas(impartiality, neutrality and autonomy) that jointly constitute the view thatscience is value free. Thus Chapters 4 and 10 provide the foci aroundwhich all the other themes are gathered. “Science is value free” has twofaces. One looks toward the sciences themselves; it deals with how valuesdo and do not, ought and ought not, interact with the making of theorychoices and with the drawing of consequences (logical and practical) fromaccepted theories, and the practical and institutional conditions in whichthey do and can do so. The other looks toward the place of science withrespect to the values we hold, and how it does and does not, can andcannot, serve the projects that express these values. Often “science is valuefree” has been treated as little more than a footnote to discussions ofscientific inference and methodology. I think it also bears on the deep valueissues and conflicts of our age. Science and values are equally at the centerof my attention, and so I embed my argument in an account of the generalcharacter of values (Chapter 2), and a detailed analysis of the role thatscientific practices play with respect to the predominant values ofmodernity (Chapter 6) and of the role they might play with respect tocompeting values. Concerning the latter I defend the cognitive credibility ofapproaches to systematic empirical inquiry that are more in tune withmaking a variety of forms of human flourishing sustainable (Chapters 8and 9).

My methodology involves two key steps. First, beginning from the ideassketched earlier in “Sources,” I offer provisional theses (Chapter 4) ofimpartiality, neutrality and autonomy. They are intended to provide aplausible reconstruction of the idea of science as value free that resonatesthroughout the tradition of modern science. The arguments of Chapters 5to 9 provide the material for a thorough critique of these theses. Then—the second step—by way of a series of revisions motivated by thesearguments I formulate (Chapter 10) new versions of impartiality andneutrality (though not autonomy) which I defend.

The provisional theses are already responsive to my account of thegeneral character of values (Chapter 2), and they are formulated on theassumption that cognitive values (rather than rule-governed accounts ofany kind) provide a satisfactory account of the character of the cognitivevalue of theories (Chapter 3). They are theses about scientific practices (andtheir theoretical products) which draw from the fusion of the Galilean andBaconian ideas introduced on p. 6. In the light of proposing that theobjective of science is to gain understanding of domains of phenomena—explanations of them and encapsulations of the possibilities they allow—Iargue (Chapter 5) that these ideas can and ought to be separated.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (33)

Then, scientific practices which draw on the fused Galilean/Baconianideas appear as the practices of one approach to science, one in principleamong many, one that deploys particular strategies—materialist strategies(Chapter 4)—(oversimplifying) that constrain the theories that areentertained to those that may represent phenomena in terms of beinggenerated from underlying structure, process and law; and that selectempirical data that may bear on such theories, especially data that reportthe outcomes of measuring and experimental operations in abstractionfrom the human and social contexts of the investigation. Following thematerialist strategies we have been remarkably successful in identifying the“material possibilities” of phenomena, those possibilities that can berepresented in terms of the generative power of underlying structure,process and law. But they do not enable us to gain access to thosepossibilities open to phenomena in virtue of their relations with humanbeings or with the social order. Perhaps (in principle) other approacheswould enable us to do so, approaches (consistent with the Baconian idea)aiming to gain systematic empirical understanding appraised in the light ofthe cognitive values.

If (in principle) there are alternative approaches, why are the variousmaterialist strategies deployed almost exclusively in the practices of modernscience? My answer in Chapter 6, consolidated with a reflection on Kuhnin Chapter 7, points not to Galilean metaphysics, but to mutuallyreinforcing interactions between adopting the materialist strategies and“the modern values of control,” most importantly the value of expandingour capability to exercise control over natural objects.8

More generally, it is not possible to pursue the objective of science(gaining understanding) except within the confines of a particularapproach, where each approach is defined by the adoption of particularstrategies which interact in mutually reinforcing ways with particular(social and moral) values.9 Are there “really” alternatives to adopting thematerialist strategies? Are there “really” alternative strategies under whichwe can gain theories that become accepted in accordance with impartiality?I offer (Chapters 8 and 9) two anticipatory alternatives: one—developedamong some grassroots movements in third-world countries—characterizedby a dialectical interplay of traditional forms of knowledge (for example, inagriculture) and materialist investigation interacts with such values as theenhancement of local well-being, agency and community, and social andecological balance; the other, which permits an essential role forintentional categories in the investigation of human cognitive capacities,interacts with feminist values. There “really” are alternatives, so thatadoption of a strategy is justified (in part) and explained by its link withvalues.

Values thus pervade the practices in which scientific understanding isgained; and, actually, the modern values of control are almost all


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (34)

pervasive. If one rejects these values, if one subordinates the place ofcontrol to other values (for example, to those of grassroots movements orto feminist ones) it is appropriate to explore alternative strategies. Valuesmake a difference to what one does in science, to what kinds ofpossibilities are grasped in the theories that are produced. Thus, howscience is practiced will have impact on the conditions and possibilities ofdaily life and experience. The phrase “science is value free” is thusmisleading, and, in practice, it functions to divert attention away from thefact that strategies (as well as theories) are chosen, and that the Galilean/Baconian approach is but one approach among (in principle) many. It isbest dropped. Nevertheless, there is much in its component ideas thatshould be refined, retained and emphasized.

The way in which values are pervasive in scientific practice is incompatiblewith autonomy. But impartiality remains an important, indeed essential,value of all approaches to gaining scientific understanding. Althoughtheories are developed under strategies whose adoption is influenced byvalues, theories should be accepted only in the light of considerations thatinvolve empirical data, other accepted theories and the cognitive values(Chapter 10). The levels of (and grounds for) strategy adoption and theoryacceptance need to be clearly separated, while maintaining that (in the longrun) links with values cannot support a strategy in the face of its failure togenerate theories that are accepted in accordance with impartiality.Neutrality, with its three components: “consistent with all valuejudgments,” “no (cognitive) consequences in the realm of values” and“evenhandedly applicable regardless of values held” turns out to be morecomplicated, with the second and third components requiring verysignificant revisions (Chapter 10). Even then, unlike in the case ofimpartiality, my final version of neutrality does not articulate a value thatit*elf would be ranked high in all value outlooks; what should be the scopeof neutrality remains controversial.

The novelty of my account derives from the identification of a set ofdifferentiated dialectical relations between scientific practices and values.10

It makes possible a synthesis in which due recognition is paid to the criticsof “science is value free” while retaining, with a measure of re-definition,core insights of its defenders.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (35)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (36)


If the sciences are value free, exactly what is it that they are free from? Inthis chapter1 I offer a general account of values and value judgments. Thisaccount underlies the understanding of cognitive values presupposed in thestatements of impartiality, and has far-reaching implications for theassessment of neutrality (Chapters 4 and 10). It also grounds subsequentsubstantive discussions about particular social values and their (potential)relevance to scientific inquiry.

The word “value” has varied and complex uses. In ordinary discourse,when we refer to a personal value, we may be pointing to some or all of thefollowing:

1 A fundamental good that one pursues consistently over an extendedperiod of one’s life; an ultimate reason for one’s actions.

2 A quality (or a practice) that gives worth, goodness, meaning or afulfilling character to the life one is leading or aspiring to lead.

3 A quality (or a practice) that is partially constitutive of one’s identityas a self-evaluating, self-interpreting and a partly self-making being.

4 A fundamental criterion for one to choose what is good amongpossible courses of action.

5 A fundamental standard to which one holds the behavior of self andothers.

6 An “object of value,” an appropriate relationship with which ispartially constitutive both of a worthwhile life and of one’s personalidentity. Objects of value can include works of art, scientific theories,technological devices, sacred objects, cultures, traditions, institutions,other people and nature itself. Appropriate relations with objects ofvalue, depending on the particular object, include the following:production, reproduction, respect, nurturance, maintenance,preservation, worship, love, public recognition and personal possession.

In practical life, beliefs and desires constitute an essential part of theexplanation of human action: one performs an action because one desires acertain outcome and believes that the action will contribute toward

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (37)

bringing about the outcome. Desires are thus among the causes of actionand, as such, they may be objects of psychological and social inquiry. Inaddition, desires are objects of evaluation. They can be judged by thepeople who hold them, and others, concerning the possibility of theirrealization and their worth in one’s own life or in a human life in general.In particular, where desires can be represented as the having of goals,agents aim, as an ideal, to have the desires that play a causal role in theirbehavior included among the positively evaluated desires—those that areconsonant with their values.

Explaining actions in terms of an agent’s beliefs and desires alwayspresupposes a broader context in which the action in question is related toother actions (including acts of evaluation) through developing networks ofbeliefs and desires, which eventually make contact with the agent’sfundamental goals and desires, that is, the agent’s values. Through suchdeveloped explanations the causal role of values in behavior becomesapparent. Ordinary intentional explanations of action thus presuppose thatvalues play a causal role in behavior.


We may think of personal values as dialectically both the products and thepoints of reference of the processes with which we reflect on and evaluateour desires.3 Holding values, then, involves second-order desires (Taylor1985), desires about the first-order desires that play and will play a causalrole in our lives; desires that only first-order desires with certain featureswill mark our lives as lives that are experienced as fulfilling and worthy ofa human being. Holding a personal value involves the second-order desire,which represents one of a person’s fundamental goals, that one’s (acted onas distinct from merely felt) first-order desires be of the kinds that lead toactions that shape or produce a life marked by a certain quality (byparticipation in a certain practice or by appropriate relationship with acertain object of value) that it is believed makes for a fulfilling (good,meaningful, well-lived) life, and that is partially constitutive of one’sidentity. The role of values as criteria of choice and standards of behaviorare derivative from this core meaning. Desires are personal. The desirecomponent of holding values points to the personal character of values,that one’s values are tied to one’s most fundamental desires and to one’sdeepest feelings. Holding a value also involves a belief component, thebelief that the quality referred to is indeed linked with the experience of afulfilled life and perhaps also the belief that a life marked by this qualitydoes not cause or rest on conditions that cause diminished lives for others.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (38)

The modes of personal values

Understood in this way, values are manifested in behavior whenever they,and their associated feelings and emotions, figure in the explanatorynarratives that are formed to understand the behavior of an agent. Valuesare woven into a life to the extent (more or less) that the trajectory of anagent’s life displays behavior constantly, consistently and recurrentlymanifesting the values. A value is expressed in a practice where conductwithin the practice is furthered by and requires behavior that manifests thevalue. Values can also be present (both felt and reflected on) inconsciousness, and articulated in words, representing a partial account ofwho one is (or would like to be or would like others to think one is), one’saspirations for the future and what one believes about human well-beingand its condition.

There will always be some measure of a gap between values-as-manifested and values-as-articulated. One comes to hold values reasonablyin the light of the desire and the commitment to narrow that gap. The gaphas various sources. On the one hand, one’s aspirations can, and oftenshould, properly go beyond current realities. On the other hand, the gapcan be a consequence of inadequate self-understanding, limited orunderdeveloped capacity for self-interpretation, the desire to appear toconform to the norms of some group and even willful self-deception.

Though values cannot be reduced to their articulation, their articulationis critically important. It is not just talk about values, as if values had akind of being separable from their articulations, making the articulation averbal representation of a separate, perhaps mental, reality (as a sentencedescribing a material object is separate from the material object). It is partof the nature of values that they be articulated. Articulation is itself anessential mode of values—part of their formation, maintenance,transformation, deepening, clarification, recognition and definition.Moreover, the very act of articulation of values may also manifest ourvalues, since to whom we articulate our values, how and with what depthwill vary according to whether the immediate audience is composed ofloved one, friends, colleagues in a movement, and so on. Such articulationis part of the practice of self-interpretation, a practice necessary for a lifewithout self-deception. It helps to define one’s aspirations. It implies notonly an anticipation or prediction about the future trajectory of one’s life,but also promises concerning that trajectory—that the values-as-articulatedwill become the values-as-woven-into one’s life. To be credibleas aspirations, values-as-articulated must go beyond values-as-manifestedcurrently. To be credible as predictions and promises, the gap must not betoo great, for one’s future possibilities are constrained by the presentrealities of one’s life. The credibility is enhanced when one is consciouslyengaged in practices that offer a well-founded possibility of realizing the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (39)

aspirations. There can be tension here and ample space for self-deception.Finally the articulation of values enables values to become objects ofinvestigation (psychological, epistemic and valuative), of reflection, ofdiscussion and of critical argument, and when one discovers—as aconsequence of articulation—that one shares one’s values with others, theycan become the basis of participation in shared practices and in theconstruction of community, the ground for living together withoutviolence. This articulation makes it possible to reason about values; and ifone does not reason about values, one will not value reason.

Embodiment of personal values in social institutions

Personal values can also be embodied (more or less) in social institutions,and in society as a whole. An institution embodies a value to a high degreewhen its normal functioning offers roles into which the value is woven,encouraging behavior that manifests it and practices that express it,reinforcing its articulation and providing the conditions for its beingfurther woven into its members’ lives. In this sense, elite universitiesembody to a high degree the value of intellectual cultivation anddistinctiveness, but not that of solidarity with the poor; and capitalisteconomic institutions embody to a high degree various egoist values, butnot sharing. A social order embodies a value to a high degree if it providesconditions that support institutions that embody the value, and especially ifits maintenance and normal functioning depend on such institutions.

The values that can be woven into a person’s life are constrainedsignificantly by the values that are embodied to a high degree in the societyin which the person lives. That is partly because articulation is an essentialmode of values, and what can be articulated is a function of the linguisticresources available in one’s society, which will reflect to some degree theconceptions of well-being that are dominant and reinforced in the society.This language may not readily permit the expression that one’s ownexperience of well-being (or diminishment) does not fit well with thereigning accounts of what constitutes well-being. For example, in a societythat highly embodies egoist values, in which persons are respected andrecognized in virtue of their possessions, the language is not readilyavailable for one to articulate the experience (if one has it) thatmanifesting such values does not produce a sense of well-being. Thence thepull is to submit one’s experience to the reigning accounts of well-being, ifonly for the sake of being recognized and respected. In this way, values arepartially constituted by the available discourse of value, and part of thereality of holding values is essentially linking one’s life to the community(and its traditions) which is the source of the language of one’s values.

The constraint due to the social embodiment of values exists alsobecause people live their lives in interaction with others. Most actions are


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (40)

also interactions, as mentioned earlier, so that one’s values will include thefundamental relations one desires to establish with others. Typically ourinteractions with others are mediated by social institutions: family, school,church, political and economic institutions, clubs, etc.—so that we interactin accordance with our institutional roles and with relations structured byinstitutions. To a considerable extent, one cannot manifest one’s personalvalues without participating in institutions that permit their manifestation.Not every institution encourages or even permits the manifestation of one’spersonal values, so that whether a personal value can be woven into a lifedepends considerably upon the availability of institutions in which it isembodied to some degree.

I have discussed personal values as articulated in words, as present inconsciousness, as manifested in action, as expressed in practices, as woveninto lives and as embodied by social institutions. Values have no realityapart from these modes. In particular, personal values cannot be reduced tomental representations or simple conscious phenomena. Ontologically,values reside in the interplay of these six modes, which together areconstitutive of values; and so necessarily values are developing and notsimply given. They may be shared, in virtue of their being expressed inpractices, articulated, and embodied in institutions—and to a considerableextent they must be. At the same time, in virtue of their manifestation inaction and their being woven into individual lives, their character retains apersonal element.

Kinds of values

There are various kinds of values. A value is held by an agent or agents.When an agent (X) holds a value (v), the fundamental expression is: “Xvalues that ø be characterized by v.” The different kinds of valuescorrespond to different instantiations of ø: for example, when ø=myself, wehave my personal values; ø=persons in general or relations and interactionsbetween persons, moral values; ø=an institution, institutional values;ø=society, social values; ø=works of art, aesthetic values; and (Chapter 3)ø=scientific theories or systematic bodies of beliefs, cognitive values.5

Values have both desire (want, goal) and belief dimensions: the desirethat ø be (becomes) characterized by v; the belief that being characterizedby v is partly constitutive of a “good” (“worthily desired”) ø.

Regarding the belief dimension, depending on the ø in question, differentconsiderations come into play. When ø=human person, the beliefs involvedconcern the characteristics of a fulfilled, flourishing, meaningful or well-lived human life; and they concern the relations among persons that foster(and partly constitute) fulfilled lives and that do not rest on conditions thatproduce diminished lives. When ø=myself, the values include characteristicsthat are partly constitutive of my personal (individual) identity (as


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (41)

discussed on p. 24). When ø=society, the values involve characteristics ofsocial structures and organization that contribute to human well-being. Nomatter what ø may be, contribution to human well-being is always the“bottom line” of value discourse.

Social values and their modes

A social order is marked by the personal values that are predominantlyembodied by it, and also by the social values that are woven into it. As inthe case with personal (and all kinds of) values, social values involve theinterplay of several modes. They are manifested in the programs, laws andpolicies of a society, and expressed in the practices the conditions of whichit provides and reinforces. These are the values that become articulated inhistories of the society’s tradition, in explaining the kinds of institutions ithas fostered, and in the rhetoric of its leadership. Again, there is alwayssome gap between manifestation and articulation, the handling of whichpartially defines positions on the political spectrum. Social values arewoven into a society to the extent that they are manifested constantly andconsistently, and the gap is quite narrow. For example, liberty, the primacyof property rights and, to a much lesser extent, equality are social valueshighly woven into US society.

Articulation of values has a special significance in the case of socialvalues, since typically there is contestation about social values amongvarious members and groups in the society. Different groups within societywill perceive and interpret the gap between values-as-articulated andvalues-as-manifested quite differently, and much of modern politicaldiscourse centers on the various competing assessments of the significanceof this gap.

There is a close link between the social values woven into a society andthe personal values a society embodies, and also between the valuesthat are articulated by the dominant institutions of a society (ideology) andthe personal values that become articulated throughout the society. Thislink need not be formal and may only become apparent as the social orderunfolds concretely over time. Thus, for example, liberty (negative liberty)and the primacy of property rights, as woven into the concrete economicand legal institutions of the US, foster the embodiment of individualistic,egoistic and competitive personal values. Indeed, the embodiment of suchpersonal values may itself be construed as a social value highly woven intothe society. Under conditions in which the link between social and personalvalues is especially tight, the personal values people hold may come to seemnatural and inevitable—so much so that they cease to be deliberated aboutas values and become construed as facts of human nature.

A social value can also be personalized when a person’s acts directedtoward the maintenance, modification or transformation of the social


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (42)

order are guided by the personal desire for a society into which this socialvalue is woven. For example, where individualist personal valuespredominate, the social values of tolerance, of relations mediated bycontract and of justice as fairness under the law tend to be widelypersonalized. This is presumably because it is believed that, given theconcrete economic and legal institutions into which they are woven, thesesocial values are among the conditions that shape a society that embodiesthe desired personal values. The stability of a society depends on thewidespread personalizing of its predominant social values.

Furthermore, if a person’s aspirations are impeded because of theprevailing predominant social values, then it makes sense to personalizeother social values and to engage in political action in order to producesocial forms in which they are manifested. Thus (Chapter 8; Lacey 1997c)if one aspires to express the value of solidarity with the poor, one will seeksocial change that would produce a social order into which positivefreedom (the availability of conditions in which all have the possibility tolive significant lives of their own choosing) and the primacy of economicand social rights are woven, and so personalize these social values. In thisway, social values—either those predominantly manifested or those aspiredto—are included among personal values.

There are, of course, differences and disagreements in the realm ofvalues, the complexities of which cannot be cut through here. Theirexistence, however, highlights the question of which values one is to holdand what the relevance is of public discussion to settling the matter. Publicdiscussion cannot be expected to result in a consensus about which valuesone is to hold. Indeed, a certain difference in the values people hold may beessential within the texture of an environment that can sustainhuman freedom. Without some diversity and tension among values, peoplecould easily come to view the values they currently hold as the onlypossible values—a result that would seriously erode the scope of humanaspiration and the possibilities for human development. But what publicdiscussion can lead to is well-grounded knowledge about what are thesocial conditions needed for holding particular values. This is significantfor it provides a causal understanding of the formation and holding ofvalues that enable arguments to be made for the modification of existingsocial institutions and structures in some directions rather than others. Inthe light of this, I will explore a number of issues relevant to explaining thevalues that people do come to hold.


Human discourse is not merely “factual”; it is not limited to providingdescriptions and explanations of the way things are or have been. It is also


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (43)

future oriented and so contains valuative aspects. While explanation relatespresent states of affairs to past ones (according to causal laws or within anexplanatory narrative), evaluation tends to relate the present to desiredfuture possibilities or to their anticipated realizations. Evaluation serves, inpart, to set the course of our lives in the light of the constraints of actualrealities. The future is neither determined by the present nor is it theproduct of voluntarist action unconstrained by the present. Rather, it takesshape in part as present realities are modified, and sometimes transformed,through intentional action. Our beliefs and desires play a causal role inshaping the future, but under powerful constraints that are not themselvessubject to modification simply in the light of our present beliefs anddesires. Values are intelligible only within this context of constraint.

Morally salient phenomena of lived experience

The context of constraint also enables the reasonably precise definition of anumber of phenomena of frequent and repeated salience in everyone’s livedexperience. To a large degree, I hypothesize, people come to hold thevalues they hold in the course of responding to these phenomena, for thereare a limited range of possible responses, each providing coherence to acomplex (ensemble, set, cluster, scheme, outlook or perspective) of valuesthat are woven into a person’s life. These phenomena, of which I will listfour, concern various gaps between aspiration and realization.

Gap between intention and effective action

The first phenomenon is related to the gap, already referred to, between themanifestation and articulation of values. It is the gap between intention andeffective action, between desire and the outcomes of action. Frequently ouractions do not lead to what we intend and our desires are not fulfilledthrough the actions they engender. Our efforts (individually or socially) tobetter the world do not always succeed in bettering it or producing asituation that is experienced as more fulfilling; sometimes the efforts justfail, other times they produce unintended (and undesired) effects. This gapreveals limits to our expressive capacity, our power to shape our own lives,our self-understanding, our grasp of what we can expect from others andour understanding of the social and material conditions of our lives. Forexample, those who propose that a life consisting of a sequence of actionsbased on spontaneous desires will bring a source of happiness andcontentment often find that it brings instead a sense of degradation,emptiness, self-contempt and shame. Those who wish to do just “what theyfeel like” are often bewildered by their ineffectualness and they oftendiscover that they fail to develop the capacities they need later on to realizedesires that then take on importance for them. While first-order desires


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (44)

may, and often do, predate second-order desires (values) in a person’s life,the continual coherence of first-order desires depends on one’s developing(more or less articulately) second-order ones. There may be social orpsychological conditions in which some people are unable to developconstant second-order desires and so are unable to hold values (even tohold one’s self to be an object of value). Under such conditions, we mightexpect to find profound psychological pathology, and even littleappreciation of the value of life, with consequent recourse to indifference toothers or to gratuitous violence.

Gap between what we experience and what we sense canbe

The second phenomenon is another gap: between what we experience to bethe case and what we sense can be the case. We experience and observesuffering of various kinds and we sense that some of it can be mitigated,that there need not be so much suffering, that things can be “better,” thatthe salience of suffering can be reduced and that more fulfilling possibilitiescan be realized. The experience of suffering, as it were, provides impetus torank, in some sort of moral order, the possibilities that may be realized inthe future. It attunes us to a sense of what well-being might be and that sensemight be heightened by the observation of lives (and interaction with them)that seem to realize more fulfilling possibilities. Our experience is infusedthrough and through with moral content. We do not experience the worldmerely as a sequence of facts, but as amenable to change caused by actionsinformed by prior deliberation. We experience it as fulfilling or lacking, asgenerative of a sense of well-being and of suffering—sometimes one,sometimes the other; sometimes in one respect, other times in another;sometimes better, sometimes worse; sometimes improved by our actionsand social projects, sometimes worsened. Thus, it is fundamental to ourexperience that the world does not have to be the way it is actually, that itscurrent state has not realized all possibilities and so we can aspire to realizeother possibilities—better ones.

Different values embodied in different institutions

The third phenomenon is that each of us is placed (early in life largely byforce of circ*mstances; later to some degree by choice) in a variety ofinstitutions, each embodying a different complex of values. Some of theseembodied values can be seen as complementary, mutually contributingtoward a fulfilled life. Others “contradict” one another, setting upconflicting tendencies in the person. In the extreme case, one might beliving the greater part of one’s life in institutions that embody values inconflict with the most central personal values that one holds, and the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (45)

values articulated in those institutions may deny credibility to the valuesthat one is personally inclined to articulate.

Gap between values-as-articulated and values-as-manifested

Finally, within each of the institutions referred to above there is often a gapbetween values-as-articulated and values-as-manifested. Althoughinstitutions exist for the sake of the values they embody and articulate, inorder to maintain themselves they are often pushed to pursue extraneous(perhaps important) values. For example, the core values of the university(for example, the pursuit of the truth) may find themselves compromisedor overridden by the values of producing professionals to serve the currentpredominant order, which the university finds itself emphasizing in theservice of the ends of funding and recruitment, without which it could notcontinue to pursue its primary values (cf. the discussions of autonomy.Chapters 1, 4 and 10). Institutions thus simultaneously create theconditions for the manifestation of certain values and also establish limitsto their manifestation. The magnitude and severity of the tension betweenthe values that justify the existence of an institution and the valuesthat enable it actually to function (MacIntyre 1981) can be expected tovary from institution to institution and from society to society. But thetension will always be present in some degree and, as such, will underliethe gap between aspiration and manifestation of institutional values.

These four phenomena are among those that engulf our lives. We cannotavoid them, though they can impinge on our critical awareness more orless sharply. They cause disequilibrium in our lives—so much so that, to aconsiderable degree, we can conceive the unfolding of a life (whichmanifests values as distinct from simply desires) as the narrative of aperson’s attempts to strive toward a satisfying or at least a tolerableequilibrium. In modern times, the gap between desire and the outcome ofaction is especially disconcerting because it reflects limits on our personalfreedom. In order to reduce the gap and produce a measure of equilibrium,one can attempt—always without assurance of success because the causesof the gap (including lack of knowledge and lack of access to requiredmaterial and social conditions) might be insuperable—either to change theshape or the social conditions of one’s life, or both.

Paths toward equilibrium

I hypothesize that the paths toward equilibrium can, to a firstapproximation, be classified into the following five kinds and that theordered, coherent, unified complex of values that a person comes to adoptreflects the path followed.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (46)


One adjusts one’s goals to the way things are, the path of “realism” thataccepts (more or less consciously) that there will be no fundamental changein the predominant institutions that shape one’s life, that there are nopossibilities for the immediate future outside of the current predominantinstitutions, that the future is framed by those institutions. Accordingly,one chooses to participate—taking into account, where one can, one’sopportunities, one’s education, one’s family, one’s talents and interests,and one’s assessment of the viability of various institutions and the furtheropportunities they might open up—in those current institutions to whichone has access so as to bring about the least tension and the greatestequilibrium. One adjusts one’s goals largely to what is realizable withinthese institutions, leading a life into which are woven values that areembodied in one’s society. Various ways of life fall under the path ofadjustment, reflecting the variety of institutions present, classdifferences and even the existence of “fringe” niches in a society. While thepath of adjustment admits of variety, within it the range of acceptablevalues is limited by those embodied in current dominant institutions (Lacey1997c)and the fact that they are socially embodied becomes de facto areason for holding them or at least the ground that makes them immune tocriticism.

Adopting the path of adjustment can be more or less conscious. Becausethe values manifested within this stance are embodied in society, normallythe question of their legitimation does not arise, and if it does, the reigningsocietal articulations of value (ideology) quickly provide answers thatprima facie are compelling. Adopting this stance, therefore, needs littlepersonal reflection and, indeed, critical reflection is not a highly rated valuewithin it, at least not critical reflection upon social structures or the kind ofreflection that leads to self-consciousness within dominant practices.Critical reflection may make one become aware of the disvalue (forexample, oppression, discrimination, domination) that may also beembodied in the structures, and thus intensify the sense that there are fullerpossibilities waiting to be realized, and create the perception of an evenlarger gap between the value socially manifested and that articulated.

The path of adjustment enables, for some, the experience of a measure offulfillment reinforced by institutional articulation (ideology) that the valueswoven into adjusted lives are those that define a fulfilled life (Lacey1991a). The more stable and highly developed the structures are, the morethey provide space for large numbers of people to live adjusted lives and tohave their “realistically” limited desires satisfied. The stability of suchstructures reflects the fact that within them the actual desires of many arebeing satisfied, and may have the consequence that these desires appear tobe fundamental and universal, reflective of human nature (Schwartz 1986).


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (47)

The predominant economic and political institutions of any societyreinforce the path of adjustment for a significant number of people and“privilege” those who adopt it. Rarely, however, can it be adopted byeveryone, for in most if not all societies such privileged lives depend uponrelations of domination, where the possibilities for the dominated to gainfulfillment are severely limited.


One comes to resign oneself to the inevitability of the social and personalconditions of one’s life, that one’s desires are inefficacious, that one’saspirations are empty, that where there is change it happens outside of theoperation of one’s will. Then, desire becomes reduced to the desire tosurvive or perhaps to make life barely tolerable; and life becomes a reactionto outside forces. Here we find the phenomena of fatalism, lack of self-valueand internalized oppression, diminished intelligence, suppressedconsciousness and conscience, and nihilism. Resignation should perhaps beconsidered a “degenerate” (in the mathematical sense) path—the pathadopted when the causes of disequilibrium turn out effectively to beinsuperable, and so where there may be little unity and coherence within thevalue complexes that come to be held (and where, for many, holding valuesbecomes reduced to simply “having” them). The path does admit variety:among other possibilities it can generate gratuitous (voluntaristic) violence,the deep involvement in religious practices that transfer one’s aspirationsbeyond the world of history, dependence on alcohol and drugs, as well ascountless lives following the daily grind of survival.

There is, of course, no sharp dividing line between the paths ofadjustment and resignation. Interpretive methods are needed to assign a lifeto one or the other type, methods that need to recognize the remarkablehuman capacity to find or create niches in which a meaningful life is possible.Nevertheless, resignation is a dialectical counterpart of adjustment insocieties structured by dominative relations (whether the structures beeconomic, patriarchal, racial, or other: cf. Chapter 9). In such societies,stability requires that both paths be followed; indeed, since dominantideology serves to disguise the structuring relations or to make themimmune to critique, for most people only these two paths will concretely beavailable for adoption. Since the path of resignation is not conceived ashaving structural sources, ideology explains the adopting of it in terms ofpersonal “defects” (laziness, lack of intelligence, etc.) woven into one’s life—hence the ready willingness of the “privileged” to accept that the lives ofthe resigned (and, no doubt, others whom it cannot distinguish from theresigned) be “managed,” and even subjected to institutional violence. Suchviolence is not seen as reflecting structural domination, but as protecting thevalue that the society should be embodying and manifesting. This


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (48)

conception is reinforced by the perception of various (though few)individuals moving from places where the resigned abound into the ranks ofthe adjusted. Unaware of structural limits, it supposes that what isavailable to some is available to all.

Creative marginality

The resigned are also marginalized and they make little contribution evento defining the shape of the margins. The adjusted adopt for the most partthe value embodied in the current predominant institutions and so they,too, do not push the limits of the margins, but live within them. Anotherpath accepts that the fundamental structures of society will (can, ought)not be changed in the near future, so that they will continue to frame viablelives, but reject many of the values they embody regarding them asunworthy of human aspiration. For example, it may reject the consumerist,possessivist values common in our society as degenerate versions of theaspiration of freedom; it may also be aware of the dialectic betweenadjustment and resignation, and react with indignation and outrage to thesuffering and misery that it maintains are grounded in the prevailingstructures. The response of this path is to push beyond the margins, tocreate spaces for the (greater) manifestation of worthier values, and forlives into which these values are woven. A number of distinct versions of thispath can be distinguished: individual creativity, communal service andpreservation of an alternative tradition.

Individual creativity: One recognizes that, within the prevailingstructures, there are possibilities for creative expression (in art, music,science, etc.). One then pursues a talent and works on it, gaining relevantskills in order to generate something new—an object of value that isrecognized in one’s culture as such—that is distinctive, expressive of theself and that expands the realm of value that can be embodied within theculture. While the link between individuality and novelty makes this pathcompelling, it is not a solitary affair; it involves participation in sharedpractices, which often are institutionalized. At times the path of adjustmentshades into this path. At least in some domains, the value of producing andrespecting objects of value is highly embodied in most societies. Indeed, itis quite common for dominant ideology to “justify” current socialstructures with reference to the objects of value generated within thestructures, implying that the generation of such value is sufficient tolegitimize the material and social conditions necessary for its generation(Lacey 1991a). One may include in this path certain feats ofentrepreneurship or creative administration. And certain corrupt practices,unusual forms of the accumulation of property and conspicuousconsumption, may be considered as degenerate versions. Its defining markis that of individual creativity in pushing out the margins, and it is not


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (49)

uncommon for one who follows it to find it necessary to contest the valuesembodied in particular institutions (for example, universities, foundations,government departments, publication houses, art museums).

Communal service: One participates in or attempts to createcommunities manifesting values that run counter to those of themainstream, such as service to the needy and marginalized, and, throughsuch communities, one seeks out ways to be less dependent on the materialand social conditions that sustain the marginalization of some people. It isdifficult to find space for this path. Sometimes it is found within parts ofreligious institutions. It is also an important factor in the women’s andcivil rights movements. Reforms in predominant structures may beinstigated from this path, as its proponents act to make the values manifestin the dominant institutions more closely approximate their authoritativearticulations, opening the path of adjustment (at least) to more of theresigned. This path also, at times, generates remarkable lives that displaythe creative power of radical love—the “saints,” whose lives exhibit rarelyrealized potentials of human nature, whom we all admire even when we donot aspire to emulate them. This path may be open more to those from themarginalized sectors of society than to those from elsewhere.

Preservation of an alternative tradition: One participates in an institutionor movement for the sake of preserving an alternative tradition (religious,cultural, ethnic). This can involve creating new spaces, and sometimeseventually new structures, in governmental, economic, educational andother institutions. Much of the current activity subsumed by the label“multiculturalism” in educational institutions falls into this category. Thissub-path also admits of conservative varieties.

The emphasis in the first three paths is on individual change oradaptation in the light of structures that are perceived effectively as givenconditions of one’s life. Each of them can recognize the possibility ofstructural reforms, and viable structures may even need to include a placefor adjusted paths committed to the administration of reforms. Theremaining two paths place the emphasis on fundamental structural change.

The quest for power

This path reflects the desire to gain power (political or economic) in orderto adjust social structures to one’s (and, no doubt, to those one thinks thatothers “ought” to hold) intentions, interests and values; to use power totransform institutional structures so that one’s interests or perceivedobligations can be satisfied. There can be corporate, military and electoralvarieties of this path.

Within prevailing structures there are roles for the exercise of power.These roles fall under the category of adjustment. The present category, incontrast, involves the use of power for fundamental structural change.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (50)

Power may be used to conserve what is, to produce reform or to producerevolution. The lines between conservation and reform, and reform andrevolution are not always easy to define or to identify; and one who gainsan office of power may, when confronted with the realities of exercising it,move away from an intended use of power in another direction.

Transformation from below

One may hold that existing structures, even under reform, cannot provideconditions in which everyone can live lives into which are woven valuesthat are plausibly considered their own, expressive of their human selves.One may also hold that the quest for power, at best, will only bring aboutchanges in who occupies the privileged places in dominative structures (orperhaps replace old structures of domination with new but no lessdominating kinds). Holding these views, one might enter intoorganizations, practices and communities (modeled on Latin American“popular organizations”: Chapter 8) whose objectives are: (1) to enabletheir members, composed largely from the marginalized groups, tomanifest values that are their own, and to practice service and cooperationto this end; (2) to expand the compass of these organizations by creatingnew ones and cooperating with others so that more and more peoplebecome included in the process; (3) to work with sectors from mainstreaminstitutions in a spirit of reciprocity so as together to open up more spacefor fulfilling options for increasing numbers of people; (4) and in so doing,to form the institutions in which values like cooperation, participation andopenness to difference can be embodied; (5) eventually to constitute theinstitutional base of new social structures in which relations of dominationwould be diminished.

This path of transformation from below has similarities to the sub-pathof communal service: it begins with addressing the needs of themarginalized but, rather than service and charity, it emphasizes personalempowerment, solidarity and cooperation. Because it emphasizes thedialectic of personal and social change, it does not rest upon the agency ofpower. It is argued that power cannot bring about the desired changes, forpower cannot make people live lives into which are woven their ownvalues. This does not imply, however, that those adopting this path may notin actual fact sometimes become allied with groups that are using violentmeans to gain state power. The gaining of power by such groups is not thedesired change. But, where oppression and repression are intense,expedient alliances may be judged necessary to remove crucial obstacles toprogress along this path of transformation from below. Ambiguity willalways be present when the followers of this path, attempting to bringabout the fuller embodiment of the values it represents, interact withinstitutions of power. Sometimes (perhaps most of the time) the conditions


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (51)

of the expedient alliance will bring about a lapse into the quest for power,where the values associated with power supplant the initially motivatingcommunity values. Nevertheless, where the path of transformation frombelow is followed authentically, only the growth of the movementsin dialectical interaction with the formation of personal values can producethe desired transformation. It rests upon a step-by-step process of change,testing each step for viability as it proceeds, a process in which there isorganic unity between means and ends. It does not evaluate each step interms of whether or not it is a means to a systematically articulated socialobjective, for such evaluation is unresponsive to the personal/socialdialectic and open to having key roles being granted to power andviolence. On the contrary, each step is evaluated in terms of its bringingabout a fuller embodiment of the values articulated by the movements, andwhat the limits of that embodiment are; thus, of whether it represents inanticipation a society that embodies the desired values adequately andprovides a ground for proceeding with fuller exploratory steps out of whichthe concrete structures of a transformed society can emerge (Lacey 1997c).

I suggest that these five paths (as “ideal types”) are the ones that areopen to people when they experience the kind of disequilibrium describedat the beginning of this section. They are not pure paths. Up to a pointeveryone shares some aspects of all the paths, but for each person aparticular path eventually comes to the fore, reflecting who the person isand what are his or her most fundamental values—the value complex thatis largely constitutive of his or her identity I hypothesize further that thefirst two paths, adjustment and resignation, are the most common in thecontemporary world, and that to adopt any of the other paths one musthave substantial motivation, because pursuing these paths often introducesnew forms of disequilibrium and disorientation which, however, representattempts to discover and realize some of the human possibilities that havenot yet been realized, and to develop critical and creative consciousness inall of its dimensions.


According to my analysis the path that one adopts provides a unity to thecomplex of values that one holds. People have various reasons, which maybe more or less articulated, for adopting and persisting in their respectivepaths. I do not suggest that one chooses one’s path as a consequence of anisolatable deliberative process, as if one deliberates considering the reasonsfor and against each path and then adopts one. Rather one makes choicesabout such matters as educational objectives, friendships to cultivate, skillsto obtain, jobs or careers to pursue, places to live, commitments to family,as well as choices about countless matters of consumption andpossessions, etc.—choices that are made possible, enhanced and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (52)

constrained by such matters as family, class and religious background.From this multiplicity and complexity of choices emerge the contours ofone’s adopted path, and the ordered, unified, integrated, coherent complexof values that largely constitute one’s identity. The reasons for adoptingone’s path become apparent not antecedently, but as one attempts tocreate, articulate or discern unity in the values that are manifested in thevarious choices and commitments that one has made, and as one looksforward to a life that displays coherence, a life into which an ordered,unified, integrated, coherent complex of values is increasingly woven. Sucha complex of values is itself subject to evaluation in the light of a numberof criteria, which often become explicit when one attempts to articulate thelegitimacy of one’s adopted path in the face of challenges. These criteriaalso play an explanatory role, at least to the extent that recognizing thatone’s adopted complex of values fails to meet one or other of them canoccasion a life change.

I have spoken of people “holding” values and of the totality of valuesthey hold constituting a “value complex”; where one holds a value (v)8 ifone desires that the relevant object (ø) be characterized by v, and believesthat ø’s being characterized by v is partly constitutive of a “good” ø, andone is committed to narrow the gap between its manifestation andarticulation. I will say that one adopts a value complex if one can defendthe possibility, given the constraints of prevailing material and socialconditions, of each value (v) in it being more fully manifested constantlyand coherently in the relevant ø (self, society, etc.) and more fullyembodied in society, and if one can (to one’s own satisfaction) defend thebelief that the relevant ø’s having v is worthily desired—where the defensewill, at least in part, appeal to a view of human nature, a view of whatconstitutes human well-being and of what lies within human potential.Adopting a value complex, thus, has “presuppositions” that render itsitems integrated and coherent.

Value judgments

Adopting a value complex involves making value judgments, of whichthere are two fundamental kinds: “v is a value” (“v is worthily desired ofø”); and “v1 is subordinate to v2”. I will say—for convenience—that valuejudgments (as well as values) belong to a person’s value complex. Valuejudgments should be distinguished from judgments of the form: “v ismanifested in ø to such and such a degree,” and in general, any judgmentsabout “measuring” (evaluating, estimating) the degree of manifestation ofvalues. Often judgments of the latter type (in view of my account of thepaths toward equilibrium) will play important roles in the arguments thatunderlie sound value judgments, and being able to make them is acondition on being able to hold values at all.9


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (53)

Making value judgments involves responsiveness to various criteria, ofwhich I will highlight two that are particularly pertinent for subsequentdiscussions of neutrality (Chapter 4).

• The possibility criterion: The genuine possibility—given the actualconstraints of prevailing material and social conditions—of the complexof values being woven consistently, constantly and coherently into aconcrete personal life (or other relevant ø), and so of being embodied insociety.

• The human nature criterion: The availability of an articulated view ofhuman nature, with some empirical support, that renders intelligible theclaim that holding the complex of values shapes fulfilling lives.

• In practice, these two criteria are deployed in conjunction with anumber of other criteria, several of which I list without elaboration anddefense.

• The formal consistency of the value complex.• The continuity of the value complex, perhaps under considerable

reinterpretation, (a) with some of the values that one has “inherited,”that one shares with some others and that appear “obvious” (forexample, rejection of murder) and also (b) with values that are actuallywoven into concrete lives (or compelling literary characters) that onerecognizes as fulfilled.

• The inclusion in the value complex of those values that are constitutiveof the intelligibility of valuative discourse in general (for example,respect for the participants in the discourse, elevation of dialogue overpower, truthfulness).

• The universalizability of the core values of the complex—or rather, thatthe material and social conditions of their embodiment are compatiblewith all participants in the discourse being able to conduct lives intowhich these values (or, more generally, those expressive of their ownindividual identities) are woven.

I have left it open whom one considers the participants in valuativediscourse to be, for that—perhaps—reflects one’s own values, rather than ageneral criterion. If everyone (in principle) is included among theparticipants then, as the final two criteria are met, one gains a greatercapacity to understand well the actions of other persons and socialmovements, even those that manifest values quite different from one’sown. Gaining such a capacity is a core value of the fifth path—the pathof “transformation from below” (Lacey 1997c). So, too, is gaining a clearconsciousness of the conditions necessary for one’s values to be woven intoone’s life.

Whatever one may think of the other proposed criteria, the possibilityand human nature criteria are of utmost importance for evaluating the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (54)

value complex that a person holds, although they do not functionindependently of attempts to gain “reflective equilibrium” (to borrowRawls’ term) with other criteria. The view of human nature underlies thecoherence and intelligibility of the complex; the genuine possibility of itsbeing held is necessary for its social salience. Frequently, these two criterialie at the heart of valuative disputes functioning in concert, for example,arguments about the possibility or impossibility of social transformationoften appeal to a view of human nature and the possibilities encapsulatedin its articulation. Moreover, if one articulates values not embodied in thecurrent social order, and it is not possible to generate social transformationof the desired kind, then that is a reason to reconsider one’s aspirations. Ifsocial transformation is possible, then that a rival value complex isembodied in the current order ceases to be a compelling argument for one’saspirations being limited by it (Lacey 1991a, 1997c). If arguments can bemounted simultaneously that social transformation is possible and thatthere is a supported view of human nature that suggests more fulfillingpossibilities in the proposed new order, then the value complex embodiedin the current order may come into crisis and, for want of coherence andsalience, the extent of its embodiment may decline.

If valuation is to guide lives and not become reduced to mere idealisticcriticism, determinations of what is possible are always very important.Formal consistency of a value complex is not sufficient for suchdeterminations. For example, community values like cooperation andsharing are formally consistent with the primacy of property rights, butarguably the material and social conditions required for the steadfastmanifestation of one precludes that of the other. More than consistency isinvolved in assessing the possibility of coherently holding a value complex.When a value complex is already highly embodied in a society, there is nofurther question about its possibility. What is actual is possible. Aparticularly difficult question confronts us when the fifth path, the path ofsocial transformation, is under consideration (Chapter 8). Unless the kindof social transformation proposed is possible, the path of socialtransformation reduces to that of communal service. In our society, thiskind of social transformation is widely believed to be impossible, so thatmost lives can be charted on the first three paths. It is widely believed thatany viable social possibilities (for the foreseeable future) will be framed bythe institutions of private property, the free market and formal, electoraldemocracy (Lacey 1997c). Why? There are various proposed answers: (a)because those institutions are arguably better (more fulfilling, moreencouraging of human freedom, more conducive to social justice), andrecognized as such, than any available and most imaginable alternatives;(b) because of their inertia and ever growing momentum; (c) because of thevirtual hegemony of power associated with them and the expectancy thatthis power will be used to maintain their hegemony; (d) because of the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (55)

belief that they effectively embody human nature, which is understood tounderlie individualist and egoist values. If (a) and (d) are soundly grounded,then the use of power, referred to in (c), will often be legitimate—but nototherwise. Much hinges, then, on an assessment of what human nature isand an assessment of what makes it that way.

It accords with the two criteria under discussion that beliefs about whatis possible and about human nature may properly influence the personalvalues that one holds and desires to be embodied in society, and the socialvalues that one attempts to personalize. Indeed, the applicability of thecriteria depends upon such beliefs. As already indicated, beliefs about whatis possible and about human nature are deeply intertwined. Our beliefsabout the values that can be woven into lives and articulated authenticallydepend (in part) on our beliefs about human nature; and our beliefs abouthuman nature draw heavily upon the values that we observe to be actuallywoven into lives. Broadly speaking, these beliefs are open to empiricalscrutiny (and, because of this, the proposed component of neutrality, thatestablished scientific theories can have no consequences for the values oneholds, cannot be sustained: Chapter 4). What kind of empirical inquiry couldthrow light on them, and can its results accord with impartiality?

Answering this question is made exceptionally difficult by the fact thatthe possibility of genuinely holding values depends not only upon whathuman nature is, but also upon the values that are embodied in actualsocietal institutions and upon the power (authority) relations that structurethese institutions. Prevailing power relations may actually prevent certainpossibilities allowed by human nature from being realized, especially inthose cases where their being realized rests upon social conditions that areincompatible with the prevailing conditions. Moreover, the human desiresthat are present at a given time might reflect not the full potential of humannature, but the possibilities reinforced in institutions. If this is so, then anempirical charting of what is actually manifested (and empirical inquirymust be based in observation of the actual) cannot result in acomprehensive account of human potential, for there may be hithertounrealized possibilities.

What kind of empirical (psychological, sociological) inquiry can prop-erly inform beliefs about the possibility (not necessarily high probability)or impossibility of realizing value complexes, such as those implicit in thefifth path, that have not hitherto been realized to any significant extent?What is the appropriate methodology? What are the appropriate theoreticalconstraints? What are the appropriate empirical phenomena to select toinvestigate? Answers to these questions are usually intertwined and reflectthe practices (and the values they express) that the research informs (Lacey1990, 1997c). If most research informs practices within our first threepaths, then the question about alternative possibilities is not likely to arise(expect at the margins). If today these paths are framed by the institutions


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (56)

of private property, the free market, etc. and those institutions tend toembody egoist values, made coherent and intelligible by an individualistview of human nature (Lacey 1997c), then we would expect to findpsychological and social theory today constrained by the individualist viewof human nature and its evidential base would draw principally from thecharacteristic phenomena of these institutions, where one observes thepredominance of egoist values (Lacey and Schwartz 1996). The prospectsfor accord with the impartiality of such investigations will be raised inChapters 8 and 10.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (57)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (58)

3Cognitive values

In the grammar of “values,” the most fundamental expression has the form:“X values that ø be characterized by v,” where “X” designates a personand where different kinds of values correspond to different instantiationsof ø (Chapter 2). When ø is instantiated by a person’s beliefs or by anaccepted scientific theory, the “v” designates cognitive values.


Cognitive values are characteristics (criteria) of “good” (rationallyacceptable, desirably held) beliefs and “good” (soundly accepted) theories.I will consider them first, in the most general way, general as desiderata ofbeliefs. Beliefs (about, for example, human nature and what it is possible toachieve) are presupposed in adopting a value complex (Chapter 2). Theyare also involved, together with desires, in the generation of action: oneacts because one desires a certain outcome and believes that the action willfurther the realization of the outcome. One’s success in adopting valuesand weaving them into one’s life, therefore, rests in part upon holding theright kinds of beliefs; not just “significant” or “relevant” beliefs, but thosethat have been gained, sorted and ranked in the light of the ideal of truth.

Not all beliefs are true, of course, just as not all desires are good. But itwould be contradictory to affirm simultaneously: “p is false and I believethat p.” Truth does not display itself as a manifest property of the beliefsone affirms to be true, just as goodness does not display itself as a manifestproperty of what one affirms to be good. One judges a belief to be true invirtue of whether one believes it to possess certain properties and relationswith other beliefs (however they may be elaborated in detail): to be wellgrounded in evidence, to follow by entailment from other true beliefs, tosatisfy rational canons, to bear the appropriate inductive and deductiverelations with other beliefs, to have a particular kind of causal history,to inform action that is successful consistently—in other words in virtue ofan appraisal of its cognitive value. The properties and relations of beliefsthat one identifies as serving this function are (to a first approximation)one’s cognitive values. Clearly most of them may be manifested in one’s

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (59)

beliefs to a greater or lesser degree; the greater the degree to which they aremanifested, then the greater is the “rational acceptability” of the belief.Judgments of degree of rational acceptability are framed by the ideal oftruth, but we have no indicator of truth other than rational acceptability.

Grammatically and logically, cognitive values have much in commonwith other values. Holding a cognitive value v, involves second-orderpropositional attitudes: a belief about beliefs, a second-order belief thatbeing characterized by v contributes to making a (first-order) beliefrationally acceptable; and a desire that ceteris paribus one’s beliefs becharacterized by v. Thus, v represents an attribute that one desires one’sbeliefs to possess and one believes that they ought to possess, in part forthe sake of being able to live a life that expresses one’s fundamental values.Then, holding a set of cognitive values implies a commitment to evaluatingthe beliefs that inform one’s actions to the extent appropriate and possiblein the light of assessments of their manifestations of the items of this set—while recognizing that there will remain inevitable ambiguities, gaps andlapses, and that the exigencies of action often curtail deliberation. Thecognitive values one adopts form an essential part of one’s total valuecomplex. For some, it may be the most fundamental or highly ranked part.Putnam, for example, maintains that the cognitive values constitute part ofour idea of rationality and of total human flourishing, so that aiming toweave them integrally into one’s life represents a value that may (ought to)be manifested and articulated quite apart from the immediate contexts ofpractical action (1981:134–6; 1990:139–41).


Beliefs are propositional attitudes that, together with desires, intentions,having goals and the like, may play causal roles in generating actions. Inthe grammar of “belief,” the fundamental expression is: “X (an agent)believes that p.” A belief is always a belief of an agent or shared amongagents. It is true if, and only if, its propositional content is true. Thus, thecritical evaluation of the belief that p is identical with the cognitive(rational) appraisal of p.

The causal role of beliefs is represented not in law-like schemata but in“practical syllogisms,” in which one’s actions are represented as following(rationally) from having certain goals (desires) and beliefs (Lacey 1996;Lacey and Schwartz 1986; and especially Donagan 1989). Atypical practical syllogism includes one’s beliefs about the character of thesituation one is in, about the means to the desired outcome in thatsituation, about the side effects (desired and undesired, intended andunintended) of the action, and about the possibility of realization of thedesired outcome. I will say that a belief informs an action if it is among thoseof the practical syllogism that serves to explain the action and that it


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (60)

informs one’s activity if it regularly informs one’s actions and/or it isamong the presuppositions that play a role in the articulation andjustification of one’s value complex. Here, I include under “belief” any ofthe cognitive attitudes that agents enunciate as beliefs, or that accompanydesires (as represented in practical syllogisms) in the causation of theiractions.

Agents beliefs themselves have causes—more accurately, that an agentbelieves that p has causes—which may, or may not, include their assessmentsof evidence and other explicitly cognitive or rational factors. The causes ofan agent’s beliefs may, or may not, be reasons to hold them. They mayinclude a wide variety of social, psychological and experiential factors; forexample, one’s immediate or reflected-upon experience, the testimony of“authorities,” the inheritance of a tradition and efforts to make intelligiblea value complex to which one is attracted.1 These help to explain the factthat often one’s actions are informed by false beliefs.

Attention to the causal role of beliefs is insufficient to evaluate them,that is, to appraise whether or not they manifest adequately the relevantcognitive values. To make this clear, I will distinguish: one has the beliefthat p if that belief informs one’s actions (including verbal acts ofenunciating beliefs). One holds the belief that p if one reflectively endorsesthat p and it informs one’s activity; and one can defend p from criticism toone’s own satisfaction by pointing to its place in a network of evidentialand logical relations, articulated by one’s adopted cognitive values. Oneholds that the belief that p is consolidated if one judges that p belongs tothe class of rationally acceptable beliefs, those that (methodologically)require no further investigation, like theories of certain domains ofphenomena (Chapter 4) which manifest all of the appropriate cognitivevalues to a very high degree. An agent may have but ceteris paribus nothold both the beliefs that p and that ¬p. Holding, rather than just having abelief, implies that one’s cognitive activities of assessment of evidence andargument are among the causes of (maintenance of) the belief. That doesnot imply that psychological and social causes are absent or of low salience.Social causes, for example, may explain the limited range of evidence onehas considered. Holding a belief requires responding to the criticisms oneencounters to one’s own satisfaction; holding that a belief is consolidatedrequires a fuller articulation of the evidential and logical network in whichthe belief is inserted, and follows from having actively sought out criticismsamong the relevant community of inquiry and having arrived at a virtualconsensus with its members concerning both the belief and the appropriatecognitive values. Beliefs that are held to be consolidated are often referredto as “knowledge.”

There should be, it seems, an identity between the beliefs one has andthose one holds; or at least, it seems, if one holds the belief that p then itwould be irrational to act in ways that were informed by ¬p. I will call this


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (61)

an “informal ideal of rationality.” But it is not, in all situations, part of theideal that there be an identity between the beliefs one holds and those oneholds to be consolidated, because in many of the contexts in which wemust act beliefs held to be consolidated are not available, and becausegenuinely creative activity and activity driven by our deepest hopes andmost fundamental values, and that which aims to realize hithertounrealized possibilities (whose realizability depends causally uponcommitted action: Lacey 1997c) will necessarily be informed by beliefswhich do not manifest all of the cognitive values to high degrees.Presumably, however, when one holds certain beliefs to be consolidated,and they are relevant given the context and immediate objectives of action,one’s actions should be informed by them. The ideal includes ceterisparibus the desirability of expanding the stock of beliefs that one holds tobe consolidated. It is not always rational to act informed by the bestavailable beliefs that are held to be consolidated; our objectives should beshaped by our values without curtailing them to be informed by (thoughthey should be consistent with) available scientific knowledge. It followsthat the evaluation of beliefs, that are pertinent to planning actions, isresponsive not only to the ideal of truth, but also to relevance orsignificance; that is, whether it has content appropriate to inform actionsaiming at the proposed goal. But considerations of significance do notrender the ideal of truth irrelevant, for alone it is insufficient to informsuccessful action (Chapter 9), and the judgment of insignificance does notimply that of falsity. If available scientific knowledge is not relevantlyapplicable for the pursuit of one’s objectives, however, there is often goodreason to attempt to gain well-founded, empirical knowledge that wouldbecome so applicable (Chapters 8 and 9).

Sometimes actions are partly caused by false beliefs; but if one acts basedon the belief that p, and then one discovers that p is false, one says, “Ibelieved that p was true but I was in error.” One does not tend to lapseinto subjectivism here, and say “p is (was) true for me.” Similarly wherepeople hold opposed beliefs which the available evidence does notadjudicate decisively, we do not relativize truth to the persons. Rather wesay that they have different opinions concerning whether theproposition affirmed in the belief is true or false (even if the significancecriterion explains one person holding the belief, but not another). In thisrespect the grammar of beliefs diverges from that of desires. That X desiresthat p, but Y desires that ¬p need not imply conflict or contradiction; thatX believes that p, and Y believes that ¬p, does.2 This is reflected inexpecting that the cognitive values one comes to hold should make arational claim on everyone.

Often it is the falsity of the beliefs that inform action, not the agent’sbelief itself, that explains some important feature of the context of action.Having false beliefs about means to desired outcomes can lead to frustrated


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (62)

desires; having them about side-effects can underlie destructiveconsequences of one’s actions; and having them about the possibilities ofoutcomes and the presuppositions of practices can be responsible for futileefforts, for desirable projects not being entertained or supported (oftenentrenching conditions of widespread suffering: Chapter 8) and forassuming self-serving objectives. These familiar kinds of facts reinforce theclose connection between the cognitive values and other values. We cannotunderstand our own and other people’s actions, when they succeed and cansucceed in producing their desired effects, and when they contribute andcan contribute to the expression of our values, unless we have effectivemethods for evaluating beliefs. Without such effective methods, thecommitment to a life into which one’s values are woven is weakened.

Holding, not just having, beliefs is essential for being able to weave one’svalues into one’s life. One holds only those beliefs that have been evaluated,and thus put into a coherent network of beliefs, articulated throughreference to one’s cognitive values. Beliefs may be held with varyingdegrees of confidence, often marked in ordinary discourse by using termslike “know,” “probably,” “sure,” “hunch,” “opinion,” “think,”“speculate,” “conjecture” and “hope”; and reflected in the distinctionbetween holding beliefs and holding beliefs to be consolidated. Held beliefsare a subset of the beliefs one has. The former generally arise from thelatter or from consideration of the beliefs of others that one has beencaused to be aware of. The strengthening of the beliefs one holds oftenhappens in the course of rejecting conflicting views. Evaluating the beliefsone brings to the planning of action, therefore, ought not abstract fromconsiderations of their causation and the causation of other candidates forbeliefs that have been rejected. In particular, it is important to raise thequestion of whether one has considered a suitable range of candidates forbelief.

Beliefs, like values, are present in various modes: they are among thecauses of action; they may be articulated in words; in virtue of which theycan become objects of evaluation and become held (or rejected).Articulation is a necessary condition for evaluation (whether responsiveto the ideal of truth or to significance); it enables a belief to be consideredin inquiry largely in abstraction from its causal roles. Rational agents aimto minimize the gap that exists between the beliefs they hold and those thatinform their actions, so that they act to the extent possible only upon heldbeliefs. The interplay between beliefs as causal factors in action and beliefsas articulated and held is characteristic of the nature of beliefs (Lacey1996). In order to identify the beliefs that people have, we must attendboth to the beliefs that we discern (on interpretive inquiry) to be the causesof their actions, and to those they verbally enunciate and present reasonsfor holding—for there may be unarticulated beliefs, and a person’s ownenunciation is insufficient to identify his or her beliefs. A clear


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (63)

identification is possible only when there are no gaps between the two; andin the case of held beliefs only when also their place in a network ofbeliefs, linked by logical and evidential relations or articulated by theiradopted cognitive values, is made clear. Of course, one may hold beliefsand properly affirm that they are well supported, even though they play norole in one’s actions (other than perhaps in verbal acts). These will bebeliefs gained in the course of elaborating the network of beliefs that oneholds. Theory extends beyond practice and one’s curiosity extends beyondthe immediacies of one’s experiences and actions.

Modes of cognitive values

As with values in general, and with beliefs, a person’s cognitive valuesfunction in several modes.

Cognitive values as manifested in patterns of beliefs

Cognitive values are manifested in the networks of first-order beliefs thatone holds insofar as the patterns among one’s held beliefs (and the kinds ofbeliefs that one considers and rejects) are intelligible only in virtue of(perhaps implicit) beliefs about such matters as the links among acceptablebeliefs, which beliefs constitute evidence for other beliefs, what theevidential relations are and which causal sources readily engenderacceptable beliefs. For example, one manifests the cognitive value“predictive power” if one holds various beliefs, because from them reliablepredictions have been made concerning other items that are believedbecause they were gained in the course of direct observation. Clearly thiscognitive value is manifested to a greater degree as the range of vindicatedpredictions in one’s belief network is greater, and where beliefs becomeheld because of the relations among beliefs that enable successfulpredictions to be made. In this respect, predictive power is typical ofmany cognitive values; it may be manifested more or less. Identifying thecognitive values that are manifested in one’s beliefs, like identifying thevalues manifested in one’s behavior, can involve complicated interpretiveactivity. To the extent that a set of cognitive values is manifested in one’sbeliefs to a high degree constantly, consistently, recurrently and more sowith the passage of time, they are woven into one’s life.

Cognitive values as partly constitutive of belief-gainingpractices

One comes to hold beliefs in the course of engagement in “belief-gainingpractices,” the objective of which is the generation, selection, evaluationand consolidation of acceptable beliefs. Cognitive values are expressed in


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (64)

and partly constitutive of such practices; one engages in them because theylead to holding the desired kinds of beliefs, those which manifest thecognitive values.

Belief-gaining practices may involve the use of observational, inferential,conversational, argumentative, listening, scholarly, interpretive, historical,linguistic, imaginative, mathematical and other skills and procedures. Theyinclude both causal (generating candidates for belief) and evaluativedimensions. They may not always be identified by those who engage inthem as belief-gaining, and generally they may not be separable in people’slives from the other practices in which they engage. They may simply be anintegral part of other practices; one may engage in them eitherspontaneously or because they provide the relevant kind of beliefs neededto inform action within, these practices. When a belief-gaining practice isexplicitly identified and institutionalized (as in the sciences), engaging in itinvolves adopting the cognitive values that are partly constitutive of it andsharing them with other practitioners. Then, in the context of this practice,if one is authentically engaged in it, the cognitive values manifested inone’s beliefs will be identical with those that are partly constitutive of thepractice.

Articulation of cognitive values

One can also articulate one’s cognitive values. Thus, they can be laid outfor public scrutiny. While it is a commonplace to articulate beliefs, usuallythe articulation of cognitive values is an indication of engaging inphilosophical reflection. Nevertheless, categories used to articulatecognitive values do play a more or less rigorous role in daily discoursewhen one uses, for example, “evidence,” “probable,” “predict,”“consistent,” “certain,” “explain” and other terms that come into playwhen asked why one holds certain beliefs and when challenging those ofothers. There seems to be considerable overlap in the cognitive values thatpeople have. I conjecture that there are few who do not have the followingunder some interpretation or other: direct origin in one’s own or atrustworthy informant’s perceptual experience; explanatory, anticipatoryand predictive power among beliefs originating in experience; inductivederivability—all, of course, prima facie and to be brought into anappropriate reflective balance. Without some such shared cognitive values,under shared interpretations, communication would be impossible, andwhere it cannot be counted on there is defective communication (Lacey1991a).


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (65)

Disagreements about cognitive values

Often there is, alongside considerable agreement, disagreement about whatshould be included in the list of cognitive values, about their relativerankings, and about how adequately some of them are manifested incertain bodies of beliefs or scientific theories. Such disagreements can becompatible with reasoned dialogue and arguments about, for example,what cognitive values ought to be held and whether one ought to deploythe same set of cognitive values regardless of the particular practice oractivity that one wants one’s beliefs to inform.

Disagreements about cognitive values obviously will be implicated inconflicts about the beliefs that people come to hold. Since there are suchdisagreements, it follows that people—even after careful deliberation—canend up with conflicting beliefs. (I do not imply that agreement aboutcognitive values ensures agreements about beliefs.) When this happens, it issimplistic to urge people to solve their differences by more careful scrutinyof the evidence. It does not follow that the disagreements indicate that atleast one of the parties’ beliefs must rest upon such things as blindauthority bias, making premature judgments, or uncritical reliance ontradition; for among the interacting factors that causally account forconflicting beliefs being held are: 1) the different causal sources (social,psychological, experiential) of the beliefs; 2) their bearers holding differentsets of cognitive values; 3) and related to this, their engagement in differentbelief-gaining practices.

The fact of such disagreements, and their persistence, does not signifythat truth ought to be relativized to the believer. Putnam argues thatholding different sets of cognitive values should be regarded as signifyingdisagreements and not simply differences, for it can lead to holdingcontradictory beliefs (1981, 1987, 1990). It is difficult to get away fromthe sense that there is a correct set of cognitive values that one ought to aspireto identify.4 The aspiration need not presuppose that the creation ofnew belief-gaining practices will not lead to surpassing, or radicallyreinterpreting, currently available lists of cognitive values, and it iscompatible with there being a dialectical interaction between holdingcognitive values and historically located belief-gaining practices (Bernstein1983; Laudan 1984). Unlike with some personal values, where holdingdifferent values represents only difference and not disagreement, sincedifferent people properly aspire to lives into which different personalvalues are woven, differences in the cognitive values that are held alwaysrepresent disagreements. One’s personal identity is not linked with havingone’s “personal” set of cognitive values. My living a life of integrity andauthentic identity seems to require that I hold not “my” cognitive values,but the right ones. The cognitive values one holds, it seems, are right or


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (66)

wrong; and so, throughout this book, I often refer to the cognitive valueswithout specifying who holds them.

Disagreements about which cognitive values to hold are very common,even when discussing the sciences, whose practices leading to theacceptance of theories are often considered to be the exemplary belief-gaining practices. To illustrate, consider the following to be the list ofcognitive values pertinent to accepting theories: accuracy, consistency,predictive and explanatory scope, simplicity and fruitfulness in generatingresearch puzzles (Kuhn 1977). Some have argued that the list ought toinclude instrumental efficacy (Putnam 1981; or “prediction and control”—behaviorist psychology), high degree of falsifiability (Popper 1959),capability to explain through a narrative what is sound and unsound inhistorically preceding theories (MacIntyre 1977), inductive derivability andno use of “hypotheses” (Newton: Principia), or certainty of fundamentalposits (Aristotle, Descartes). Others argue that it ought not includesimplicity or explanatory scope (van Fraassen 1980). These disagreementsare open to argument; they are disagreements, not simply differences intaste. For example, one might reject instrumental efficacy maintaining thatit is a social value not a cognitive value (McMullin 1983). Certainty,despite its obvious appeal for ancient and early modern science, has beenunanimously dropped from contemporary lists because it is clear thatscientific practices cannot produce theories that manifest it; and it has beenargued that the explanatory scope is a value derived from pragmatic ratherthan epistemic interests (van Fraassen 1980). These illustrate the kinds ofargument that can be mounted when there is disagreement (details followin Chapter 5).

In addition to disagreements about items on the list, there can bereasonable controversy about rankings (Kuhn 1977; McMullin 1983;1996). For example, is explanatory scope more important than simplicity,or accuracy than fruitfulness? There are also disputes abouthow adequately a particular cognitive value is manifested in a theory. Forexample, was Copernican theory sufficiently fruitful in view of its weakconsistency with the physical theory of its time, or does Skinner’s theory ofbehavior have sufficient predictive power in view of its non-falsifiability(Lacey 1974)? These disputes are open to rational dialogue and thedeployment of interpretive skills, though eventually one’s answers restupon practical judgments. In light of these phenomena, it is clear thatagreement on a list of cognitive values does not ensure agreement injudgments about acceptable theories. Rationality does not guaranteeagreement. This only reinforces the importance of articulating one’scognitive values.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (67)

Embodiment of cognitive values in social institutions

In all the preceding respects, the grammar of cognitive values parallels thatof values in general, and the parallel continues: cognitive values can also beembodied in social institutions and in society as a whole. An institutionembodies a set of cognitive values to a high degree when it highlights andprovides support for those belief-gaining practices that are partlyconstituted by the set. Scientific institutions, for example, embody somecomplex of cognitive values containing items like those on the list (p. 53),no doubt with some variation of the items and their interpretations frominstitution to institution. The variation is within quite narrow limits. Noneof these institutions embody such once held cognitive values as certainty,and consistency with biblical revelation or with dialectical materialism, orproposals such as intelligibility to the general public or dialectical linkswith the traditional local knowledge of cultures in the Third World.5 Sincebelief-gaining practices require material and social conditions for theirpursuit, and since such conditions typically are made available ininstitutions, the widespread embodiment of a set of cognitive values will bea significant factor in reducing potential disagreements about them, thoughnot necessarily about the beliefs that are held. It can also be a source ofambiguity about the identification of cognitive values, because institutions,even those that exist for the sake of embodying a set of cognitive values(for example, if truth is their motivating value), also embody otherpersonal and moral values and manifest certain social values. Thus, whereembodiment is an important factor in there being agreement aboutcognitive values, it may remain open to investigation to what extent theagreement is to be attributed to the role of these other values (and toconsiderations of “significance”), rather than to the outcome of “purelycognitive” dialogue and argument.


I have introduced cognitive values as criteria for holding beliefs in generalto be consolidated, and then—in order to evoke a sense of continuitybetween ordinary attitudes of belief and certain stances taken towardsscientific theories—taken them also as criteria for accepting scientifictheories. But, the extension to scientific theories needs clarification,elaboration and qualification. It might be thought at the outset, however,that granting cognitive values this role in connection with scientific theoriesis simply to reject outright the view that science is value free, for doing sohas values (cognitive values) in play essentially in making fundamentalscientific judgments. But the idea of science as value free is concernedprimarily with the content of scientific theories and the characteristics and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (68)

consequences of soundly accepted theories. Thus it depends upon scientifictheories being able to be properly evaluated, to be appraised—for theirvalue as scientific theories—according to the proper criteria (Scriven 1974).Some scientific theories are better than others and we accept or reject themon the basis of such appraisals; whether (depending on one’sepistemological viewpoint) one expresses the “goodness” or the nature ofthe cognitive value of a theory in the terminology of “confirmation,”“probability,” “corroboration,” “verification,” “evidential warrant,”“verisimilitude” or whatever. The language of value pervades the cognitiveattitudes involved in choosing theories. Indeed, I have said that it is a valuethat science be value free. No irony or paradox is intended. The valuedesired to be reflected in sound theory choice is not moral, personal, socialor aesthetic value, the sort of value from which it is said that science isfree. It is cognitive value. The idea of science as value free requires thatcognitive and other forms of value be distinct in the sense that answers to“What are the criteria of a theory manifesting high cognitive value?” notdepend on answers to “What constitutes a good human life or a goodhuman society?”—that cognitive value be identifiable separately fromsignificance.


In the preceding section I pointed out that there is dispute about whichitems should be included in the list of cognitive values. There is also a morefundamental dispute about the character of the criteria of cognitive valuefor assessing scientific theories. Most agree that the criteria involve certainrelations obtaining between theories and available empirical data (andamong theories themselves), so that a theory is acceptable to some degree ifthe appropriate relations obtain. The dispute is about the character of therelations, especially those between theories and data. It has been widelyheld that (under idealization) the relevant relationships obtain if the relataare connected in a way that can be established by the application of a finiteset of formal rules, like the rules that (under idealization) underliemathematical proof; or, if the relationship’s obtaining is a matter ofdegree, that the degree can be calculated by the application of a set ofrules. High cognitive value accrues to those theories that can be representedas (or whose high degree of confirmation can be calculated from) theoutcomes of (appropriate) formal rule-governed operations. Then, whichscientific theories are soundly accepted rests solely on the data, otheravailable theories and the rules. The criteria of cognitive value, appropriatefor assessing theory choices, can be represented, so to speak, as governedby formal rules. The intersubjectivity of the data combines with the formalcharacter of the rules to ensure that there can be no place for any kind ofvalues in soundly made theory choices.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (69)

Among logical empiricists, the rule-governed character of the criteria ofscientific value was often taken as the mark of the objectivity of scientificevaluations. For them, the rules stood in sharp contrast to the criteria of(for example) social value, which could be deployed only in the light of thejudgment—“subjective” judgment—of a person. Rules, unlike judgments,lead to determinate, unambiguous, non-arbitrary, shared and obligatoryoutcomes of scientific evaluation. “Objectivity,” in this usage, is linkedwith precision, lack of ambiguity, being able to be measured or calculated,determinateness, and output independent of human variability; and“judgment” with such usages as “it’s only a value judgment,” and “it’s justa matter of judgment, yours against mine,” where judgments are held to besubject to endless contestation. “Judgment” can also be used in a morerobust sense, as when we refer to the judgment that a theory is the bestavailable one. In this latter sense, according to logical empiricism, soundjudgments are the outcomes of the interplay of data and rules, often ofcalculations of the degrees of confirmation (or inductive support) oftheories; so that properly made “scientific” judgments differ in characterfrom “mere” value judgments. In science, it seemed clear to them, highconfirmation was the prioritized objective; and the degree of confirmationof a theory was settled by the deployment of rules. Things, however, werenot actually this way: not all, for example, accepted that high confirmationwas the prioritized objective (Popper 1959); and appeal to rules in factsettled little—in part because there was little agreement about what therules are—or even about whether they are inductive, deductive,hypothetico-deductive, formalizable within the calculus of probabilityandin part because confirmation theory remained underdeveloped. This didnot seem to matter: objectivity was linked with rules; subjectivity withvalues. The contrast, objective/subjective (rational/non-rational) furtherreinforced that values should be kept out of deliberations about theorychoice and theory acceptance. The idea of impartiality has been stronglyinfluenced by the appeal of rule-governed accounts of cognitive value.

Cognitive values

The idea of impartiality, however, need not include that the relevantrelationship between theory and data be reflected in rules, but instead canbe specified in terms of the appropriate cognitive values being adequatelymanifested in theories (Chapter 1; see the theses of impartiality stated inChapters 4 and 10).7 I neither have a decisive argument that it is impossibleto produce rule-governed accounts of the grounds for accepting scientifictheories, nor claim that rules play no role at all in concert with thecognitive values. The appeal of rules is obvious enough (and reinforced bycognitive science approaches to cognitive processes), so that on-goingefforts, including those that deploy Bayes’ Theorem, to overcome the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (70)

current difficulties facing rule-governed accounts have merit (Salmon,1983). Nevertheless, the efforts to produce rule-governed accounts seem tome to have remained mired in intractable controversy and stasis; thealternative approach fits well with certain phenomena of the history ofscience: that disagreement is commonplace in the scientific community, theunderdetermination of theory by data, versions of the Duhem-Quine thesis;it fits better with certain philosophical accounts of the nature of scientifictheories, for example, the semantic model account; it is more readilyapplicable in contexts where the relevant research is not exclusively orfundamentally quantitative; and it does not share the philosophicalbackground which denies that values can be the objects of critical, rationaldiscussion; and so it does not link the rejection of rule-governed accountsto the denial of the possibility of objectivity.

Instead of taking the criteria of the cognitive value of scientific theoriesto be rule-governed, then, I consider them to be cognitive values with allthe features laid out in the preceding section. Thus, cognitive values may bemanifested in theories, more or less, where the adequacy of theirmanifestations and their relative rankings and interpretations are mattersof reasonable controversy, open to critical dialogue within and with therelevant communities of inquiry and in principle open to “objective”resolution through such dialogue (Hempel 1983a, b; Bernstein 1983;McMullin 1993; Sankey 1997).


I will now review fairly extensively candidates that have been proposed forinclusion in the list of cognitive values. Only on pp. 89–95 (Chapter 5) willI address how one identifies which of these items is and is not a cognitivevalue, how one settles disputes that may arise about this, and how onedistinguishes cognitive values from other kinds of values.

There may not be a single, definitive list of cognitive values. Science isvariegated; it admits of a variety of approaches with a variety of aims andforms, and with its historical unfolding new cognitive values may beidentified and items, once held to be cognitive values, becomereconsidered. The list contains no surprises. Defending its adequacy andcompleteness, providing a ranking of its items or a general account of howtheir respective “weights” might be balanced, working out how (and howwell) the roles of the various items may be represented in Bayesian models,and even exploring the general consistency of the items—all importantmatters in need of fuller exploration—lie beyond the scope of this work. Iintroduce the list here for the sake of concreteness of discussion only; mycentral theme—the critical evaluation of “science is value free”—does notrequire a definitive list of cognitive values.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (71)

Empirical adequacy. This refers to “the quality of fit between theory andempirical data” (McMullin 1996). What sort of fit? Let T be a theoryunder consideration for acceptance, and E be the class of relevant empiricaldata that is available. Minimally, empirical adequacy involves:

1 T is consistent with items of E.2 T has implications that coincide, to a sufficient degree of accuracy with

items of E and with statements derived by induction or statisticalanalysis from items of E.

3 T, via prediction, contributes (or has contributed) to the expansion ofE.

4 E includes any unrefuted data fitted by antecedent theories of therelevant domain.10

Some clarifying comments: “implications” in (2) does not reduce to“entailments,” for typically a theory makes contact with the data only inconjunction with an array of auxiliary assumptions. That is why I have notincluded falsifiability in the account of empirical adequacy or as anadditional cognitive value. The formulation of (2) also highlights that oftencarefully restricted empirical generalizations or data that have beenstatistically analyzed, rather than individual items or sets of data, arewhat is brought into contact with theory. Whenever an experiment isreplicated, such generalizations, derived inductively, are available where“induction” means either induction by enumeration, by elimination or bycurve-fitting.

E refers to the set of available data, items of which must satisfy therequirement of intersubjectivity. The items of E (causally) derive fromobservation, often of phenomena encountered or created in the course ofexperimental and measurement practices. A particular theory is expected to“fit” (2–4) only certain classes of data, those derived from specifieddomains of phenomena. This, though quite obvious, is important toemphasize. We need the “right” data to deal with our theoreticalquestions; we do not just collect data and then ask which theory fits them.Typically, a theory is held of a certain domain of phenomena, or of objectsunder certain boundary conditions, or of objects insofar as they may becharacterized in a certain way; it is not just held in an unqualified way.Empirical adequacy, except perhaps with respect to (1), thus, should berelativized to classes of data (Chapters 7, 9 and 10). The causal origin ofdata in observation does not imply that putting an item in E is withoutpresuppositions (Chapter 7), that the language used for its description istheoretically neutral or necessarily constant across theoreticaldevelopments, or that it may not be revised (or even rejected) in the courseof attempts to make T fit with E.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (72)

When there is a misfit between T and E, in principle, adjustments maycome from either end. Judgments about the appropriate adjustments aremade in the light of the full play of all of the cognitive values. What goes intoE is neither arbitrary nor (at times) uncontroversial, partly for the reasonjust stated, and partly because there can be disputes about who is to beincluded among the subjects of the “intersubjectivity” (Chapter 8).Particular approaches to science (Chapter 5) may add further requirementsupon the items of E, or upon the language in which they should bedescribed, for example, by granting special salience to experimental andquantitative data and materialist categories (Chapter 4); to precision(exactness); or to data that reflect the richness, complexity and variation ofordinary experience (Chapter 9).

Explanatory and unifying power. This is open to two interpretations(Chapter 5): (1) wide-ranging explanatory power. explain and unifyphenomena (as described in the items of E) in a wide range and variety ofdomains, perhaps in virtue of the theory offering reductionist explanationsor more generally because of its explanatory depth (Bhaskar 1986);consilience: unifying a diverse range of phenomena and of other theories(McMullin 1996, interpreting Whewell); (2) full explanatory power.account for all the aspects and the dimensions of phenomena, all theircauses and effects, responsive to particularity, concreteness anduniqueness. Clearly this item needs to be elaborated with a full-blowntheory of explanation and explanatory ideals.

Power to encapsulate possibilities. Open to a domain of phenomena,including to identify novel, low probability and unrealized possibilities. Itincorporates the predictive power implicit in the statement of empiricaladequacy, and generally implies being able to define the bounds of atheory’s predictive power and sometimes having the power to predict novelpossibilities.11 Successful prediction reflects the power of a theory toanticipate possibilities within certain bounds or in certain settings. As suchit cannot be considered a measure in general of a theory’s power toencapsulate possibilities, for this power can be expected to extend beyondthese bounds. This item, in combination with the previous one, alsoincorporates informativeness (van Fraassen 1980) or amount of content(Feyerabend 1975).

Internal consistency.12

Consonance (McMullin 1994), connectivity (Ellis 1990), holism (Nelson1995): having appropriate relations with other well established theories,for example, consistency, integration; inter-theory support: explaining orbeing explained by other theories (Newton-Smith 1981); enabling theirbounds of application to be specified; and intermeshing through mutualdeployment of and dependence on one another’s results.

Source of interpretative power. enables the development of aninterpretive narrative of the successes and failures of its precursor theories


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (73)

(MacIntyre 1977; Tiles 1985; Taylor 1995) and a specification of thebounds within which they were soundly accepted-sometimes byincorporating (either by explaining or representing as special cases) any oftheir unrefuted sub-theories; or by observational nesting (Newton-Smith1981), incorporating the observational successes of precursor theories.

Puzzle solving power (Kuhn 1970): the capability to solve both empiricaland theoretical puzzles (problems); open to extensions that enableproblems to be solved (McMullin 1996).

Simplicity. Notoriously this means different things to different people. Inone sense it connotes harmony, elegance and other aesthetic qualities. I donot take these to represent cognitive values. In another sense it suggestsparsimony, economy (of formulation, of technical devices); efficiency in usefor explanatory, predictive and other “scientific” purposes; deployment ofthe “simplest” available mathematical equations; conceptual clarity,“clearness and distinctness” (Descartes), intelligibility; idealization whichprovides a benchmark, departures from which can be convenientlyexplained; having appropriate analogies with other theories (Campbell1957)13 and formaliza-bility. In yet another sense it involves coherence(McMullin 1994), appropriate- ness, smoothness (Newton-Smith 1981),epistemic conservatism (Ellis 1990) or the rejection of ad hoc features.14

Most of the commonly cited cognitive values appear on my list, but a fewdo not, for example, fertility (fruitfulness, fecundity). It is worthy of moreattention, though I am inclined to think that either it is reducible to somecombination of items on the list, or it pertains more to such stances as theprovisional entertainment of theories rather than to their acceptance (andto the adoption of “strategies”: Chapters 5 and 7). My list is intended toprovide criteria for the acceptance of theories. Also missing is inductivelyderived; or something like using only an observational vocabulary or notinvolving hypotheses about unobservables, as in some interpretations ofNewton’s “Hypotheses non fingo” or in Skinner’s (1972) rejection of“theories” (Lacey 1974). This is not to deny the importance of induction inscience. As pointed out on pp. 58–9, theories often come into contact withempirical data by way of the intermediary of inductively derivedgeneralizations. It is a cognitive value of empirical generalizations to bederived in accordance with sound canons of inductive inference. It is not onthe list of cognitive values for soundly accepted theories because, generally,being inductively derived is incompatible with the significant manifestationof such cognitive values as explanatory and unifying power and being asource of interpretative power. Induction also plays a role at a “meta”level. Accepting a theory assumes that it will continue to manifest thecognitive values to a high degree, a meta-inductive inference.

Terms like “being highly confirmed,” “being highly corroborated,”“having a high degree of verisimilitude,” or “true” do not refer to cognitivevalues. They are cited in attempts to express the nature of the cognitive


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (74)

value of a theory; they are not criteria whose manifestation in a theory ispertinent to judging its acceptability. The cognitive values are criteria forappraising—depending on one’s epistemological orientation—(forexample) a theory’s confirmation, corroboration, verisimilitude, or truth.Among candidates that have been proposed as cognitive values, thathistorically have appeared on other lists or that have actually functioned ascriteria of theory choice, I definitely exclude the following: certainty,provable (as in “mathematical proof”), consistency with materialistmetaphysics, consistency with biblical interpretations, popularity, fittingwith “common sense”, consensus about what counts as “legitimate”discourse, and Baconian utility.

Baconian utility (generative of “prediction and control”; Tiles 1986,1987) may be considered an ambiguous case, for usually it implies bothsuccessful application and the service of certain interests. Being of service tocertain interests may be endorsed as a social value, but it does not providereason to accept a theory, as distinct from judging it significant. But that ithas been successfully applied in (technological) practice is relevant. Forsuccessful application of T is prima facie evidence that T possesses bothexplanatory and predictive power in the setting of the application. Aftersuccessful application of T, its explanatory power has been demonstratedto be more wide-ranging, ranging over applied as well as experimental (andperhaps some natural) spaces. So, successful application (and thusBaconian utility) does not constitute a further cognitive value. Rather, itcontributes to the fuller manifestation of some of the cognitive values, forexample, the power of T to encapsulate the possibilities open to the objectsthat have been successfully brought under practical control, but not ofothers of them. Similar remarks apply to control, as it is used in the phrase“prediction and control of behavior,” commonly used in the psychologicalliterature. Control of behavior is not an indicator that the theory, whichinforms the practices of control, encapsulates widely the possibilities opento human behavior (Lacey 1979).

Standards for estimating the degree of manifestation ofthe cognitive values

Just as personal values may be more or less manifested in a person’sbehavior, so cognitive values may be manifested more or less in theoriesproposed for acceptance of various domains of phenomena. Accepting atheory involves judgments that the cognitive values are sufficiently wellmanifested. How do we estimate (“measure”) the degree of manifestationof the cognitive values in a theory? According to what “standards” is itmeasured? If sufficiently demanding standards are not in play, theories willnot be submitted to rigorous empirical scrutiny and testing. I suggest thatat least the following standards should be in play.15


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (75)

Standards pertaining to empirical adequacy

• Representativeness. The items of E are representative of the possibledata that could be obtained by observing (often after constructing)phenomena of the domain(s) of which one is considering accepting T.The items do not reflect “bias”; they include data pertaining to all“relevant” characteristics of the phenomena.

• Pertaining to characteristic phenomena. The available data includereports of characteristic phenomena of the domain(s). It is important toconsider differences and relations among phenomena occurring invarious kinds of spaces: experimental, applied (technological), natural,and those of the realm of daily life and experience (Lacey 1984).Phenomena occurring in experimental spaces, for example, may notappro priately resemble those occurring in natural spaces (so that, forexample, a theory may encapsulate what we can do when we act uponthings rather than represent the way the things normally behave). So,attention to this matter is crucial for identifying the domains of which atheory can be soundly accepted.

Applying this standard can be both controversial and value laden. In afamous controversy, Skinner and Chomsky maintained opposedconclusions about the range of domains of which radical behavioristprinciples were soundly accepted. At the core of the dispute is Skinner’sview that characteristic human phenomena are the most common andgenerally observed as opposed to Chomsky’s that they are unique anddistinctive (Lacey 1980).16 Clearly value judgments are involved. Moregenerally, there are strong dialectical links between considering what are thecharacteristic phenomena of a domain and judging how it is appropriate tointeract with them. This standard is particularly important whenconsidering accepting a theory of a domain of phenomena that are ofsignificance in the realm of daily life and experience (Chapters 7, 9 and10).

• Relevant to critical confrontation with competing theories. E includesdata of relevance to the potential rejection of T; data pertinent toputting T into critical confrontation with competitors and to definingclearly the bounds within which T is soundly accepted. This standard (inconjunction with comparative testability) involves a generalization ofthe considerations lying behind Popper’s (1959) esteemed proposedcognitive value “high falsifiability”

• Reliability. The items of E have been reliably obtained (and they meet themost rigorous standards of numerical precision permitted by availablemeasuring instruments), and empirical generalizations derived from themreflect sound inductive and statistical canons. Here a variety of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (76)

considerations can come into play, including empirical ones. While Edoes not constitute bedrock, it is the point of reference for acceptingtheories. Its items derive from experience, but subject to the requirementof intersubjectivity, usually met by replicability. Often, however, theexperience is of highly complex, subtle, short-lasting phenomena createdby humans with the aid of expensive (and perhaps not well understood)advanced technology, available only to those with specially cultivatedobservational skills, and difficult to replicate (see Hacking 1983 on the“creation” of phenomena in the laboratory). From whose experience dowe obtain the items of E? Which experiential reports should we take asreliable? Often, we are not placed to replicate them ourselves, andbecause of expense it may not be possible for any replication to takeplace. Part of the answer to this involves answering: Whose reports arereliable? This is a question about evaluating testimony and authority(Coady 1992), and the sincerity, trustworthiness, observational acuity,rationality and other virtues of the observer. These issues can becomemore complicated in the human sciences. Can we expect widespreadintersubjective agreement on experiential items regardless of the valuesof investigators? Or, does observing certain human phenomena andrecognizing them for what they are, require a certain moral character orthe adoption of certain values? In the natural sciences the notion that oneis trained to exhibit the virtues of the “scientific ethos” is widely seen asplausible (Chapter 1). In the human sciences does one have to expresscertain moral virtues as well? (Taylor 1971; Lacey 1991a, 1997c).

Empirical adequacy is more highly manifested in a theory the more thedata it fits meet these conditions, and the more the theory fits the availabledata that meet the conditions; that is, the fewer anomalies it exhibits.

Generally applicable standards

• Comparative testability. T’s degree of manifestation of the cognitivevalues has been “measured” against, and shown to be greater than, theirmanifestation in a “sufficient” array of (though much less than “allpossible”) rivals.

Have the “right” array of theories actually been generated to make for asevere test? T comes to be accepted following the unfolding of a researchprocess. In earlier stages of the process, T (or its anticipatory precursors)has been provisionally entertained and adhered to, and these are stancesthat may be taken towards T in ways that may unobjectionably be partlysupported by value commitments. The present standard needs to be met sothat any values driving the research process become filtered out, so that the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (77)

judgment that T is soundly accepted does not tacitly, and withoutawareness, depend on the possibility that certain rivals were notinvestigated because conditions for their development were not madeavailable because of their potential links with other values. So, it suggestsraising such questions as: have theories that could in principle be appliedby those who hold rival value complexes been developed among the array?Do the prevailing social conditions permit the relevant data to begenerated (obtained)? It draws explicit attention to the social context of theresearch process, to the availability of the material and social conditionsneeded for a research project to develop, and to the values of theparticipants in the research (Chapters 4 and 10). The matter gains greaterimportance in those cases where the neutrality of T is questionable, andwith respect to appraising how well manifested is the cognitive value,“power to encapsulate possibilities” (Chapter 8).18

Theories are accepted of particular domains of phenomena. Thedomains, with respect to which rival theories compete, are also relevant tocomparative testability, leading me to propose the following two furtherstandards (both of which figure in Chapter 7) to complement the previousone:

• Comparative comprehensiveness. If two theories, T and T′ conflict, andif T manifests the cognitive values well of a domain D and T′ manifeststhem well of D′, where D′ is contained in D, then ceteris paribus Tmanifests the cognitive values to a higher degree than T′; even more soif, from its perspective, the success of T′ with respect to D′ can beexplained.

• Comparative local strength. T manifests the cognitive values highly ofD, where D has sub-domains [D1,…, Dn] only if for each sub-domain D1it manifests them more highly than any competing theory of Di.

Finally, in addition to comparative testability and its complements, I alsoinclude the following:

• Comparison with the most well entrenched theories. The manifestationof the cognitive values in T compares favorably with their manifestationin the most soundly entrenched theories, those considered to belong tothe stock of knowledge.

• Response to criticisms. Criticisms that T does not adequately manifestthe cognitive values have been adequately responded to, especiallycriticisms that make explicit what would count as more adequatemanifestation. The degree of manifestation of the cognitive valuesincreases under conditions of strong criticism.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (78)

Clearly these (and any other) standards will have to be deployed within amode of interpretive understanding, related to the context and practices ofgenerating and collecting empirical data and of developing and testingtheories. Their being deployed adequately may rest upon certainsocial conditions obtaining, upon the personal values of participants inscientific practices and upon the diversity of values held among theseparticipants. Moreover, the push to more rigorous interpretation of thestandards may come from those who adopt particular value complexes(“Rudner’s argument”: Chapter 4) for they—seeing the acceptance andpractical adoption of certain theories as contrary to their interests—maysometimes (though other times the opposite) more rigorously evaluate thecredentials of supposed authorities, and seek out the conditions that wouldenable a wider array of theories to be developed and tested.

The key idea of impartiality maintains that soundly accepted theories arethose that manifest the cognitive values to a high degree as estimated inaccord with the most rigorous available standards. We turn now to thediscussion of this idea.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (79)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (80)

4Science as value free: provisional theses

The sources of the idea that science is value free have been discussed(Chapter 1). Drawing from them I will now motivate and state provisionaltheses of impartiality, neutrality and autonomy, which will later becriticized and revised (Chapter 10).


Impartiality is a view about properly accepted theories. A theory (T) isaccepted of a domain (D) of phenomena, things and their possibilities.1 Toaccept T (of D) is to judge that it is so well supported that it needs nofurther investigation, that it should be included in the stock of consolidatedbeliefs. The core of impartiality is that T is “soundly” or “properly”accepted only if such judgments are based solely on assessments of thecognitive value of the theory, regardless of any considerations ofsignificance; so that to accept T (of D) is (ideally) to make the judgmentthat believing T (of D) has high cognitive value. The cognitive values arethe criteria of cognitive value. Most of them involve relations betweentheories and available empirical data and relations among theories. Thepossibility of impartiality depends upon cognitive values being distinct anddistinguishable from [other] values.

The seemingly innocuous phrase “available empirical data” hidesvarious complexities and potential controversies. For convenience I will usethe term “empirical datum” to denote both an observational report andwhat is reported in an observational report (an “observed fact”). Not allempirical data of phenomena of D are relevant for the assessment of T. Fora theory of projectile motion, for example, the height the arrow reaches inits flight is relevant but not its color. But we cannot “measure” the degreeof manifestation of the cognitive values in T until the class of empiricaldata to which T is expected to be related in the specified way has beenselected. How is the relevant class of empirical data selected?

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (81)

Materialist strategies

The Baconian idea (Chapter 1) introduces intersubjectivity as a condition,and consequently grants high salience to experimental data. From theGalilean idea (Chapter 1) follows the further restriction that the descriptivelanguage in which the data are expressed contain only what I will call“materialist” terms: generally quantitative and mathematical, applicable invirtue of measurement, instrumental intervention and experimentaloperations—the kinds of terms that apply to phenomena considered asgenerated from underlying structure, processes and laws rather thanconsidered as an integral part of daily life and social practice. Limiting thelexicon of the relevant data to contain only materialist terms is a muchstronger condition than intersubjectivity. The phenomena, described by thedata, can be characterized in indefinitely many ways. Many observedphenomena described with non-materialist terms (like many described onlywith materialist ones) can be grasped intersubjectively in observation, forexample, reports of actions described with intentional terms. The Galileanidea leads us to select data that describe phenomena abstracting from theirplace in any human practice or their relation to human experience; datathat have been stripped of all links with value.

The data are selected in this way because they are the kind of data thatcan bear as evidence upon theories that have been provisionally entertained.Corresponding to this condition on the selection of data, then, there is aconstraint on the kind of theories that may be provisionally entertained:that theories only deploy categories capable of representing the appropriatekind of underlying structure, process and law. Thus, for example, theoriesthat deploy teleological, intentional or sensory terms are not to beentertained. Prior (logically, not necessarily temporally) to anyconsideration of the acceptability of a theory, following the Baconian/Galilean ideas, what I call the materialist strategies are in play. Developingtheories under the materialist strategies2 makes it possible to bring theoriesinto contact with data in such a way that the degree to which they manifestthe cognitive values can be “measured.”

Later (Chapter 5), I will propose that the materialist strategies constitutejust one among several kinds of strategies that can be adopted whilecontinuing to emphasize the intersubjectivity of empirical data. Then (inChapter 6) I will discuss more fully the materialist strategies and theirgrounds,3 and reject the view that the referents of materialist terms have aspecial ontological status. Influenced by Baconian and Galileanideas, however, most of modern science proceeds under materialiststrategies. Note the plural: “strategies.” Modern science admits ofconsiderable variety. Some of it emphasizes systematic description ratherthan explanation; some structure rather than law; some phenomenologicalor statistical regularities rather than underlying law; some finding law


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (82)

through inserting an object into a larger, encompassing structure. Not allscience is conducted under “physicalist” strategies: constraining theories tobe open to reduction to physics, or deploying categories derived fromphysical or chemical theory—though reductionist tendencies are widespread—and where they are not present it tends to be a requirement that theoriesbe left open in principle to eventually being subsumed under laws inconjunction with posits about underlying (possibly developmental)processes and (possibly emergent) structures.4 Sometimes materialiststrategies may include further constraints, (for example) restricting laws tobeing determinist, processes to being mechanical, and (in behavioristpsychology) the independent variables of laws to being instantiated byenvironmental events. Materialist strategies exhibit multiplicity andvariability.

The materialist strategies indicate the kinds of data that are needed inorder to estimate the degree of manifestation of the cognitive values in atheory (of a domain), and to which any theory which is a candidate foracceptance of the domain is to be held accountable. Following thestrategies one seeks to obtain the data, constructs the experimentalphenomena, the observation of which can occasion gaining relevant data,makes the relevant observations and records their outcomes. Both logicallyand temporally, the class of available data is open-ended, and it neverexhausts the class of possible relevant data. What data are available varieswith time; and individually and collectively, in principle, they are subject torevision, even (rarely in practice) fundamental reconstitution. Which onesare available at a given time, especially concerning experimentalphenomena, is a function of our interventions and foci of interest, andoften of available technology or resources. Because of these contingencies,the data available at a given time (despite being of the required kind) maynot enable a judgment to be made about the cognitive value of a theory andits competitors; then judgment should be suspended pending furtherinvestigation.

A thesis of impartiality

In the light of the preceding considerations, I offer the followingprovisionally as a thesis of impartiality:5

I′ 1 The cognitive values are distinct and distinguishable from otherkinds of values.

2 A scientific value (T) is accepted of a domain of phenomena (D) if,and only if, T (of D) in relation to the available empirical data (E)manifests the cognitive values highly according to the mostrigorous available standards; and to a greater degree than doesany rival theory—where T meets the constraints of, and the items


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (83)

of E have been selected in accordance with the materialiststrategies.

3 T is rejected of D if, and only if, another theory (T′) is accepted ofD, and T and T′ are inconsistent.


4 Values, and assessments of a theory’s significance, are not amongthe grounds for accepting and rejecting theories.

5 Only the materialist strategies generally constrain theories thatmay be provisionally entertained and put requirements upon thekinds of data selected.

I′(2) is the key component. Is it a fact that theories in modern science areaccepted in accordance with it? No doubt it often is, but there are plenty ofexceptions (Chapter 9), cases in which agreement among members of thescientific community to accept a theory is explained only by their sharingparticular values. It is easy to understand how there can be such exceptions,for it is consistent with (5) being an articulated value of scientific practicethat particular members of the scientific community— having an interest(consciously or not) in consolidating theories that are significant for theiradopted value complex—may actually employ further constraints on thekinds of theories they wish to come to accept (and reciprocally on thekinds of data they select). According to I′ adopting these further constraintsis not a ground for rejecting (denying the cognitive value of) theories thatdo not fit them. It may be a good reason, however, to deem these theorieslacking in significance. Accepting a theory does not imply its significance;and deeming a kind of theory to be lacking in significance does not implythat theories of that kind may be rejected. It may sometimes be a goodreason not to investigate them; but at other times not to investigate themleads to accepting theories without their having been tested against asufficient range of competitors. Vigilance is necessary to avoid slippingfrom insignificance to rejection, and then from rejection of its rivals toacceptance of a theory.

The existence and intelligibility of exceptions to (2) and (3) do not seemcommonly to lead to the denial of these items among members of thescientific community. They seem to function as an ideal; so that “accept” istaken to connote “soundly accept” or “properly accept.” The exceptionstend to be seen as lapses, not as characteristic of occasions of acceptingtheories; and, when they are clearly identified, acceptance is usuallywithdrawn. I′(2)/(3) expresses a value of scientific practice, a value whosemanifestation is diminished by departures from it in fact—and (1) is apresupposition of its adoption. Those who adopt this value consider it a


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (84)

regulative ideal of scientific practice that all theoretical conflicts can beresolved in accordance with (2)/(3) consequent upon suitable growth andtransformation of the available empirical data and the rival theories, andthey affirm that throughout the modern scientific tradition many conflictshave been so resolved—leaving us with a solid and expanding stock ofsoundly accepted theories.

Rudner’s argument

When we accept T (of D) we never gain certainty. Nevertheless accepting T(of D) in accordance with I′(2) carries with it (pragmatic) certitude, wellvindicated by the great successes of the practical application of the bestaccepted theories of modern science. Pragmatic certitude is not the samething as epistemic certainty; but when we have certitude of T (of D), itseems clear that no cognitive barrier remains to its application. Even so, inall sorts of unanticipated ways the body of available data may change or ahitherto unthought of theory might pose a challenge so that it alwaysremains possible that T (of D) may come to be rejected. Were it to becomerejected, we might expect that there would be consequences of havingpractically applied it that would have negative moral value.

In a famous article, Rudner (1953) argued that when we accept T (of D)we are committed to the judgment (which I paraphrase into myterminology and call ‘Rudner’s condition”):

(R) T manifests the cognitive values to a sufficiently high degree (of D)so that the legitimacy of its being applied in practical activity is not tobe challenged on the ground that, if T (of D) were to turn out to befalse, consequences of negative moral significance (consequencesundesired for one’s value complex) might follow from so applying it.

We might put it: we need sufficient certitude to compensate for our necessarylack of certainty; but gaining that certitude is implicated in the valuejudgment that the undesirable consequences of applying T in practicalactivity T, should T actually be false, do not warrant (pending furtherinvestigation) withholding on the application of T. We need sufficientcertitude to compensate for morally significant risks that would be takenby applying a false theory in practice. Rudner goes on: “The scientist, quascientist, makes value judgments.” If he is right, denying (R) concerning T(of D) seems to imply denying I′(2) concerning it. Does not (R), therefore,conflict with (4)—so that I′ does not even express a coherent ideal thatcould be embraced as a value?

Can we affirm I′(2), but deny (R)? For what type of reason? It wouldhave to be that the cognitive values were not sufficiently manifested in T(of D). But, by hypothesis, T manifests the cognitive values to a very high


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (85)

degree according to the highest recognized standards of evaluation—somuch so that the scientific community concurs that further investigationceteris paribus would serve no cognitive interests. So, the denial of (R) wouldinvolve questioning the stringency of the recognized standards ofevaluation. For example: Was T adequately tested against data that arerepresentative of characteristic phenomena of the domains in whichpractical applications of T would have consequences, in particular the onesthat would be undesired should T be false? Was T tested adequatelyagainst rival theories which, should they be true, would involve fewer suchundesirable consequences if adopted? Is the community of scientistsadequately constituted to warrant deference to consensus achieved in it? Inprinciple, these questions can always be raised because the domains oftesting (especially experimental) and application do not coincide, becausetheory is underdetermined by the data, and because the criteria of whatconstitutes scientific competence may implicitly be tied to particular values.I am interested in the questions here, however, only insofar as they lendthemselves to answers in practice, for example, when the allegedcharacteristic phenomena are specified, a rival theory outlined, or evidencepresented that the composition of the scientific community makes it proneto “bias”; for instance when they point to inquiry relevant to answering thequestions.

Questioning the standards in this way implies that there are not adequategrounds to accept T (at least of the domains relevant when considering itspractical application). Thus, the same ground that serves to deny (R) alsoserves to deny (2). Then, ceteris paribus affirming (2) implies affirming (R).Accepting T (of D) implies the judgment that all the testing relevant foraffirming (R) has been conducted; so that it involves making both of thejudgments (2) and (R). The key link between them is provided by thestandards for “measuring” the degree of manifestation of the cognitivevalues in a theory. A value judgment may be appropriately, and essentially,involved in assessing these standards. Thus, holding a particular valuecomplex may lead one to push for higher standards; and differences in thevalue complexes held—and thus possibly different judgments regarding (R)—may lead to different standards being deployed, and thus differentjudgments regarding (2) actually being made. But the dispute in thesesituations is not about I′; it is about the standards. In any case, the differentjudgments are “to accept” and “not to accept.” If one makes the former,one judges that T (of D) manifests the cognitive values highly according tothe most demanding available standards; it is irrelevant to making thatjudgment that values may influence our sense of what are sufficientlydemanding standards. If one makes the latter, it remains that no theory hasbeen accepted or rejected with values among the grounds; the values pushone to adopt more rigorous standards and thus to engage in further


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (86)

investigation, not to reject the theory. There is, therefore, no conflictbetween (4) and (R). Rudner’s argument does not threaten I′.

While Rudner’s argument does not threaten impartiality, and so doesnot imply that values and science “interpenetrate,” it does point toimportant aspects of the “touch” (or shall I say “constant rubbing”) ofscience and values—referring back to Poincaré’s metaphor (Chapter 1).Holding certain values may push in the direction of adopting moredemanding standards for estimating the degree of manifestation of thecognitive values; so that impartiality may be furthered more constantlywithin some value complexes than others. Where a value complexunderlies making the value judgment of denying (R), it may open up fieldsof empirical inquiry effectively closed by those who affirm (R)—perhapswithout their being aware of doing so or perhaps because they are drivenby economic interests and they cannot market the products of a discoverywhile (R) is denied; but it does not put constraints on the content ofacceptable (as distinct from significant) theories.

I emphasized “practical” questioning of the standards, questioning thatpoints explicitly to further (novel, not simply replicative) inquiries, basedon the proposed more demanding standards, which (depending on theoutcome of the inquiries) could vindicate the point of the questioning or putan end to it.7 Pending the outcome of the inquiries, it is appropriate todeny (R) and therefore (2); even though—by the consensus of the scientificcommunity—according to the hitherto recognized highest standards (2)would be affirmed. The very fact that the standards are being called intoquestion breaks that consensus (at least temporarily), and the very fact thatthe proposed standards are more demanding is prima facie reason foradopting them. Here, several difficulties can arise. In the first place,practical, material or social conditions may not be available to carry outthe inquiries, or may actively pose barriers to them. The higherstandards may be recognized (by some), but not be effectively available inpractice. Under these conditions, it will be appropriate to continue to deny(R), and therefore (2).8 Second, the new standards may be recognized bysome, but not by others (perhaps even the majority of the scientificcommunity). They may not be understood; this is not implausible if the newstandards derive from reflections on what are the characteristic phenomenaof the domain of intended application (especially in the human sciences),and if the participants in the scientific community represent a limited rangeof value complexes. Or, they may be disputed and the dispute may not beresolved. Concerning some domains of inquiry, consensus regarding thestandards may not be reached. In such cases we cannot expect that theorieswill be accepted in accordance with I′. That would not be a ground toreject impartiality as a value, but only to recognize that the conditions forits realization are not always available. In these cases we only have thepulls and pushes of opposed opinions (where what is held to be possible


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (87)

tends to be derived from what is desired, and where acceptance of theoriestends to be confused with judgments of their significance) with power asthe ultimate “authority,” and where the possibility of neutrality does notarise.


Impartiality concerns the sound acceptance of scientific theories and itsgrounds. In this section, for the purpose of stating a working provisionalthesis, I will consider neutrality to concern the logical implications andvarious consequences of accepting theories.

Neutrality is important to the self-image of the modern scientifictradition. Think, for example, of the rhetoric of science as a universalcommunity, a universal culture, and a unifying humanistic force cuttingacross barriers of culture, race, religion and gender, generating theoreticalproducts that are of benefit to all and threatening to none (Bronowski1961). Regardless of culture, race, religion, gender—moral and socialvalues in general—science (its accepted theories and its practices of inquiry)is always an object of value, so that one does not have to resolve valuedisputes in order to value science. Science, it is said, has (or aspires tomanifest) features in virtue of which, for every value complex, it is anobject of value; it belongs to the common patrimony of humankind; it is apublic good.9 I will state a thesis that attempts to express what those(aspired to) features of science are, building upon the three components tothe idea of neutrality (introduced in Chapter 1). The rhetoric also suggeststhat not only is science an object of value for all value complexes, but alsothat it is an object of value partly because it is neutral, not justbecause piecemeal it serves some particular interests of each valuecomplex. This suggests that, for any value complex, not only is science anobject of value but also the value of neutrality, articulated in a defensiblecoherent thesis of neutrality, is (should be) endorsed: science is neutral andneutrality is a universal value. We will see on pp. 79–82 that it is difficultto hold both.

The three components to the idea of neutrality (Chapter 1), adapted tothe context provided by the proposed thesis of impartiality, may be put asfollows:

1 “Consistent with all value judgments”: Any value judgment isconsistent with a scientific theory accepted in accordance with I′moregenerally, accepting a theory in accordance with I′ implies (logically orrationally) no commitments about values to be held.

2 “No value consequences”: Accepting a theory in accordance with I′ hasno consequences for the fundamental values one holds; it neithersupports nor undermines the holding of a value complex.10


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (88)

3 “Evenhandedness in application”: For any value complex, theories,accepted in accordance with I′, may be applied in principle tosignificant phenomena and in practical activities in ways that further(and do not undermine) the interests grounded in the complex; or: Forany value complex, theories accepted in accordance with I′ aresignificant to some non-trivial extent—in principle, they can be put atthe service of any values, explaining valued phenomena, illuminatingthe realm of the possible, informing means to ends and the attainabilityof ends.11

“Consistent with all value judgments” is true; it has been “made true” bythe adoption of the materialist strategies to represent phenomena inabstraction from all contexts in which values are manifest. The othercomponents confront serious difficulties. “No value consequences” doesnot follow from “consistent with all value judgments” (see p 77); and todefend “evenhandedness,” it seems that considerable emphasis has to beput on the “in principle,” for modern scientific knowledge has beenpractically applied far more readily for the sake of advancing technology toserve economic, military and other projects in the advanced industrialworld than it has been in association with any other values (cf. Harding1998).

The idea of neutrality, with these three components, clashes withanother facet of the self-image of the tradition of modern science—science’s special service to progress. Progress is not neutral; it cannotcoexist with the traditional values of numerous cultures. This latter facet ofmodern science’s self-image often celebrates the falsity of the “no valueconsequences” component, as in the following classic passage:

Science as an institutionalized art of inquiry has yielded varied fruit.Its currently best-publicized products are undoubtedly thetechnological skills that have been transforming traditional forms ofhuman economy at an accelerating rate. It is also responsible formany other things not at the focus of present public attention, thoughsome of them have been, and continue to be, frequently prized as themost precious harvest of the scientific enterprise. Foremost amongthese are: the achievement of generalized theoretical knowledgeconcerning fundamental determining conditions for the occurrence ofvarious types of events and processes; the emancipation of men’sminds from ancient superstitions in which barbarous practices andoppressive fears are often rooted; the undermining of the intellectualfoundations for moral and religious dogmas, with a resultantweakening in the protective cover that the hard crust of unreasonedcustom provides for the continuation of social injustices; and, moregenerally, the gradual development among increasing numbers of a


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (89)

questioning intellectual temper toward traditional beliefs, adevelopment frequently accompanied by the adoption in domainspreviously closed to systematic critical thought of logical methods forassessing, on the basis of reliable data of observation, the merits ofalternative assumptions concerning matters of fact or of desirablepolicy.

(Nagel 1961:vii)

As an unrestricted generalization it is false that accepting a theory has noconsequences for the fundamental values one holds. In some historicallystriking cases, alluded to in the quotation, the consequences of accepting atheory have indeed included undermining certain fundamental values.

Accepting a theory and consequences for value judgments

The Galileo case is perhaps the most celebrated in which a theoreticaldevelopment had among its consequences the undermining of traditionalvalues. We may summarize the logic underlying this case as follows. InAristotelian science, the domain of the planets had two descriptiveframeworks: one composed of geometric/kinematic categories, the otherrelating the planets to the cosmos in virtue of which their arrangements andmovements were explicable teleologically. The centrality and immobility ofthe earth figured prominently in both frameworks. The teleologicalaccount was directly relevant to a value complex. In medieval ideology theorder of the cosmos grounded the predominant value complex, especiallythat pertaining to the social values which were to be expressed. Thegrounding goes roughly:

1 Proper social relations, those which manifest the desired values, areGod-ordained.

2 God-ordained social relations mirror the order of the universe.3 The hierarchical order of the Aristotelian cosmos is the best account of

the order of the universe.4 The Aristotelian cosmos reveals the earth as stationary and central.5 Therefore, the earth is stationary and central.6 Feudal social arrangements mirror the hierarchical order of the

Aristotelian cosmos.7 Hence, in the absence of alternative candidates, feudal social

arrangements are God-ordained.8 Thus, it is impossible to transform the feudal order into a better order.

A crucial part of the argument for (3) rested on the theoretical context andempirical evidence for (4). With the refutation of (5) (see Chapter 7), (3) isalso refuted and the argument for (7) dissolves, since no alternative


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (90)

grounding of the feudal value complex was able to be constructed. In thisway a judgment of Galilean science is incompatible with the feudal valuecomplex, being inconsistent with its presuppositions, though not with anyof its value judgments. According to impartiality, the feudal value complexmust adjust. (Confronted with the incompatibility Galileo’s persecutorsopted to reject impartiality.) In the process, the argument for (8) alsodissolved, and with that dissolution came the removal of a barrier topursue the interest of extending human control over nature (which wouldnecessitate new social relations) rather than remain confined by obligationsto retain the old social order (which was implicated in a stance ofadaptation or attunement with nature, Chapter 6). Thus, scientificdevelopments not only removed a pillar of the old value complex, but alsowere part of the process that would show that a new value complex couldbecome manifested. The Copernican revolution achieved the former; theacceptance of theories confirmed experimentally achieved the latter bydisplaying that the possibilities of control over nature far outstrippedprevious anticipations (Chapters 6 and 7).

I have cited an historical example, as well as an aspect of the self-understanding of the tradition of modern science, to cast doubt on “novalue consequences.” It might seem that “consistent with all valuejudgments” implies “no value consequences”; then doubts about the latterwould suffice to also cast doubt on the former.

But it does not. I have maintained that a value complex consists of anintegrated ensemble of values and value judgments rendered coherent byvarious presuppositions about human nature and about what is possibleand not possible. Since a scientific theory provides encapsulations of thepossibilities allowed by a domain of phenomena (Chapter 5), it may implyto be impossible (possible) what a value complex presupposes to bepossible (impossible), or it may contradict a posit of a conception of humannature. So theories may be inconsistent with the presuppositions involvedin adopting value complexes. Suppose that T, a theory accepted inaccordance with I′ is inconsistent with a presupposition (p) of adopting avalue complex (c). (For example, p may be (5) in the Galilean example, andc the feudal value complex.) Then, accepting T implies: 1) cp [ceterisparibus] rejecting p, and thus 2) cp ceasing to adopt c which (in turn) willlead 3) cp to the rejection of c. In this sense, accepting T undermines or isincompatible with adopting c; though, reflecting the cp in (2) and (3), sincec may be newly rendered coherent under different presuppositions (p′), Tremains formally consistent with the value judgments contained in c.12

Revision of “no value consequences”

The growth of scientific knowledge, the consolidation of more and moretheories in accordance with impartiality, thus, can lead to challenges of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (91)

value complexes (and thence their social sustainability) simply byundermining key presuppositions of their legitimation. Impartiality,therefore, actually challenges a key component of the common idea ofneutrality rather than being the opposite side of the same coin; though forwhatever measure of neutrality there is achieved, impartiality constitutes anecessary condition. Moreover, if “no value consequences” is false, itwould not be surprising if there is little room to apply the theories(accepted in accordance with I′) in practices valued from the perspective ofthreatened value complexes, at least without the applications becoming theoccasion for these value complexes to become more threatened; and so“evenhandedness” would also be false.

Thus, it is not plausible to think of neutrality as cutting across all valuecomplexes, but only across the “viable” ones. Rather, we might takeimpartiality—I′(2) and (3)—to define the limits of application of neutrality.I will say that a value complex is viable if its presuppositions are consistentwith the body of theories which have been accepted (of the relevantdomains) in accordance with impartiality. Some value complexes are notviable, as Nagel celebrates. But there remains an array of viable valuecomplexes, the array of modern pluralism. When we limit considerationonly to this array, accepting scientific theories neither supports uniquely norundermines any value complex. Theories pertain to values only where theyilluminate the workings of significant phenomena and deliberations aboutmeans to ends and the attainability of ends; and in principle, and to a highdegree in practice, do so regardless of what the values are. To itsproponents, neutrality pertains among the items of this array. I will dropthe “no value consequences” component in favor of the “range of viablevalue complexes” component. The “evenhandedness” component ofneutrality can now be restated: For any viable value complex, theories,accepted in accordance with I′, (in principle) are significant to some non-trivial extent.

A thesis of neutrality

Against this background, retaining “consistent with all value judgments,”13

transforming “no value consequences” into “range of viable valuecomplexes” and restating “evenhandedness in practical application,” Ioffer provisionally the following statement of a thesis of the neutrality ofscientific theory and theory acceptance:

N′ 1 No scientific theory (accepted in accordance with I′) has valuejudgements among its logical implications; and accepting atheory in accordance with I′ implies rationally no valuecommitments.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (92)

2 There exists a range of viable value complexes; then, acceptingtheory neither undermines nor supports the holding of any viablevalue complex.

3 A theory accepted in accordance with I′, in principle, is significantto some non-trivial extent for any viable value complex.

N′ states a modest thesis, some would object too modest and that, inparticular, it does not capture well enough the way in which theapplicability of theories is evenhanded. It is not just, the objectors wouldsay, that for any viable value complex an accepted theory (in principle) issignificant (but with possible great variations of significance across thecomplexes), but that (in principle) all viable value complexes are equally ableto gain from the application of accepted theories in practical activities, orthat accepted theories are available to be applied to serve the good ofeveryone, regardless of their adopted value complexes. If such a strongerversion of “evenhandedness” could be established generally, it wouldfollow that gaining and adopting scientific knowledge would be itself a(social) value within all viable value complexes–a view, as pointed out atthe beginning of this section, widely associated with neutrality. Let us, then,consider replacing (3) by (3a):

3a Every viable value complex gains, in principle, a fuller realizationfrom the practical application of theories accepted in accordancewith I′, and none are better suited than others to apply them for thesake of furthering their own realization.


3b In principle, the interests of some value complexes are not especiallyfurthered, rather than those of others, by the gaining and applicationof accepted scientific theories.

The difference between (3) and (3a) is important: (3) requires only that, foreach accepted theory, application is possible (in principle) within thecontext of any viable value complex. Compatible with that, in practice, theextent of applying scientific knowledge (and the capability to apply it moreextensively) may vary markedly within the contexts of different valuecomplexes—from application pertaining principally to marginal (orisolatable) activities to it shaping the central productive acts and contexts ofdaily life and experience. In the latter cases de facto the application ofscientific knowledge has a more important role in the socialimplementation and consolidation of the value complex. Often, this will benot merely de facto, but also “in principle,” for in the former cases the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (93)

more extensive application of scientific theories may require conditions,and generate consequences, that undermine the value complex. For certain(many?) particular theories, (3a) will be defensible, but some complexes aregenerally more suited than others for the application of theories, contraryto (3b). Compare the role of the practical applications of theories (acceptedin accordance with I′ and thus developed under materialist strategies) in aLatin American peasant culture with those in advanced industrialcountries. Both provide instances of viable value complexes, yet clearly theone whose central activities and experiences are informed by scientificknowledge is more suited to apply scientific theories than the one whosecentral activities are not.

In the extreme case a value complex may be viable (consistent withtheories accepted in accordance with impartiality), but not historically andsocially sustainable, for instance the social and material conditionsnecessary for its component values to be manifested are not available andits central practices, largely uninformed by scientific knowledge, are unableto resist the thrust and expansion of “development” practices which arewell informed by scientific knowledge (Chapter 8). Moreover, whereverthere have been clashes between modern and (viable) pre-modern orgrassroots alternative value complexes, applied science has always servedthe interests of the modern. (3) does not imply (3a), and (3a) representsneither a fact, nor even an idealization of historical fact. If (3a) represents avalue, it is far from being highly manifested in the practices of modernscience, and it is not endorsed by those who hold certain viable valuecomplexes (Chapter 8). So I will not include it in the statement of N′.

I am attempting to formulate a clear thesis in which neutrality cutsacross viable—not socially sustainable—value complexes, across those valuecomplexes that are cognitively admissible not just across those permittedby current arrangements of power. In doing so I wish to remain faithfulboth to the initial idea of neutrality and to the historical record thatscientific advances have had among their (cognitive) consequences theundermining of certain traditional value complexes, and thus to be able toreconcile endorsing both neutrality and progress. The upshot has been thatthe proposed thesis, N′, does not capture a robust sense of“evenhandedness” across value complexes. Perhaps I have strained toomuch to be inclusive by defining “viable” only in terms of consistency withtheories accepted in accordance with impartiality. Others may want todefine it more stringently. Neutrality, it might be proposed, is across valuecomplexes that include the value of “rationality,” supposedly a universalor a cognitive value, where rationality requires not simply consistency withsoundly accepted theories, but also engaging in practices of centralimportance to society in a way that is informed by theories accepted inaccordance with I′. Restricting “viable” to those value complexes thatinclude rationality understood in this way, while it enables the expression


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (94)

of a robust kind of evenhandedness of application, does so at the price ofmaintaining that neutrality holds only across value complexes that includethe high salience of a particular social value (not a cognitive value), namelychoosing one’s actions and practices so that they can be informed byknowledge gained under materialist strategies.

Still others might wish to restrict neutrality to hold across those valuecomplexes whose presuppositions are consistent with materialistmetaphysics, not just consistent with theories that are accepted inaccordance with I′ and thus accepted under materialist strategies. Thisproposed restriction, like the one in the previous paragraph, generates akind of paradox. The paradox arises because the best grounds foraccepting materialist metaphysics are linked with adoption of “the modernvalues control” (Chapter 6), so that to restrict viable value complexes inthis way is equivalent to restricting them to those that include the modernvalues of control. Then, neutrality across values holds only whereparticular values are held. This tends not to be seen as paradoxical becausethe adoption of the modern values of control is widely taken to be universal,an essential part of the self-definition of modernity, where not to adoptthem is seen as somehow to misconstrue the way the world is and to bemaladaptive to finding one’s way around in it.

Later (Chapter 10), I will consider an alternative to N′ (N) in which (3)will be replaced by:

3′ for any viable value complex, in principle, there are some theoriesaccepted in accordance with I′ that are significant.

(3′) is a weaker thesis than (3). It does not imply that any accepted theorycan, in principle, serve interests of all viable value complexes. One mayendorse (3′), yet not hold the view—that partly motivated consideration of(3a)—that in general the unqualified gaining of scientific knowledge is avalue. We do not get “evenhandedness” in a robust form from (3′): notevery theory accepted in accordance with I′ (and thus under the materialiststrategies) need be applicable in every (viable) value complex; but for each(viable) value complex there are theories accepted in accordance with I′that are significant to some degree. Finally (Chapter 10) I will introduce afurther version (N) which I will endorse. This version of neutrality is notrestricted to the context of research conducted under the materialiststrategies. It attempts to capture a measure of “evenhandedness ofapplication,” and expresses neutrality as a thesis about the character ofscientific inquiry rather than simply about the implications andconsequences of accepting theories in accordance with impartiality.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (95)


The idea of autonomy concerns features of the processes and practices ofscience that are conducive to generating theoretical products that manifestimpartiality and neutrality to a high degree; so that it is subordinate tothem. It proposes that the processes of science are driven by purelycognitive considerations, by the goals of collecting more empirical data anddeveloping more and more encompassing theories that become accepted inaccordance with I′, and not by “outside interferences” that are irrelevant tothe sound acceptance of theories. Among “outside interferences” I include:values and (for example) metaphysics, power, personal ambition, popularappeal, government, economic interests, law, religion, the military,ideology; the “will of the majority” and special interests of any kind.

Autonomy is a value—of course not always (or even often) manifested. Itis often supported by pointing to historical cases in which outside factorshave in fact interfered with the process of science, and where the process,aiming to gain products that accord with impartiality and neutrality, hasbeen distorted, threatened, decelerated or compromised. The idea ofautonomy may also build in the further proposals that outside interferencesare more likely to be discerned and resisted if the members of the scientificcommunity have been formed in the practice of the “scientific ethos”(Chapter 1), and if scientific research is conducted by an “autonomous”(“self-directed”) community working within “autonomous” institutions.

I offer for consideration the following thesis designed to capture the coreof the idea of autonomy:

A′ 1 Scientific practices aim primarily to gain more, deeper and moreencompassing theories that are accepted in accordance with I′ andwhose acceptance accords with N′.

2 They are conducted without “outside interference” by thescientific community which, in order to ensure the furtherance of(1): a) defines its own problems, questions, priorities, anddomains of phenomena to be investigated; b) has uniqueauthority with respect to matters of method, theory acceptance,and standards (both cognitive and moral) of scientific conduct; c)determines who, and the qualifications of whom, will be admittedinto the scientific community, and what counts as competenceand excellence; d) shapes the form and content of scientificeducation, and the structure and activities of scientificinstitutions; e) forms its members in the practice of the “scientificethos”; and f) exercises its responsibility to the public fully byacting in accord with items a)–e).

3 The scientific community conducts its investigations in self-governed institutions which are free from “outside interference”,


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (96)

but provided with sufficient resources in order to conduct itsinvestigations efficiently and reliably.

A′ is riddled with tension. Scientific institutions and research practicesrequire financial, material and social conditions which they do not generateby their own activity, and so they must be provided by “outside sources”(generally government, business and universities). There cannot but be“outside influences” on scientific institutions; autonomy depends oncurtailing the “influence” so that it does not become “interference.” Howcan this be done? There are not unlimited resources, and provisions forscientific research must compete with other legitimate interests. It isproperly a matter for public deliberation, not an internal scientific question,to consider how scientific research (and in what areas) ranks in importancewith such other social values as alleviating hunger, maintaining economicgrowth, improving education and sustaining the environment.

Whatever balance is reached, the outcome will involve limits upon theresources available for research, and require adjustment of its direction andemphases. Thus, it appears, “compromises” must be made with (2) as wellas (3). Can they be made in such a way that the outside “influences” playalong with (1) and not in opposition to it?

Any compromise will involve adjustment to some of the items of (2). An“acceptable compromise” might qualify items (a) and (d) of (2) (and onlythem) by adding “in collaboration with the appropriate ‘outside influence’—government agency, biotechnology company, etc.” to the beginning ofeach of them, and “where the collaboration is not to such an extent that itinvolves a practical denial or subordination of (1)” to the end. Scientistscannot investigate everything; there is always a choice about whichphenomena and domains of phenomena to investigate, and about thepriorities of research. To some extent no domain is devoid of scientificinterest, although clearly the value represented by (1) is better served byengaging in research in selected domains. But investigating any domain cancontribute to expanding the compass of scientific knowledge. There is,then, some arbitrariness (even a considerable amount) while remainingconsistent with (1) in choosing what domains to investigate, leaving plentyof room for the choice to be influenced by personal and institutional valuesand outside factors. So long as the influence of these factors is limited tothat of domain of investigation I′ will not be threatened, for choices madein accordance with I′ are about theories of specified domains. So long assome space is left open for “fundamental research,” that directly concernedwith the gaining of deeper and more encompassing theories, that outsidefactors play some role in these choices may de facto not subordinate (1) totheir goals.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (97)

Such compromises, it may plausibly be argued, are necessary for theprogress of science. Without them the conditions needed for the conduct ofscientific practices will not be provided, so that without them there can bevirtually no progress of scientific knowledge. With them, however, theoriescan come to be accepted of certain domains of phenomena, even if (often?)the domains are chosen more for their practical interest than for howinvestigation of them may contribute to fundamental research.Furthermore, since grasping the underlying order of these domains requiresthe grasp of more fundamental laws, interest in investigating them maysupport to some extent also the interest in fundamental research. Pressuresto compromise on (2) and to settle on a qualified version of it as a value ofscientific practice derive in large measure from the need for funds toconduct research, for it is rare that funding is provided without stringsattached, and without it being open to cuts for reasons that have little to dowith scientific schedules and interests. No doubt some sources offunding (for example, universities) may permit a closer approximation to A′ than others (for example, corporations) do. In general, funders would beexpected to desire that research be conducted with the aim of generatingtheories acceptable in accordance with I′ that could be applied to servetheir particular interests well. Thus, the more compromises concerning (2)are entered into, the less likely it is that N′, particularly N′(3), would bemanifested, and thus the strong manifestation of A′(l) is called intoquestion.

As indicated, gaining theories that accord with I′ is not per se threatenedby entering into the mentioned compromises. Neither is it threatened bythe fact, which is consistent with A′, that individual scientists, whendeciding which theories to entertain provisionally and to adhere to, maymake different choices (or follow different hunches)—in ways that reflecttheir personal or institutional values—about how to further (1). But it maybecome threatened (at least in some fields of science) where there arestrong identities of interests or values (apart from the interest to actaccording to (1)) among members of the scientific community, scientificinstitutions and the relevant outside factors. Then the shared values maylead to constraints being put (but perhaps not noticed) on the theories thatare provisionally entertained in the scientific community, and thus theexclusion from entertainment of certain kinds of theories simply becausethey do not fit these constraints rather than because they cannot stand upto empirical test. This suggests that I′ might be more constantly manifestedif item (e) of A(2) were replaced or, preferably, complemented by:

(e′) ensures that widely diverse value complexes are held among themembers of the scientific community.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (98)

This would also be conducive for movement towards the fullermanifestation of N′(3), and for countering the residue of values that maybe left (unnoticed) from the provisional entertaining of and early adherenceto theories.

The autonomy of the processes and practices of science is a value that isoften articulated by the scientific community, though in the hands of thespokespeople of scientific institutions it may be a compromised versionthat is intended. A′ expresses a view about how scientific practices are(ought to be) cognitively self-contained and self-governed by thecommunity of its practitioners. It includes (3), a view about the“autonomy” of scientific institutions—free from outside interference whileprovided with appropriate resources. Now, an institution’s values need not(fully) coincide with those expressed in the practices whose conduct itsupposedly enables (MacIntyre 1981: Chapter 14), and in (for example) acorporation’s research unit the direction of research may have little to dowith the general interests of the scientific community. Nevertheless, it isnot uncommon for references to the autonomy of science to emphasize (3)(or not to clearly distinguish (2) and (3)), perhaps because it is assumed thatinstitutional autonomy is a prerequisite for the high manifestation of (1).That, of course, is a matter for historical and sociological investigation.

Clearly A′ is a value of relevance to the interaction between science andthe general public (as well as government, business and other powers).Why should the public endorse it? Particularly, why should it not onlyabstain from interference with the internal workings of science, but alsoprovide it with the positive conditions needed for its pursuit? Why, for thatmatter, should the public accept that A′(1) is furthered more fully by acommunity, institutionalized as in (3), behaving as in (2) rather than in thecourse of a broad based dialogue with groups representing a variety ofsocial, cultural and moral interests? Since the products and effects—cognitive, technological, ideological (for example, Lewontin 1993; Longino1990; Harding 1993)—of scientific institutions have major impact on thepublic, the public’s interest in the autonomy of science is tied to“autonomous” scientific practices serving its more general interests. As arecipient of public support and a significant source of the conditions thatframe and transform contemporary lives, science’s claim to autonomy thushas to accompanied by a responsibility to the public.14

Item (f) of (2) holds that the responsibility of the scientific community tothe public is appropriately and adequately exercised in effect by followingits own methods, and consistently and constantly producing theories whoseacceptance manifests both impartiality and neutrality. Now, I think it istrue that it is in the public interest to support scientific practice insofar asits accepted theories manifest impartiality and neutrality—both of them;and thus to endorse A′, provided that there is reason to believe that thepursuit of high manifestation of (2) furthers that of (1). The other side of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (99)

responsibility is accountability. Autonomy does not excuse science fromaccountability to the public. Part of that accountability, I suggest, involvesempirical (historical and sociological) investigation of such questions as:Does scientific practice conducted in accordance with (2) by an“autonomous” community in “autonomous” institutions lead to theorychoices that are impartial and neutral? Do these choices serve the publicwell, or do they serve principally certain special interests? Allowing forpossible variation from field to field, does item (e) or (e/e′) of (2) bettercontribute to generating neutral products?

It does not serve public interest for outside factors to use their influenceor power to undermine that theories be accepted in accordance with I′. Itmay, however, depending on the outcomes of the empiricalinvestigations just mentioned, serve public interest to have (e)supplemented by (e′) in (2), if there were reason to hold that impartialityand neutrality would thereby be enhanced. Since autonomy is subordinateto them, the responsibility of the scientific community would require that itmake the appropriate adjustment in (e). If it did not, there would be noobjection, deriving from the priority of impartiality and neutrality, to thepublic withholding its support for research pending the adjustments beingmade. Public intervention into scientific practices and the institutions inwhich they are conducted, of course, is full of risks, for the line betweenmatters pertaining to item (a) and those pertaining to (e, e′) in A(2) can bea fine one in practice. It is not called for to the extent that the scientificcommunity is exercising its responsibility adequately

The public, then, may have an interest in a variety of value perspectivesbeing represented among the members of the scientific community becauseof its interest in impartiality and neutrality—and also for reasons of socialjustice to open up social spaces that have largely been closed to members ofcertain groups. Either way it seems appropriate to extend the domain ofacceptable compromise to include matters pertaining to (e/e′) as well as (a)and (d). I do not suggest, however, that (e/e′) be granted general priorityover (c), but that the scientific community be restructured over time so thatthe two become compatible and mutually supporting.

The above discussion has been premised on the assumption thatimpartiality and neutrality are adequately expressed in I′ and N′respectively. The issues just discussed, especially that of the membership ofthe scientific community, become more complicated and pointed when thisassumption is rejected. Then autonomy will have to be reconsidered(Chapter 10).

The central objectives of this book are to produce elucidations of theview that the sciences are value free and of its three component ideas, andto assess exactly what is and is not defensible in them. We will see(Chapter 10) that the provisional theses stated in this chapter, in view ofthe arguments accumulated in the intervening chapters, cannot be


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (100)

sustained, but revised versions of impartiality and neutrality in differentways (though not autonomy) can be defended.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (101)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (102)

5Scientific understanding

Impartiality is the “rock-bottom” component of the idea that the sciencesare value free. Neutrality and autonomy both presuppose it. It, in turn,requires that the cognitive values be distinct and distinguishable from otherkinds of values. I have not yet addressed how to identify cognitive valuesand deal with controversy about their identification. To remedy this lack Iwill draw upon an idea of “scientific understanding.” I have alreadyreferred to the cognitive values as “criteria of cognitive value” (Chapter 3);now I will portray them as “indicators of sound scientific understanding.”

The provisional statement of impartiality (I′) presupposes that theoriesfit the constraints of the materialist strategies. Is fitting these constraintsitself an indicator of having gained sound scientific understanding? Is it acognitive value? If not, what is its ground? If it involves the play of (non-cognitive) values, does its requiring fit with the materialist strategiescontradict the claim of I′ that values play no role in the accepting andrejecting of theories?

Modern science is marked in fact by the almost exclusive adoption ofmaterialist strategies in its investigations. I will argue that this rests onmutually reinforcing interactions between the materialist strategies andwhat I will call the “modern values of control.” I introduce the argument inthis chapter and subsequently develop it in detail (Chapter 6 and parts ofChapter 7). At the same time, I will argue that adopting the materialiststrategies defines only one, among in principle many approaches-twoexamples of which will be sketched in Chapters 8 and 9-to gainingscientific understanding, in all of which the cognitive values ground thesound acceptance of theories.1


The grammar and logic of cognitive values parallels in many ways that ofother kinds of values (Chapter 3). A cognitive value appears in variousmodes: (including) manifested in theories, partly constitutive of thescientific theory-choosing practices in which it is expressed, articulated inwords, and embodied in social institutions. Thus, there can be

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (103)

discrepancies between the cognitive values that are manifested and thosearticulated. Adopting a set of cognitive values implies a commitment notonly to reduce and (if possible) eliminate such discrepancies, but also thecapability to defend that it is the “right” set; and thus, paralleling theadoption of values in general (Chapter 2), to defend that manifesting thesecognitive values is possible, and that ceteris paribus they are indicators ofcognitive value—indicators of having gained scientific understanding.

Disagreements about cognitive values

Removing the discrepancies can involve modification of the cognitivevalues either as manifested (and thus changes in theory-choosing practices)or as articulated. Those expressed in theory-choosing practices, and thencemanifested in the theories that are chosen, may need to be modified in thelight of argument at the level of articulation; and sometimes thosearticulated will need to be modified in view of reflection upon actualtheory-choosing practices and the possibilities they admit for themanifestation of supposed cognitive values. By way of this two-wayinteraction between the modes of manifestation and articulation, cognitivevalues become “reflectively endorsed” (Anderson 1995a).2

Modifying one’s own adopted cognitive values may be furthered orhindered by which ones are embodied in social institutions. A fundamentalshift of cognitive values within a tradition of inquiry (as occurred, forexample, in the scientific revolution of the seventeenth century) requires anaccompanying fundamental transformation of the institutions that supportthe inquiry. Personal adjusting of cognitive values will always beconstrained in the light of those that are actually embodied, preciselybecause one’s cognitive values partially constitute theory-choosingpractices that one shares with some community. So we would expect thatthe cognitive values that are adopted will be susceptible to variation withthe theory-choosing practices to which one is exposed in virtue of thehistorical moment, culture, class and personal background. This providesthe point of entry for some of the radical arguments, which broadly fallunder “social constructionism,” that in reality cognitive valuesare subordinate to moral and social values, and that institutional inertiaand power are the principal causal factors for the scientific beliefs that wehold.

Given the bipolar sources of the modification of cognitive values—thatpeople (and communities) remove the discrepancies between the manifestedand articulated systematically in different ways and thus actually adoptdifferent sets of cognitive values—is neither surprising nor per se a threat torationality. Personal (and group) judgment, and thus a measure ofcontestation, cannot be eliminated from the use of value words. In comingto adopt cognitive values, however, it seems wise to seek out the sources of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (104)

disagreements, to take alternative traditions seriously, and even to entertainthe possibility that the social conditions for the resolution of disagreementsmay not presently exist (MacIntyre 1988). Encouraging criticalconfrontation between articulately developed alternatives, as part of theprocess of coming to adopt cognitive values, is not a denial of rationality;nor is exploring anticipatory alternatives or attempting to reclaim thecontributions of cultures that have been suppressed. On the other hand,rationality is denied when it is held that theories, and the practices in whichthey are gained, are not subject to critical discussion; when theory-choosingpractices no longer are profiled against the horizon of the ideal of truth.

This point has become clouded in some recent discussions, connectedwith multiculturalism and diversity, where it has been claimed that anyattempt to submit cognitive values to critical discussion—to show how theyare partly constitutive of rationality—is really an attempt to impose aparticular set of cognitive values that is reflective of the interests of aparticular culture or ideology. Ironically, reasons are usually given for suchclaims. Yet an important point is being hidden by this paradoxicalrhetoric; it is a critique of the authority and cognitive (epistemic) privilegethat has been granted to modern science and, to a lesser extent, certainmodes of philosophical, historical and political discourse.3 Science (asconducted under the materialist strategies) has been widely taken toprovide the exemplary expression of rationality, so that the cognitivevalues thought to be expressed in its theory-choosing practices, under theinterpretations they gain in them, have become identified with rationality.It is not paradoxical to challenge rationality so conceived and its deep andprioritized embodiment in institutions committed to the gaining ofknowledge, for example, to challenge that rationality so conceived inimplicated in the false identification of “fitting with particular strategies”(for example, materialist) as a cognitive value, which then serves to disguisea link between this conception of rationality and specific values.4

Cognitive values are constitutive of “acceptable” scientific theories. It isin virtue of their manifestation that, according to I′, we (should) maketheoretical choices. In order to be identified as a cognitive value, I suggestthat an attribute of theories should meet two conditions:

1 It be needed to explain (perhaps under idealization or rationalreconstruction) theory choices that are actually made, and thecharacter of controversies engaged in by the community of scientists.

2 That it is a criterion of cognitive value—an indicator of soundscientific understanding—be well defended.

McMullin (1983; but cf. 1993) proposes that only the first condition needbe met. Commenting on the following passage from Kuhn: “Though theexperience of scientists provides no philosophical justification for the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (105)

values they deploy (such justification would solve the problem ofinduction), those values are in part learned from that experience and theyevolve with it” (Kuhn 1977:335), he says: “This is to take theHumePopper challenge to induction far too seriously…. The characteristicvalues guiding theory-choice are firmly rooted in the complex learningexperience which is the history of science; this is their primary justification,and it is an adequate one.” (McMullin 1983:19) There is no good reason tohold, however, that the values guiding theory choice that are rooted in thiscomplex learning experience are necessarily all cognitive values. Thecommon practices of the tradition, at least in particular epochs, may leadto non-cognitive values also playing a role in theory choice in ways thatmay not readily be recognized, perhaps because of widely sharedmetaphysical or value assumptions within the scientific community (andwithin the social institutions that support its work). I argue on p. 107 thatfitting the materialist strategies is an instance of such a value. The secondcondition is needed to separate values like it from the cognitive values.

Cognitive values must carry both explanatory and normative burdens.They play their roles in a context that not only makes genuine contact withscientific practice, but also recognizes that scientific practice is open and, inthe long run, responsive to rational criticism.

Criteria actually used in making theory choices

In drawing up the list of cognitive values, then, the first task is interpretive:to reconstruct key episodes of theory choice and controversy in order toidentify the criteria that can reasonably be held to have been deployed bythe participants in these episodes. In doing so, relevant matters to take intoaccount include:

• The criteria that scientists, who innovate, make decisive choices orengage in important controversy, claim to be using.

• Appraisals of how well the proclaimed criteria explain the choices toaccept theories actually made, and of whether there are gaps betweenwhat the scientists claim and how they act in practice (Laudan 1984);and if there are gaps, plausible proposals about how to fill them.

• The criteria appealed to (for example, in textbooks and critical reviews)in support of a theory (of certain domains of phenomena) becoming heldto be part of the stock of knowledge.

• The assent of working scientists to proposed criteria of theory choice.• Variations of criteria across fields, episodes and epochs, and the reasons

given for the variation.

In short, the relevant reconstructions should be grounded in detailedinterpretive historical and sociological studies, and interaction with the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (106)

critical reflections of working scientists. The list of cognitive values(Chapter 3) is generally well grounded in this way, since it draws heavilyon Kuhn (1970, 1977) and McMullin (1983, 1993, 1996) who derive theirlists from detailed interpretive historical studies; and thus it generally hasthe ring of plausibility. Even so, as a list of the criteria actually used intheory choice (in many fields of science), I think that it is incomplete—in away that makes a difference.

In particular, empirical adequacy seems to function in interaction withanother criterion that helps to interpret it. Empirical adequacy concerns“the quality of fit between theory and empirical data.” This “quality of fit”is enhanced by restrictions upon both theory and data that seem to betaken for granted, restrictions that may be construed as showing thatanother criterion is being brought into play in reaching a balance of thecognitive values. In Chapter 4, I assumed that such restrictions were inplace and that they derived from the materialist strategies. I did not includethem in the statement of “empirical adequacy” (Chapter 3), so that I couldhighlight that fitting the materialist strategies serves as a further criterionfor accepting theories, or at least for rejecting them (since they permit onlytheories that meet the constraints to become candidates for acceptance).That fitting the materialist strategies functions as such a criterion leaves itopen whether it constitutes an additional cognitive value, or whether itgains this function from some other source, such as metaphysical beliefs orshared social values.

Which criteria are cognitive values?

I turn now to the second task proposed on p. 91. How do we determinethat a criterion actually (or said to be) used in choosing theories is, or isnot, one of the cognitive values? How can we determine whether or not itis a criterion of a theory’s cognitive value?

Broadly speaking, there seem to be four relevant kinds of considerationsthat have been entertained:

From general theories of knowledge.From evolutionary naturalist and cognitive psychological theories of

knowledge acquisition and appraisal.From arguments about the possibility, or impossibility, of the concrete

manifestation of the proposed criterion in a theory.From whether or not it serves the objectives of science.5

Considerations of the first kind have sometimes sustained attempts toground rule-governed accounts of scientific rationality; they have also beenused in the context of defending realist interpretations of science(McMullin 1993). Those of the second kind may build upon accounts ofhuman cognitive faculties (Ellis 1990), or come to propose that scientificrationality is fundamentally social (Solomon 1992, 1994). Those of the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (107)

third kind account for the absence of certain items from the list, forexample, “certainty,” either of the Aristotelian kind (“necessity” or“intuitive”) or the Cartesian (“a priori”), since the character of ourscientific practices does not permit us to anticipate (or even to recognize)the concrete manifestation of such a value. I (without wishing to minimizethe importance of the other considerations) will concentrate exclusively onconsiderations about the objectives of science.

Here complications abound (Laudan 1984). It is a difficult interpretivetask to discern objectives, and there is pretty intractable disagreement aboutthem. Moreover, we cannot discern them without attempting to explainwhy we use the criteria we actually use (and to identify which ones weshould use) in choosing theories; so that identification of objectives andpicking out which criteria are cognitive values are deeply intertwined tasks.Moreover, objectives may vary with field, epoch and even school of thought.

While recognizing this, let us consider, as a candidate for being an (the)objective of science, the following:

O1 The objective of science is to represent phenomena (in rationallyacceptable theories) in terms of their being generated fromunderlying structure, process and law, and thence to discover novelphenomena.

Something like O1 has often been considered the objective of science withinthe tradition of modern science, and it is rooted in the influential Galilean(metaphysical) idea (Chapter 1). Given it, a case can readily be made(though I will only hint at it here) that most (not all) of the items on the list(Chapter 3) indeed are cognitive values, characteristics of theories whoseacceptance serves O1. Explanatory power, for example, is such acharacteristic-provided that an explanatory ideal is assumed: to explain isto represent phenomena as generated from underlying law, process andstructure. O1 seems to carry with it an explanatory ideal: explanation ismaterialist and (often reductionist), but it is not, for example, teleologicaland it need not be deterministic. Choosing theories informed by the itemson the list (perhaps, as in the case of explanation, under particularinterpretations) serves O1.6

Following a similar argument, adopting the materialist strategies alsoserves O1. Then, fitting the materialist strategies would appear to be anadditional cognitive value, one perhaps that significantly affects the seemsto vindicate that I′ provides a suitable statement of impartiality.interpretation of all the other, particularly “empirical adequacy.” O1, thus,Under the materialist strategies, scientific theories represent phenomena(objects, entities, beings, things, events, fields) simply in terms of theirhypothesized structures, processes and components interacting with oneanother in a way that can be represented in mathematically-formulable laws


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (108)

—abstracting the objects from any value they may be bearers of, or anyplace they may have in human practices or social structures. From pursuingO1, nothing follows directly about the relevance of theories, and of thephenomena discovered in the course of scientific practice, to humanpractices in general and to the objects of ordinary experience.

Why do we attempt to gain understanding of natural objects throughcognitive practices that abstract them from the contexts of human practicesin general, from their role in ordinary experience and from the possibilitiesopen to them in these contexts? (Why are O1 and its variants often taken toarticulate the objective of science?) To address this usefully, we need toconsider generally what it is to gain understanding. This will open up spacefor proposing various kinds of objectives that might be entertained forscience. It will also provide the context for addressing sharply the questionof whether fitting the materialist strategies represents a cognitive value.


What are natural objects and phenomena, and what is it to understandthem? The answers vary with context: focus of interest, practice beingengaged in, background knowledge, and participants in the discourse.Regardless of context, however, understanding things (events, states ofaffairs, phenomena) involves the following interacting components:

• An account of what they are: of the kind of thing they are; of theirproperties, behavior, relations and their variations with time.

• An account of why they are the way they are: why they have theproperties, behavior and relations that they do; why they have varied inthe way in which they have; an account of their origin.

• An account of the possibilities (including hitherto unrealized ones) thatthey allow in virtue of their own powers to develop and by means oftheir interactions with other things.

Understanding reality involves grasping the “what?,” the “why?” and the“what possible?” of phenomena.

Each of the three components is open to an array of interpretations.Regarding “what?”, an object may be considered as: an object ofexperience; an object of a human practice, something acted upon orinteracted with in practical life for the sake of personal, social orinstitutional ends; and an object which manifests causal relations withother objects—whether in virtue of lawful connections among thephenomena in which it participates and others, or of its place in a structureor a system (for example, ecological, social), or (in the case of humanbeings) of intentional and communication relations.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (109)

Clearly, depending (contextually) on what an object is considered to be,the answers to the “why?” and “what possible?” questions will takedifferent forms and reflect different interests. To illustrate, consider theseed (for example, of wheat) and ask what possibilities are open to it(Chapter 8). The seed can be considered to be many things, of which I willpick out two: an object which generates, on cultivation, crops withquantifiable yields; and an object which is integrally a part of socialprocesses. Considered the first way, seeds may become hybrids andgenetically engineered such that, when cultivated under specific conditions,crop yields increase significantly. Considered the second way, in becomingthe generator of such higher yields, the seed also becomes a commodity—something produced and cultivated in capital-intensive enterprises andbought and sold on the market—rather than an object produced annuallyas a routine part of the crop. It thus becomes a different object ofhuman practice related differently to the social order. What is aquantitative enhancement when the seed is considered one way is identicalwith a fundamental social change when considered another way.

Understanding, then, can take different forms and, in so doing, beresponsive to the different interests of different practices. Understandingsought under the materialist strategies is just one form. Why it should begranted primary emphasis needs explanation. I turn now to elaborate thisbrief introductory statement, and to give a systematic characterization of anumber of different forms of understanding.

Forms of understanding: wide-ranging and full

What is it to understand a phenomenon (event, state of affairs, thing,object in the broadest sense: including material object, person, socialinstitution)? What do we attempt to identify when we aim to understandan object? As indicated on p. 95, what we identify varies with theparticular context or focus of the quest for understanding, and so with theparticular way in which the question is put. Understanding an object alwaysinvolves being able to explain it and to identify the possibilities open to it.It also involves meeting additional cognitive (epistemic) conditions.Understanding is offered systematically, for example, in (or with the aid of)theories, or in narratives, that manifest the appropriate cognitive values tothe appropriate degree.

Much of the argument of this book deploys (explicitly or implicitly) adistinction between wide-ranging and full understanding. In order tointroduce it, I offer first a list of kinds of items that may appropriately beoffered (in some contexts or other) as components of understanding anobject, or kind of object, recognizing that some amount of provisionalclassification (some provisional answer to “what”) is a prerequisite togaining understanding.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (110)

When aiming to produce understanding of objects we identify (posit)some of the items on the following list:

1 Their components, how they are structured, and the kind of processesthey may be part of.

2 Their properties and relations, present and past.3 The principles with reference to which their movements, variations and

interactions can be explained; and particular conditions whose causalroles are represented in the principles.

4 The possibilities open to them to:

a change (to exhibit different properties and relations; to develop into);b affect as causal agents;c become upon decomposition; andd become upon becoming constituents of other objects, environments,

practices or systems;and the conditions under which these possibilities are realized.

5 The conditions which brought them into being, and (where applicable)those which sustain their existence.

6 The other objects or systems with which they share the sameexplanatory principles, and the particular conditions (often instancesof variables of the principles) that account for different variations inthe different systems.

7 Their relationships with the environments—physical, ecological,human, social, (sometimes) spiritual—in which they are located, andthe sort of reciprocal interactions they exhibit with one another,including how they relate to us (human beings) and what we can dowith them, and so their places:

a in the realm of daily life and experience;b in various human practices, and in the institutions in which they are

conducted;c specifically in the practices of investigation: their interactions within

these practices and the similarities and differences with theirinteractions in other contexts—paying attention to the material andsocial conditions of the investigatory practices and the institutionsthat provide them, the broader human practices that can beinformed by the outcomes of the investigatory practices (and theirhuman, social and ecological consequences), and identifying whatcan be done with them following investigation.

Not only does gaining a comprehensive form of understanding, thatsystematically offers items of all these kinds, seem to lie beyond our


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (111)

powers, but also the items are open to various interpretations and differentemphases. It is this that makes possible that there be different forms ofunderstanding. Think of “principle” in item (3). Principles can havedifferent characters: lawful, intentional, teleological, functional orspecifying location in relation to some whole. Not all can besimultaneously in play. Depending on the character of the principles,different items will have greater or less significance, and sometimes differentitems might be in tension. Yet probably there is no form of understandingin which there is not at least a hint of every item. I will identify someimportant forms of understanding.

Wide-ranging understanding

The understanding, gained under the materialist strategies, is an instance ofwhat I call “wide-ranging understanding,” a form of understanding thatgives special emphasis to item (6). Ready responsiveness to this item ismost available when principles are interpreted as laws, which express(possible) relations among quantities. Then, it highlights quantities in (2),and—in order to be able to represent the lawfulness of phenomena morecompletely—it may offer posits about the underlying components of theobjects (characterized quantitatively), about how they are structured andabout the processes into which they enter and the laws they reflect. It aimsto consolidate principles which, when deployed in conjunction with positsabout underlying components, structures and processes, produceunderstanding across the widest range of experimental, technological andnatural spaces. The same principles play the key explanatory roles acrossall of these spaces, where the different phenomena in the different spacesare accounted for in terms of variation of the values (boundary and initialconditions characterized quantitatively) of variables of the principles(Taylor 1970).

These principles, together with the accompanying posits about theunderlying order, encapsulate the material possibilities of these spaces,those that can be identified in terms of being generated from underlyinglaw, structure and process, given the appropriate boundary and initialconditions. The material possibilities do not exhaust the possibilities ofthings in the spaces. In the case of the seed, the increased crop yields areamong the material possibilities encapsulated in the relevant biologicaltheories, but not the seed becoming a commodity. The latter possibilitycannot be identified apart from the context of social relations. Also theobtaining of the boundary and initial conditions typically cannot beexplained without reference to this context. Wide-ranging understandingabstracts from the human, social and ecological characterizations that alsofit these spaces; it largely ignores item (7). It aims to be context-freeunderstanding. Abstraction from context and wide-rangingness of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (112)

application go hand in hand. Understanding can be more or less wide-ranging.

Full understanding

There is another form (more accurately, set of forms) of understanding thatgrants high salience to (7). I call it “full understanding.” It seeks, to theextent possible and deemed worthwhile, to understand objects in all (thefullness) of their dimensions, aspects, concreteness, wholeness andparticularity; and, to the extent possible, identify all their significantpossibilities, including those that the process of gaining understandingopens up. It does not dissociate gaining understanding of an object fromunderstanding what one is doing—what possibilities of the object (not justmaterial ones) are being identified, and what are their links with thematerial and social conditions of the process and with the institutionsproviding them—when engaged in the understanding-gaining process. Fullunderstanding does not ignore (6). In principle, it attends equally to thematerial possibilities of spaces (and so may be informed by wide-rangingunderstanding), and to the human and social characterizations of theirboundary conditions, to the human, social and ecological consequences ofprocesses within the spaces, and to the human and social possibilities thatmay be hidden in them.

But identifying (6) and (7) can be in tension. Not all the possibilities ofan object can be realized simultaneously; and obviously the realization ofsome of them may preclude the realization of others. Equally sinceinvestigation requires significant material and social conditions, theinvestigation of one class of possibilities may effectively preclude that ofanother. Investigating the material possibilities of the seed may,contextually, preclude investigating the seed’s possibilities under socialrelations where it does not become a commodity; pursuing wide-rangingunderstanding of the seed may, contextually, conflict with attempting tograsp its place and its possibilities under a desired set of social relations(Chapter 8). Where there is such tension, a balance is to be sought. Priorityis not automatically granted to (6); it may be subordinated (Chapters 6, 7and especially 8). Thus, while full understanding may draw freely from theresults of wide-ranging inquiry; it may query the general significance of thedrive, that tends to motivate scientific institutions today, to keep expandingwide-ranging understanding (Chapter 10).

Such tensions show that full understanding is necessarily bounded. Oftenwe cannot aspire to identify all the possibilities of the objects ofinvestigation. Then one must choose which class(es) of possibilities toinvestigate. (Investigating only the material possibilities represents achoice.) The choice will reflect one’s place in prevailing social relations and,no doubt, one’s values. The ideal behind full understanding pushes for the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (113)

recognition of multiple possibilities, including “lost possibilities”(Chapter 8), for the multiple-facetness of all objects and of theconsequences of their uses and interactions, and for the fact that theachievement of any possibility can always be described with differentcategories, where the different descriptions will be more or less pertinent tothe valuation of the achievement. In the case of the seed, augmenting thecrop yield is (contextually) the same achievement as turning the seed into acommodity. While fullness is bounded, full understanding aims for anarray of characterizations of objects that is sufficient to enable reflectivevaluations of the possibilities of objects that become grasped in the courseof the investigation.

Scientific (systematic empirical) inquiry

A number of objections have been made to my account of fullunderstanding. It has been said that science is, by definition, conductedunder the materialist strategies,7 so that the discussion of fullunderstanding has little to do with science. It is, however, also of thenature of science to involve systematic empirical inquiry. My point is thatthe latter need not be conducted exclusively under the materialiststrategies. Whether or not we call other efforts to engage in systematicempirical inquiry “science” is of little moment. The question of why are thematerialist strategies adopted still remains; it just gets turned into: whyengage in “scientific” rather than some other form of systematic empiricalinquiry? In order to maintain a reasonably concise terminology, I will treat“scientific inquiry” as equivalent to “systematic empirical inquiry,” so that‘modern science” (conducted under the materialist strategies) is consideredone approach to scientific (systematic empirical) inquiry. Then, I will use“theory” to refer to any “systematic empirical body of posits,” so that abody of traditional knowledge about varieties of plants and their ecologicalrelations will be called a theory, as well as a mathematically formulatedtheory of physics and a narrative of the evolutionary development of hom*osapiens.

A second objection is that, although objects are multifaceted, we havelearned as modern science has developed that the most effective way toinvestigate them is from the perspectives of a variety of autonomousdisciplines, each of which investigates the possibilities of objects from itsown particular perspective, so that, for example, in the case of the seed it isentirely proper to separate the possibilities of higher yields and theeconomic and sociological possibilities. While there is a large measure oftruth in this point, it ignores the tension between identifying (6) and (7)discussed on p.99: the social conditions for realizing the possibilities forhigher yields from the hybrid varieties may eliminate the social conditionsin which the seed might not become a commodity. It may not be possible to


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (114)

investigate the “two” possibilities separately. In the concrete context of thegreen revolution, for example, the “two” possibilities may be realized inone and the same achievement. Then it would not be appropriate ordesirable to attempt to demarcate the material possibilities from the fullrange of possibilities; rather the explanation of the achievement shouldreflect the inseparability (in the specified social context) of the economicand the materialist causal factors9.

A related objection is that I have conflated the distinction betweenfundamental and applied research, and that full understanding isappropriate only in the context of applied research. It is true that appliedresearch is unintelligible outside of its social and ecological locatedness.However, when conducted under the materialist strategies it tends not todeal adequately with the multifacetness of things, and in particular to thecontextual incompatibility of realizing some different classes ofpossibilities, and thus to the issue of “lost possibilities” (Chapter 8). Itasks: How can we realize the material possibilities of things that we havediscovered in useful ways (in general or in service of a specified goal)without also generating undesirable side effects? The side effects consideredgenerally are those that themselves can be understood as generated fromunderlying law, structure and process; others tend to be considered in an adhoc way. It does not tend to investigate the effects of applications onprevailing social arrangements (though in recent years there has been muchpressure to investigate “environmental impact”), and to ask, for example:what possibilities do we need to identify in order to strengthen a certainsocial arrangement, to preserve an ecological system, and to avoidundesirable consequences for human lives (Tiles 1987)? Asking questionsof these kinds we presuppose neither that there is a sharp separationbetween fundamental and applied research, nor that wide-rangingunderstanding (alone) will be able to play a useful role.

Full understanding abstracts the science neither from the sociology andthe ecology, nor from the practices and institutions that generate thescience. Full understanding thus has a critical component as it addresses thedominant scientific practices, and a positive investigative component that isexplicitly informed by social values, posing such questions (that thetradition of modern science usually presumes to be settled) as: Is wide-ranging science genuinely context free? And, can the material possibilities ofspaces be explored and charted generally when one abstracts from thesocial arrangements and practices that shape these spaces?


I suggested on p. 99 that adopting the materialist strategies makes clear senseif one considers the objective of science to be given by O1 (to representphenomena as generated from underlying structure, process and law). Why


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (115)

adopt O1 as the objective of scientific (systematic empirical) inquiry? Well,one might say: if not O1, then what? Here is an alternative proposal:

O The objective of science is to gain understanding of phenomena.This includes to encapsulate (reliably in rationally acceptabletheories) possibilities that are open to a domain of objects, and todiscover means to realize some of the hitherto unrealizedpossibilities.

O is more encompassing than O1 So, too, are the following:O′ The objective of science is to provide (in rationally acceptable

theories) a literally true account of what the world is like.O″ The objective of science to provide (in rationally acceptable

theories) the best explanatory account of natural phenomena.

Adopting O1, I indicated, may be partly motivated by Galilean (materialist)metaphysics. O, O′ and O″ may not. They can be interpreted to mean moreor less the same thing, but they connote different images. O′ suggests thatthe objective is to gain a detailed, accurate map of the spatio-temporaltotality that is the world. O suggests, rather, that the world be consideredan inexhaustible well of possibilities to be probed over and over again, butnever exhausted. O is motivated by emphasizing that what is the case(what has been, and what is actually realized, and the projection into thefuture of its regularities and structures) cannot be identified with what canbe. There are possibilities that fall outside of currently actualizedregularities and structures, so that an empirical charting of what is, and ofwhat its regularities are, does not suffice to sum up what can be or whatgenuinely could have been. The real is not exhausted by the actual; itincludes also the genuinely, as distinct from merely logically orimaginatively, possible (Bhaskar 1975; Lacey 1997c). O″, unlike O′, makesclear that explaining as well as describing phenomena lies within thepurview of science and it is not tied to any one explanatory ideal.Moreover, being able to explain the current condition of phenomena is anecessary condition for encapsulating reliably the possibilities that may beopen to them, including the novel possibilities that may remain hiddenwhen we attend only to current regularities. O contains O″. Both because ofits generality and because the links with values (that I will discuss) drawattention to alternative classes of possibilities, I will consider O as thesalient alternative to O1.

O is more general than O1, but following O1 is consistent with followingO. Following O1 we do encapsulate possibilities, but only those that can berepresented as generable from the underlying order. Following O beyondwhere O1 takes us, we are open also to those possibilities that can only bedescribed when we do not abstract objects from their human, social andecological contexts. When our focus is upon spaces in which human agency


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (116)

is not relevant as a causal factor (Chapters 6 and 10), the success ofmodern scientific inquiry suggests that O reduces to O1; so O retainscontact with scientific practice that accords with the Galilean idea.Departures from O1 tend to concern the realm of daily life and experience,and currently to be important principally in the human sciences and inthose areas of biology that impinge upon human capabilities and socialarrangements (Chapters 8 and 9). One is drawn to adopt O when onekeeps in mind that scientific knowledge is there to be applied, accepted forthe end of informing practical projects as well as cognitively; and when oneasks: What kinds of systematic empirical understanding are needed toinform all of the great variety of human projects?

Different approaches to science

Adopting O permits that there may be systematic empirical inquiries inwhich relevant data are not selected to fit the materialist strategies. So,under O, fitting the constraints of the materialist strategies will not beconsidered a cognitive value. Indeed, if we adopt O, we might consider O1as the objective of a particular (very important) approach to scientificinquiry—the materialist or the Galilean/Baconian approach-rather than theobjective of science as such In the next chapter, I will justify defining thisapproach in terms of a link with social values, the “modern values ofcontrol” (Baconian utility) by arguing that O1 bears a close affinity with anobjective that can be put as follows:

O1′ The objective of the Baconian approach to science is to encapsulate(reliably in rationally acceptable theories) the possibilities of adomain of objects that would serve well the interests of the modernvalues of control, and to discover means to realize some of itsh*therto unrealized possibilities.

Parallel to this, there could be defined a class of approaches to scientific(systematic empirical) inquiry, where each approach (Oi) bears a closeaffinity with an objective which is an instance of the following schema:

Oi′ The objective of the (…) approach to science is to encapsulate(reliably) the possibilities of a domain of objects that would servewell the moral/social project (…), and to discover means to realizesome of its hitherto unrealized possibilities.

Filling in the “…” with “grassroots empowerment” we get O2′, linked withan approach (O2) whose strategies involve links with traditional, localknowledge in impoverished countries (Chapter 8); with “feminist” we getO3′, linked with an approach (O3) that deploys strategies that constrain


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (117)

theories in the direction of “complexity, ontological heterogeneity,interaction and non-reductionism” (Chapter 9). In general, which strategiesare adopted will follow readily from the approach to science adopted.12

Not only O1, but also all of Oi (in principle) may define concreteperhapscomplementary—ways to further O, where in each approach a different classof possibilities is investigated. I am tempted to replace “possibilities” in Owith “the full range of possibilities.” Then O1 and O2, for example, coulddefinitely be seen as complementary. But the pursuit of a particularapproach, such as O1, has material and social conditions, which mayeffectively preclude the pursuit of another approach, such as O2. Forinstance, the exploration of one range of possibilities may effectively, in thelight of necessary material and social conditions, preclude the explorationof another (Chapters 7 and 8). The perceived complementarity may only beabstract and not open to effective realization under concrete socialcirc*mstances.

Why adopt the materialist strategies?

O is a very general encompassing objective, a critical reference point fromwhich to discern the partiality of approaches that deploy particularstrategies, but not a positive source of research guidance. Only theadoption of a strategy provides positive direction to research, but O doesnot guide us towards any particular strategy. The materialist strategies arewidely adopted, and it is often held that they ought to be adopted virtuallyto the exclusion of other strategies, that somehow the objective O is bestfurthered by following the approach O1 Why? Why is research conductedunder the materialist strategies considered exemplary? Why is inquiryaiming to encapsulate the material possibilities of things consideredexemplary? These questions need to be answered by the scientificcommunity at large and the body of its institutions, and not just byindividual scientists.

Historically three kinds of answers have been given to these questions.The first appeals to materialist metaphysics: the world really is such thatall phenomena are generated from underlying structure, process and law.Thus, a theory that soundly represents the underlying structure, processand law of a domain of phenomena will ipso facto represent relevantly itscausal structure, and it will suffice to encapsulate its possibilities for they willbe exhausted by the material possibilities of the domain (Chapter 6).

A second answer is that adopting the materialist strategies serves theinterests of Baconian utility: understanding gained from following them,and effectively only them, includes that which enhances the humancapability for exercising control over nature. A variant of this would bethat it provides the kind of understanding needed to grasp the dominantobjects and leading practices of modern practical life and lived experience,


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (118)

and thus to live effectively in the modern world. To account (rationally) forthe virtual exclusivity of adopting O1, we must add that the interests ofBaconian control are generally compelling and in conflict with the valuecomplexes implicated in the definition of the other approaches.

The sheer intellectual interest of theories that represent underlying order,combined with the fact that we have a very successful “track record” inestablishing theories of this kind, is sufficient for a third answer. Theories,gained under these strategies, often manifest the cognitive values to a highdegree with respect to a broad array of experimental data, supplementedby the data of successful practical applications and of some naturalphenomena. Here, one might say, the stock of knowledge is increasing.Moreover, there are good reasons to hold that following the samestrategies will enable the stock of knowledge to continue to grow; and, atthe present time, there do not appear to be plausible alternative strategiesto explore if we wish to add to the stock of knowledge, especially if wewish to continue to investigate certain kinds of phenomena (for example,astronomical or optical ones). O1 defines “the only game in town.”

This answer may be reinforced by arguments of Putnam (1981, 1990)that the activities and virtues involved in the gaining of scientificknowledge are partly constitutive of human flourishing; that commitmentto the cognitive values is “part of our idea of human cognitive flourishing,and hence is part of our idea of total human flourishing, of Eudaimonia”(Putnam 1981:134). Then, we adopt O1 for want of alternative ways topursue O in the institutions actually available for the conduct of inquiry inadvanced industrial societies. How powerful the reinforcement is needs tobe assessed by balancing Putnam’s claim with investigating thecontribution of science to “humanitarian” ends (Feyerabend 1979) or to“wisdom” (Maxwell 1984), and in the light of raising the question ofwhether the kind of human flourishing that Putnam has identified can beachieved in a way that is compatible with furthering social justice. It mightbe further reinforced by combining Putnam’s argument with Kuhn’s(1970) analysis that historically an old paradigm has (must have?) a uniquesuccessor; if we want to play the “game” of science we would expect thatthere would be only one “in town” (Chapter 7).13

My own view is that the best answer is provided by a suitably refinedBaconian one—the second with important input from the third (cf. Tiles1987). It is difficult to separate the second and the third answers (even ifappeal to neutrality supports that they are distinct), for satisfying theintellectual interest also tends to satisfy the interests of Baconian utility;and social institutions support the projects of O1 principally because oftheir contributions to Baconian utility. Pragmatically, especially in theawareness of the individual researcher, some version of the third answer isusually stated in support of the adoption of O1. Nevertheless, it is notsufficient. At best it is an argument to consider O1 worthy of adoption; but


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (119)

it does not provide support for not adopting some other Oi (at least bysectors of the scientific community), even if the practices of that alternativehave still to be developed in detail and the institutions that might supportthem still have to be created. Being the “only game in town” is not a goodreason not to attempt to bring other games to town, unless there are goodreasons why they should not be brought to town. The absence of analternative “scientific game” in contemporary universities and researchinstitutions might reflect only that currently hegemonic values have ensuredthat the necessary material and social conditions for development havebeen denied (through structurally maintained mechanisms) to alternativeapproaches that deploy different strategies. The “only game in town”argument is not value neutral if the lack of alternatives is a consequence ofdenying the conditions necessary for an alternative to develop. That it is notvalue neutral, as such, does not imply that it is a bad argument; it dependsupon whether the relevant values can be defended.

I suggested that, whatever the motives of individual scientists may be,societal institutions provide conditions for scientific research because of itscontributions to Baconian utility. It is quite common to say that it is theobjective of science to understand “the world we live in” or “the worldaround us.” These phrases tend to be used ambiguously. Sometimes theyare intended to refer to the underlying material order and vast materialcomplexity of things in which our lives are located, evoking wonder,puzzlement and perhaps a sense of our insignificance in the overall order ofthings. But “our world,” the world we experience in day-to-day living, hasalso been shaped historically, so that most of the objects we deal with in dailylife and experience bear the mark of human history. Key objects that weencounter in the course of daily life exist (causally) because of their role inimplementations of social values, such as Baconian utility and the socio-economic values with which it tends to be co-manifested. Thus, thepresence of objects whose behavior is generated from certain specific kindsof underlying law, structure and process—or the underlying order that ispertinent to the general conduct of our lives—will sometimes in part be theoutcome (causally) of the adoption of social values; so that the full range ofpossibilities open to these objects cannot be identified without reference tothe social values.

These objects are simultaneously objects of materialist understandingand objects of social value. Any ready extrapolation from the investigationof them to objects of daily life and experience in general runs the risk ofimperceptibly importing this social value into other contexts. It can be agood reason to adopt O1 (at least as a subsidiary objective), that keyobjects and practices of the world of our lived experience requirematerialist understanding. It does not follow that values are not crucial inmaking that adoption, for the possibilities open to the objects are notexhausted by those realizable where these values are predominant, or those


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (120)

that can be predicted under current social constraints. The “world” ofdaily life and experience is always structured (where the structures havecausal implications) in ways that partially represent the embodiment ofsocial values. But the possibilities open to the objects found in this “world”extend beyond those realizable (and perhaps predictable) under theprevailing (or any other) structures. There may be powerful institutionallimitations upon investigating possibilities that could not be realized withinprevailing structures, precisely because these structures do embody highlycertain social values. To confine investigation just to the possibilitiesrealizable within these structures is, therefore, implicated in values(Chapter 8; Lacey 1997c).

I will motivate my view by exploring in the next chapter the relationshipthat exists between following the materialist strategies (adopting O1) andthe distinctively modern attitude towards the control of nature.

Does adopting the materialist strategies represent acognitive value?

Where does this leave the question of whether or not fitting the materialiststrategies represents a cognitive value? Certainly satisfying the constraintsof the strategies commonly functions as a criterion of theory choice(typically as a ground for rejecting certain theories) and, if O wereconsidered identical to O1, it ought to. One way to regard it may not be asan additional cognitive value (since it cannot be grounded in O byitself), but as a “constitutive value” of the approach O1. This suggestion isreinforced by observing that the materialist strategies operate as a criterionof theory choice at a different level from those items that I have listed ascognitive values. Antecedently, it constrains the class of acceptable theoriesand selects the class of relevant data; then the cognitive values play theirrole in choosing among the theoretical candidates. As such a constitutivevalue, it may be considered grounded either in the general features of thechosen object of interest for science (underlying structure, process andlaw), or in the interest of Baconian utility to explore only the materialpossibilities of things.

Looked at in this way, the materialist strategies function as a criterion oftheory choice in virtue of an interest in the underlying law, process andstructure of things14 and in their material possibilities. They define the classof theories and the classes of possibilities of interest. That interest, and thesocial values that nourish it, has no implications regarding the specifictheoretical posits investigated and confirmed, and regarding the specificmaterial possibilities that are encapsulated. After the play of the strategies,the play of the cognitive values sorts out such specifics in the light of thedata that are selected and obtained. Indeed, values could have noimplications in regard to these matters. Neither can values even ensure that


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (121)

adopting a strategy will lead to success in generating theories that manifestthe cognitive values highly. No matter how much interest one might havein encapsulating a kind of possibilities, there is—prior to the outcome ofthe research—no assurance that there are any such possibilities to beencapsulated. Thus, the great success of generating theories under thematerialist strategies that manifest the cognitive values highly vindicates -after the fact—that adopting these strategies indeed is partly constitutive ofan approach that contributes to realizing O; but the objective of realizing Ois not served uniquely by following O1—and thus I′ cannot stand as asatisfactory explication of impartiality. Different values could lead tointerest in different (more encompassing or intersecting) classes ofpossibilities, the investigation of which might require strategies other thatthe materialist ones. As long as the role of values in theory choice is playedout in the context of the definition of the classes of possibilities of interest(and thence at the level of choice of strategies) and does not extend to thechoice of specific theories, nothing paradoxical or logically problematicneed arise.

Nothing follows a priori from reflection on the objective O regardingwhich strategies might contribute to the further realization of it. Anything,in principle, might be tried, for only after the fact, after having attemptedinvestigation under a strategy, does it become apparent whether adoptingthat strategy serves to further O. It is a condition upon a strategy thatcan be partly constitutive of an approach to furthering O that it actuallyproduce theories that manifest the cognitive values highly. Thus, forexample, we no longer attempt to follow the strategies of Aristotelianscience (Chapter 7).15

Strategies as constitutive values of approaches to inquiry

The cognitive values are the features desired of theories in virtue of theirbeing generated for the sake of furthering the objective of scientific(systematic empirical) inquiry, O. There remains dispute about what theexact list of cognitive values is. I highlight: empirical adequacy,explanatory power (both wide-ranging and full), power to encapsulatepossibilities, internal consistency, consonance, source of interpretivepower, and rejection of ad hoc features. Agreement on the list is notimportant to the argument that follows for the remainder of the book, butthat fitting a particular strategy is not a cognitive value is central. Fitting aparticular strategy, however, is a constitutive value of an approach tofurthering O, which typically plays its role (logically) prior to the roles ofthe cognitive values being played. The cognitive values involve anassortment of desiderata for theories, not all of which may be able to bemanifested highly in the same theories. Often, for example, there will be a“trade-off” between wide-ranging and full explanatory power. How the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (122)

cognitive values are to be ranked may vary with the strategy adopted, aswill also the interpretation of empirical adequacy, since its meaningremains unspecified until a class of empirical data has been selected(Chapter 10).

The cognitive values do not suffice to pick out any one approach,orprior to the conduct of inquiry—to identify those approaches that cancontribute to furthering O. Every scientific (systematic empirical) inquiry isconducted within a particular approach, partly constituted by the adoptionof particular strategies. I will argue that, generally, settling on a particularstrategy is linked with its mutually reinforcing interactions with particularsocial values, and that the conditions of realizability of the possibilitiesidentified under the strategy include social (institutional) structures thatembody these values. This point can be obscured where the social values inquestion seem to be “obvious,” or (as in the case of Baconian utility) arepart of the deep self-understanding of a culture, for then the highmanifestation of these values may readily be taken as a natural (universal)rather than as a historical (culturally specific) phenomenon. The obscuringcan be even greater when there are no ongoing counter research programsthat offer concrete results challenging such alleged universality. In the nextchapter I will argue that adopting the materialist strategies (followingO1) involves mutually reinforcing interactions with the modern values ofcontrol. Later (Chapters 8 and 9), I will illustrate other strategies and thesocial values with which they enter into mutually reinforcing interactions.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (123)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (124)

6The control of nature

While it is part of human nature to exercise control over natural objects, inmodern times exercising control has gained distinctive features (Leiss1972): its extent, its pre-eminence and its centrality in our lives, the highand virtually unsubordinated value granted to it, the dissociation ofconsiderations of control from those of the meaning and value of ouractivities and social arrangements, the intense efforts to expand andimplement our capabilities of control, and the conviction that these effortswill be at the heart of projects to meet human needs and wants even astheir embodiments continually generate new needs and wants.Consequently, certain values connected with the control of nature rankespecially highly in modern value complexes. I will argue, as anticipated inthe previous chapter, that the nearly unanimous adoption of materialiststrategies in modern scientific practices becomes intelligible largely in virtueof its mutually reinforcing interaction with these values.


By nature, human beings are reflective, “embodied and active in the world”(Taylor 1982:101); they are also social and cultural beings. They are agentswhose interactions with material objects and other human beings requireintentional explanation, in which action is portrayed as following from anagent’s beliefs and objectives, and thus understanding. We exercise controlover objects when we deliberately and successfully, informed by our beliefsabout them, submit them to our power and use them as means to our ownends. Not every effective intentional interaction with the world is aninstance of control. Perhaps in some cultures few of them are, but in allcultures under some conditions control is practiced and valued. Control iscontrasted with such stances as reciprocity, mutuality and respect, wherethe value of the object interacted with reflects a measure of integrityaccorded to it, and is not reducible to instrumental value for the agent.

The exercise of control is obviously served by practical understanding,understanding of the effects of our actions on things and their effects onus. There exists systematic empirical practical understanding in all cultures,

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (125)

as well as the ongoing interplay between interaction with material objectsand practical understanding. Successful interaction provides the decisivetests of this understanding, whose forms may vary, reflecting differentways in which interaction with material objects may be related tointeractions with other beings.

Interaction with nature may be circ*mscribed by its fit with a social,ecological or cosmic order, and by a particular conception of humanflourishing. It may, for example, in a given culture, take place withinnatural rhythms, with a limited set of ends and means defined bytraditional practice, where the tradition ensures that, save for unexpectedcirc*mstances, there is harmonious and reciprocal interaction with nature.Then, human control of the natural environment is balanced bynourishment and maintenance of it, so that human relationships with andwithin it can be permanent, and the preservation of the environment setsbounds to acceptable ends. Such a constancy, punctuated only rarely byperiodic or occasional variations, can provide the basis for a stable socialorder where there is deep interlocking of social and cosmic visions. Wherethis is the case, the form of practical understanding will reflect howinteraction with nature contributes positively or negatively to the desiredorder, and it will explore the possibilities of nature in relationship to thosethat this order allows. It will be constrained to grasp things with categoriesrelated to the social, ecological and cosmic order, and to serving theparticular ideal of human flourishing pursued in this order. Where suchconceptions and practices are present, the distinctive human stancetowards nature is well captured with notions such as attunement,adaptation, harmony, and participation. Control is subordinate to theserelations, and limited in scope, valued only to the extent that it contributesto ends circ*mscribed by the desired social order and ideal of humanflourishing. Exploring the possibilities of control beyond these bounds hasno moral (or rational) intelligibility. Important echoes of these conceptionsare to be found in certain ecological, feminist and third world popularmovements today (Chapters 8 and 9).

The place of control in modern value complexes

The distinctive modern attitude towards exercising control over thingsrejects in general the subordination of control to other stances towardsnature and to particular social values and ideals of human flourishing. Itlooks for the regular expansion of the scope of effective controlsthroughout the activities of practical life. That expansion has occurred sosuccessfully that practical life has become shaped pre-eminently in thecourse of implementing novel and far-reaching possibilities of control,principally by way of technological advances. Whether they are connectedwith energy, transportation, medicine, agriculture, communications, or


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (126)

education, practical problems and social questions increasingly arebecoming considered to be properly open primarily to technologicaladdress. Consequently the realm of daily life and experience has becomedominated by the products of our exercising control over things, and itssocial institutions transformed and adapted to accommodate and serve theresulting forces, needs, wants and interests of practical life.

Although an historical fact of major importance, the enormousexpansion throughout modernity of the successful control of natural thingswas not inevitable; and, despite the current “realities” of the globalizationof the market, reasonable aspirations for the future need not be limited tothose that depend upon its continued expansion (though they must contendwith the social forces that engender it; Lacey 1997c). Gaining control overmaterial things has become a very highly rated social value—notsubordinated in a general way to any other values, but also notunambiguously granted ascendancy over them. Where its interests andthose of other prominent social values clash, there is no systematic way toallocate priority or to define compromises. Sometimes control is taken tobe a value for its own sake, a power whose exercise is the exemplaryexpression of human rationality. More commonly and more defensibly, viatechnological advances, it is taken to be able to serve all sustainable socialvalues and ideals of human flourishing, and to serve to enhance humanwell-being in general and in the long run.

Thus exercising control over things has come to be considered largely inabstraction from links with other values, and efforts to further itsexpression can proceed with relative autonomy so that matters of viableand desirable social arrangements, and meaningful cosmic order, havetended to become subordinate to the value of control. Not every instanceof exercising control is valued, both because human well-being remains arelevant evaluative standard and because in some instances it may conflictwith other highly rated social values, for example, those of the market.Consistent with this, it remains that exercising control over things istaken to be the key to enhancing human well-being. It is a centralorganizing principle of modern society, a principal way to approachproblems, accompanied by a confidence that further developments of ourcapabilities to exercise control will be able to deal effectively with any newproblems and undesired side-effects that its exercise might create, as well asto introduce hitherto unthought of possibilities. Implementing a newtechnology may never be entirely free from controversy in the realm ofvalues concerning, for example, whether its consequences will threatencivil liberties or produce environmental devastation. Nevertheless,modernity has unfolded with the confidence that risks are worth takingwith technological advances, and that any problems that arise can and willbe taken care of as the technological project itself progresses.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (127)

In the modern viewpoint technological advances as such, rather than, forexample, the social relations under which technologies are implemented,are seen as the key to enhancing human well-being. That partly explainsthe acceptance of disruptions of social relations, and the imperative toconstruct new social arrangements, following the movements andimplementations of technology. This gets at the sense in which exercisingcontrol over things has become considered as a value unsubor-dinated toother social values. The most distinctive modern value concerned withcontrol, however, seems to be that of expanding the human capability toexercise control over nature.

The modern values of control

Values are held in integrated structured complexes (Chapter 2), theconcrete manifestations of the items of which tend to reinforce one other.Modern value complexes include a set of distinct values about control,which I will call the modern values of control. Among the modern valuesof control, expanding human capabilities to control material objects rankshighest. Exercising control over material objects (as a characteristic activityof practical life so that, wherever possible, problems become redefined ashaving a technological solution) and especially implementing novel forms ofcontrol also rank high among these values. Thus technological objects andtheir products tend to be considered objects of value, at least some of themfor some amount of time; and natural objects tend to be considered objectsof value largely for their instrumental value. These values being held in agenerally unsubordinated place, so that to their concrete manifestations theprojects and institutions which express competing values must in largemeasure defer and adapt, may also be regarded as among the modernvalues of control. They gain reinforcement from the fact that they tend tobe manifested in the same institutions as other social values (for example,private property, the market, and expanded options for choice) that arehighly rated in modern value complexes; their manifestation also reinforcesthat of these other values.

The interests connected with the modern values of control can clearly befurthered by a form of understanding that enables us to encapsulatesoundly what are the possibilities of control and the means to realize themin a way that abstracts them from their connection with lived experience,practical life, social arrangements and ecological and cosmic structures.This kind of encapsulation enables us to deliberate about control withoutbeing encumbered by considerations related to other social values, and toseparate the question “Can it be done and how might it work?” from“What value does it have and is it worth implementing?” If thesepossibilities are to be implemented, however, they must ultimately besoundly represented also as functions of variables that can be directly


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (128)

manipulated by our action, under conditions which are within our powerto bring about or to maintain, or which we have reason to accept will bepresent and maintained because of the way the world (or society) is. Thelimits of possible control will be inseparable from the limits of thisunderstanding. These limits, while they depend on human ingenuity ininvestigation and interaction with the world, cannot go beyond those set bythe world. The world has been amenable to the expanded exercise ofcontrol that we have seen in modern times. But that leaves open whetherthe world may yet impose limits to the expansion; and under what socialconditions such expansion may or may not occur. It also leaves openwhether the expanded capability to control things (and its accompanyingform of understanding) provides the key to enhancing human flourishing inall of its ideals under prevailing historical conditions.


Modern Western culture understands itself as the foremost bearer ofrationality, and this self-understanding rests upon the twin pillars ofscience and technology. From one perspective, that in which knowledgeclaims are recognized as the primary location for rational evaluation,science looms larger. In its light the posits of theories developed under thematerialist strategies, more than any other form of understanding, gainsupport from rational evaluative canons. They offer the best account wehave of the nature and ways of material things, and consequently providethe theoretical underpinning of technological success and advance. Fromanother perspective, that in which rational evaluation pertains in thefirst instance to actions designed to enhance the exercise of our designsupon the world, technology is in the foreground. Then science, conductedunder the materialist strategies, gains rational precedence as a form ofunderstanding because it provides the theory that furthers technologicalpractice. Movement back and forth between the two perspectives is easyand common, since they are mutually reinforcing: the hegemony oftechnological practice is often grounded on its being informed by(materialist) scientific theory which is said to offer a superiorunderstanding of the world; the virtual hegemony of research under thematerialist strategies and the massive social and material investment in itare often legitimated in terms of its contribution to technologicaldevelopment. Either way the appeal to modern science carries anunmatched authority.

There is, in modernity, a mutually reinforcing relationship betweengaining understanding under the materialist strategies and adopting themodern values of control, marked by mutual dependence, many sharedinterests and conditions, intertwined causal dynamics, virtual inseparability


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (129)

of the social appeal and force of one from that of the other, considerableoverlap in the institutions in which they are pursued—but less thancomplete identity. The materialist strategies and the modern values ofcontrol came into historical prominence together early in the seventeenthcentury, but they had distinct, earlier anticipations. Moreover, in theseventeenth century Bacon, expounding an early version of the modernvalues of control, advised (though he did not always practice) the use ofinductive rather than materialist strategies; and Descartes, an earlyexponent of materialist strategies, did not justify their cognitive or rationalmerits (as distinct from their social importance) by appealing to their linkswith control. As ideas they are distinct and their historical dynamics have ameasure of independence. Thus, for example, not every theory that issoundly accepted of a domain of phenomena under the materialiststrategies can inform the expansion of our capability to control nature; andfundamental theories provide understanding of some phenomena thatbelong neither to technological nor to experimental spaces, those in whichparadigmatically we exercise control over natural objects. Conversely, notevery technological innovation reflects the input of (materialist) scientificunderstanding.

While the materialist strategies and the modern values of control aredistinct, in the modern context it is difficult to separate them. As we willsee, following the materialist strategies in research also contributes to thedeeper manifestation of the modern values of control; and adopting thesevalues with commitment motivates and depends upon furthering researchunder these strategies. The contributions are variegated and go both ways,not in every individual case, but as a solid, constant pattern, and notalways in ways that are foreseen and antecedently intended. The strategiesand the values have mutually reinforced each other so effectively that theirinterests largely have become identical under concrete historical conditionsthat make their simultaneous and reinforcing presence highly likely inleading social institutions.

The mutually reinforcing interaction between the materialist strategiesand the modern values of control was anticipated by Bacon:

I am laboring to lay the foundation…of human utility and power….For the end which this science of mine proposes is the invention not ofarguments but of arts;…the effect…being…to command nature inaction.… nor can nature be commanded except by being obeyed. Andso those twin objects, human knowledge and human power, do reallymeet in one… [The Great Instauration]…now their understanding isemancipated…; whence there cannot but follow an improvement ofman’s estate and an enlargement of his power over nature [The NewOrganon].

(Bacon 1620/1960:16–9, 29, 267)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (130)

It may be summed up in the following propositions:

1 The furtherance of the modern values of control is dependent on theexpansion of understanding gained under the materialist strategies.

2 The pursuit of materialist understanding fosters an interest in the fullermanifestation of the modern values of control.

3 Materialist understanding is gained from the perspective of control.4 Any values furthered by the pursuit of materialist understanding (for

example, those associated with “fundamental” research) aremanifested today as part of value complexes that also include themodern values of control.

In the next four subsections respectively I will elaborate these propositions;and then, in the following one, I will discuss links between the mutuallyreinforcing interaction and materialist metaphysics.

The modern values of control and the need for materialistunderstanding

A value complex cannot be manifested to any significant degree inpractices and institutions unless the world is a certain way; in particular, themodern values of control cannot be manifested unless the world isamenable to being controlled by human action. What must things be likeif they are to become objects of possible control for us? How should weaim to understand them if we wish to encapsulate extensively thepossibilities of control of things open to us?

Suppose that we want to act upon an object, X, in order to bring about astate of affairs, S, where S is characterized abstracting from its place inhuman experience and practical life. To do this we will need to possesscertain kinds of knowledge and certain skills and abilities. Concerningwhat we need to know, the following requirements (requirements forcontrol) must be satisfied. We can identify conditions, (C1, C2,…, Cn), suchthat given them: 1) S’s occurring can be represented as a function of X’sgaining a property, P; 2) we can make X gain P by direct action; 3) each Ciis such that either a) its occurrence or maintenance is within direct humanpower, b) it can be controlled by a similar but independent process to thatinvolved in bringing about S, or c) we have good reason to accept that it isa standing condition in the context; 4) S’s occurring being a function ofX’s gaining P may reflect a lawful connection (or an empirical regularity)between S and X’s being P, perhaps mediated by several lawful connectionswhich may only be recognizable when underlying structure and process areconsidered, or that X’s becoming P initiates a process that, barring furtherintervention, will eventuate in S.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (131)

S’s occurring is within our capabilities of control if there are objects thatwe can manipulate in accordance with some set of requirements forcontrol; and X is within these capabilities if there are states of affairs thatwe can bring about deliberately by manipulating X. The requirementsinclude knowledge of regularities: given C1, C2,…Cn, X’s gaining P willeventuate in S; and identifications of what one can bring about by directaction, the immediate effects of one’s bodily movements. The latter dependon personal know-how and ingenuity, cultural traditions, and the materialand social organization of one’s society. Only those establishedregularities, in connection with which we have the relevant control over theCi and direct control over X becoming P, occur in sets of knownrequirements for control. Whether or not a regularity can inform ourcurrent capabilities to control material objects, however, it can alwaysserve as a focus around which to explore extensions of these capabilities byindicating relevant conditions and things which, if brought under ourdirect control, would enable us to control further specified states of affairs.The totality of established regularities, thus, represents the limit of ourcapability to control things at a given time; so that a form of understandingthat enables us systematically to derive and subsume regularitiesencapsulates (in the limit), by representing how to bring them torealization, the possibilities of things insofar as they are under our control.

Within the many forms of traditional knowledge, various sets ofrequirements for control have been established, so successfully that, forexample, they informed the practices that produced the varieties of seeds,without which many current developments in agricultural technologywould be impossible (Chapter 8). They have been derived inductively in thecourse of engaging in practices and skillfull activities in a familiar localewhose particularities, regularities and complexities of relationship havebeen charted through repeated observation and carried through thegenerations; but these sets have been limited in number, focus,systematicity and generality.

Understanding gained under the materialist strategies generates anexpanding array of regularities with ongoing novelty of focus. A regularitymay be obtained, directly in the course of experimental investigation, byinduction from phenomena observed in experimental situations; fromposing questions about how to extend our current powers to control; ormore commonly by derivation from consolidated posits about underlyingstructure, process and law, posits which serve to explain, under thespecified conditions, the connection between X becoming P and S. Often,therefore, objects as grasped under the materialist strategies are grasped inthe way they need to be grasped to be included among those that lie withinour capabilities of control, and they become objects of control mosteffectively in experimental and technological spaces, those spaces in whichwe initiate events under boundary conditions which close off other


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (132)

interferences so that the consequences of the initiated event all, as it were,flow out of the underlying order.

So if things are, or can become like, the way they are represented underthe materialist strategies, they can become objects of control–provided thatwe can directly manipulate the relevant initiating events of regularitiesencompassing them, and ensure that the relevant boundary conditionsremain in place, either through our own direct control or because they arestanding conditions of nature as attested to in soundly accepted theories.The more general the laws are and the more wide-ranging the soundlyaccepted theories, the greater are the number and range of regularities, and(provided appropriate social arrangements prevail) of sets of requirementsfor control, which can be expected to be derived. I leave aside for nowwhether or not the sets of requirements for control established intraditional forms of knowledge can all be rearticulated within materialistunderstanding, and whether or not traditional forms can be developedthrough research so as to generate further such sets (Chapter 8). Whateverthe case may be, materialist understanding leads us to sets of requirementsfor control that far transcend traditional constraints, so much so that inmodern practical life, shaped as it is by the modern values of control, it hascome about that objects, insofar as they are to become objects of control,tend to be considered as objects of materialist understanding.

Thus, holding the modern values of control brings with it an interest inthe pursuit of understanding under the materialist strategies and of noother forms of understanding of material things. Under the materialiststrategies we come to grasp material possibilities of things, which include(but go beyond) their possibilities as objects of control. Sinceunderstanding gained under the materialist strategies is wide-ranging(Chapter 5), and the ranges of which it provides understanding includephenomena that are within our direct control, grasping more of thematerial possibilities of things leads almost inevitably to grasping more ofthe possibilities of things as objects of control. It follows that the interestsof the modern values of control are served generally by investigationconducted under the materialist strategies (“fundamental research”), andnot just by that addressed to immediate practical problems (“appliedresearch”).

Materialist understanding (and, in the mainstream of the advancedindustrial countries, largely it alone) grasps objects as they must be graspedin order to become objects of control. Since it abstracts from therelationships of objects to experience and practical life and to the social(and cosmic) order, it also encapsulates possibilities in a way particularlywell suited for expanding our capabilities of control without subordinatingthe modern values of control to other social values. While it does informpractices that serve other social values (for example, of the market or themilitary), it does so only in virtue of particular instances of control


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (133)

becoming embedded in those practices. Under the materialist strategiesphenomena are grasped only as generated from the underlying order or assubsumed under regularities, and so they are represented as of the samegeneral character regardless of how they may be valued in the light ofvarious competing value complexes. This is a mode of representation thatis (by design) indifferent to differences of valuation, and so one in whichmoral reserve about certain practices of control cannot be expressed. Itcontributes to sustaining the sense, widespread in contemporary society, ofthe inevitability of someone (some corporation) bringing to realizationwhatever is shown to be possible (for example, connected with cloning andgenetic engineering), and thus of technological “progress” into ever morerealms of life and the ready tolerance of risks that generally accompaniesit.

Reciprocity of theoretical and technological interests

The pursuit of materialist understanding fosters an interest in the fullermanifestation of the modern values of control. This interest derives in thefirst place from the facts that theoretical developments under thematerialist strategies depend in crucial ways upon technologicalinnovations which are only available where the modern values of controlare deeply manifested, and second that where those values are deeplymanifested it is virtually assured that theories established under thematerialist strategies will be significant, needed to understand importantobjects in the realm of daily life and experience and to inform practicalactivities.

Developments under the materialist strategies depend upon theavailability of technological innovations, increasingly of the most advancedand sophisticated kinds (themselves often products of applications ofmaterialist understanding) to provide the necessary instruments andequipment to conduct relevant empirical (for example, experimental in thecase of subatomic physics, observational in the case of outer space) andtheoretical (for example, concerning computational needs) inquiries.Sometimes technological innovations are made in the first instance for thesake of furthering materialist scientific investigation. Then we may expectpractical “spin-off” (for example, from computer software developments)simply from engaging in the research efforts as well as from subsequentpractical applications (if there are any) of theories (for example, of highenergy physics) that are consolidated in the course of the research—clearwitness to the reciprocity and the dynamic interaction of the interests ofmaterialist understanding and the modern values of control.

Technological advances today not only provide essential means for theprogress of materialist understanding. They also may open up access tohitherto unknown, uncreated or inaccessible phenomena, or offer models


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (134)

(for example, the mechanical clock in early modern physics, the digitalcomputer in contemporary cognitive science), without which certainphenomena would remain intractable to investigation, and so provide theoccasion for defining new theoretical problems. In these situations,technological objects are an integral part of the research; they are amongthe objects of investigation.

Because of the dependence of its theoretical upon technologicaldevelopments, materialist investigation can be conducted today at its mostadvanced levels only where the modern values of control are deeplymanifested. Furthermore, as these values have become progressively woveninto the predominant institutions of society, the more our practicalactivities and our daily life and experience in general become dominated byobjects which are products of technology and thence explicable in theirworkings by materialist understanding. Materialist understanding hasinformed the social practices, expressive of the modern values of control,that have shaped the “world” of daily life and experience many of whosekey objects can only be grasped and successfully dealt in its light. Thus, thesignificance of theories consolidated under the materialist strategies isvirtually assured within these institutions.3 Furthermore, for wont of theirexercise in these institutions, our sensibilities become dulled to other formsof understanding, including those that might legitimate subordinatingcontrol to other social values, so that increasingly materialistunderstanding appears as the only form of understanding, as in principlewithout competitors, ensuring even more its significance.

It is when we attend to technological objects and to their place in themodern “world” (Chapter 7) of daily life and experience that thereciprocity of the interests fostered respectively by pursuit of understandingunder the materialist strategies and by commitment to the modern valuesof control is most apparent. Simultaneously, in this “world” these objectsare objects of materialist understanding whose existence in many cases is acausal consequence of practices informed by materialist understanding, andobjects of high value from the perspective of value complexes that containthe modern values of control. It is little wonder, then, that in manyresearch institutions and widely shared public conceptions the interests ofthe scientific and the technological tend to be considered effectivelyidentical.

Materialist understanding: that grasped from thepractices of control

Soundly accepted theories manifest the cognitive values to high degrees inrelation to the appropriately selected set of empirical data. Typically, inresearch under the materialist strategies, these data are obtained fromobserving phenomena in the course of experimental practices, which are


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (135)

exemplary practices of control. We regularly anticipate, therefore, beingable to generalize from them to further practices of control (for example,technological ones); and it is not uncommon for experiments to beperformed for the sake of exploring technological possibilities and theireffects, when the experimental space serves as a kind of a mini-prototypeof a proposed technological innovation.

In saying that experimental phenomena occur within practices ofcontrol, I am describing them as products of human intentional agency,and as having humanly relevant consequences. Within the experimentalpractices, however, we describe the phenomena with categories drawnfrom the materialist lexicon, as we also describe the boundary conditionsof the spaces in which they occur. Intentional agency stops short withfixing the boundary conditions (which may involve creating a complex andsophisticated space) and intervening to bring about the initial conditions; itthen picks up again in order to observe and measure the experimentaloutcomes. This enables the phenomena of interest to be describedadequately in materialistic terms and explained well in terms of theunderlying order. Again, we anticipate being able to generalize fromphenomena in such spaces, grasped materialistically, to similar phenomenaand spaces including natural ones many of which are not and cannot beobjects of human control.

Under the materialist strategies, making such generalizations is crucialfor gaining adequate understanding of phenomena in natural spaces. Whilethese phenomena may (and must) be grasped initially by observation andmeasurement, as well as a modicum of order arrived at through inductiveinference or statistical analysis, in order to represent them in theories withsignificant explanatory power they must be represented with categoriesthat have been shaped in the course of experimental (and measurement)practices and the theoretical efforts to make sense of the phenomenaencountered in them. The language and posits of theory, even incosmology, draw from experimental (and measurement) practices.

The heart of modern science is experimental. Experiment, as it were, sitsbetween technology and natural spaces, providing for both a basis fromwhich to generalize, a model of the ways things are, and a context forcritical testing. Like technology, experiment is a human practice of control.Like phenomena in certain natural spaces, experimental phenomena cancharacteristically be portrayed as generated from underlying structure,process and law. In experiment, we come to identify or to confirm thepowers of nature that we are able to deploy for the exercise of control overthings.

Thus, even though materialist understanding extends well beyond therealm of control, and many scientists may value it largely for this reason, itis proper to identify it as understanding gained from the perspective ofcontrol. It is understanding of objects of the world insofar as they can be


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (136)

grasped from the perspective of practices of control. That understanding, itturns out, provides a sound grasp of the causal structure of numerousphenomena in spaces where human agency is not relevant, and it is broadlyan empirical question how far such understanding can extend. Thepractices of control, in a many layered way, provide essential viewpoints,means and conditions for the pursuit of materialistic understanding.

“Fundamental” research and the modern values of control

There are often good reasons to engage in research in particular fieldswhere there can be no reasonable general expectancy that establishedtheories will become practically applicable. Clearly not all worthwhileresearch is motivated by the quest for practical applications, or concernedthat any regularities it consolidates become items in sets of requirementsfor control. Historically, some phenomena have been the focus ofspontaneous, culturally widespread and persistent interest because theypertain to reflections about “our place in the universe” or about recurrentand striking features of objects that impinge universally on humanexperiences. Any theories offering posits about the underlying order ofthese phenomena are normally assured of being highly valued, not becauseof their potential role in practical life, but because of the sheer interest ofthe domains. Consider cosmology. Here, one’s interest is likely to be“knowledge for its own sake”, knowledge of features of the world ratherthan knowledge intended for application in practices of control. Furtheringthe general objective of science, encapsulating the possibilities open to adomain of phenomena and discovering how they are realized is a valueeven if usually (Chapter 5) furthering it is bounded by the values thatsupport adopting the strategies of a particular approach. In the case ofcosmology, however, one adopts the materialist strategies for the sake ofgaining understanding of phenomena in cosmological domains; adoptingthem is subordinate to the value of gaining understanding of these domains—and this approach has proved itself unrivalled in being able to generatetheories that highly manifest the cognitive values of these domains.

In the previous chapter, I raised the question: Why are the materialiststrategies adopted among the community of scientists virtually to theexclusion of any other strategies? Leaving aside the variety of individualmotivations there may be for adopting these strategies, there could be anumber of relatively distinct answers all pointing in the same direction.One answer, it will now be clear, rests upon commitment to the modernvalues of control: its interests are well served by research conducted underthe materialist strategies as a whole, and not just by research directedimmediately to informing particular controls.

There is also a second answer: the materialist strategies are adoptedbecause they, and apparently they alone, enable the furthering of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (137)

understanding, sought for its sheer intellectual interest, of certain domainsof phenomena (for example, cosmological) or in “fundamental” research(for example, in particle physics, biochemistry, genetics,developmental biology or neurophysiology) that directly aims to grasp theunderlying law, structure and process of phenomena in a deeper and moreencompassing way? And does not this answer provide a more satisfactoryexplanation of the adoption of the materialist strategies since it fits betterwith the self-understanding of the scientific community, whose commonarticulations maintain that the cognitive and theoretical interests of sciencefar transcend any links with the practical, and that any such links areeffects of the successful and autonomous pursuit of these interests? Thoseattracted by the second answer might go on to suggest that the first is reallyan answer to a different question; not to: “Why has the scientificcommunity adopted the materialist strategies virtually exclusively?”, but to“Why do the relevant social institutions provide support for researchconducted under the materialist strategies?”

The second answer cannot stand firmly on its own. It remains that anyunderstanding gained under the materialist strategies, of any phenomena orof the underlying order, is gained from the perspective of control; andadopting the strategies requires certain material and social conditionswhich are products of implementations of the modern values of control.Within a value complex, understanding of the kinds of phenomenadiscussed on p. 125 may be held as an object of value; but any valuecomplex which does include it as an object of value, assuming that it alsoaccords value to the research practices from which the understanding isgained, includes also the modern values of control. That object of value,the understanding of these phenomena, can only come into existence wherethe modern values of control are highly manifested. Unless the value of theunderstanding gained is dissociated from the values expressed in thepractices in which it is gained, and unless social values in general aresubordinated to it, the interest of a domain of phenomena—considered apartfrom the mutually reinforcing interaction of the materialist strategies andthe modern values of control—cannot provide a distinct ground for thealmost exclusive adoption of the materialist strategies.

Nevertheless, the second answer does significantly supplement the firstby pointing to how values, not reducible to the modern values of control,are also furthered by research under the materialist strategies. This helps toexplain that, when we attend primarily to the immediacies of practicalapplication or even to the general possibilities of control, the impulsebehind much materialist research cannot be grasped—whether that impulsebe connected with an intellectual interest in certain domains of phenomenaor with the desire to create and consolidate “fundamental” (materialist)theories that manifest the cognitive values ever more highly as they unfoldand when they replace each other. The grounds for the almost exclusive


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (138)

adoption of the materialist strategies leave open the focus of researchactivities under those strategies—whether it be “applied,” focused onquestions of immediate practical pertinence; or “fundamental,” driven bythe interests of obtaining theories that manifest the cognitive values evermore highly, and theories that are consolidated of particular favoreddomains. Either way it serves the interests of the modern values of control.My claim is that the mutually reinforcing interaction between thematerialist strategies and the modern values of control explains theirvirtual inseparability in modern societies. But there remains a distinctionbetween them, so much so that in “fundamental” research attention to thevalues has no role in the detailed play of the strategies. (One is apprenticeddirectly into the play of the strategies without necessarily gaining a clearawareness of the conditions that sustain the play, the values linked withtheir widespread adoption, and the way in which there might bealternatives to them.) Major scientific advances made under the materialiststrategies (including those that are accompanied by radical switches inversions of the strategies, for example, from determinist to probabilisticversions) need not be occasioned (and generally are not) by any reference tocontrol or to the values of control, but simply by the desire to gain theories(fitting the materialist strategies) that manifest the cognitive values moreand more highly, or that manifest them of domains of phenomena ofspecial interest.

Commitment to the modern values of control is the key to explaining thevirtually exclusive adoption of the materialist strategies in modern science.Research under these strategies serves the interests that spring from thesevalues—in general—and not only when it is addressed immediately topractical questions of control. In addition, any other values that areimplicated in modern research activities or embodied in researchinstitutions must, under modern historical conditions (whether or notarticulated by the individual scientific investigator), co-occur invaluecomplexes along with the modern values of control.

The relevance of materialist metaphysics

The mutually reinforcing interaction between adopting the materialiststrategies and holding the modern values of control is further strengthenedwhen the adoption is considered to be grounded in the acceptance ofmaterialist metaphysics. Materialist metaphysics affirms that the world“really is” such that all the objects in it (including human beings) are fullycharacterizable by materialist (perhaps ultimately physicalist) propertiesand relations and all phenomena in terms of being generated in accordancewith underlying structure, process and law; and the possibilities of thingsare exhausted by their material possibilities. Then, in principle, followingthe materialist strategies could give us a complete account of the world. In


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (139)

principle, no possibilities would be left out. I argued on p. 120 that gainingaccess to the material possibilities of things contributes to expanding ourcapability to exercise control over material objects and states of affairs. If,essentially, there are no other possibilities then gaining understanding ofthe world per se contributes to expand this capability. In addition,regardless of what projects of control we may incorporate material objectsinto, their nature remains unchanged; not being objects of value per se, it isopen to us to accord them whatever value we desire and to deny them anybut instrumental value—control cannot be opposed with the argument thatit changes the natural character of things.

The very nature of the world, then, seems to underlie the exercise ofcontrol as the characteristic activity to engage in when relating to materialthings; and ceteris paribus gaining understanding of it expands the range ofpossibilities for its exercise. Furthermore, not only do the categoriesdeployed under the materialist strategies not provide a barrier to movingfrom the possibility to the legitimacy of introducing particular controls, but(in principle) they are the ones appropriate to grasp the world as it is,implying that categories (value ones) that one might use in attempts to barthe move have no grip on the world. It also supports that, where practicesof control bring with them undesirable or unexpected side-effects, inprinciple they can be dealt with through further controlling interventions.

Although this story has a certain appeal there are tensions in it. Like allstories it is told in intentional not in materialist idiom. Our understandingof the gaining, consolidating and applying of materialist (includingphysicalist) understanding is expressed in intentional categories, as is ourunderstanding of human action in general. This mode of understanding islinked with the value that human beings not be treated as objects ofcontrol. Agents are not, by nature, objects of control, though (bydiminishing their agency) they can be made to approximate them (Laceyand Schwartz 1986; 1987). The exercise of control (by humans) overhumans diminishes their agency, an essential aspect of human nature (cf.Chapter 9), and this underlies a general moral objection to establishingrelations of control among human beings (Lacey 1979; 1985; 1990). Butagents are part of nature and so it is appropriate to expect that a generalview of nature be formulated with categories apt for representingagencyespecially since agency is both a phenomenon of lived experienceand practical life, and a presupposition of scientific practice.

The content of materialist metaphysics derives from extrapolating thecategories of theories, well established under the materialist strategies, toall phenomena and states of affairs in the world. There is no compellingreason, however, to hold that the very activity that produced the theoriesin the first place can be adequately represented within the theories’ owncategories (no matter how they may be generalized and abstracted).Perhaps it can be; but it has not (yet) been. Meanwhile, I see no serious


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (140)

difficulty in treating intentional understanding effectively as neitherreducible nor replaceable, though often needing to be supplemented. I saidthat accepting materialist metaphysics strengthens the mutually reinforcinginteraction between adopting the materialist strategies and holding themodern values of control. Early in the modern scientific tradition attemptswere made to ground versions of materialist metaphysics a priori. Mostagree that these attempts failed though their residue remains, and manysuggest instead that it is grounded dialectically as both an extrapolationfrom and a presupposition of the success of modern science.

But materialist metaphysics is not a presupposition of the remarkablesuccess of modern science in generating theories that come to be soundlyaccepted in the course of following the materialist strategies. It is enoughthat there exists a wide (and, in principle, unlimited) array of spaces—many of them created by human experimental and technologicalintervention—in which phenomena can be well represented as generatedfrom the underlying order. Parts or aspects of the world must be this way ifwe are to come to accept the theories we do accept. And so, too, must webe intentional agents. Nothing more need be presupposed; certainly notthat human action can be understood in terms of the same kinds ofprinciples deployed to understand phenomena in these spaces.

What can be extrapolated from the success of science? Certainly thatmore phenomena in more spaces will fall under the grasp of materialistunderstanding, that the laws represented in widely encompassing theoriesmay represent universal tendencies of nature (though not that universallythey are highly salient explanatory factors), and that many of the entitiesdiscovered as a consequence of experimental activity have important effectsboth in the natural world and in the “world” of daily life and experience.We can also extrapolate that increasingly material possibilities of thingswill be encapsulated, so that increasingly things will become objects ofpossible control. But there is no sound inference from can become toalready is, or to cannot become otherwise. The content of materialistmetaphysics can be extrapolated from the most soundly accepted, widelyencompassing theories, but an argument to endorse this metaphysicscannot be.

What, then, explains the appeal of materialist metaphysics, thecommitment or certitude—going well beyond what can currently beestablished by evidence or argument—displayed by those who adopt it, andtheir confidence that difficulties will be overcome and that particularrefuted arguments of theirs will readily be replaced by better ones? Perhapsit is the allure of a unitary world view. Perhaps it is a sense that unlessthe world can be grasped under the materialist strategies, we cannot havetheories manifesting the sort of “clearness and distinctness” needed tobring them into a decisive meeting with the empirical data, and thus tohave genuine knowledge. Relatedly, perhaps it is a sense of the ultimate


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (141)

unintelligibility of alternatives; and that despite the difficulties thatintentionality imposes for materialism, there remains an on-going programthat seems to be making progress in dealing with them. Perhaps it derivesfrom the mutually reinforcing interaction between the materialist strategiesand the modern values of control, that the way of understanding reflectedin the interaction has so come to dominate our consciousness in practicallife that no other mode of understanding seems comparable in power, to beintelligible, or even to be worthy of exploration.

Whatever the explanation may be, it falls short of providing acompelling argument for the adoption of materialist metaphysics. I inclineto the last proposal. Then, materialist metaphysics does not provide anargument for the adoption of the materialist strategies that is independentof the one rooted in its mutually reinforcing interaction with the modernvalues of control.4

This conclusion accords with my general view of the relationship ofmetaphysics with science. Like the empiricists, I do not think that science,defined by the general objective to gain understanding of things (O,Chapter 5), is committed, except heuristically and temporarily (Hesse 1977),to any particular metaphysical view—either as presupposition of itspractices or as implication of its established results. Like the scientificrealists, I think that there are posits of the underlying structure, process andlaw of certain spaces that are so well confirmed as to be placed in the stockof uncontested knowledge (cf. McMullin 1998:378). An approach toscience proceeds under particular strategies, whose source need not be inmetaphysics but can be, as I am arguing, in values. For instance, it need notbe in considerations about the general nature of things, but may be inconsiderations about the general possibilities of interest for our (stances of)interacting with the world.

Looked at this way, we can recognize the most general categories of thelexicon borne by the materialist strategies to be derived not frommaterialist metaphysics, but from responding to the question: “How mustwe think of material objects if we want to further the manifestation of themodern values of control?” They appear to be derived from materialistmetaphysics—and the lexicon appears to be the lexicon of science ratherthan of a particular approach to science (Chapter 5)—when commitment toit is unproblematized and the modern values of control are uncontested.Then the ceaseless expansion of our capabilities of control also appears tobe ensured by the metaphysics. When we attend to the lack ofsolid grounding for materialist metaphysics, however, my conclusion thatit* grip comes from the mutually reinforcing interaction of the materialiststrategies and the modern values of control becomes compelling.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (142)


My answer to the question: “Why adopt the materialist strategies to thevirtual exclusion of other strategies?” is neo-Baconian. Control of nature isthe key. Not that the aim of science is the control of nature or that(fundamental) research projects are shaped by immediate practicalconcerns; but that the modern values of control have become deeply woveninto modern society and its most powerful institutions, and that theyinteract in mutually reinforcing ways with research conducted under thematerialist strategies, whose theoretical products are generally significant inthe same institutions as those into which the modern values of control arewoven. Consistent with this, it remains a necessary condition for theadoption of these (or any other) strategies that under them are producedtheories that manifest the cognitive values highly But, the interest toproduce theories that manifest the cognitive values highly cannot explainthe virtual exclusivity of adoption of the materialist strategies. Cognitiveinterests underdetermine the choice of which strategies to adopt, and socialvalues take up the slack. I emphasize that cognitive and social values donot play their roles at the same level. The social values provide animportant part of the reason to adopt a strategy, but theories developedunder the strategy are properly accepted in virtue of the manifestation ofthe cognitive values. My account preserves impartiality as an ideal ofscientific practice, though it cannot remove neutrality from ambiguity(Chapter 10).

My answer to the question in the previous paragraph clashes with theidea that science is value free (with autonomy and perhaps neutrality) sothat those who articulate the modern scientific tradition will be reluctant toendorse it. Their reluctance will be assuaged neither by my preservation ofimpartiality, nor by my claim that providing special support for values,widely assumed to be universal and partly constitutive of rationality, is thekey ground on which neutrality appears to be violated. The reluctancederives from the conviction that modern science, the project that has vastlyexpanded our common stock of knowledge, is driven (ideally) by purelycognitive interests and that these suffice to explain not only the properacceptance of theories, but also the sound adoption of strategies.

The reluctance is compatible with recognizing that there is the mutuallyreinforcing interaction between adopting “the strategies of modern science”and the modern values of control, that this interaction may provide themotivation for individual scientists to engage in research, and that it servesto explain the ready availability of the material and social conditionsrequired for research in the advanced industrial countries. Then it might beobjected that all I have succeeded in explaining is the accelerating pace ofscientific development and the widespread social support for the activities


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (143)

conducted by the scientific community, not the grounds for the almostunanimous adoption of the materialist strategies within the scientificcommunity. Perhaps, the objection might go, I have rushed too hastily toconclude that cognitive interests underdetermine the rational choice ofstrategies. In particular, that I have not sufficiently considered that ourcapabilities to control material things have been successfully enhanced (andthus the mutually reinforcing interaction between the strategies and themodern values of control has become possible) because research conductedunder the materialist strategies enables the consolidation of theories, moreand more of them encompassing ever larger domains of phenomena, withgreater cognitive credentials than understanding gained from otherstrategies. If this is so, then could we not explain the almost unanimousadoption of the materialist strategies in terms of them being the strategiesunder which we can consolidate theories that manifest the cognitive valuesmost highly of whatever domain of phenomena is chosen for investigation—and would not that explanation stand on its own regardless of whateverreinforcing social explanations there might also be?

Is this so? It is true that theories consolidated of certain domains underthe materialist strategies are commonly considered to be exemplars ofitems properly included in the stock of knowledge. But that is not enoughto sustain the objection that the cognitive values do not underdetermine thechoice of strategies. It would have to be argued that in principle theoriesconsolidated under the materialist strategies manifest the cognitive valuesmost highly, and not just in actual fact, for the prevailing social conditionsmight account for the underdevelopment of alternatives (Chapters 7 and8). How might such an argument unfold? There is a hint in the suggestionjust made that our capabilities to control material things have been vastlyenhanced because research conducted under the materialist strategies hasconsolidated theories with superior cognitive credentials. Control, wemight say, has been enhanced because we have gained more and betterknowledge of material things; enhanced control is a symptom of thecognitive superiority of theories consolidated under the materialiststrategies. An argument of this kind has been proposed by Taylor (1982).5

I will now offer a version of it.6 In it “comprehensiveness” is the keycognitive value appealed to; understanding gained under the materialiststrategies is maintained to be inherently the bearer of the mostcomprehensive understanding of the “material world” or the “physicaluniverse.”

The value of materialist understanding

To understand a thing, according to Taylor, is to have a “rational grasp”of it, to have an “articulation” of it in which its various features aredistinguished and presented in “perspicuous order,” so that the cognitive


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (144)

or rational credentials of an articulation derive (provided that it isconsistent) from the perspicuous order that it lays out (90). Taylor issensitive to the intelligibility of alternative forms of understanding(Chapter 5) conforming to a variety of ideals of “perspicuity.”Nevertheless, he proposes that greater perspicuity derives from “a broader,more comprehensive grasp on things” (ibid.), and it is such a grasp that isprovided by understanding gained under the materialist strategies.

How is it that materialist understanding provides a broader, morecomprehensive grasp of things? I distinguish this question from: Why is itthat materialist understanding has displaced (for the most part) hithertoexisting forms of understanding (pertaining to material objects)? Myaccount of the grounds for adopting the materialist strategies answers thesecond question. For Taylor, in contrast, it is the comprehensiveness ofmaterialist understanding that principally accounts (rationally) for itshaving displaced earlier alternatives, though comprehensiveness functionsin concert with an account of how materialist understanding and earlierforms of understanding are incompatible, how they cannot both generategenerally significant products in the same social/historical/cultural nexus.

In an earlier phase of Western civilization there was a form ofunderstanding in which “understanding” and “attunement” wereinseparable, and in other cultures there are forms of understanding inwhich the modern Western separation between practical activity andsymbolic expression cannot be made (Taylor 1981:209). Taylor maintainsthat they are incompatible with materialist understanding. It is not thattheir products are formally inconsistent (Chapter 7); rather theincompatibility derives from the ways in which they inform human actionand the practices from which they are produced. We may say that to eachform of understanding there corresponds a characteristic activity, orpredominant stance towards nature, which not only does the form ofunderstanding illuminate, but which also contributes towards thegeneration or consolidation of the form of understanding. For example,where understanding and attunement are inseparable there is thecharacteristic activity of adaptation to nature, and corresponding tomaterialist understanding there is that of exercising control over natureframed by the modern values of control. While, in any culture, elements ofthe various characteristic activities may be present (at least marginally), aspredominant stances those of control and, for example, adaptationmutually exclude each other. They do not fit together or complement eachother; they “are rivals; their constitutive rules prescribe in contradiction toeach other…. [They] cut across [each other] in disconcerting ways” (98–9).They are not merely different; they cannot be adopted together, and theattempt of different groups to adopt them simultaneously in the samesocial space must lead to conflict. The very success in exercising controlchanges the environment in which one lives, yet adaptation presupposes a


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (145)

more or less constant environment, subject at most to periodic rhythms.Conversely, the predominance of adaptation precludes the kind of exerciseof control necessary for the conduct of research under the materialiststrategies (Chapter 7).

Where the stance of adaptation has been adopted as the predominantone of a culture its associated form of understanding is taken both to haveempirical support and to articulate the value of the prevailing social(ecological, cosmic, and possibly spiritual and theocentric) order; it makessense of a good deal of daily life and experience, and outlines the path toattunement. It also delimits the class of the possible: it may lack theconceptual resources to encapsulate the possibilities that may derive fromanother stance, and its posits will not have been tested against those thatmight be put forward by a form of understanding associated with adifferent stance; and it denies the possibility of attunement if one engages ina different activity. There is a sort of self-enclosure that neither recognizesnor permits much space for adopting alternative stancesexcept on themargins—and possibilities that are not identifiable and realizable within it,even if recognized, are not valued. Similarly, the modern Western“enclosure,” defined by the shifting perspectives of (materialist) scientificunderstanding and technological control, and thus not dependent onstability, but on constant innovation and change, leaves little space foralternative activities and for the exploration and identification ofpossibilities that are not realizable within its structures (Chapter 8; Lacey1997c).

In order to challenge the demarcation of the range of possibilities that aform of understanding associated with adaptation admits, one would haveto engage in activity under an alternative stance—but that would threatenattunement. Therefore, one cannot rationally challenge the demarcation ofpossibilities without also challenging the valuation of them. One mustengage in that activity (which is negatively valued within the prevailingform of understanding) prior to gaining empirical evidence that thereare genuine (and valuable) possibilities that are unrecognized within theself-enclosure. In doing so, one might threaten the viability of the stance ofadaptation for the culture as a whole, since the natural world in whichadaptation takes place could be significantly changed in the course ofattempting to establish one’s claims. For the alternative activity to getunder way, it seems, there would have to be doubts about attunement. Atleast on the margins there would have to be embryonic grounds to suggestthat the alternative practice may produce betterment for human beings.Otherwise, the self-enclosure would seem to be unbreachable, unless itsuccumbed to a natural disaster or to a powerful outside intrusion.7

The promise of human betterment was a major theme for virtually all theimportant contributors to the scientific revolution of the seventeenthcentury It has been ignored in many discussions of the superiority of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (146)

modern to ancient science, being held to be of no cognitive significance.Certainly it has no cognitive bearing on whether a certain theory bestarticulates a given domain of phenomena. But it does bear upon whetherthe pursuit of materialist understanding is desirable and whether theunderstanding gained is significant. In the process of the scientificrevolution, the notion of “human betterment” like that of “theoreticalunderstanding” was transformed. Taylor says that, in older traditions, notto be attuned to nature is “to be in misery and confusion.” Through thisnegative side there is retained a thread of contact between the old notion ofattunement and the modern sense of betterment.

Materialist understanding and the displacement of earlierforms of understanding

Human activity in all cultures involves some measure of exercising controlover natural things. In a broad sense, all living things intervene in naturethrough various mechanisms of assimilation and accommodation. Thehuman distinctiveness is that the intervention is purposive and planned,expressive of stances which are illuminated by forms of understanding. Thedifference between the stances of adaptation and control is that in theformer, but not the latter, the exercise of control is subordinated to suchvalues as ecological and social stability and bounded by ends and meansdefined in traditional practice (Chapter 8) or, in Taylor’s terms, theactivities in which control is exercised are not articulated as disjoined from“expressive activities” (Taylor 1981:209). In contrast, when the stance ofcontrol is adopted the exercise of control is framed by the modern valuesof control; ends are not circ*mscribed as a matter of course by a naturalenvironment which is to be nurtured or a social order to be maintained,but only by the possible and the power to implement it.

The differences are important. Nevertheless the forms of understandingof older traditions do have components that inform certain practices ofcontrol of material objects (in agriculture, engineering, medicine, etc.); theyhave a certain amount of systematic empirical knowledge of various sets ofrequirements for control; or they have understanding which informs our“ability to make our way around in [the world] and deal with things in it”(101). This enables the transcultural recognition of effective practices andtheir comparison and, thus, the transcultural recognition of the muchgreater practical efficacy of technological practices in which the stance ofcontrol is adopted. The scope, power and effects—penetrating, as theyunfold, into the lives of almost all human beings—of these practices expandthe horizons of the possible (and the historically actualized) in a way thatfar surpasses anything entertained in traditional forms of knowledge(Taylor 1995). Because of their continuity with the goals of manytraditional practices of control, because of their greater effectiveness and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (147)

efficiency in serving some of them and thus (at least piecemeal) their beingwelcomed as replacements for some traditional practices and sources forthe betterment of the human condition, and because of their loomingpresence as influences confronted in daily life and experience, theachievements of technological control cannot be ignored. The mechanismsthat account for this impact of technology, that cannot be ignored, can becomplex and various and those that were important in the West have notgenerally been recapitulated elsewhere. Sometimes desire for the“betterments” promised by modern technology is a key factor, andsometimes the interests of powerful elites; at other times colonial forceplays a role, but Taylor is right to emphasize the continuity of technologicalwith traditional practices so that members of a culture can embrace atechnological practice because it performs a traditional task moreefficaciously.

Any form of understanding, which cannot encompass the possibilitiesand explain the material workings of technological controls, becomesunable to inform daily life and experience adequately, and so its theories(posits) are rendered insignificant. A form of understanding which cannotdo this will, as the impact of modern technology grows in the daily life andexperience of a culture, be displaced by one that can. Displacement occurs,I suggest, when a culture’s contact with technology has been sufficientlyprolonged that technological activity has become either highly salient orpace-setting in the practical life of the society. It does not seem to be possiblein the contemporary world for a culture to isolate itself significantly fromthe encroachments of technological advance, so much so that technologicalactivity has become a major factor in shaping the future of every culture,even if only to the extent of threatening to destroy the space within which aparticular culture thrives.8

It is materialist understanding that encompasses the possibilities andexplains the material workings of modern technological objects. This leadsto an “inner connection” between understanding of “the world” andcontrol which “rightly commands everyone’s attention” (101, 103). Thus,materialist understanding displaces earlier forms of understanding, unlessthey are able to develop so as to enter into rich dialectical relations with it,as forms of understanding associated with the Western pre-seventeenthcentury stance of adaptation did not.

The displacement argument relates to considerations ofcomprehensiveness ambiguously. Its conclusion is not that materialistunderstanding can in principle encompass all phenomena of which otherforms of understanding offer accounts. Materialist understanding does notexplain, for example, the social forces and the sources of the social valuesthat have come to be woven into social forms along with the modernvalues of control that must also be grasped in order to fully understand theachievements of technological controls. Rather, it encompasses key


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (148)

phenomena and possibilities, those of control characterized in abstractionfrom their social and cultural context, with a comprehensiveness unrivalled(and unapproachable) by its pre-modern competitors. It is a kind of“bounded” comprehensiveness that is involved—bounded by the interest ofcontrol. Without historical success in transforming the world of daily lifeand experience which derived from adopting the modern values of control,the displacement argument would have gained no footing.

The displacement argument, therefore, does not challenge myexplanation for the virtually exclusive adoption of the materialist strategiesin the modern scientific community. On the contrary, it deepens thereflection on the reciprocity of interests between research conducted underthe materialist strategy and adoption of the modern values of control. Thephenomena and possibilities of control simultaneously are objects graspedwithin theories consolidated under the materialist strategies and objectshighly valued within complexes that contain the modern values of control.The more salient their place becomes in a society the more powerful is thedisplacement argument, for a form of understanding must offer an accountof salient objects in the world of daily life and experience; that is the test ofthe significance of the theories it produces.

This leaves open, however, that there may be good reasons—linked withsubordinating control to other social values—to explore sources ofalternative strategies in which the understanding of material objects doesnot involve abstracting them from their relations with human and socialfactors, and which thus explore possibilities that are notencapsulated under the materialist strategies. Such alternative strategiesmay meet the just-mentioned test of significance by absorbing materialistunderstanding into a subordinate place within a form of full understanding(Chapter 5) in which its limits of applicability (including desiredapplicability) would be identified, and in which more generally an accountof materialist understanding and its place would be constructed,accounting for its trajectory within currently dominant social forms andcharacterizing what its trajectory would be in the alternatives to it. Nodoubt these strategies would compete with the materialist strategies forsocial space in which to develop, but not (normally) with them asgenerators of theories that encapsulate the material possibilities of things.Understanding gained under them would not challenge that theories thatencapsulate the material possibilities of things regularly and progressivelyare consolidated under the materialist strategies, and that they represent aremarkable contribution to knowledge. It queries, not the knowledge, butit* significance, and thence the value of gaining (without subordination tospecified values) further materialist understanding. These conjecturedalternatives may be continuous with older forms of understanding: theymay result from the development and radical redeployment of older formsin the context of modern daily life and experience, involving dialectical


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (149)

syntheses of old strengths and new gains. There might even be importantresidue to rescue from the older forms concerning particular materialobjects and practices, something that would not be surprising given thatmany older practices presumably survived because they were soundly basedempirically (see references in Chapter 8). The displacement need be neithera wholesale obliteration nor a barrier to the emergence of alternatives.

Technological objects and their gaining an important role at the “cuttingedge” of Western society significantly pre-dated the capability ofmaterialist understanding to anticipate novel material possibilities and touncover the means to realizing them. Nevertheless, for well over a centurynow materialist understanding has contributed enormously to furthering thetechnological project and thus to bringing about a world, key objects ofwhich have made recourse to materialist understanding indispensable andapparently ever more so, as technologies spread into more and moredomains of life. That it has become indispensable in this way helps toexplain why there are plentiful resources available to pursue research underthe materialist strategies, but it does not provide a general reason (asdistinct from one grounded in endorsem*nt of the modern values ofcontrol) to adopt them in research rather than any others—except bydefault. The mark of historicity remains. The socio-historical world had tochange for the materialist displacement to occur and to be rationallygrounded. It is the success of a socio-historical project (itself informedby materialist understanding), the ever deeper manifestation of the modernvalues of control, that has made the materialist strategies indispensable forgrasping “the world we live in” today.

In a similar way, an alternative form of understanding may come toinform a socio-historical project, perhaps articulated around the ideals ofecological soundness and social justice, that may become realized; so thereis no less reason to explore such alternatives now than there was to explorematerialist understanding at the very beginning of modern science. Thereare powerful social forces posed against the success of such nascentalternatives (regardless of whether there are good reasons to explore them),so that power—mediated through its being exercised on behalf of theinterests of the modern values of control—may be an important part of theexplanation for the dominance of the materialist strategies. That isconsistent with my viewpoint, but not with those who maintain theunqualified cognitive or rational superiority of the products of researchunder the materialist strategies.

The comprehensiveness of materialist understanding

The fact that it is materialist understanding that encompasses thepossibilities and explains the material workings and effects of moderntechnology is sufficient to account for the kind of displacement just


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (150)

described. Attempting to move beyond displacement, Taylor goes on toargue that materialist understanding is able to displace earlier forms ofunderstanding because it offers a broader, more comprehensive grasp onthings; “it has greatly advanced our knowledge of the material world”(103). Modern technological success, he maintains, depending as it does onour having gained understanding that encompasses the possibilities andexplains the material workings and effects of technological objects, reflectsthe greater comprehensiveness of this understanding; the success oftechnology is “proof” of this. That it enables higher manifestations of acognitive value, “comprehensiveness”—not that it successfully informstechnological practices—it is thus proposed, grounds the cognitive orrational superiority of materialist understanding; technological success is asymptom not a ground of this cognitive superiority. Then, one mightargue, it is the quest for this kind of cognitive superiority that explains thealmost unanimous adoption of the materialist strategies within thescientific community. We might put it: adopt the materialist strategiesbecause research under them can provide a more comprehensive grasp ofthe material world rather than because it can provide a better account ofa range of possibilities (the material possibilities of things) that areconsidered especially valuable.

Does materialist understanding offer a more comprehensive grasp ofthings? Taylor says: “modern science represents a superior understanding ofthe universe, or if you like the physical universe…. [It] achieves greaterunderstanding at least of physical nature” (102–3)…. It is infinitely superiorfor understanding the natural world. Our immense technological success isproof of this” (Taylor 1981:209). Does materialist understanding offer amore comprehensive grasp of the material (physical, natural) world? (Whatis the “material world”?) Does the fact that it explains technologicalsuccesses and anticipates novel technologies-more generally, that it enablesthe encapsulation of the material possibilities of things in an ever moreencompassing way—reflect that it offers a more comprehensive grasp ofthe “material world?” Is there an argument, independent of technologicalsuccess, for this greater comprehensiveness? Or is the argument simply thata condition of the possibility of technological success is that theunderstanding which informs it must manifest such comprehensiveness?

Materialist understanding is wide-ranging; but it is not full (Chapter 5),since it abstracts from the human, social and ecological dimensions andconsequences of phenomena within the spaces of which it providesunderstanding, and from the human and social possibilities that may behidden in them. It encapsulates the material possibilities of things, thosepossibilities that can be identified from the generative power of underlyingstructure, process and law, abstracting from the place of things in humanexperience and practice. Only in spaces where human factors are causallyirrelevant could it be seriously entertained10 thatthematerial possibilities be


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (151)

considered identical with the complete range of possibilities permitted bythe space and the arrangements of phenomena in it; and only in thosespaces where ecological and organism/environment interaction factors areirrelevant can one plausibly be confident of the identity.

Modern technological successes and anticipations of novel technologicalinnovations presuppose only that the understanding that informs theworkings of technological objects is wide-ranging and that the spaces overwhich it ranges include experimental ones. Technological objects, likeexperimental phenomena, fall within the domains of phenomena of whichtheories, consolidated under the materialist strategy, manifest to a highdegree an array of cognitive values, including wide-ranging explanatorypower. Modern technological success—not technological advance per sesince it predated sophisticated developments of materialist understanding –attests to success in having gained wide-ranging understanding, and to thehistorical superiority of materialist understanding as an instance of wide-ranging understanding. Equally, however, the success of research under thematerialist strategies presupposes that we have successfully exercisedcontrol in numerous experimental spaces and made certain kinds oftechnological advances. Materialist understanding and our capabilities toexercise control develop together in mutually reinforcing interaction in anunfolding spiral of development, so that each moment of development ofthe one presupposes relevant moments of development of the other.

Materialist understanding encapsulates the material possibilities ofthings, and helps to generate sets of requirements for control, in a moreencompassing way than rivals. But it does not generate full understanding:it not only does not treat things as cultural objects (of course some thingsare not cultural objects) or their effects on human lives or the humancausal factors which influence the boundary and initial conditions of theirmotions and changes and which may be essential for their very existence,but also does not deal centrally and sometimes not at all with the side-effects of technological interventions, including their social andenvironmental consequences. Materialist understanding deals with thingssolely under descriptions that relate them to underlying law, structure andprocess; a mode of understanding which underlies the power, given certainboundary conditions (which it characterizes without reference to thehuman agency and social conditions that may be necessary to bring theminto being) to predict and control, and thence the power to uncover novel(material) possibilities of nature. Nevertheless, its mode of dealing withmost of the objects we encounter in daily life involves abstraction. Theoriesconsolidated under the materialist strategies manifest well the cognitivevalue “wide-ranging explanatory power,” but not “full explanatorypower.” The gain is more or less unimpeded technological advance. Oneproblem is that technological objects are commonly introduced without afull understanding, and this is reflected in current social and ecological


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (152)

crises. Another problem is that the mutually supporting interaction ofresearch under the materialist strategies and the modern values of controlremains implicit, without the sharp articulation that it might gain within aform of full understanding, thus disguising that much of materialistunderstanding is not neutral, that its general significance is assured onlywhere the modern values of control are deeply manifested.

Perhaps I have side-stepped from the central thrust of Taylor’sargument. In it such phrases as “the material world,” “the physicaluniverse” and “the natural world” frequently recur apparently assynonyms. It is of the “material world” that materialist understanding issaid to give a more comprehensive account. Materialist understandingcomes to the fore when we dissociate activities of control from expressiveand valuative activities. That dissociation is historical, but when we makeit we gain superior access to the “material” (or “natural”) world whichapparently is not considered as constituted within human history. SinceTaylor subscribes to neither materialist nor Cartesian metaphysics, I find itdifficult to interpret these phrases. Human beings are part of nature or “thenatural world.” But then there are parts of this world, those where humanphenomena are pertinent, which are not well grasped under the materialiststrategies; the comprehensiveness Taylor speaks of does not embrace thesephenomena.

“The material world” and its synonyms are intended to designate theobject of inquiry in the natural sciences, but it seems that it cannot beidentified with “the natural world” that we are part of in all its fullness.This object of inquiry (Taylor says) “exists independently of us humanpercipients” or consists of bits of the natural world and how they affecteach other “even when we aren’t on the scene or we aren’t playing a role”(32, 47). In this “world” human causal agency is not relevantly a factor,and its components and their properties (unlike those investigated in theinterpretive human sciences) are not partly constituted by theinterpretations of human beings engaged in various practices. Can weproperly characterize the object of research conducted under the materialiststrategies in this way? Any answer to this question, I suggest, must beconsistent with the following conditions:

1 Our understanding of objects is mediated by available conceptualresources and (in scientific practice) by the lexicons of the adoptedstrategies.

2 We are part of the natural world, causal agents in it; we cannot graspthe world except in virtue of our causal agency in it.

3 We are able to grasp segments of the natural world, in which humanagency is not a relevant causal factor, through practices that relatethem to segments (usually experiments) where human agency isrelevant.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (153)

4 We are able to extrapolate this grasp to spaces even further removedfrom human experience and agency, including to times before therewere any human beings and, in principle, the extrapolation may extendto the coming into being of human beings with their distinctivecapabilities, with the extent of the extrapolation remaining an openquestion, often begged by premature commitment to materialistmetaphysics.

Taylor holds that the lexicon deployed within the materialist strategiesreflects that the object grasped in inquiry under them is properlycharacterized as independent (ontologically) of human observers. Hedistinguishes “subject-related” or “anthropocentric” (Taylor 1985:2;1995) and “absolute” terms. Subject-related terms designate subject-relatedproperties, those which a thing has in virtue of being an object ofexperience of human subjects, its situation within the context of humanaction, interest or concern, and how it affects human beings. Absolute termsare non subject-related terms; they designate properties that are not subject-related, properties that things have regardless of their relations with humanbeings. According to Taylor, the lexicon deployed within the materialiststrategies consists entirely of absolute terms. So, we might put it, the“material world” is the totality of objects designated by the lexicons usedto articulate the theories consolidated under the materialist strategies. Thismove shifts our focus to the terms contained in these lexicons. When wemake it, however, I am led to a conclusion that diverges sharply fromTaylor’s. Taylor offers as examples of subject-related properties desirabilitycharacterizations of people, and the secondary qualities, those qualities(according to him) that things have in virtue of their powers to causecertain types of experience in people. The terms used in the materialistlexicons indeed have different features from these examples and, asdeployed in theories, they show no manifest “sign” that they are subject-related too, and they are by design dissociated from value terms. Are theyabsolute?

Two (interrelated) considerations need to be brought to bear. First, theseterms are predicated of things in the course of scientific (theoretical,measurement and experimental) practices conducted under the materialiststrategies; and which terms are predicated (and which of alternative largelyincommensurable lexicons they are part of) varies with changes within thematerialist strategies. Many of them are predicated successfully; forinstance, they enable the articulation of theories that highly manifest thecognitive values of a wide range of domains. This much is clear. There isnothing about the scientific practices, however, that enables us to concludethat the terms designate independently existing properties of things. Thepractices are compatible with there being no more to affirming that thingshave certain properties than that the relevant terms have been successfully


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (154)

predicated. Then the terms of the materialist lexicon would not be absolute.If so, and if the “material world” is that which is represented in our bestscientific theories, then the things of the “material world” and theirproperties are partly constituted by the practices (and their associatedlexicons) which enable their articulation.

Second, the terms of the materialist lexicon are for the most partquantitative. Quantitative terms might be taken to be paradigmaticallyabsolute; a quantity apparently reflects no values, and its applicabilityfollows from procedures whose outcomes are intersubjective and sharedamong a wide variety of human practices that express manydifferent interests. Things are not so simple. That a procedure leads todesignating the same numerical value of a quantity to a thing regardless ofwho conducts the procedure does not guarantee that essential links withvalues are not implicit. The use of quantities in the social sciences makesthis clear. (Think of the controversies about IQ and the index ofunemployment.) A quantitative term is applied to particular things viasome measuring operations and usually an array of mathematicalinferences and calculations that are theory-dependent. Measurementrequires instruments and bringing the thing to be measured intoappropriate relationship with the instrument, so that it concerns a relationbetween the thing and the instrument, which is a human artifact,theoretically articulated, that we insert into the space of investigation(Chapter 7). The meaning of a quantitative term cannot be dissociated fromthe operations of instruments; nor can it be reduced to them, since it is alsoimplicated in the theoretical articulations of the construction and uses of theinstruments. Quantitative terms are not absolute; they do not refer toproperties that things have regardless of their relations with human beingsand of their role in human practices.

In the light of these two considerations I do not think that a relevant,ontologically grounded distinction between absolute and subject-relatedterms can be sustained. Where does this leave the “material world?” I thinkthat appeal to it is largely the residue of Cartesian metaphysics. Leavingaside Cartesian or materialist metaphysics, one might consider it to bewhatever is represented in the best theories consolidated under thematerialist strategies, but that does not serve Taylor’s purposes. Even so, itis a useful way to think of it. The “material world” is the world as graspedin terms of what can be generated from underlying law, process andstructure by means, not of absolute terms, but of categories suitable forarticulating underlying law, process and structure. This is effectivelyequivalent to the world as grasped from the perspective of the stance ofcontrol. Alternatively, and again equivalently, we may think of the“material world” as the totality of the material possibilities of things.Theories consolidated under the materialist strategies clearly provide abroader, more comprehensive grasp of the “material world,” so


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (155)

understood. However, this is not a sign of general cognitive superiorityover competing forms of understanding, for it is a “world” (Chapter 7) notfully cohabited by proponents of competing forms of understanding, whichare associated with stances towards material objects in which control issubordinated to other social values. The material possibilities of things donot exhaust the possibilities deemed valuable in these other “worlds” andmany of them themselves lack value in them.

When one speaks of the “material world” without elaboration itsuggests something that is shared and with which, regardless of valuecommitments, we all interact as a matter of course. Then understanding ofit, the most encompassing possible, seems to make a claim on us all, as if itcould be a sort of constant in our practical dealings with the world. Whenone speaks of the “material world” as the totality of the materialpossibilities of things, in contrast, the suggestion of universal interest retreatsinto the background. It is not that, for reasons connected with particularinterests, the genuineness of possibilities encapsulated in theoriesconsolidated under the materialist strategies is denied, but that the value ofsome (many) of them may be denied and so the interest in furthering theirencapsulation is diminished. A competing form of understanding may wishto identify a class of possibilities that only intersects with that of thematerial possibilities, for example, those possibilities of things consistentwith and supportive of ecological stability and a particular conception ofsocial justice. Then, whether one opts to participate in developingmaterialist understanding or the alternative form of understanding cannotbe grounded in the kind of comprehensiveness claimed for materialistunderstanding for it does not encompass the alternative class ofpossibilities. In valuing this alternative class of possibilities above thematerial possibilities and pursuing research to identify them, one does nottherefore deny a claim made on us all that properly follows from oursharing an independent “material world.” Similarly, in ranking thecognitive value “wide-ranging explanatory power” above “full explanatorypower” (or vice versa) one does not draw upon the general features of anobject of inquiry shared by us all, but on the values that interact inmutually reinforcing ways with one’s adopted strategies.

Taylor also refers to materialist understanding providing “recipes formore effective practice,” increased “ability to make our way about andeffect our purposes,” increased “practical ability” and “increased practicalcapacity” (Taylor 1995:48). Perhaps, then, we might think of the “materialworld” as the world as it must be grasped for the sake of most effectivepractical activity. But we cannot separate what counts as effective practicalactivity from the social world in which it is conducted and the conditionsmade available in it. Materialist understanding does not, in general asdistinct from in a subordinate role, serve practical activity especially well ina”world” which aspires to attunement, or in one in which ecological


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (156)

stability and certain ideals of social justice are high aspirations. It does notenhance practical activity in general, only practical activity linked centrallywith the stance of control of nature.

Technological (and experimental) success bears testimony that the worldis amenable to grasp under the materialist strategies. As long astheories with wide-ranging explanatory power continue to be consolidatedunder the materialist strategies, technological innovations are assured(provided that the relevant social conditions remain in place), as also isgaining understanding of numerous domains of phenomena in whichhuman causal influence is not relevantly present. Theories consolidatedunder the materialist strategies provide wide-ranging understanding of anextraordinary array of domains. But for many of the objects in thesedomains, including technological and experimental objects, it does notprovide full understanding. We cannot, as it were, form a totality of all theobjects in all the domains that materialist understanding ranges over andcall this totality the “material world,” for many of these objects play a rolein human experience and practice, even though under appropriatelyformulated boundary conditions their behavior is well grasped inmaterialist terms. Concerning these objects, to engage in research aiming togain wide-ranging rather than full understanding (or vice versa) does notrest upon purely cognitive grounds.

We do not find in comprehensiveness, therefore, a cognitive factor thatcan take us beyond displacement. As argued on p.135–6, materialistunderstanding properly displaces earlier forms of understanding because it,and not they, can grasp features of objects (particularly technologicalobjects) that have become central in the “world” of daily life andexperience. Displacement alone, however, does not provide, except perhapsby default (Chapter 7), a ground for the virtually exclusive adoption of thematerialist strategies. Comprehensiveness does not provide it either, sincematerialist understanding is more “comprehensive” only in the sense of“wide-ranging,” and cognitive interests do not suffice to prioritize wide-ranging to all versions of full understanding.

My earlier conclusion remains intact: it is their mutually reinforcinginteraction with the modern values of control that explains the virtuallyuniversal adoption of the materialist strategies within the scientificcommunity. We may say that they are adopted for the sake of grasping the“material world.” But, if I am right, the most viable sense of “the materialworld” is that of the world as grasped under the materialist strategies, a“world” not consisting of objects that are ontologically independent ofhuman observers, but partly constituted in practices that have mutuallyreinforcing interactions with the modern values of control. In this way Iremove from Taylor’s idiom traces of the suggestion that materialistunderstanding is generally (as distinct from selectively) applicable acrossperspectives of engagement, and that its pursuit has a purely cognitively


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (157)

grounded role in all cultures and epochs. My view is sufficient to accountfor technological success (and it recognizes that the material possibilities ofthings can continue to be grasped in an apparently limitless way), and itis unencumbered by the gratuitous universality that comes with taking “thematerial world” at face value, where the apparent absoluteness of theterms in the materialist lexicon disguises the link with the modern values ofcontrol. It also leaves open that adopting values that clash with the modernvalues of control may support the adoption of strategies in research whichmay provide bounds to the realm of materialist research that is consideredworthwhile, or which may relocate the place of materialist researchsubordinating it to where it is relevant to a form of full inquiry.

My argument has left open that there may be alternatives to thematerialist strategies, and it has indicated how commitment to values thatchallenge the modern values of control would render intelligible the questto develop such strategies. But it has been objected that there really are noalternative strategies to be explored (McMullin 1999). Is this so? Certainlythere is no alternative today institutionalized as materialist inquiry is anddeveloped with comparable sophistication, systematicity, general credibilityand power. Clearly, moreover, if there are no alternatives, the thrust of myargument is very much weakened. If alternatives cannot be identified, muchof the force and social salience of the mutually reinforcing interactionbetween research under the materialist strategy and the modern values ofcontrol is diminished. If there are no alternatives then, although themutually reinforcing relationship remains intact, the question aboutadopting the materialist strategies rather than some others is simply moot.But, if there are no alternatives, why is this so? I have already expresseddoubts about arguments rooted in materialist metaphysics. Could it be thatprevailing social forces do not permit or foster alternatives?

My suggestion is that in contemporary consciousness the categories ofscientific materialism, reinforced by the hegemony of the modern values ofcontrol, have become so dominant (functioning as a virtual a priori) thatthe categories in which alternatives could be articulated have beenmarginalized or deemed pertinent only to expressive rather thaninvestigative activities. Following this suggestion we might look foralternatives (or perhaps anticipations or residues of them) at the margins ofthe advanced industrial world or among minority approaches inmainstream institutions. That is why (Chapter 8) I will address efforts insome third world countries to integrate materialist knowledge andtraditional local knowledge, and (Chapter 9) discuss alternatives that havebeen proposed from feminist perspectives. I do not look for one, but adiversity of more modest, less encompassing alternatives, complementingeach other and materialist inquiry, reflecting the diversity of social andcultural values which have been adopted in different places. These arenot alternatives in which social values will trump cognitive values, but ones


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (158)

in which social, cultural and material factors are considered together incomplex, interacting causal networks (in projects of systematic empiricalinquiry, whose products are subject to appraisal in view of the cognitivevalues), and in which the quest for those possibilities that may furtherhuman flourishing is foremost. The arguments developed in Chapters 8 and9 will further illustrate how social and cognitive values play their roles atdifferent logical moments.

Before proceeding to consider anticipatory alternatives, in the nextchapter I will consider, in the course of a critical reflection on Kuhn, howsocial and cognitive values played their respective roles in the process of thetransition from pre-Galilean to Galilean science. I will rebut Kuhn’saccount (one that, unlike Taylor’s, does not rest upon seeing technologicalsuccess as a symptom of cognitive superiority) of how the cognitive valuesalone account for the rationality of this transition, and in doing soreinforce my conclusion about the grounding of the materialist strategies inthe link with the modern values of control. This will further open up thespace for entertaining alternatives to the materialist strategies.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (159)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (160)

7Kuhn: scientific activity in different


I have maintained that social and cognitive values both play importantroles in scientific activity, but the roles are played at different logicalmoments. Social values can have a legitimate role concerning the adoptionof a strategy, but a theory is properly accepted of a domain of phenomenaonly if it manifests the cognitive values to a high degree. Consistent withthis there may be alternative strategies (to which the materialist strategiesare subordinate or with which they are incompatible) under whichtheories, manifesting the cognitive values to a high degree, could bedeveloped. In this chapter—looking for insight into how to make soundtheory choices when conflict extends across strategies, and thus laying thegroundwork for a defensible account of impartiality in Chapter 10—I willreflect on the historical moment at which the materialist strategies began tocome into their dominant position, prior to which (in Europe) thestrategies of Aristotelian science largely framed scientific activity. At thismoment, there were alternative strategies in competition, and I will arguethat social values actually played key roles in supporting the adoption of therespective strategies, and a crucial role in bringing about the virtuallyuncontested adoption of the materialist strategies (further consolidating theconclusion of the previous chapter). Here I disagree with Kuhn, whomaintains that the Galilean (materialist) strategies were adopted principallybecause of their demonstrably greater fruitfulness in generating theoriesthat manifested the cognitive values to a high degree.


With his famous aphorism: “…though the world does not change with achange of paradigms, the scientist afterwards works in a different world”(Kuhn 1970:121), Kuhn points to an important feature of transitions, likethe one from Aristotelian to Galilean science. In what sense didAristote lian scientists work in a different ‘world’ from the one Galileanscientists work in? How do these ‘worlds’ exclude each other? Why workin one ‘world’ rather than another, and when is the scientist confrontedwith that choice? Are there cognitively (rationally) compelling grounds to

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (161)

work in the Galilean ‘world?’ What changes, and what does not change,with a fundamental change in the strategies that frame research activity?1

Can paradox be avoided if one accepts Kuhn’s further claim that scientists,working in different ‘worlds,’ observe different (and incompatible) thingswhen looking at the same object? It is a cluster of questions like these that“the new-world problem” (a term coined by Hacking 1993; see also Kuhn1993) designates.

The world, scientific ‘worlds,’ social “worlds”

There is only one world, the repository of all possibilities, or the totality ofthings, events and phenomena that constitute the causal order of whichhuman activities and experiences are a part. The world, characterizedabstractly in this way, can be thought of, for example, as the Aristoteliancosmos (shaped by teleological causal principles) or the modernspatiotemporal totality (structured by quantitative laws). The world that“does not change with a change of paradigms” is, for Kuhn (I take it), thecausal principles and fundamental constituents of the world. BothAristotelians and Galileans agree that they do not change, though theydiffer about what they are.

Any talk of “worlds,” therefore, is essentially metaphorical and, incolloquial idiom, it tends to move among multiple layers of meaning.Following the metaphor, a “world” is a kind of self-contained totality, as itis grasped, interacted with and articulated by its “inhabitants”; unlike theworld, a “world” does not exist apart from the practices, modes ofinteraction, self-understandings, and articulations of its human inhabitants.In colloquial usage, “worlds” may overlap and be contained in other“worlds,” so that their limits and dependencies, and that there arealternatives to what is taken for granted in them, may not be recognizableas such from within; and they all manifest historicity. The self-containmentand all-inclusiveness is only more or less, allowing that a “world” may besusceptible to “outside influences” (coming from the world or other“worlds”), which sometimes explain important happenings in it, but whichthe categories deployed in its commonly shared articulations may be ill-suited to grasp. The historical sustainability of a “world” depends upon itsbeing able to gain or to maintain its high degree of self-containment andall-inclusiveness in the face of conflict and competition for resources withother “worlds.” Ultimately, then, to understand well what happens ina “world,” one needs to gain access to an explanatory scheme thattranscends the provinciality of one “world” and to be able to compare“worlds,” recognizing how they interact with one another in an ever moreencompassing movement, so that one will recognize layers of “sub-worlds”contained in larger “worlds.” It is doubtful that there could be a “world”


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (162)

that contains all “worlds”; but the larger a “world” is the more relativelyself-contained it is likely to be.2

What Kuhn means by “the world in which the scientist works” has beeninterpreted in many different ways, and what truth, if any, it may beattempting to disentangle about scientific change remains a matter ofcontroversy. Some (for example, Sankey 1994, 1997), who considersemantical incommensurability to be an important (though limited)phenomenon, tend to dismiss the aphorism as simply inflated rhetoric onthe ground that it ignores that different theories may refer to the same“mind-independent objects” even though their respective referring termsare not inter-translatable. Others (for example, Hoyningen-Huene 1993)have provided a neo-Kantian interpretation, identifying scientific ‘worlds’with “phenomenal worlds,” and the world with the epistemologicallyinaccessible “world-in-itself.” Still others (for example, Rouse 1987)emphasize that a ‘world’ is linked with a “form of life,” its required skills(habits, expectations and sense of what is possible), its organizingstructures, and its ways of actively engaging in research3

Developing the last line of interpretation, I take a scientific ‘world’ to beconstituted in practices that are shared among a community ofscientistsdata gaining, theory formation and appraisal, and theoryapplication practices—and in the shared beliefs and categories deployed tomake the practices, their conditions and their outcomes intelligible,communicable and effective. A ‘world’ is the set of objects that becomesarticulated in and by the characteristic practices shared within thiscommunity, and that are interacted with (in characteristic modes) andwhose possibilities are investigated (and sometimes brought to realization)in these practices.

According to Kuhn, scientific activity is necessarily carried out within a‘world,’ and any understanding we have of the world is always gained withina’world,’ for we can only investigate the world by investigating objectsthat are appropriately characterized and appropriately interacted with.Within a’world,’ theory developing and data gaining practices arereciprocally related by strategies that they deploy: strategies that constrainthe kind of theory that may be entertained and lead to the selection andseeking-out of empirical data with certain characteristics—such as thestrategy to entertain only theories that can be constructed with thespecialized vocabulary of a chosen lexicon, and to select data that areexpressed in categories appropriate for being put into relationship withsuch theories for the sake of appraising how well they manifest thecognitive values. We might say that the object of scientific inquiry (or anapproach to scientific inquiry, Chapter 5) is that which can be graspedunder the adopted strategies.

Historically there have been changes of ‘worlds’ in which scientificactivity is carried out. A change of strategies accompanies (and partly


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (163)

causes) a transition from theory to theory—like the one from Aristotelianto Galilean theory–where the theories on either side of the transition areincommensurable: the lexicons in which they are respectively formulatedcontain key terms that cannot be intertranslated; and empirical data,expressed without implicating one or other of the lexicons, generally arenot available as a decisive ground for choice among the fundamentaltheories. The historical reality of these transitions challenges—and Kuhn’saphorism suggests how deep the challenge is—the common (logicalempiricist) accounts that transitions from theory to theory can beexplained, and ideally shown to be rational, in terms of relations betweenthe theories and a body of empirical data that all of the competing theoriescan be expected to encompass.

The challenge requires that the question of the rationality of theorychoice be relocated so that it is linked with that of the rationality ofadoption of a strategy. Thence the new-world problem. I will give specialattention (later in the chapter) to the question: Why (or why not) engage inthe scientific practices which deploy a new strategy? This question cannotjustifiably be reduced to one about which fundamental theory to accept, orto any questions about soundly held beliefs. Change of strategy doesinvolve change of lexicon (whose expressive power defines the limits of thepossible of the new ‘world’), and thus of the categories with whichadmissible theories are formulated. This, in turn, may involve change ingeneral beliefs about the object of inquiry, its general features, andfundamental causal principles, as well as of the theories that are accepted.But the new strategy is not adopted as a rational response to the newbeliefs, for the beliefs, theories, and strategy develop together in complex,on-going interaction within the new and developing patterns of scientificactivity. (That is why, after the transition is complete, the gulf between thenew and old ‘worlds’ can be so vast.) Even if the general beliefs about theobject of inquiry and fundamental causal principles are considered as(metaphysical) beliefs about the world, at least for a period of time the newstrategy cannot be rationally grounded as the one especially suited forgenerating concrete knowledge of the world (since, pending the unfoldingof the strategy, that the new metaphysical beliefs about the worldconstitute knowledge, could not be rationally grounded). Consequently, atleast during this period, support for adopting the strategy and for acceptingany accompanying metaphysical beliefs, must partly come from othersources. One source might be the (social or moral) value attributed to themode of interaction with things which is prioritized in the new scientificpractices. In that case, the source would be in the social “world” in whichthe competing scientific ‘worlds’ are located.

Kuhn’s discussion of the question about the adoption of a strategyconcerns only the world and scientific ‘worlds,’ but not social “worlds.”But a scientific ‘world’ is always part of a social “world”; scientific


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (164)

practices have social conditions, and scientific theory is applied in theprevailing social “world”; and changes of scientific ‘worlds’ and social“worlds” often accompany each other. Bringing social “worlds” into thepicture, we will see, helps to relieve some of the paradox that may seem tothreaten Kuhn’s position.

The new-world problem incorporates also “the old-world problem”:What are we to say about the old scientific ‘world’ and, above all, about theobjects that are said to constitute it and the kinds of objects said to beobserved in it? Aristotelian theory contains posits about natural motions(motions directed towards natural ends) and the motions of the heavenlyspheres. Are we to say that these objects existed in the Aristotelian ‘world,’but no longer exist in ours? Or, are we to say that these objects do not, andnever did, exist? The first alternative has the ring of absurdity. But, giventhe incommensurability of Aristotelian and our science, how could werationally justify opting for the second?

Kuhn (1970) describes the theory transitions under discussion asrevolutionary, making an analogy with revolutionary historical transitionsbetween incompatible political orders or political “worlds,” such asabsolute monarchic and democratic “worlds.” Does this analogy provideguidance for the old-world problem? It is clear enough what we are to sayabout the old political “world”: it once existed, and now it does not for itwas overthrown and replaced by the new “world”; monarchic institutionsand the roles they nurtured once were real, and now they are not.Moreover, we are comfortable in explaining political revolutionarytransitions largely in terms of such factors as material and economicconditions, personal and social values, and power. From the perspective ofa social science theory of the new democratic “world” we may be able toexplain the incompatibility of democratic and monarchic “worlds” andweaknesses in the monarchic “world” that may not have been apparentfrom within, we may reject beliefs that were essential to the maintenance ofmonarchy, we may value monarchic institutions differently, but we will notdeny the existence of these institutions and their constitutive roles.

Kuhn, unlike Feyerabend (1989), shrinks back from embracing theradical implications of following the political analogy wholeheartedly.In the Aristotelian ‘world,’ as portrayed in its theories, there are naturalmotions and the like, and the practices (including observational ones) ofAristotelian science are characterized by its practitioners in terms ofinteractions with and relations to such objects. In contrast, in the Galilean‘world’ there are lawful motions (motions that fit differential equations,undirected to ends). If the political analogy carried through in detail, wemight be led to conclude: once there were natural motions, now there arenot; now there are only lawful motions. It also might suggest that, as withmonarchic institutions, we have excluded natural motions from our‘world’ for reasons of social value or socio-economic power, and thus that


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (165)

the theory transitions are explicable primarily in terms of social factors.Here Kuhn makes the crucial break with the analogy: the theory transitionscan be explained and justified in cognitive (epistemic or rational) terms.Because of that we properly conclude that natural motions do not andnever did exist (see pp. 168–70); Aristotelian scientists believed that theydid exist and, given their epistemological horizon, their beliefs were well-founded, but they were mistaken. Across the gulf of the theory transition,not only are the old and new theories incommensurable and products ofincompatible practices, they also are cognitively incompatible. In somesense both the old and new theories are (in part) about the same things(objects in the world?), and (in some sense) in some respects the newstrategy enables us to grasp certain phenomena and spaces moreadequately. It is not that we have excluded natural motions from our‘world’ because we deny a certain kind of value to them; we deny that theworld contains them. There is no counterpart of this in the politicalanalogue.

Kuhn’s metaphor of ‘worlds’ now appears as a metaphor of a metaphor!As ‘worlds’ problems multiply, one might wonder if these problems are

artifacts of using the metaphor, rather than problems driven by thephenomenology of scientific change. Why not, one might ask, simply saythat the Aristotelians were fundamentally mistaken about the world? Whysay that they conducted their scientific activity in a different ‘world’ whencentral objects with which they attempted to organize and explain theirexperience do not exist? They said that they were investigating theproperties of natural motions, but they were not doing that for there are nonatural motions to investigate. So they were radically misdescribing whatthey were doing and what they were observing. We and they both investigatethe one world, and we have evidence that our theories offer better accountsof it. This is clear enough. Why confuse the matter by piling metaphorupon metaphor?

Kuhn resists the impact of this line of questioning, in part because hedoes not think that it captures the magnitude and (above all) the rangeof dimensions of the transition. It is also in part because he rejects what isoften an undercurrent, that we (modern scientists) have got the broad linesof the causal principles and fundamental constituents of the world right, sothat we should not be considered to be doing science in a ‘world’ ofcomparable status to the Aristotelian ‘world.’ For Kuhn, as pointed out onp. 150, any understanding of the world is gained within a ‘world,’ whosestrategies (and thus the character of the practices in which they aredeployed) partly constitute the object of scientific inquiry. In this (and onlythis) respect Galilean and Aristotelian science are on an equal footing. Fromour viewpoint, we can affirm justifiably that certain objects of Aristoteliantheory do not exist in the world, and that the Aristotelian lexicon reflectedmore the character of the practices of Aristotelian science and the values


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (166)

linked with them than features of the world. But our viewpoint does notground that the Galilean lexicon reflects features of the world more thanthe character of the practices of Galilean science, and the values linked withthem. We have no ironclad assurance that our ‘world’ will not change,even go the way of the Aristotelian ‘world,’4 and that our strategies mightnot come to be supplanted too. There is nothing about scientific practicesthat entitles us to affirm that theory offers understanding of the world, orobjects in it, independent of their relations with human beings. Thatdoesn’t mean that we don’t gain knowledge of the world; we are part ofthe world. We grasp it against the background of an essentially historical,structured lexicon. That is our ‘world,’ the world of which we haveknowledge, a ‘world’ constructed in the course of scientific practices.

What is clear is that our strategies are working very well in solvingpuzzles today, just as (though with much greater effectiveness than) theAristotelian one did for a time in its own way in its own ‘world’ before(according to Kuhn) dissolving in the face of internal crisis and theemergence of a more compelling competitor. Whatever its weaknesses maybe diagnosed to be from the viewpoint of the Galilean ‘world,’ it remainsthat, in its own time, Aristotelian theory provided of a significant body ofexperience what were reasonably taken to be illuminating explanationsthat were superior to available competitors. In order that doubts could beformulated in a probing way about whether its core categories referred tothings and kinds in the world, a new kind of competing ‘world’ needed tobe in play. A new kind of competitor, deploying a new strategy, could sowsimilar doubts about our current lexicon, though it would leave intact(even if no longer relevant for theoretical or practical ends—cf. thediscussion of alternative strategies in Chapters 8 and 9) a considerablebody of empirical knowledge. It is fair to say that our best theories, insofaras they apply to certain specified domains, have been submitted to far moresevere testing than their predecessors.


According to Kuhn, a “good” theory, regardless of the strategies underwhich it is developed, is related to the relevant empirical data, and to othertheories, in such a way that the cognitive values are manifested to a highdegree (Chapter 3). The empirical data—the observed facts—are beliefsgained properly through acts of observation and interactions of humanswith objects which fix beliefs. Beliefs reflect intentionality. They are relatedwith other beliefs with which they exhibit logical, rational and evidentialrelations, and as such can be expressed as having propositional content(Chapter 3). Beliefs, thus, presuppose a lexicon. For Kuhn, there is no“observational lexicon” or special vocabulary that does not draw from


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (167)

theoretical categories that is especially apt for reporting the observed factsrelevant to evaluating scientific theories. I will now illustrate this byshowing in detail the way in which much of the empirical data, relevantwithin scientific activity, bear the mark of strategies, and thus cannot servegenerally to resolve issues of theory choice that cut across the strategies. Thisdetail will throw light on how ‘worlds’ are incompatible and will pointtowards solutions of the new-world problem.

Compare formulations of physical theories of Aristotelian (A) andGalilean (G) types.5 On the one hand, they aim to express understanding ofobjects of the world. On the other hand, they express different ontologies,and refer to different natural kinds. The intentional objects of A and G aredifferent. A is about such things as the natural motions of things and theterrestrial elements, fully characterized in qualitative terms; whereas G isabout the underlying law, structure and process of things, characterized inmathematical terms. Neither theory has the lexical resources to express inits own terms the full ontology of the other. This fits with the metaphor: Aand G are about objects in ‘different worlds.’ And the differences go all theway down to the observational data. Exploring the implications of hisaphorism, Kuhn added: “when Aristotle and Galileo looked at swingingstones, the first saw constrained fall, the second a pendulum…. Apendulum is not a falling stone… Galileo saw the swinging stonedifferently” (Kuhn 1970:121–3). Different things are observed in the two‘worlds.’ Yet A and G make their observations while “looking at” the samething:

the proponents of competing paradigms practice their trades indifferent worlds. One contains constrained bodies that fall slowly, theother pendulums that repeat their motions again and again….Practicing in different worlds, the two groups of scientists see differentthings when they look from the same point in the same direction…this is not to say that they can see anything they please. Both arelooking at the world, and what they look at has not changed. But insome areas they see different things, and they see them in differentrelations one to the other.

(Kuhn 1970:150)

Although the theories of A and G are incommensurable to a significantextent, their theoretical categories figure respectively in observationalreports made while looking at objects in the one world. But, in the lexiconsof A and G there is no way to state that the reports are about the samething, for the referring terms (inhibited free fall and damped pendulum) areas different as the referring terms used to describe the institutions of twosocieties separated by a revolutionary divide. And there are noobservational reports that do not use theoretical terms from one or other


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (168)

of the theories that both A and G would not only assent to, but also deemrelevant to their theoretical deliberations.

A and G are different. They are incompatible only if, at least in part,they are about the same objects. To settle whether they are about the sameobjects, we need to go beyond the conceptual resources of the twolexicons, to go beyond inter-and intra-theoretical reflection and, I willargue, pay attention to the role of theory in application, for instance inproviding understanding of phenomena in the realm of daily life andexperience, and guidance in carrying out practical activities. Thesephenomena may be observed within common practices in which both ofthe competing theorists participate in their daily lives. Consider a child’sswing.

Observing things from the perspectives of differentpractices

There are countless (logically) distinct observed facts about any thing. Gand A observe many of the same facts about the swing: that it is a swing,that it has a certain spatio-temporal location, that its chains have a certainlength, that it needs painting. Which observed fact about a thing isconsidered at a given time depends on context, purpose, audience, andactivity: playing with a child, giving instructions to a repair-person,engaging in scientific activity. Being able to identify the same thing acrosspractices and contexts, and so being able to refer to the same thing usingexpressions with different meanings, is a normal part of learning themeaning of basic referring expressions. It is one and the same swing thatmy child is playing on, that I bought in the store, that wasrepaired yesterday—and to whose movements I am trying to apply thescientific theories I endorse.

Outside of scientific activity, G and A agree on most of the observedfacts. That is how we know that they are looking at the same thing, andthat what they are looking at does not change with the change of strategy.There are social practices in which there is interaction with objects towhich the theories on either side of the transition are both applied and thatdo not change with the theory transitions. Using their vocabulary, objectslooked at by both theorists can be described in a way to which they bothassent. That is enough to assure that A and G, regardless of the differencesin what they observe when engaged in scientific activity, are looking at thesame thing; and it consistent with Kuhn’s view that we cannot grasp theworld, except from within a “world.” That “the world does not change”plays no operational role in providing this assurance. Within theirrespective scientific activities, A and G report different observed factsabout the swing when they look at it—different and also incompatible.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (169)

At one level, the difference is like that present when a coach, player,commentator and spectator each describes a particular episode in a soccergame. They observe different things because of their different perspectives,modes of engagement, interests, available vocabularies, and the capabilitiesto make the discriminations necessary to deploy the relevant vocabulary.But these descriptions need not be incompatible; and, if they are, with“correction” compatibility can readily be obtained. On the other hand, A’sand G’s observational reports are not incompatible in the way in whichreports of a thing that it is completely red and completely green areincompatible, where at least one is straight-forwardly erroneous, andwhose resolution requires, for example, further observation and carefulattention to what is being observed. I will return to the question of what sortof incompatibility is involved.

Observation and its goals within an explanatory practice

Let us consider an observation to be an action of looking at a phenomenonwith the immediate goal to produce a belief (expressed in an“observational report”) that is appropriate to the activity being engaged in.Then, the immediate goal of observation is subordinate to the objectives ofthe activity at hand.

When the activity is science, the objective may be, for example, relatingthe phenomenon to the causal order in which it is produced and hasconsequences, or encapsulating the possibilities open to it (Chapter 5).Then, categories will be used in scientific observational reports thatenable them to be connected with proposals about the causal order, and sothey may be drawn from the specialized vocabulary of one’s adoptedlexicon. Moreover, which observations are made may vary with views (thatmay be implicit in the lexicon) about the fundamental causal order andabout the best available general theories. These, in turn, are connected withdifferences in scientific practices, including disciplined, shared and stablenorms of observation. What a phenomenon is observed to be derives (inpart) from practices and presuppositions, and involves its place in thepractices. What is observed—the phenomenon, under the description of theobservational report—is constituted (in part) by human intentionality(McDowell 1994), and it is a product of an interaction: both a way oflooking at the phenomenon or at it under certain conditions or at certainsequences of its states, and looking at it in relation (a product of ourinteractions) to certain objects. A observes that the swing (in motion) is inthe state of inhibited free fall; G that it is a damped pendulum.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (170)

The temporal interval of an observation

Any observation takes place over a period of time, so that what is observedmay vary with the sequence of events held to be constitutive of thephenomenon being looked at. For A, in our example, the relevant sequenceends when the swing comes to rest as close as it can get to the earth’ssurface. That it is observed to come to rest there is critical for linking thephenomenon to the general causal order, as represented in A’s physicaltheory. For G, the relevant sequence is much shorter, including a sufficientnumber of the back and forth movements to enable the period to bemeasured (Kuhn 1970:119, 123).

Both A and G, no doubt, would assent to the following twoobservational reports: that the swing eventually comes to rest, and that itsmovements have a period of approximately (say) two seconds; just as theywould both recognize such other facts as the color of the swing, the rustingof its chains, and the sounds of pleasure coming from the children playingon it. Scientific theory is not expected to “fit” all observed facts, only theones appropriately selected in the light of scientific objectives. A and Gagree that none of these other facts are pertinent for the objectives ofscientific practice; but they disagree on the pertinence of the first tworeports. For A, the periodicity of the swing’s motion has little morerelevance than the rust on the chains; for G, that the swing comes to rest isa fact to be explained, but not one critical for grasping the causal order ofthings.

Each of A and G looks at the sequence he deems relevant and draws fromthe categories of his adopted lexicon; then A observes “inhibited free fall,”and G “damped pendulum.” Within their relevant communities, suchobservations are routine, spontaneous, stable, marked by intersubjectiveagreement, and not inferred from reports made in a purely “observationallanguage”; and the phenomenon is replicable. In the present case, A and Gboth could have observed differently; they could have producedobservational reports to which both would assent, and had they beenengaged in another activity together (for example, consulting with a repair-person) no doubt they would have. When pursuing scientific objectives,however, characteristically they do not.

One might query: where we confront two stable but incompatibleobservations of the same phenomenon, should we not restrict the relevantlexicon for observational reports to that which is shared among thedisputants? Then, would it not be clear that the disputing parties aresimply looking at different sequences of events, each of which is nowdescribed in a way that all can accept? Is not G’s sequence simplycontained in A’s, with A and G attending to different aspects of it: A thatthe swing come to rest; G that the movements are approximately periodic?


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (171)

Then, should not the theories of both be judged in terms of their “fit” withthe common set of data?

Observation and what is looked at; observation andmeasurement

In fact, however, G does not just look at some of the movements in asequence, while A looks at all of them. A may not look carefully at theback and forth movements at all. More importantly, G looks at therelationships of the movements to various instruments, such as clocks. Gmeasures rather than simply observes. G’s and A’s “less theory-laden”observed facts are, to a large extent, distinct. They need not be totallydistinct. G and A compete. G, too, observes that the swing comes to rest,and he offers an explanation of it; but, that it comes to rest, is irrelevant asan explanatory factor of the motion. They both look at the swing. But thatunderdescribes what each looks at directly, and misses the differences inthe ways they look at it. G’s way of looking at the swing is to observe itsrelationship with instruments, which he himself makes part of thephenomenon. G, through measurement, instrumental and experimentalinterventions “creates” many of the phenomena that he looks at (Hacking1983); A looks at “natural” phenomena. For A, G’s observations are of“artifacts,” not of “nature”; investigating them can illuminate humanobjectives, but not the natural order. For G, A’s reports cannot gain usaccess to the causal order of the world, which is marked by quantity andlawful relations among quantities.

As indicated already, A may assent to G’s (less theory-laden) observedfacts (for example, that the period is two seconds) while regarding them asirrelevant to the objectives of scientific practice. Although such facts mayelude the predictive and explanatory scope of A’s theory (like the rusting ofthe swing), they do not contradict it. With the developing sophistication ofG’s practices, however, his observed facts may only be recognizable bysomeone skilled in the relevant instrumental interventions andexperimental manipulations, much as an unskilled spectator (or a skilledrugby coach) cannot observe all that a soccer coach observes in the courseof a soccer game. The phenomena they describe come into being only in thecourse of a practice involving such skills (Rouse 1987).8 In these cases, G’sobserved facts make no claim that A, given his adopted strategy, need takeinto account. A has no reason to consider the testimony of G authoritativefor contexts outside of those of G’s practices, for G’s practices concernanother realm of possibilities, not those of nature; just as a soccer coachhas no reason to consider the testimony of a rugby coach authoritativeoutside the context of rugby. A and G are “playing different games.” Likesoccer and rugby, they cannot be played on the same field at the same time(Taylor 1982)—but which playing field is the swing in?


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (172)

Observational reports and how they are theory-laden

Any observational report involves classification, representing thephenomenon as of a kind, and thus treating it as like certain otherphenomena. skillfull observers, in the act of observing, make the relevantclassifications spontaneously without engaging in inference. In observingthe swing, the phenomena that A and G (both skillfull observers) considerthe swing’s motion to be like are different.

A relates the swing’s motion to that of freely falling bodies, where“relate” means first to affirm a certain likeness, and second to imply that itis deviation from free fall that needs to be explained. In scienceobservations may be made for the sake of displaying a phenomenon as partof a causal order; then, we observe in a way that facilitates that objective. Arelates the motion of the swing to a “natural” motion. In contrast, Grelates it to the motion of a pendulum without retarding forces. For G, it isdepartures from the pendulum’s motion (represented ideally) that need tobe explained. To understand the swing’s motion, G relates it to aphenomenon which is an experimental artifact, a product of skillfullhuman activity, more exactly, to an idealization of this phenomenon.

In this way each observational report is theory-laden, for instance itpresupposes a theory and deploys its categories, respectively A′s theory ofnatural motions and G’s of the pendulum. For A, G’s mode of relatingis doubly problematic: such pendula do not exist, but rather they areidealizations; and even as approximations, they do not exist “in nature,”but only as constructed in experimental spaces, as human artifacts.

G’s observation, thus, presupposes working in a certain sort of ‘world,’one in which the practices of skillfull experimentation and measurement,and mathematical idealization have been developed among a community ofscientists. A theory of the pendulum’s motion is grounded in phenomena,aspects of which are measured, in experimental spaces whose boundaryconditions, as well as the initial conditions of sequences of events in them,are humanly controlled. Thus, it is grounded in observed facts ofphenomena that are the causal consequence of planned human intervention.Again, given A′s outlook, while he may assent to many of these facts, theyhave no relevance to (his notion of) the objective of science, and the lexiconupon which his theories draw contains no categories which could usefullyredescribe these facts. They may be relevant to a skilled artisan’s practice(dealing with, for example, clocks, musical instruments, pumps orweapons), but not to science.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (173)


A’s and G’s data gaining and theory development practices are significantlydifferent, involving different immediate objectives, skills and patterns ofobservational discrimination, so different that their core lexicons cannot beinter-translated. Are they just different “games”—each with its own lore,styles, skills, possibilities of achievement, rules and mode of discoursethat,like soccer and rugby, could coexist in the same social “world” at the sametime (though not on the “same field”)?

We sense, however, that there is not just difference, but deepincompatibility. It seems that the only way for A and G to coexist (in thelong run) would be if they concerned completely distinct domains ofphenomena. But, despite the differences discussed in the previous section,each offers an explanatory account of the motion of the swing and of othersignificant phenomena in the realm of daily life and experience, forinstance of those things, events and phenomena that are confrontedcommonly in daily life in the social “world” in which the scientific ‘worlds’are located, and which structures such activities as those directed tomeeting basic needs and to the ends of economic production. How are A′sand G’s observations of the swing incompatible? Kuhn is right that it is notin virtue of inconsistency expressed within the lexicons deployed by A and/or G. Could the observations be considered complementary, as thedescriptions of the swing as a child’s plaything and as a commodity for salein a store are?

Practical and cognitive aspects of incompatibility

Two features seem to mark the incompatibility. One is practical orpragmatic. It is that the scientific practices (like those of soccer and rugby),including the data gaining practices, of A and G are incompatible: onecannot simultaneously observe a natural (in A’s sense) phenomenon andmeasure it or observe it under human intervention; one cannotsimultaneously relate a phenomenon to its place in the cosmos and relate itto an idealized model of an experimental phenomenon.

The second feature ensures cognitive, and not just practical,incompatibility. It arises when we apply either of the theories to theswing’s motion. Then, we make use of vocabularies that extend beyondthose of the two theoretical lexicons, and in effect include items from thelexicons among the categories of daily practical activity. We thereby gaingreater expressive resources with which we can derive a contradiction fromthe two observational reports: from “the swing is in a state of inhibited freefall” we infer “it is moving towards its natural end, the center of theuniverse”; from “its movements are those of a damped pendulum” we infer


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (174)

“it is moving in accord with mathematical laws”—but, if it is moving inaccord with mathematical laws it is not in a state of natural motion, andthence a contradiction is derived. To derive the contradiction we need thegreater expressive resources of the lexicons of daily practical activity, whichcan serve as a metalanguage of both theories. A theoretical lexicon can belooked at from two perspectives. On the one hand, it is learned and theskills for its deployment are gained during an apprenticeship, in the courseof practical engagement, using its categories in connection withobservational (measurement, experimental) and theoretical tasks. Onesimply learns the lexicon in use. On the other hand, it can be discussed in ametalanguage, compared with other technical lexicons, and the reasons forits adoption investigated. According to Kuhn, the former can happenwithout the latter and most of the time (“normal science”) it does. As longas the lexicon remains a fruitful means for resolving puzzles so that itscomparison with other lexicons is not an immediate issue, this need causeno tension. Then, contradiction that can only be formulated at themetalinguistic level remains invisible, and to practitioners who have notbecome flucnt in the metalanguage the difference between A and G wouldappear to be similar to that between soccer and rugby, having as muchimpact on their consciousness as cricket has in the USA.

These two features of incompatability are not specific to the case of theswing; they have systematic sources. Neither A nor G makes anobservational “error” in observing the swing—as G does, for examplewhen he mismeasures the period of the swing, or A when he reports risingsmoke as an instance of inhibited free fall. Such errors can easily becorrected without disruption of theory. The “error” in the present caseresides in the acceptance of a theory and ultimately in the adoption of astrategy (and its associated skills and practices). There is no relevant“correction” of either of the two observational reports that can be madewithout abandoning the strategy under which the observation was made.The incompatibility of the observations derives from a systematicincompatibility of the strategies of A and G, which has a cognitive (logical)dimension in that, as just argued, when embedded in an appropriatemetalanguage, reporting observations with the categories of the respectivetheoretical lexicons leads to contradiction. It also has a pragmaticdimension: the data selection strategies are parts of (locally) incompatiblepractices. From the cognitive dimension of incompatibility, if follows thatone cannot choose both of A’s and G’s theories of domains of phenomenathat contain the swing’s motion. But the practical dimension precludesempirical adequacy grounding a decisive choice between them.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (175)

Incompatibility pertaining to application

There is also a third dimension of incompatability, pertaining to the realmof daily life and experience: the competing theories cannot be appliedtogether to objects in this realm, so that in the long run the broader social“world” cannot contain as “sub-worlds” both of the scientific ‘worlds’constituted respectively by the practices of A and G (even, as it were, on“different fields”).

Certain constancies in the realm of daily life and experience, maintainedthrough the period of theory transition, ensure—I have argued— that (inpart) A and G are about the same things, and thus incompatible. The swingcannot be both an instance of impeded fall and a damped pendulum, but itlends itself easily to either interpretation, so that its role in social practicesremains largely unaffected by the outcome of the conflict between A andG. On the other hand, during this period the realm of daily life did changein significant ways, as the historical processes of shaping modernity wereunder way. The characteristic stance towards natural objects, as manifestedin mechanical innovations and expressed in the leading articulations of thetime, moved more to that of control and away from adaptation orattunement. Through this there came about a striking change in the objectsthat became central to the structuring of daily life, or at least to setting itstransforming trajectories. Before the transition, A applied most easily tomany of them; after it, to an increasing extent, G did. After the transition,the scientist worked not only in a new scientific ‘world’, but also in a newsocial “world.”

A and G do not apply in the same way, and they cannot be appliedtogether coherently. Upon development, they provide different anticipa-tions of what is possible in the realm of daily life and experience, differentaccounts of the limits of the possible, different guidance about means toends and the consequences predicted to follow from actions, and differentexplanations of how things work. Each theory, upon application, partlydefines possibilities of a social order shaped by a particular conception ofhuman well-being or flourishing, and a particular characteristic stance ofhuman beings towards material things—A that of “adaptation”, G that of“control” (Chapter 6). Thus, these differences may also be accompanied bydifferent assessments of the significance (value, worth) of the kinds ofpossibilities anticipated, for instance different judgments about the value ofrealizing them in the realm of daily life.

The possibilities, that A and G respectively highlight, cannot be realizedtogether systematically in any historical “world.” A social “world”increasingly shaped by relations with mechanical objects (to which G easilyapplies) undermines the conditions of stability and relative constancy(where A easily applies) required to articulate coherently daily life andaspirations towards flourishing in terms of attunement to nature or to the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (176)

cosmos; conversely, commitment to maintain such stability poses obstaclesto projects of technological development10.

After the period of the theory transition G, but not A, was widelyapplicable, at least in the practices of the ascendent powers that werereshaping the social “world” in the direction of mechanical objectsbecoming of major salience in daily life and productive practices. Before it,A applied well within the “world” of medieval Christendom as this“world” was represented in its dominant articulations; whereas, evenretrospectively and counterfactually, G would not be much applicable inthis “world,” except to phenomena that were relegated in the dominantarticulations to the status of artifact rather than of nature.

A applied well to numerous objects of salience in the realm of daily lifeand experience, and its practices were cultivated, first in the historical,social “world” of ancient Greece and later in that of medieval Christendom.A’s ‘world’ was a “sub-world” of these larger “worlds” or cultures, fittingneatly with them and the self-interpretations borne with them because ofseveral mutually reinforcing features: the ubiquity of teleology (andmeaning); emphasis (which became articulated theologically with thebacking of the Church’s authority and power) on hierarchical structurein the cosmos and in society, with the two structures considered asordained by God to mirror each other (Chapter 4); attention to the many-sidedness of things and interest in full understanding (Chapter 5); thecharacteristic stance towards nature articulated in terms of notions like“attunement” linked with “contemplation” or “being in touch with thenatural course and rhythms of things”; the sense (though not always theactuality, especially as the “world” of medieval Christendom entered intofatal crisis!) of a stable order, or at least that there were no significant ordesirable new possibilities to be discovered; and a strong distinctionbetween natural objects (responsive to natural ends) and artifacts(responsive to ends imposed by human beings).

No “world” is ever fully self-contained, all-inclusive and withouttendencies that run counter to what is articulated in its reigning self-understandings, and these counter-tendencies can cause crisis in it andbecome the source of a reshaped social order. Immediately before theperiod of the theory transition, the “world” of Christendom was, ofcourse, in crisis. Within it, projects linked with ascendent values andpowers that were to displace it involved increasingly a place for mechanicalobjects (to the workings of which G, but not A, could be applied easily);and the more they found a place in the realm of daily life and becamerecognized as a key part of the transforming trajectory of the social order,the manifestation of the cognitive values in A of the key objects in the realmof daily life diminished. I reiterate that those possibilities of G that wererealized at the time of the transition were (with few exceptions) notproducts of applied science in the contemporary sense, but objects of which


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (177)

G, but not A, could give an interesting explanatory account. What I amcalling G’s possibilities includes (extending far beyond those actuallyforeseen in the seventeenth century), on the one hand, those that can begrasped from successfully establishing theories that represent them lawfullyand, on the other hand, those that are opened up through adopting thestance of control as the characteristic one towards material things. Thosepossibilities that fall under the second description in general also fall underthe first (Chapter 6); but not always conversely (the planets).

As Kuhn has made us well aware, there were numerous anomaliesconfronting A, addressing some of which led to its late medievalmodifications, including projectile and circular motions, free fall,quantitative treatment of motions and forces, the discord betweenPtolemaic astronomy and Aristotelian physics. The anomalies were wellknown, pertinent to phenomena encountered in the realm of daily life andexperience, not ignored, but typically dealt with either by introducingauxiliary (including ad hoc) hypotheses or by instrumentalistinterpretations. Since G came to shed light on these phenomena, it istempting to say, counterfactually, that G could have solved problems facedin the earlier social “worlds.” But G could not have done so withoutcreating discord in those “worlds” and their self-understanding (as it laterdid)—unless G’s terrestrial science was interpreted as pertaining to thedomain of artifacts only, and the Copernican theory interpretedinstrumentally (as in Osiander’s preface to Copernicus’ DeRevolutionibus). It could not have illuminated the characteristic modes ofinteraction with material objects and with other people in these worlds. Inthe prevailing self-understandings, the anomalies were marginal to thesecharacteristic modes. Then, inquiry that focused on the objects encounteredin the characteristic activities of these “worlds” would have no place forG’s type of inquiry, since the latter is not a mode of inquiry that seeks fullunderstanding, but rather abstracts the objects of inquiry from their social,value-related and ecological dimensions. It is not a matter of ignoring theevidence or shutting off alternative modes of inquiry, but of focusinginquiry on key objects of the realm of daily life and experience prevailing inthe “worlds.” Theories, even our best established ones, cannot be graftedonto any social order.

My suggestion is that the applicability of G (or of A)—both as offeringaccounts of important (in the self-understanding of a “world”) objects inthe realm of daily life and experience, and as providing knowledge thatinforms important social practices—depends on what are the characteristicactivities of a social “world.” One cannot enter the ‘world’ of G withoutexiting the “world” of medieval Christendom, and so without entering anew social “world” that cannot coexist with that “world.”

In summary, I have identified three dimensions of systematicincompatibil-ity between A and G. First, their respective theories are


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (178)

developed under incompatible strategies. Second, within a language,sufficiently rich to articulate the realm of daily life and experience, whichfunctions as a metalanguage of both theories, a contradiction can be drawnbetween their theories. Third, social “worlds” in which the theories arerespectively systematically applicable are incompatible.

There is also, accompanying the theory transition, not only a change ofscientific ‘world,’ but also a change of social “world,” from one where Aeasily applies to one where G applies well within the ascendent practices.This is well known. What are the connections between the theory and thesocial changes? Kuhn would acknowledge causal ones, for example: that theemerging social “world” provided material and social conditions neededfor G to develop. Or, in the case of some later profound theory transitions,for example the chemical revolution of the nineteenth century (Hacking1993), the new social “world” might be partly brought aboutby technological applications of the new theory. But Kuhn does not thinkthat the connection has anything to do with explicating the rationality ofthe theory transition or of the change of strategy adopted. For Kuhn thetrajectory of the tradition of scientific inquiry is essentially autonomous.While, for example, the conditions, sources of support, institutionalizationand acceptance among the general public of the new strategy clearly needsocial explanation, the rationality of transitions from one scientific ‘world’to another can be assessed principally in cognitive or epistemic terms, interms of factors “internal” to the tradition—without taking into accountthe place of the scientific ‘worlds’ as “sub-worlds” of larger social“worlds,” the applicability of theories in them, and historical transitionsbetween them. I am not so sure.


According to Kuhn, one rationally adopts a strategy because it is morefruitful than its competitors. A strategy’s fruitfulness is assessed by itsenabling theories to be developed that manifest the cognitive values to ahigh degree in the light of data selected (and generated) in its data gainingpractices. After a certain time (by the time Newton made hiscontributions?), the greater fruitfulness of G’s strategy became widelyuncontroversial. This is compatible with incommensurability, forfruitfulness is judged within the framework of a given strategy. Within G’sframework, one judges that theories are being generated that aremanifesting the cognitive values increasingly to a higher degree; within A’s,one judges that, for example, anomalies are multiplying. So, on Kuhn’sview, no non-cognitive values are deployed in these assessments ofcomparative fruitfulness.

A, by his own lights, did not have to take into account all the new dataaccumulated by G, but he did have to address many of the criticisms made


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (179)

by Galileo in Dialogue Concerning the Two Chief World Systems in whichGalileo appealed to phenomena (for example, the falling of a lighter andheavier body joined by a string) that were in A′s purview, but which Acould only deal with by use of auxiliary hypotheses that lowered the degreeof manifestation of most of the cognitive values. That A faced anomalies(dealing with which lowered the degree of manifestation of the cognitivevalues) was well known. Galileo’s criticism showed that they could bemultiplied vastly. This is criticism conducted within A′s own terms ofdiscussion. That G also applied better to key objects of increasingimportance in the realm of daily life and experience, as such, did notchallenge that A applied well to the objects dealt with under its strategy,but it did put pressure on A by contributing to a growing sense ofthe insignificance of A. G also applied better to certain phenomena (forexample, the swing, projectile motion, the movements of the planets)—under descriptions selected by those working under A’s strategies,sometimes supplemented by categories deployed in common socialpractices-by incorporating them into a theory which manifested thecognitive values more highly of a domain that contains them. It is becauseA and G competed as theories of certain shared domains of phenomenathat the increasing fruitfulness of G’s strategies was inevitably accompaniedby decreases in that of A′s.

Aristotelian and Galilean accounts of planetary motions

I have emphasized that theories are accepted, not in an unqualified way, butwith respect to specified domains of phenomena; and (Chapter 3) Imaintained that if two theories conflict, choice between them cansometimes be made by appealing to the “comparative comprehensive”standard. If one of them manifests the cognitive values well with respect toa domain that includes that of the other, then ceteris paribus it is moreacceptable, even more so if, from its perspective, the success of the otherwith respect to the smaller domain can be explained. Let DA and DG be thedomains of phenomena of which A and G respectively manifest thecognitive values to a reasonably high degree. They only overlap; neither iscontained in the other (so “comparative comprehensiveness” does not helpto settle the dispute between A and G). Let D be the domain of overlap.The movements of the planets, moon and sun belong to D.11 Furthermore,data concerning these phenomena, observational reports of the varyingangular positions, brightnesses and other appearances of these bodies asobtained with naked-eye vision are selected to fit theories developed underboth strategies. Attending to significant areas of incommensurabilitybetween A and G should not obscure that these theories are expected tomanifest the cognitive values highly with respect to some of the same data.Both A and G wanted to “save the phenomena” of planetary motion, and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (180)

there is considerable plausibility in those accounts that speak of the“empirical equivalence” (or equal empirical adequacy) of Ptolemaic andCopernican astronomy. How does G manifest the cognitive values morehighly of D than does A?

It is well known that in A′s tradition, Eudoxan was replaced byPtolemaic astronomy largely because of the greater empirical adequacy ofthe latter (in relation to data describing the planetary phenomena of D)—even at the price of inconsistency with the fundamental posits of A’sphysical theory and thus significant reduction in explanatory power andinternal (organic) unity, which is reflected in the common recourse toinstrumentalist interpretations of Ptolemaic theory. Given that A′s physicaltheory is fundamental, that Ptolemy’s theory has the degree of empiricaladequacy it has is a mystery. How can posits strictly inconsistent with thefundamental posits of the cosmos predict better than posits (Eudoxan) thatare consistent with the fundamental ones? And there are puzzling details:Why do the Ptolemaic constructions of each of the planets havecomponents that are clearly linked with the sun’s annual motion? Andmuch appears to be ad hoc: the order in which the planets are arrangedfrom the earth as center, which combination of geometrical devices(deferent, epicycle, epicycle on epicycle, eccentric, equant) works best ineach case.

MacIntyre points out that from the perspective of the Copernican theory(certainly in its Newtonian version) all the mystery is resolved and thepuzzling details are explained (1977). With the resources of the new theory(interpreted realistically), we can construct a narrative that explains howPtolemaic theory could be empirically adequate within certain limits ofprecision, how those limits of precision could be improved upon, and whythe theory needed the features it had. But from the perspective of thePtolemaic theory we cannot construct a converse narrative. MacIntyre hasconvincingly argued that “source of interpretive power” (Chapter 3), beingable to explain the strengths and weaknesses of a rival theory in a narrativeis a highly ranked cognitive value, one that would be among the cognitivevalues endorsed by A (as well as G). Where it is highly manifested in a newtheory (together with other key cognitive values), not only is it evident thatthe new strategy is more fruitful than the old (for dealing with certaindomains of phenomena), but it is also explained why it is more fruitful.Therefore, a reason is provided to hold that the old will not be able to berejuvenated in a way that might regain ascendancy over the new.

Consistent with the extent of incommensurability that obtains, and notapplying standards for estimating the degree of manifestation of thecognitive values that A would not normally apply, we can affirm insummary:

1 G’s strategies became more fruitful than A.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (181)

2 G’s theory came to provide a better account of certain phenomena(and to fit shared data about them) that are of considerable importanceto A.

3 G’s theory came to apply better to certain objects of increasingimportance in the realm of daily life and experience.

For Kuhn, (1) and (2) provide sufficient reason to accept G’s theory and toreject A′s, and to discard A’s strategies. I agree, though it is not a reasonto accept that G’s strategies have the power to generate theories that willencompass the phenomena, excluding those in D, to which A appliedsuccessfully. The argument appeals to the “comparative local strength”(Chapter 3) standard: since G manifests the cognitive values more highly ofD than does A, ceteris paribus A is not acceptable of DA.12 That does notmean that A has been falsified. Through suitable adjustments of auxiliaryhypotheses, A can be maintained consistent with its selected data; though,since these adjustments tend to be ad hoc, they generally bring about muchlower manifestations of such highly ranked cognitive values as explanatorypower. This logically permits the possibility that A may continue to adopthis strategy, and to accept theories developed under it of D, justifying hischoice on grounds (social values, metaphysical commitments, religiousfaith?) to which fruitfulness—and ultimately any form of empirical inquiry—is considered subordinate. This is a logical possibility to which A, theAristotelian scientist, can take resort only at the price of changingfundamentally the character of Aristotelian science as an activity, rooted inexperience, seeking to discover and consolidate posits that manifest highlythe cognitive values according to the most rigorous available standards oftheir estimation. It always remains possible to subordinate fruitfulness tonon-cognitive values, but within the practice of science fruitfulness hasprimacy; to subordinate it to other values is to opt out of the “game” ofscience.

Fruitfulness: a necessary or a sufficient condition foradopting a strategy?

Kuhn maintains also that not only do (1) and (2) justify rationallydiscarding A’s strategy, but also they justify adopting G’s strategy ratherthan searching around for other strategies (which, if developed, mightcompete with G)—until such time as G itself may enter into crisis. For thesake of getting on with the “game” of science this makes a good deal ofpragmatic sense: work in a ‘world’ where there is a community engaged inshared practices, where one can virtually be assured of progress rather thanflounder around speculating in the darkness.

Attending only to pragmatic considerations, however, may obscure otherpertinent matters, and uncritically presuppose answers to questions such as


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (182)

the following: Is G’s strategy more fruitful than any other strategy thatmight have developed as a successor to A’s? Was it (and only it) developedprincipally because of the historical exigencies of investigation? Did it (andonly it) gain the opportunity to develop because disproportionate resourceswere devoted to its development, perhaps because of its potential interestof the bearers of certain social values? Were potential alternatives activelysuppressed? Must there be a unique successor to A’s, rather than a varietyof competing strategies, each displaying fruitfulness— but developingtheories which respectively become acceptable of different domains ofphenomena that may only overlap one another? Might it be that the rangein which G’s strategies can be fruitful is essentially bounded (even ifeffectively unlimited) such that they produce theories acceptable ofdomains of phenomena (including those of D on p. 168) within the bounds;but that phenomena outside the bounds require for their investigationdifferent strategies, but ones which grant an essential role for G’s strategiesin dealing with phenomena inside the bounds? Is it enough to respond tofruitfulness alone, and not also to take into account what are the domainsof phenomena that can in principle become encompassed within G’sstrategies?

These questions are all obviously connected if not equally penetrating,but they may appear to be odd and strangely speculative. Instinctively onemay want to respond: if there are alternatives to G’s strategies, producethem and show what can be done with them; if there are bounds to theirfruitfulness, define them and produce the arguments; otherwise, get on withthe work of science. That would be the end of it if, in historical reality, the“game” of science were conducted in independence of its applications. But,in fact, science produces theories that are applied; and I doubt that thevalue of scientific activity can be articulated coherently without taking thisinto account. If I am right that the applications of G are particularlyattuned to serve particular values, then the questions point to matters notonly of social but perhaps also of cognitive significance.

The questions may be deemed inappropriate for other reasons too. Onederives from the grip that materialist metaphysics has had on the self-interpretations of modern science, for it implies that material objectssimply are such that they can be completely grasped with the categories ofG’s strategies. I have rejected this view in Chapter 6, where I suggest thatthe grip derives from the relationship of this metaphysics with the modernvalues of control. A second reason is that G’s strategies have beenextraordinarily fruitful; and, while particular fundamental theories thathave enabled the general strategies to be interpreted concretely have beensurpassed, the fruitfulness of the general strategies is confirmed by repeateddevelopment of fundamental theories that build upon the successes of theirpredecessors and resolve their anomalies. Fundamental theories havebecome remarkably wide-ranging, ranging across both experimental and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (183)

“natural” phenomena and also the operation of the instruments whichbecame essential in G’s data gaining practices. That their range includedthe planetary motions has been of great importance. It demonstrated that aform of understanding could deploy categories derived from experimentalpractice and yet properly serve explanatory ends concerning naturalphenomena. This kind of fruitfulness, repeatedly confirmed, may appear tosuggest that any talk of bounds or alternatives is merely skeptical. But G’sstrategies abstract objects from their human, social and ecologicaldimensions, and so its explanations do not instantiate full understanding. Ifthere are bounds and alternatives, they might arise from these dimensionsof objects. That the strategies are fruitful in the domains of astronomicaland cosmological phenomena does not counter this, for these phenomenahave no relevant human, social and ecological dimensions.

The underdetermination of later by earlier strategies

My questions are not so odd after all. They arise in the context of Kuhn’sopinion that conflict among strategies occurs (and should occur) only aftera dominant strategy comes into crisis, as reflected in its decreasingfruitfulness and the multiplication of its anomalies; and that a new strategyis (and should be) accorded hegemony in the relevant scientific communitywhen it demonstrates its power to incorporate anomalies of theoriesdeveloped under the old strategy into the domain of which its theoriesmanifest the cognitive values highly

Part of this opinion, with appropriate qualifications, can be readilydefended. If, under the new strategy, the previously anomalous phenomenacan be represented in a theory which manifests the cognitive values highly,there would be no serious point in seeking an alternative strategy for thesake of investigating the anomalies. This supports that the new strategy is aviable candidate to frame further investigation, but it supports only ahegemony limited to the domain containing the anomalies, and perhapsconservative extensions of it. Does the historical record support the factualcomponents of Kuhn’s opinion about the key role of anomalies ininstigating and consolidating change of strategies, rather than mattersconnected with application doing so? Does it support that decliningfruitfulness occasioned by the multiplication of anomalies, rather thaninsignificance of theories consolidated under the prevailing strategy, is thekey to change of strategies? (A strategy may be fruitful, but the theoriesconsolidated under it insignificant.)

In the present case the record is ambiguous because, while A’s strategy wasin crisis, from the perspective of ascendent values (including the modernvalues of control), its theories were also insignificant; and the developmentof G’s strategies led, in one seamless movement reflective of their mutuallyreinforcing interaction with the modern values of control, both to the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (184)

resolution of some of A′s anomalies and to applications to objects thatplayed central roles in the projects associated with these values. Asindicated in the previous paragraph, resolution of the anomalies does notsupport granting hegemony to G’s strategy. Claiming its significance coulddo so. Conversely, affirming its insignificance (more likely its highlycirc*mscribed significance) provides a ground to seek out other strategies,though not one to deny its fruitfulness or to claim that it is in crisis. Havingsuch a ground to seek out other strategies is, of course, no guarantee thatthey will be formed; fruitfulness (in the middle to long run) remains adesideratum of the strategies that are adopted.

My point here is not that the history of science might have beenotherwise. That is a truism. Rather it is that there is nothing inherent to thequest to gain theories that manifest the cognitive values highly that ensuresa unique successor to a worked out strategy, or that prohibitsthatparalleling the thesis of the underdetermination of theories by data—there is underdetermination of later by earlier strategies. A new strategycould gain a foothold by achieving success in attending to anomalies of theold strategy, or in attending to significant phenomena that do not fallwithin the purview of the old. Furthermore, even if one focuses on theanomalies, the strategies deployed directly in resolving them may not havethe power to provide full understanding of the phenomena to which thetheories, developed under the new strategy, apply. In the present case, Gencapsulates well the material possibilities of the phenomena to which itapplies, but one may query how well it encapsulates their possibilities whenwe do not abstract them from their human, social and ecologicaldimensions and, in particular, how well it charts the unintended side-effects of expanding and restructuring the place of these phenomena in theprevailing social “world.” The centrality of resolving the anomalies of theprevious strategies does not by itself explain that strategies, which abstractin this way, are considered adequate to guide research whose theoreticalproducts are applied the realm of daily life and experience; the significanceof the theories–deriving from the social value attributed to grasping moreand more of the material possibilities of things—is also relevant to thisexplanation.

I do not think that there are resources within Kuhn’s framework tocounter my conclusion that significance as well as fruitfulness is needed toexplain and justify the adoption of a unique strategy. It takes time for anew strategy to develop and demonstrate its fruitfulness. During this time(“revolutionary science”), a variety of competing approaches may beentertained. Then, according to Kuhn’s account of the rationality ofscience, virtually “anything goes” as a new strategy is being sought; heprovides plenty of historical documentation that, during revolutionaryperiods, many approaches are tried. Even so, it remains that G would nothave developed if the necessary material and social conditions had not been


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (185)

made available for its development, and they may have been madeavailable because of G’s potential significance for the practices favored bythe ascendent social values connected with control that I have described.On the same grounds, a potential rival strategy—either one applicable to adifferent class of phenomena deemed significant in the light of values inconflict with the insubordination of control, or one that incorporated G’ssolution of A′s anomalies but limited the role of G’s strategies to spaceswhere abstraction from human, social and ecological dimensions wasjudged appropriate—might not have received the necessary conditions todevelop so that its fruitfulness could be displayed. Put another way, whichnew strategy (or strategies) comes to the fore may be essentially linked withthe values of the social “world” which provides the material and socialconditions for its deployment. Demonstrating that a strategy is fruitful, andin a particular socio-historical context uniquely so, does not imply that ithas shed the residue of the values that may have nurtured its developmentprior to the demonstration of its fruitfulness.

Since Kuhn supposes that a strategy has a unique successor, he does notexplicitly consider this possibility. On his supposition there can be onlycontingent, causal links between the development of a strategy and theprevailing of a particular social “world.” Some social “worlds,” but notothers, may provide the necessary conditions; but, once an old strategy haslapsed into crisis, which new strategy will emerge (if one does emerge) isnot a function of prevailing social “world” and the values it embodies. ForKuhn, the trajectory of the scientific tradition remains autonomous’, valuesdo not play a role in the making of consolidated judgments of theorychoice or strategy adoption. Fitting this picture, social values play no rolein supporting the judgment that G is fruitful, and A not—so that if that isthe key judgment at times of strategy change, values are not among thegrounds for strategy change.

This picture, however, leaves as a merely contingent matter the especiallysalient applicability of G within projects in which the modern values ofcontrol are deeply embodied. Thus, while it expresses autonomy explicitly,it implies implicitly the lack of neutrality, ultimately an internallyunresolvable tension. My alternative picture relieves the tension, since itrecognizes that impartiality (though not autonomy) can coexist with lackof neutrality. Theories are evaluated in accord with impartiality, whileproduced under strategies adopted because of their significance withrespect to particular values, as well as because of their (potential)fruitfulness. Given the possibility I have described on p. 174, values (and/ormetaphysical commitments) would be an essential factor in strategy change.Only appeal to them would ground limiting systematic empirical inquiryexclusively to that conducted under an adopted strategy, in the case underdiscussion limiting the focus of inquiry to the material possibilities of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (186)

things rather than also addressing their possibilities when we do notabstract from their human, social and ecological dimensions.15

This might suggest that the scientific community, if motivated by thevalue of neutrality, ought to entertain the simultaneous development ofconflicting strategies so as to avoid that scientific research at a given timebecome essentially linked with particular social values. Conflictingstrategies would be competitors, however, not only for adherents but alsofor material and social resources. In order to be deployed fruitfully, anystrategy must gain its appropriate material and social conditions from asocial “world,” generally one in which the theories developed under thestrategy are applicable. But the conflicting strategies might be dependentupon “worlds” which exclude each other, as do the “worlds” of medievalChristendom and modernity, and the “worlds” of contemporaryglobalization and the free market and Latin American popularorganizations (Chapter 8). Then, Kuhn might be right that coexistingcompeting strategies are impossible; not because the objective of science (toproduce theories that manifest the cognitive values highly) requires that itbe so, but because the social “worlds” upon which implementing thestrategies are dependent mutually exclude each other. This would confirmthat the fruitfulness of G’s strategy is only a necessary condition for itsrational adoption in the scientific community; and that power (and relatedsocial values) are necessary for its exclusive adoption, so that the exclusiveadoption of a strategy reflects not only historicity, but also power—mosteffectively power whose associated values are articulated in a dominantconsciousness that lacks the conceptual resources to render an alternativeintelligible.

Fruitfulness and applicability in the realm of daily lifeand experience

Let us return to the applicability of A and G in the realm of daily life andexperience. This realm changes with history, partly under human causalagency as human beings often radically modify the material and socialconditions they inherit, so much so that people in different epochs think ofthemselves as living in different “ages” (“worlds”). With the movements ofhistory, there come into being different realms of daily life and experiencewhose objects, phenomena and significant modes of interaction areconsiderably (though far from completely) different. Which theories areapplicable may depend on the realm of daily life in question; and changes ofrealm may engender changes of applicable theory. Moreover, thedevelopment of theories and the practices of applying them—activitiesin the scientists’ ‘worlds’—require conditions from the realm of daily life.The scientist’s ‘world’ is always inserted in a realm of daily life andexperience, in a social “world.”


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (187)

G’s ‘world’ could not coexist (except as a minor fragment) with the“world” in which A applied easily, if only because that “world” did notmake available the technological conditions for the increasingly necessaryinstruments to be manufactured. The instruments, products of the newtechnology, were not just external conditions for the development of G’sscience (like funding and political support). They became an integral partof the object of experimental investigation (see p. 159–60; also Chapter 6).G intervened with their aid, and observed the interventions and theiroutcomes. So, if G’s supporters in practical life had acted on the maxim,“do not pursue the implementation of G’s possibilities further until theissue between G and A has been settled decisively by scientific inquiry,” theempirical data needed for the eventual, decisive support of G would nothave been obtained. Neither G’s scientific project nor the technologicalproject could develop significantly in a “world” in whose realm of daily lifeA’s theory was considered to apply satisfactorily to its most importantpractices and phenomena. In such a “world,” key objects needed for G’sexperimental program (that is, objects that belong integrally to its domainof inquiry) do not come to the fore, and the realization of G’s possibilities(except incidentally and in individual instances) is generally consideredundesirable. At the time of the transition between A and G, the “world” ofmedieval Christendom was breaking apart, and increasingly the “world”was coming to contain objects of growing importance in the structuring ofdaily life of which G, but not A, could provide some understanding (forexample, canons, mechanical clocks, the mechanisms of the printing press,pumps, optical and meteorological instruments). Nevertheless, those whodeemed G’s possibilities generally desirable could not yet appeal to thedecisive empirical support of G to ground the genuine possibility of furtherdevelopments and implementations of these possibilities. During the periodof the transition it was important, for the eventual consolidation of G, thatthe technological project develop, driven both by its values and by theinductively-derived confidence that the current developments did notrepresent the limit of development, even if the limits of the possible werenot very clear and A implied that they were highly circ*mscribedhow thepractice turned out could settle that! G, I suggest, was inconceivable,lacking internal conditions and not just external material and socialconditions, apart from the developing technological project. And the morethe technological project rose successfully to prominence in a society, theless would be the space remaining for the application of A in the realmof daily life, so much so that A has practically no applicability in the realmof daily life today.

On the other hand, even retrospectively and speculatively G would havelittle applicability in a “world” where A applies easily. (It is not just that Gcould not develop in such a “world.”) First, concerning the characteristicphenomena of the realm of daily life of such a “world,” those highlighted


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (188)

in the modes of life where the dominant articulations of the “world” werewidely shared, it is doubtful that one could plausibly affirm that, as awhole as distinct from incidental particulars, G explains them better thandoes A —especially when we remember that here the control of nature wasconsidered subordinate to such relations as attunement, that the world(cosmos) was conceived as teleological through and through, and that thecharacteristic phenomena were considered concretely in their multifacet-edness not abstracting from some of their aspects. Second, with limitedtechnology and not much interest in it in the dominant modes of life, thisrealm lacked the objects and practices where the strength of G in informingactivity is located. Retrospectively, G could be applied to the astronomicalphenomena which A aimed to grasp. This seems to be the one clearexception to the generalization stated at the beginning of this paragraph.Moreover, G manifests greater empirical adequacy with respect to thesephenomena. But it is difficult to see how that greater empirical adequacycould have been demonstrated without developments of G’s researchprogram, including its deployment of optical instruments, and certainly thedemonstration that G manifests the other cognitive values highly withrespect to these phenomena requires the embedding of G’s account of theplanetary motions in a broader theory that manifests the cognitive valueshighly with respect to experimental and mechanical phenomena that wouldnot figure prominently in a “world” where A applies easily.

With respect to the totality of phenomena to which A applied easily, Amanifests the cognitive values more highly than did G, so that in a “world”where A applied easily, G could not supplant A. But G’s manifesting thecognitive values to a higher degree of an important sub-domain (theplanetary motions) of this totality does undermine the reasonableness ofaccepting A as a theory of the totality. With the changing social “world,”many of the phenomena of this totality ceased to be significant in the realmof daily life and experience. So there was little reason left to attempt toshape a strategy which would aim to generate successor theories to A thatmight manifest the cognitive values to a sufficiently high degree of the oldtotality (in the light of the higher standards of appraisal brought about bythe competitive situation that came into being with the introduction of G).It does not follow that G’s strategy—although it is demonstrably fruitful -can encompass all the phenomena in the emerging “world” that may be ofinterest in the light of sustainable social values that conflict with thedominant or ascendent ones.

I have suggested that the factor of applicability, in addition tofruitfulness, is a key consideration in the widespread adoption of astrategy. And applicability is essentially linked with social values. Initially,G’s strategy gained the social and material conditions to develop becauseof its promise linked with the modern values of control, and its capabilityto provide understanding of phenomena of special interest to the ascendent


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (189)

groups, which began to embody these values. (Because of that link, andthus the threat it posed to the prevailing social/religious/political/economicstructures, it also had to face efforts by the then dominant powers tosuppress its development.) Eventually the social “world” became sodominated by the phenomena that G could explain well that A lost its rolein practical life. G’s strategy is “rationally” applied in the “world”; to alarge degree, it is not possible to live (function efficaciously) in this“world” unless one’s practice is informed by G—though, consistent withthis, its deployment may be limited within identifiable bounds and granteda subordinate role in relation to other strategies.

Which “world” is the swing in? Depending on its socio-historical place,it could be in any one. It is also in the world. But, given the generalcharacter of human experience, we have no way to grasp objects of theworld, except insofar as they are also in our “world.” In our “world,”objects are objects of value, objects understood in relation to ourexperience and practice, valued more or less and in different waysdepending on their place in the realm of daily life and experience. Weadopt strategies, subject to fruitfulness as a necessary condition, (in part) invirtue of their capability to develop theories that provide understanding ofthose objects that we consider to be exemplary objects of value. Kuhn’saccount disguises this by mistaking a necessary condition for adopting astrategy (one articulated fully in terms of the cognitive values) for asufficient one.

This argument reinforces my view of the mutually reinforcing interactionbetween the materialist strategies (G) and the modern values of control. Itexplains the virtually exclusive adoption of this strategy, while opening upthe possibility that there are rival, fruitful strategies to be founddialectically linked with opposing values. It also provides richer content toKuhn’s view that A and G observe in different ‘worlds,’ different social“worlds” that structure the realm of daily life and experience, as well asscientific ‘worlds’ grasped under different strategies. We turn now, in thenext two chapters, to explore concretely the prospects of two anticipatoryalternative strategies.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (190)

8A “grassroots empowerment” approach

That prioritizing the materialist strategies needs explanation has been atthe core of the discussion of the previous two chapters. I have argued thatthe relevant explanation cannot be derived simply from appeal to thegeneral objective of gaining understanding of phenomena. Instead, itfollows from mutually reinforcing interactions that exist between inquiryconducted under the materialist strategies and the modern values ofcontrol. In the course of the argument, the suggestion has repeatedlysurfaced that there may be other strategies, alternatives to the materialiststrategies, with mutually reinforcing interactions with value complexes thatclash with the modern values of control, that under the appropriatematerial and social conditions might frame fruitful scientific (systematicempirical) inquiry. Clearly the argument would be strengthened if somealternative strategies were to be concretely identified and if, at least in ananticipatory way, their potential to produce theories that manifest thecognitive values to a high degree were displayed.

Not only would this strengthen the argument it would also haveconsiderable social importance. The products consolidated under thematerialist strategies are not neutral, being especially significant for valuecomplexes containing the modern values of control. Pursuing inquiry underalternative strategies might generate knowledge of significance for rivalvalue complexes and, in doing so, open up the possibility for the aspirationto be regained that the general practices of science, considered as includinga place for a variety of strategies, might come to manifest neutrality in amore robust way (Chapter 10).

The general objective of science, to gain understanding of phenomena (O),does not lead immediately or (in principle) uniquely to adopting thematerialist strategies (Chapter 5). Under these strategies there areencapsulated possibilities of phenomena that can be identified in terms ofthe generative power of underlying structure, process and law and thus,largely equivalently (Chapter 6), in terms of their potential value forvalue complexes that contain the modern values of control. But there is noa priori or empirically well supported reason to consider these possibilitiesto exhaust the possibilities of phenomena. Thus, I proposed considering the

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (191)

materialist strategies—the strategies under which there are encapsulatedpossibilities of special significance where the modern values of control areadopted—to define a particular approach to scientific inquiry (the Galilean/Baconian, O1/O1′, approach) which I rephrase as follows:

O1 The objective of the Galilean approach to science is to represent (inrationally acceptable theories) the order (structure, process andlaw) that is posited to underlie phenomena, to represent phenomenaand possibilities in terms of being generated from the positedunderlying order, and thence to discover novel phenomena.

O1′ The objective of the Baconian approach to science is to encapsulate(reliably, in rationally acceptable theories) the possibilities ofdomains of phenomena that are of potential value for projects thatexpress the modern values of control, and thence to discover meansto realize some of the hitherto unrealized possibilities.

Recognizing that the Galilean and Baconian approaches together constitutea unity is helpful. It precludes objecting to a proposed alternative strategysimply on the ground that adopting it is motivated by holding a particularset of values, and it opens up that O1/O1′ may be considered as an instanceof a whole cluster of alternative approaches (Oi/Oi′), where each Oi′ wouldbe an instance of the schema:

Oi′ The objective of the (…) approach to science is to encapsulate(reliably, in rationally acceptable theories) the possibilities ofdomains of phenomena that are of potential value for projectsresponsive to the value complex (…), and thence to discover meansto realize some of the hitherto unrealized possibilities.

Is introducing the Oi′ more than a formal exercise? Are there reallyalternatives to O1/O1′? Can strategies be identified (1) which are notreducible to the materialist strategies (though they may complement orsubordinate them), (2) which interact in mutually reinforcing ways withvalue complexes that conflict with the modern values of control, and (3)under which theories can be developed and come to be accepted in virtueof their manifesting the cognitive values to a high degree? An alternativestrategy must have all three of these characteristics, and the second does notguarantee the third. I have suggested (end of Chapter 6) that one mightlook for alternatives at the margins of the advanced industrial societies andamong minority movements in mainstream institutions. In this chapter, Iwill follow up this suggestion with detailed consideration of an approachthat I call a “grassroots empowerment” approach (O2/O2′); and, in thenext chapter, I will explore a feminist approach (O3/O3′). The upshot willbe vindication that there are alternative strategies, albeit with vastly


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (192)

different degrees of development and numerous ambiguities about theirpromise for further development. Nevertheless, they possess sufficientcognitive credentials to infuse a certain moral urgency and complicationinto the question: Which strategies to adopt? Which approach to follow?


Let us return to the question of human flourishing, which remains—in myopinion—the touchstone of all practice and inquiry. In the modernviewpoint, the modern values of control are said to enhance humanflourishing. But that is contested, for example, by many feminist andenvironmental perspectives, and especially by numerous popular,grassroots perspectives throughout the impoverished sectors of the world.

I will work towards the definition of a “grassroots empowerment”approach to engaging in scientific (systematic empirical) inquiry in twosteps. The first is to discuss different notions of “development,” for thecontestants challenge the social values of the leading institutions of“development,” seeing them more as among the causal contributors thanas remedies to the vast suffering and misery experienced by theimpoverished. This is the background to their rejection of the modernvalues of control. They do not agree that expanding our capability tocontrol nature is able to address properly the reality they face; or that itsassociated forms of understanding in general (as distinct from whensubordinated within an appropriate form of full understanding) enable theidentification of possibilities that might contribute to a socialtransformation that would serve their communal ideals of humanflourishing, and further the manifestation of such values as cooperation,participation, assuming responsibility for the future, solidarity, self-reliance, respect for nature and the dialectical unity of means and ends. Theprojects of the contestants need to be informed by understanding—formsof systematic empirical (scientific) understanding. I will then, the secondstep, elaborate this point by considering various ways of conductingresearch on “the seed.” All the issues under discussion come to a focus init. An additional aim of the discussion of the seed is to illustrate furtherhow research conducted under the materialist strategies does not produceneutral theoretical products, thus laying the groundwork for revising theformulation of neutrality (Chapter 10).

Concepts of “development”

“Development,” like its predecessor “progress” and like “freedom” and“democracy,” is a key term in the contemporary lexicon of legitimation.These terms are deployed in order to legitimize social goals, and to affirm


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (193)

the rationality, the practicality and the realism of their adherents’ practicesand policies (Lacey 1991a). The meaning of all of them, however, iscontested and so they can be sources of ambiguity, misunderstanding, futilecontroversy, and paradoxically rationalizations for maintaining anddeepening conditions of ecological and social devastation. Fundamentalterms of the lexicon of legitimation are contested because their logical,theoretical and methodological force derives from the complex interactionof three factors:

1 moral ideals;2 concrete embodiments of the ideals, for instance actual strategies,

processes, institutions and policies which embody the ideals only moreor less; and

3 theoretical idealizations.

The contestation can sometimes be obscured, even suppressed, because ofthe power associated with a hegemonic interpretation.

In its hegemonic interpretation, “development” represents such moralideals as individual freedom and the overcoming of poverty, as well as thefurther manifestation of the modern values of control. It is regarded asembodied to an acceptable degree in the advanced industrial societies,which, in turn, are typically characterized under the theoretical idealizationsof democratic capitalism. Clearly, at the present historical moment, theforces that bear this interpretation have gained unprecedented power.

Nevertheless, the contestants remain (Fabián 1991), some evenchallenging that there is an acceptable ideal of development (Escobar1995). They challenge the predominantly individualistic ideal on groundssuch as that the embodiments offered by the advanced industrial societiescannot be universalized, that it represents the value of autonomy withoutadequate balance with that of solidarity, and that it abstracts individualfrom cultural identity. They also challenge the theoretical idealizationsmaintaining that they do not provide adequate explanatory models ofcurrent economic and social realities, proposing instead various versions ofdependent capitalism (Lacey 1985). Thus they identify certain capitalistorientated development projects in the impoverished countries not asmeans to overcome poverty, but as causes of underdevelopment.Contestation of “development” is probably inevitable at the present time.On the one hand, the aspiration to share the way of life characteristic ofthe advanced industrial societies has obvious appeal; but, on the otherhand, its institutions are (and probably will remain) implicated in the deepand multi-dimensional sufferings of perhaps the majority of people in theimpoverished countries. “Development” is the site of on-going tensionbetween aspiration and what is possible, a tension that can be relieved inthe light of the outcomes of empirical research and the achievements of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (194)

concrete practices and movements, but one that ensures that there cannotbe a sharp separation of fact and value at this site.

The issue of development demands attention because very large numbersof people in the Third World and elsewhere experience the condition of theirlives in all of its dimensions—material, social, agentive, spiritual,psychological, cultural—as diminished, distorted or even intolerable; andthey have come to realize that it need not remain that way. Developmentrepresents both a negation of their present condition, and a process oftransformation. The contestation of “development” concerns theappropriate goal of the process of transformation, that is to say, theappropriate specific form that the negation of their present conditionshould take. How is this goal chosen? Who chooses it and who ought to?Once the goal is chosen, what is the appropriate process of transformation?What is the relationship between process and goal?

A distinction between “modernizing development” and “authenticdevelopment” is at the core of the contestation of “development. These twoideal types differ in how they represent the negation of the presentcondition of impoverished peoples that expresses the goal of development,and in how they identify the processes of transformation. The negationmay be variously understood depending on whether the state ofdevelopment or the current condition of the impoverished is taken to bewell defined.

For modernizing development it is the state of development that isconsidered well defined: it is represented by the institutions and valueshegemonic in the advanced industrial countries, and the processes ofdevelopment involve economic growth, industrialization, moderntechnology transfer, integration into the world capitalist economy, etc. Thecurrent condition of impoverished peoples, then, is characterized as “under-developed.” Development is the negation of underdevelopment (thenegation of the negation of development). “Modernizing development” iswidely understood and needs no further elaboration here. “Authenticdevelopment” is not widely understood, so I will elaborate it in somedetail.

Authentic development

For authentic development, what is taken to be well defined is not the stateof development, but the current condition of the impoverished, which canbe empirically charted and theorized in terms of such notions as oppressionand dependency. For it, development gradually gains definition bynegation, through political and social action and organization, of thevarious dimensions of suffering experienced by the poor. Authenticdevelopment is meant to be a response to the concrete andmultidimensional sufferings of large numbers of people, especially in the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (195)

impoverished countries, and to provide a means towards their negation.These sufferings obviously involve a material (bodily) dimension whoseintensity often overwhelms one’s consciousness of the other dimensions.They also have a social dimension as people experience the disruption oftheir families and communities, the necessities of migration, and a sense ofisolation—to which may be added special racial and gender components.On the cultural dimension, there are the sufferings derived from perceivingone’s traditions, cultures, languages, histories and ecologies beingdestroyed. Then there is a sense of powerlessness and helplessness,sometimes demoralization and depression, the threat of nihilism, and therecognition of the unfreedom of one’s life: that one is subject to the pushesand pulls of forces outside of one’s control and often understanding(because of the cultural destruction and the denial of education), that one’sown perceptiveness, values and agency play no role in the unfolding ofhistory. The sufferings may also involve the early and painful deaths of one’schildren, the woes of unemployment and unstable intermittentemployment, daily confrontations with drugs and violence, the devastationof being driven from one’s land, experiencing the contempt (and fear) ofthe powerful and well-off, and the experience of violence exercised againstthose who organize for change (Lacey 1991b).

On articulation, the negation of the multidimensional sufferings of thepoor provides a conception of a full or flourishing life, and supports theaspiration for a social order in which free and flourishing lives can be ledby as many as possible. It does not require a radical rupture withtraditional cultures, and, in many cases, it draws from them conceptions ofsocial justice that seek to embody such values as cooperation, widespreadparticipation, self-reliance and respect for nature. It is at ease with culturaldiversity and the expectation that from different cultures different positivedefinitions of development will emerge. Its measure, thus, cannot bematerial progress or economic growth per se, and relatively independentand unhindered technological innovation cannot be its driving force.Rather it seeks to integrate economic growth with the poor claimingtheir human agency and with the unleashing of their capabilities forexercising responsibility in shaping the conditions which structure theirlives. The goal of authentic development includes its means: to generate aprocess of negating current impoverishment, which integrally includesmaterial and economic dimensions in balance with others, informed by theunderstanding of the poor themselves and with their active participation.The possibilities, hoped to be realized in this process, clearly go beyondthose encapsulated in current first-world institutions and subsumed underits regularities. They include fuller manifestations of values that fituncomfortably with those linked with modernizing development (Lacey1997a): solidarity in balance with individual autonomy, social goodsranked above private property and profits, the well-being of all persons


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (196)

above the market, the strengthening of a plurality of values in place ofcommodification, human liberation in balance with individual liberty andeconomic efficiency, the rights of the poor above the interests of the rich,taking responsibility for the future instead of resignation in face of theprojects of the powerful, democracy enriched with participatorymechanisms and not limited to formal democracy, and the proper balanceof civil/political and social/economic/cultural rights. In summary, thevalues of enhancing local well-being, agency and community—or of“grassroots empowerment.” Reflecting this, their central practices arelocated in the social movements of the poor themselves.

Popular organizations

In Latin America, the relevant social movements are often called “popular[or grassroots] organizations”2 and collectively they are referred to as “thepopular movement.” Popular organizations (as ideal type) arecharacterized, in the first place, by practices (forms of “struggle”) that bothderive from and reinforce the group identity of, for example, women,landless families, indigenous peoples, workers, or refugees. They arestruggles which tend to focus on concrete objectives that address the self-identified needs and interests of members of the organization, concerningfor example: self-reliance, education, health, housing, land tenure and basicrights (articles by Martín-Baró in Hassett and Lacey 1991). Second, theyare grassroots organizations with grassroots leadership, styles, visions,values, language, knowledge and cultural forms; and at the same time theyare open to coalition politics with elite organizations and their members,recognizing that any significant social change will have to involve a role forsuch organizations and will have to involve a refocusing of them.

Third, especially when they enter into broader (municipal, regional,national, international) deliberations, the popular organizations tendto articulate their objectives in the language of human rights.3 Indeed, theHuman Rights’ Movement and institutions for the defense and propagationof human rights have become an important part of the popular movement(Lacey 1991b). They emphasize everyone’s right, not just to life, but to alife worthy of a human being, opening for ongoing reflection, for example,the issue of the relative ranking of the rights to property and to meaningfulwork. They do not articulate their objectives in terms of a particularpolitical or economic system, capitalist or socialist. Struggle, for them, is tobring about the fullest possible embodiment of human rights, notnecessarily, for example, to usher in socialism. Bringing that about, ofcourse, entails structural transformation. What the desired social structureswill be that adequately embody human rights, however, is not definedtheoretically in advance; but will emerge from the necessities and creativityof the struggle, reflecting the inputs and actions of those engaged in it.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (197)

Within the popular organizations, justice tends to be defined in terms of anadequate embodiment of the full range of human rights, and democracy isabove all respect for human rights—a respect usually held most likely to bepresent within structures of representative government and separation ofpowers.

Finally, popular organizations represent an “organic” unity betweenmeans and ends, and between ameliorative action and structuraltransformation. They represent in anticipation the values that they desire tohave embodied throughout society. Their movement towards newstructures involves: a) growth whose claims and appeal are grounded in thepartial realizations already actualized in the organizations (Lacey 1997c;and b) a keen sense of the dialectic of personal development and socialchange. The desired new structures cannot be defined or created fromabove or from outside (though they cannot be created without allianceswith and support from other groups); nor can they be created by a violentcataclysmic event. The mistake of many revolutionary movements has beento hold that liberation comes from an armed struggle (which at most canremove obstacles to liberation) rather than from the practices of thepopular organizations in all of their variety. The mistake of themodernizing elites is that liberation comes from the imposition of thestructures of “developed” societies.

Development, science and technology

The place of science in modernizing development is clear enough. Suchdevelopment requires the availability of modern (materialist) scientificknowledge, scientific institutions, and the dialectic of science and advancedtechnology. Moreover, materialist scientific understanding itself becomesa value for it, even where it transcends possible applications (Chapter 6).This clarity, however, sits side by side with an ambiguity when we address“developing” societies—whether, while developing, they need their ownindependent institutions for generating (as distinct from transmitting andapplying) scientific knowledge. Here neutrality is often appealed to,suggesting that the place of origin of scientific knowledge is irrelevant tothe assessment of its cognitive credentials and to its significance inapplication; then, in the name of efficiency and ready availability ofresources, arguing for the concentration of basic scientific research in theadvanced industrial societies or institutions supported by them. Theconsequence, then, is that science may become yet another instrument forentrenching dependency (Bunge 1980; Lacey 1994).

There are good methodological reasons to distinguish betweenfundamental (basic, pure) and applied scientific research. Nevertheless,since scientific research is conducted in institutions, pure and applied arenever fully separated. When we emphasize fundamental research, we are


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (198)

pointing to the definitions of research problems and the realms ofpossibilities to be explored, which arise from the internal unfolding ofresearch programs (that are actively being followed under the direction ofparticular strategies), aspects of scientific practice that are linked with theideas of impartiality and neutrality. Here research is conducted inabstraction from immediate concern for application, but that does notimply the absence of real, mutually reinforcing interactions between thatresearch and applied interests. This heightens the interest in the question ofwhat alternative forms science (including fundamental science) might takein impoverished countries so that research could be conducted understrategies consistent with the ideals of authentic development rather thanexclusively with those that maintain mutually reinforcing interactions withthe modern values of control.

Scientific understanding gained under the materialist strategies, theexemplary instance of wide-ranging understanding, is especially apt forinforming advanced technology, the core instrument of the modern valuesof control. A version of full understanding (Chapter 5) may also inform atype of technology, appropriate technology. By “appropriate technology,”I mean any technology that serves the interests of authentic development.5

Appropriate technologies consist of material objects (together withassociated techniques and bodies of “know-how”) that have been made byhuman beings for the sake of increasing those forms of human control overnatural objects which will contribute to the greater well-being orflourishing (in all dimensions) of all human beings, but especially of thepoor who constitute the majority in the impoverished countries. Whereappropriate technologies are sought, the value of augmentingour capability to control natural objects is subordinated to values linkedwith widespread human flourishing. This implies that the users, incommunity, of an appropriate technology have control over its productionand use, and over its material conditions (such as the raw materials neededto make and operate it, and the services needed to maintain it), so that itsproduction and use are directed towards meeting the needs of thecommunity. An appropriate technology, thus, interacts dialectically withrelations of production that encourage universal participation.

Appropriate technology differs from the dominant technology in tworelated ways. First, it is characterized by social relations that dialecticallyfurther the well-being of the poor majority, rather than dialecticallygenerate class inequalities and the tendency to privilege the interests of suchgroups as the rich and the military. Contrary to common viewpoints, thedominant technology is not neutral (Tiles and Oberdiek 1995), since thecapabilities it generates to exercise power over natural objects empowersome human beings at the expense of others. Both the dominant andappropriate technologies are linked directly and conceptually with social


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (199)

values and interests; they differ in the specific values and interests withwhich they are linked.

Second, appropriate technology is explicitly informed by versions of fullunderstanding, often involving the interplay of “technical” and localknowledge, or the systematic development of local knowledge. It respondsto questions like: “How can we produce food so that all the people in agiven region will gain access to a well-balanced diet?” rather than to: “Howcan we maximize food production under ‘optimal’ material conditions?”Answering such questions requires a mode of investigation that, unlikematerialist research, is not restricted to analyzing food production as afunction largely of quantitative variables, and that does not considerseparately the technical, the biological and the sociological, or productionand distribution variables. The investigations that inform appropriatetechnology may and often will be informed by the results of materialistresearch, though interest in and pursuit of that research will remainsubordinated to the objectives of authentic development, and it also willrespond to specific questions about the links of crop yields and the varietyof products gained from the crops with a range of socio-economic andlocally specific variables (as detailed on pp. 194–7). Fullunderstandinginsofar as efforts to gain it follow strategies that interact inmutually reinforcing ways with the values of authentic development—tendsto produce not general theories, but local profiles, structures andnarratives, with generalizations often sharply bounded in application to thelocal domain. Furthermore, its generation and consolidation may requireparticipation from the local grassroots in interaction with the“expert” practitioner; but, as understanding, it is neither subordinate tomaterialist understanding, nor (in principle) inferior to it in status.


Earlier (Chapter 5), I considered the seed in two ways: as an object whichgenerates, upon cultivation, crops with quantifiable yields; and as an objectwhich is integrally part of social processes. Suppose we ask: How can wemaximize the production of wheat under “optimal” material conditions?This question abstracts from the conditions of daily life and experience andthe prevailing practical activities of the producers and consumers of thecrop. It can be addressed as part of research under the materialist strategies(applying basic physical, chemical, genetic, biochemical and other scientificknowledge), in which crop yield is investigated as a function of suchvariables (which are open to quantification) as the use of fertilizers,insecticides, water, machinery and strains of seeds. Such research is typicalof the “green revolution” and its biotechnology successors, whose practicesare among the foremost bearers of the modern values of control.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (200)

The green revolution

The green revolution is rooted in the phenomenon that low-yieldingvarieties of wheat (or other crop) may be grown in isolation, and fromthem may be produced hybrid seeds that generate (under the “right”conditions) high-yielding plants, whose yields are very much greater thanthose obtained from plants that are grown from regular, field-fertilizedseeds. (The low-yielding, “pure” varieties are obtained by trial and errorseparation out from field-fertilized seeds, which have been developed overthe years informed by traditional knowledge.) The phenomenon has beenwidely replicated, and its applications have been widely acclaimed.

Green revolution practices generally increase crop yields (and, often,national exports and corporate profits) at least in the short run. They alsohave significant costs especially as calculated from the perspective of thosewho give high ranking to the social values of social and ecological stabilityand to the enhancement of local well-being. In the first place, productionhas become vastly more capital intensive and it requires expensive inputs:hybrid seeds, water (irrigation), fertilizers, pesticides, herbicides, machineryand the energy to run it. Second, there have been side-effects of productionwith negative environmental and social impact. Environmentally, withvariation from case to case, there has been depletion and poisoning ofsoils, loss of diversity in the gene stock of crop seeds, disruption of streams(and other negative effects of dam construction), desertification, increaseddependence upon fertilizers, herbicides and pesticides, and reducedquantities of other outputs of traditional crop production. Socially, small-scale farming has declined, leading to migration to cities withaccompanying unhygienic and psychologically threatening living conditionsand increased homelessness, increased unemployment andunderemployment, deepening of dependence on international capital (bothfor imports of fertilizers, technology transfer, etc., and for markets)inshort, social disruption and consolidation of market economies that tendnot to cater to the needs of the poor majorities, and that have beenespecially destructive for women and children (Shiva 1989).

Regarding one of the most famous implementations of the greenrevolution, Shiva has summarized the costs as follows:

Instead of abundance, Punjab has been left with diseased soils, pest-infested crops, water-logged deserts, and indebted and discontentedfarmers…conflict and violence…. [E]cological and ethnicfragmentation and breakdown are intimately connected and are anintrinsic part of a policy of planned destruction of diversity in natureand culture to create the uniformity demanded by central managementsystems.

(Shiva l991:12, 24)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (201)

Shiva (1988) has generalized these remarks and affirmed that what shecalls “reductionist science” (roughly: science conducted exclusively underthe materialist strategies) contributes (together with other factors) toproducing a four-fold violence. First, there is violence against “thebeneficiary of knowledge,” for example, farmers in rural India. Second,against “the subject of knowledge”: granting “monopoly” to knowledgegained under the materialist strategy devalues the knowledge of the bearersof other forms of understanding and the activities informed by them, andposes no barriers to social and economic projects which render these formsinsignificant and thus diminish the agency of their bearers. Third, andrelatedly, against “knowledge itself”: when in the name of sound “scientificknowledge” traditional knowledge is not only devalued, but alsosuppressed and distorted.8 And finally, there is violence against “the objectof knowledge”: when, for example, a project informed by reductionistknowledge destroys “the innate integrity of nature and therefore robs it ofits regenerative capacity” (Shiva 1988:232–5), or destroys the geneticheritage of a region.

Commodification of the seed

Perhaps the most striking change brought about from adopting inagricultural practice the knowledge gained from green revolution research,and one that exacerbates all the alleged costs listed on p. 191, is that theseed once the common heritage of humankind—a biological entitynormally generated each year as part of the crop—has tended to become acommodity (Shiva 1991, 1993; Kloppenburg 1988). Indeed, thesignificance of the research products, and thus the provision of theconditions for carrying out the research, are largely tied to this tendency(Rouse 1987). Although the seed’s tending to become a commodity is botha condition and a consequence of the practical application of greenrevolution knowledge, the possibility that it become a commodity is notencapsulated in theories generated under the materialist strategies, sincebecoming a commodity involves the social relations of the seed. Thistendency is being furthered through the use of patents and appeal tointellectual property rights that have marked the new biotechnologyrevolution (Suárez 1990; Kloppenburg 1988; Brush and Stabinsky 1996;Shiva 1997). The scientific research that has informed the green revolution—conducted under strategies needed for investigating the relationshipbetween the magnitude of crop yields and physical and chemical inputs—has led, when adopted in practice, both to increased yields in the short runand to the commodification of the seed.

Furthermore, with the commodification of the seed third-worldagriculture has become more inserted into the international economy inways that serve the special interests of agribusiness, a sector of land owners


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (202)

and some related industries, bearers of the modern values of control whoare linked integrally with particular socio-economic values. Given thematerial and social conditions of the research, it could not have beenotherwise; it has served the interests of the market rather than other(opposed) social values; it could not be made to serve the interests of allvalue complexes. Theories, soundly accepted within the strategies deployedin green revolution research, are not generally significant across viablevalue complexes. I take this to be a fact, not per se a point of criticism ofthese theories. Criticism depends upon adhering to a value complex forwhich they lack significance; and then the criticism is that they lacksignificance, not that they lack the proper cognitive credentials—that greenrevolution knowledge is not neutral.

Is green revolution knowledge neutral?

Those who defend neutrality might counter that, in the light ofcontemporary realities, the value complexes for which these theories lacksignificance are not viable (and not just presumed to be not sociallysustainable: Chapter 4). Certainly the adherents of the modern values ofcontrol look to further technological (for example, biotechnological)developments to reverse or prevent further damage, and regard theconsequent social reorganization as necessary for development; for themdevelopment (modernization) always requires deeper implementations ofcontrol. They may elaborate their objection by maintaining that I haveignored the obvious and urgent fact that the increases in crop yield were(and are) necessary to avert hunger in a world with a rapidly increasingpopulation. Therefore, they would insist, the green revolution hascontributed to meeting the most basic human need, a universal value,which must be a part of any viable value complex. This is not animplausible view. Put like this, however, the objection is simplistic, for itdoes not take into account the relationship between the mechanisms andinstitutions of production and those of distribution (and those of research).It has been claimed that, while the green revolution has providedconditions for many more people to be fed, it has also produced aredistribution of the hungry (Shiva 1991), and it has not produced socialmechanisms that are adequately responsive to the basic human needs oflarge numbers of people (or respectful of their rights). But, however oneputs the objection, it rests upon a crucial presupposition: there are nomechanisms to increase crop yields in the needed quantities outside of thegreen revolution technologies (and their biotechnology successors). Inparticular, it presupposes that traditional agricultural mechanisms cannotbe improved through research so that they become capable of increasingyields significantly; it presupposes that they cannot approach even remotelythe efficiency introduced by the green revolution. Put another way, the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (203)

presupposition is that the green revolution has occasioned no relevant lostpossibilities.

Although “no lost possibilities” is apparently a “factual”presupposition, the policies and practices of the green revolution haverarely reflected any empirical investigation of it. Research conducted at thepredominant scientific institutions on these matters reflects antecedentlymade policy decisions, rather than vice versa. That this is not widelyrecognized reflects how deeply this presupposition is ingrained in thethought prevailing in the advanced industrial countries. It is part of the apriori thinking that sees development (modernization), and thus the furthermanifestation and embodiment of the modern values of control, as the“solution” to all problems. But, one might object, is it not empiricallysupported by the clear fact of recurrent shortfalls in food production in thepost Second World War years? Does not this fact speak definitely to theinadequacy of traditional agricultural technologies? Shiva suggests analternative explanation of this fact in the commercialization of agriculture(a product of colonialism), and in the breaking of the tie between thetraditional technologies and their social relations (1991:26). Theimplications of this suggestion become clearer in the light of the distinctionshe uses between two kinds of poverty: “poverty as subsistence” and“misery as deprivation” (Shiva 1989). The deprivation, from whichdevelopment is meant to be the solution (underdevelopment), Shivamaintains, is itself caused by the process of development. Moreover,serious investigators have argued that there is some evidence that indicatesthat the presupposition of “no lost possibilities” is false.10 Some of themmaintain that the hybrid seeds were not necessary to produce significantlyhigher yields (Shiva 1991; Levins and Lewontin 1985). According to thesame underlying genetic theory that informed the hybrid seed research,they maintain, comparable yields could be obtained from appropriatelyselected “pure” (non-hybrid) varieties and used in ways more compatiblewith social and ecological stability, but contrary to the interests ofmodernizing development. Pending further investigation, it appears that theobjection that the value complexes, from which criticism of the greenrevolution is marshalled, are not viable rests on a presupposition which hasnot been accepted in accordance with impartiality. That objection thereforefails.

“Lost possibilities”: agroecology and the green revolution

Those, often connected with grassroots movements, who adhere to valuecomplexes in which, for example, enhancing local well-being, agency andcommunity are highly rated values, consider the practices of the greenrevolution to lack relevant value and they have little positive interest in theknowledge that informs them. They do not doubt that the green revolution


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (204)

is informed (at least in the short run) by soundly accepted theories, butthey query the significance of the theories because they do not encapsulatepossibilities relevant to the realization of their fundamental values. Theyare interested in the lost possibilities of the green revolution, and how theymight be brought to realization; and they wish to nurture the social valuesof social and ecological stability. Research, that might serve their interests,aims to study natural objects explicitly in terms of their relations to thesocial and ecological orders (and not simply as objects whose possibilities aregenerated from the underlying order where their role in the social order isonly implicit), thus poses a question like: How can we produce wheat sothat all the people in a given region will gain access to a well-balanced dietin a context that enhances local agency and sustains the environment? It, incontrast to the question posed at the outset of this section: “How can wemaximize wheat production under ‘optimal’ material conditions?”, doesnot abstract from the conditions of daily life and experience and prevailingpractical activities of a region, and it does not presume that questions ofsocial order are subordinate to the implementation of novel controls. Itdoes not consider the biology, ecology and sociology (or production anddistribution) separately, and it locates questions about crop yields amongquestions of the following kinds: What are the socio-economic conditionsand social effects of agricultural production? Who controls the product?What use is made of it? How is it distributed? How do the socio-economicconditions of production affect those of distribution, and vice versa? Whatare the effects on health and ecology? Thus, crop yields are investigated notonly as functions of broadly materialist variables, but also of the social andother variables of which the materialist variables are themselves a function.This approach turns attention to the local and the particular: to local soilconditions, strains of seeds, ecologically sound methods, availability of“natural” pest controls, traditional practices; and to local socio-economicrelations, needs, aspirations and histories. These are the kinds of mattersthat must be investigated if it is sought to reshape the world of daily lifeand experience so that control ceases to be hegemonic, but contained bythe exigencies of the values highlighted by authentic development.

Some researchers connected with grassroots movements have proposedthat relevant understanding can, in part, be gained from improvements(based in empirical research) of traditional practices and understandingforms of research which follow strategies that do not abstract phenomenafrom the ecological and social relations that they exhibit.11 They challengethe claim to possess a monopoly on knowledge (soundly accepted,systematic empirical knowledge of material and biological objects) oftenmade on behalf of materialist science, by pointing to the empiricallyvindicated strengths of local, traditional, people’s knowledge which,because of its locality, assumes numerous, diverse forms. Informed by suchknowledge, traditional farming has developed, in some cases, practices that


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (205)

are ecologically sound (maintaining, for example, soil that has remainedfertile for millennia, and pest and disease controls that function throughappropriate arrangements and combinations of crops), selection processesthat have generated a richly diverse gene stock, and modes of socialorganization in harmony with natural processes.

Aware that in dealing with these matters (in medicine as well asagriculture) one can be tempted to lapse into nostalgic romanticism or“new age” enthusiasms, I emphasize that “fitting with tradition” is acultural, not a cognitive value. A “privileging” of knowledge gained underthe materialist strategies should not be replaced with a generalized anduncritical privileging of traditional knowledge forms and claims.Regardless of strategies deployed, knowledge claims are to be assessed invirtue of how well they manifest the cognitive values, not just in virtue oftheir potential significance for an adopted value complex. A cultural valuecan motivate adopting certain strategies, but cannot ensure that the worldwill be amenable to grasp under these strategies. My interest is empiricallygrounded. First, there is the empirical record that current developmentpractices are failing to satisfy the basic needs, to cultivate the humancapacities, and to respect the human rights of vast numbers ofimpoverished peoples. Second, the references (Note 10) provide evidencethat, at least under some cultural conditions and in some locales, traditional,ecologically sound and socially strengthening agricultural technologies arepotentially sustainable (though clearly they cannot be developed further,and may not be able to survive, under the hegemony of policies ofmodernizing development). Grasping and deepening the understanding thatunderlies these technologies, of “agroecology,” and separating out thatwhich is empirically sound from that which is not, has become a majorconcern for those attempting to implement conceptions of development thatare opposed to modernizing development. They push for empiricalinvestigation of the “lost possibilities” question. They oppose bothpresuming that there are significant lost possibilities, and laying aside ordismissing the question because of its inconvenience in the light of thosepolicies that foster the spread of agricultural practices that are implicated inbiotechnology and expanding agricultural exports, and that further themanifestation of the modern values of control. The mode of research thatthey advocate does not dispense with materialist understanding, but ratherproposes a dialectical interaction between traditional and materialistapproaches, with the interest of materialist results subordinated to theirrelationship with the social values of the alternative conceptions of“development.”

Shiva has referred to the relevant traditional knowledge as dealing with“preserving and building on nature’s processes and nature’s patterns,” with“repairing nature’s cycles and working in partnership with nature’sprocesses,” and with “subtle balances within the plant and invisible


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (206)

relationships of the plant to its environment.”12 This knowledge concernsthe relationships of plants (and animals) with local physical conditions,with other biological organisms, and with people and social organization.Clearly, over the centuries, indigenous seeds (obtaining from freepollination in open fields) have been improved as a result of theselection practices of local farmers. Their knowledge can become an objectof systematic investigation in which it is articulated, systematized,empirically tested and further improved. Traditional agriculturalknowledge abstracts from the molecular structures and “internalmetabolism” (Kenney 1986) of plants, and so cannot by itself encapsulateall the possibilities identified in research under the materialist strategies.Nevertheless, it has produced the genetic materials, which are thefundamental prerequisite of all research with hybrid and geneticallyengineered seeds. It is, thus, a body of knowledge with impressiveempirically vindicated credentials. There is no clear reason why it shouldbe incapable of further development.

Paradoxically, modern scientific research both presupposes the results oftraditional knowledge practices, and acts to destroy them or at least todeny their products any legitimacy as knowledge (Marglin and Marglin1990; Shiva 1988, 1997). The alternative approach, proposed by the criticsunder discussion, is to develop a dialectic between traditional andmaterialist approaches to gaining knowledge, to pursue a version of fullunderstanding that engages not only the scientific “expert,” but also thepeasant practitioner.

Dialectic of traditional knowledge and research under thematerialist strategies

How might a dialectical interplay of traditional knowledge (for example, aform that has a mutually reinforcing interaction with the values ofenhancing local well-being, agency and community) and materialistinvestigation work? Or, how might investigation be conducted under thematerialist strategy, but subordinated to other values? There are relativelystraightforward ways: the values might prioritize research on the socialpossibilities and consequences of applying theories accepted under thematerialist strategies, or point materialist research in a direction so as tofocus upon particular phenomena deemed of special social value. Can thebiological (and the material) always be demarcated from the social in thisway: where social inquiry complements and partially directs the biologicalinquiry, but where the biological possibilities (in the agricultural case underdiscussion) are discerned only in relation to the fundamental underlyinggenetic theory? Or sometimes may there be such a profound interactionbetween “natural” and social variables that an adequate encapsulation ofpossibilities could not derive from research that draws upon “adding up”


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (207)

results gained from the standard disciplines?13 So, minimally, adopting analternative strategy (if the answer to the first question is “yes”) may lead toa richer interdisciplinary approach, but (if it is “no”) it opens upthe possibility that we may need to address certain questions in ways thatcut across the standard disciplinary lines.

The discussion of these issues has entered a new stage with the advent ofthe biotechnology revolution. Some have argued that not only do the newbiotechnologies promise the development of higher-yielding varieties ofcrops, but also that they will not have (and may even reverse) the negativeecological effects encountered with the green revolution. And others havesuggested that they can realistically serve the interests derived from valuesof social justice in the impoverished countries, especially those which arethe source of the greater part of the genetic resources of humankind.Consider: “As a society, I think that we would like to use our enhancedcapacities for manipulating the genetic code to develop and deploy newplant varieties in ways that are economically productive, socially equitable,and ecologically benign. Will we be able to do so?” (Kloppenburg 1988:xiv). Putting the question this way accepts that the science of geneticengineering will develop of its own internal dynamic, so that we areconfronted with the need to find ways to use it to serve popular interestsrather than those of capital. This science is just a reality to be coped with,hopefully constructively; it frames questions about agricultural policytoday. It will have some role; the question is: what role? Following thislogic, the poor countries would need scientific institutions that engage inspecifically targeted applied research in biotechnology, or perhaps also infocused “fundamental” research linked to those applied areas.Kloppenburg, however, is sensitive to the “lost possibilities” consideration,and to the potential for improvements in traditional technologies; and sohe urges research in this area too. His solution seems to be a “both…and…”rather than a dialectical interaction. Basic biotechnological research ispresumed to be significant, with the specific focus of its applications to bejudged (and compared with traditional methods) in the light of economic,sociological and ecological considerations. But he does not evaluate thesignificance of the research program of biotechnology in the light of theinterests of authentic development (as one seeking a form of fullunderstanding would).

There is something appealing about the “both…and…” position, as itattempts to be responsive both to the realities of power and to the rights ofthe poor, and it needs fuller consideration. Nevertheless, unless theinstitutions of biotechnological research were to be radically changed andthe research relocated within a program of full understanding related withthe values of social justice, biotechnology will continue for the most part toserve the interests of expanding capital-intensive agriculture. Geneticengineering furthers the process of turning the seed into a commodity, since


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (208)

genetically engineered seeds may be patented, and the dominanceby agribusiness of the market can create pressures for farmers to use thenew seeds. According to international trade agreements, the use of suchseeds is restricted in accordance with the laws of intellectual propertyrights. Given the cost of biotechnology research, it is difficult forcommunities and institutions in poor countries to claim such rightseffectively for themselves. This is clearly disadvantageous for them. It notonly entrenches dependency and inequity in trade relations, but alsoweakens the ability of the poor nations to utilize for their own ends one oftheir most valuable natural resources, the richness and variety of theirgenetic resources.

It remains for those with the appropriate technical competence toexplore the promise of any efforts to develop a dialectic betweentraditional knowledge and materialist science. In doing so, it should bekept in mind, paralleling my discussion of the materialist strategies(Chapter 5), that alternative strategies being adopted in view of theirmutually reinforcing interactions with certain social values is neithersufficient for accepting “theories” developed under them, nor per se anobjection to engaging in research under them—as long as the roles of thecognitive and the social values are kept separate. The role of the cognitivevalues is essential. Perhaps the world will not be amenable to grasp underthe alternative strategies! But it is always relevant to ask, when researchunder certain strategies is facing difficulties, if the difficulties derive fromthe way the natural world is or from opposing social forces.

Working out alternative forms of understanding needs to be done subtlyand realistically, with full awareness that the dominant structures of powerare cast against the alternative conceptions of “development” with whichthey have mutually reinforcing relations, and with full awareness of thesocial relations of scientific research and development. In particular, itrequires awareness that modern (largely materialist) science, like capital, isnot under the control of agencies in the impoverished countries, and that itis dialectically linked both with the modern values of control and withcapital. While some of the possibilities, uncovered by materialist science,may be pertinent to the interests of authentic development, its generalimpact can be expected to be linked with the programs of modernizingdevelopment. In the light of this, it will be difficult to obtain theinstitutional conditions in which to investigate sharply whether difficultiesfacing alternative projects stem from the way the natural world is or fromopposing social forces. It is frequently said today that entertainingpossibilities that cannot be realized within the structures of modernizingdevelopment is “unrealistic” (Lacey 1997c). If so, is the source of the lackof “realism” from the natural world, or from social forces (including theuse of power)?


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (209)

A “grassroots empowerment” approach

Drawing upon the discussions of the seed and of authentic development,we may entertain as the objective of an approach to science (O2) one thatadopts strategies—involving a dialectical interplay between the methods ofgaining traditional knowledge and research conducted under thematerialist strategies—that enable us to identify those possibilities whoserealization would further the interests of authentic development and thefuller manifestation of its prioritized value, enhancing local well-being,agency and community. The strategies to be followed (and their variationwith locale) remain to be specified in detail, but the approaches in whichthey are adopted are effectively equivalent to a “grassroots empowerment”approach:

O2′ The objective of a grassroots empowerment approach to science isto encapsulate (reliably) the possibilities of domains of objects (forexample, objects relevant to agricultural practices) that are ofpotential value for projects responsive to the value of enhancinglocal wellbeing, agency and community, to discover means torealize some of the hitherto unrealized ones, and to preserve thosealready realized.

Clearly if one’s value complex includes the modern values of control, thenone will adopt O1. The point of O2 will be apparent to those whose socialvalues make them critics of the modern values of control, who maintainthat the embodiments of these values require practices and institutions thatcan only be maintained in an economic order which inherently hasundesirable consequences, such as social and ecological devastation,unacceptable inequalities, patriarchal relations, alienated labor, orclassbased relations of domination. Thus, the critics will ask: How shall weinteract with nature so as to serve the coming to be of an alternative socialorder (authentic development), which (for example) enables grassrootsempowerment to be enhanced (or which embodies some different view ofsocial justice)? For this end, what will be the characteristic way (ways) ofinteracting with nature? What kind of strategies (alternative to thematerialist strategies, or to which the materialist strategies are subordinate)should be brought to bear in order to gain empirically groundedknowledge that would serve that end? As we have seen, there may beinteresting mixtures of pre-modern and (post)modern answers to thesequestions.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (210)

9A feminist approach

There is no objection (in principle) to identifying and engaging in approachesto systematic empirical inquiry that deploy strategies distinct from thematerialist strategies (Chapter 5); and I have explored the motivation,prospects and implications of one alternative, the “grassrootsempowerment” approach (Chapter 8). In this chapter, I explore anotheralternative approach, one whose strategies interact in mutually reinforcingways with “feminist” values. Again, provided that the distinction of theroles of the values and the cognitive values is respected, this alternativeneed not involve bringing values to bear on scientific claims in any waythat is logically different from that involved when the materialist strategiesare adopted.


Feminism is a political movement with implications for restructuring thewhole gamut of social institutions. Few seriously query the propriety offeminism to have an interest in adopting a critical stance aiming to uncoverbiases that may be at play in current scientific practices, in advocating thatmore research be conducted on specifically “women’s problems,” forexample, in the health sciences or in investigating possible barriers to theadmission and advancement of women in scientific institutions andproposing ways to eliminate them. There is strong resistance, however, tothe idea that there may be a feminist approach to science which drawspositive direction from feminist values. Geertz expresses the resistanceclearly:

The worry is…that the autonomy of science, its freedom, vigor,authority, and effectiveness will be undermined by the subjection of itto a moral and political program—the social empowerment ofwomen -external to its purposes…[namely] the knowledge-seekingones of science, the no-less-impassioned effort to understand theworld as it, free of wishing, “really is.”

(Geertz: quoted in Lloyd 1996)

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (211)

The worry is that any scientific practice that bears the label “feminist” willnot be value-free in ways widely thought to be essential, at least asaspirations, to scientific practices. It concerns the very idea of a feministapproach to science, not the results of its actual practice and not objectionsto feminist values per se.

This worry is misguided. The autonomy to which Geertz appeals doesnot hold even of research conducted under the materialist strategies(Chapter 10). A feminist approach to science may extend “knowledge-seeking” practices into hitherto neglected domains, though how fully it canbe implemented remains a matter for both further empirical investigationand political activity, and the significance of such extensions can becontested. While a strategy may be adopted because it interacts in mutuallyreinforcing ways with particular values (whether the modern values ofcontrol or, for example, feminist ones), adopting the strategy does (and can)not commit one to accept any concrete theory; and, in the long run, astrategy ceases to be (cognitively) worthy of adoption if it fails to generatetheories that become accepted (Chapter 10). I belabor this point becausebarriers to its recognition are strong, for in the mainstream scientificcommunity (and much of the philosophical) it is rarely seriously queriedthat autonomy is in fact realized to a tolerable approximation. This view,in turn, both gains support from, and reinforces, the further view thattheories consolidated under the materialist strategy broadly characterizethe world as it is rather than that primarily they characterize objectsgrasped from the stance of control. It is the grip of materialism that putsall incipient alternatives at an additional disadvantage. They, but notresearch under the materialist strategies, have to defend themselves in faceof the charge of the “intrusion of politics.” In view of the way in which themodern values of control are among the values highly expressed in thedominant contemporary economic, social and political institutions, so thattheir claim appears to be universal, this may be inevitable. Nevertheless,there is no sound argument for singling out the alternatives in this way.That they are so singled out, antecedent to the outcomes of research underthe alternative strategies, I guess we can say, is “ideological.”


The idea of a feminist approach to science, which I discuss in this section,derives from reflecting on recent writings of Helen E. Longino (especiallyLongino 1990).1 Longino states her version of “feminism” as follows:

Feminism…is at its core in part about the expression of humanpotentiality. When feminists talk of breaking out and do break out ofsocially prescribed sex roles, when feminists criticize the institutionsof domination, we are thereby insisting on the capacity of


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (212)

humansmale and female—to act on perceptions of self and societyand to act to bring about changes in self and society on the basis ofthese perceptions.

(Longino 1990:190)

Lying behind feminism, so understood, is a conception of human nature:human beings have the capacity to act informed by their own values in thelight of their assessments of current realities, and to act efficaciously to“bring about changes in self and society on the basis of those perceptions.”Human beings are agents, with “capacities for self-consciousness, self-reflection, and self-determination,” and whose intentional states areefficacious. This conception of human nature serves to ground suchprioritized values as “liberty, autonomy and responsibility.” It also is seenas “partly a validation of our…subjective experience of thought,deliberation, and choice.”2

Agency is a human capacity. Its exercise can gain positive conditionsfrom one’s relations with others and one’s places in social institutions, or bediminished by them. Historically, diminished agency has been common,but affecting various groups of people differently. In addition to theconception of human nature, Longino’s feminism endorses the possibility ofenhancing agency where it has been diminished, a possibility whoserealization itself depends upon concerted action which will depend uponidentifying and eliminating the causes of diminished agency.3 Expandingthe exercise of agency becomes the central objective of a political movement.What kind of approach to science might serve this objective? Or, asAnderson puts it, what are the “scientific practices, which incorporate acommitment to the liberation of women and the social and politicalequality of all persons” (Anderson 1995a: 51)? Or, noting that factorspertaining to gender are among the causes of fundamental inequalities,what kind of approach to science might serve the narrower objective thatsometimes Longino focuses upon: “to reveal gender,” that is, to reveal theaction of “an asymmetric power relation that both conceals andsuppresses the independent activity of those gendered female…, [a]relation… sustained by social institutions and symbolic practices…itselfmade invisible as a relation of power by…naturalizing models in the lifeand behavioral sciences of sex and gender differences” (Longino 1996:50).

Feminism emphasizes enhancing human agency, and so it subordinatesthe modern values of control. In principle, feminist values can be linkeddialectically with alternative strategies for pursuing the objective ofencapsulating the possibilities of domains of phenomena. We can see thisbest by developing further a contrast with the materialist strategies. When Iconsidered the reasons for adopting the Galilean/Baconian approach (O1/O1′), the focus of attention was human interactions with material objects.There (Chapter 6), in effect, a question like: “How must material objects


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (213)

be thought of, if control is to become the characteristic human stancetowards them?” was posed. The answer—summarily: as grasped under thematerialist strategies—frames an approach which produces theories thathave come to inform practices which bring about a fuller manifestation ofcontrol as a social value within the predominant social institutions. Morefully manifested control is evident in the creation and multiplication oftechnological objects, which become the center and the cutting edge of theproductive process; and so it has implications concerning the shape,possibilities and structuring of human lives. Understanding of the materialworkings of technological objects is gained by adopting O1. This form ofunderstanding treats these objects in abstraction from their social (andecological) contexts. It attends to their material products and to theprocesses underlying their production. But it lacks the resources to offerunderstanding of the social institutions of production, distribution and use,the values and interests they represent, and disparities they may introduceand require concerning enhanced or diminished human agency.

Materialist understanding does not answer: “What do people becomelike (and with what kinds of variation) when understanding, gained underthe materialist strategies, is applied within modern institutions?” Thenatural science is dissociated from the sociology. This seems to be acorollary of focusing almost exclusive attention on the question posed in theprevious paragraph—where we consider the modern values of control inabstraction from the conditions required for their becoming highlymanifested in social institutions, as if their universality or desirability couldbe taken for granted. Control can indeed be distinguished from such othervalues as meeting people’s material needs, profit, efficiency, serving the freeenterprise system, or developing socialist or feminist consciousness. Butwithin actual institutions, it will be manifested along with the highmanifestation of some and the low manifestation of other values. Howone ranks control as a social value, thus, can be a function of rankings ofthe other social values from which (contextually, historically) it cannot beseparated.

Now, consider these questions:

1 How must human beings be thought of if feminist objectives are to befurthered?

2 What kinds of knowledge could inform the furtherance of feministobjectives?

In posing questions like these first, the feminist approach under discussionis the opposite of the one I attributed to materialism. It begins with whathuman beings are, not with our relations with material objects and notwith what material objects are. Suppose that those stressing the modernvalues of control also asked: How must we conceive of human beings if


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (214)

they are successfully to adopt control as the characteristic stance towardsmaterial objects; and if they are to apply scientific knowledge to furthersuch control? It seems that two answers would have to be given. First, humanbeings are agents, capable of acting in the light of their knowledge for theirown ends. Second, human beings are adaptable to the roles (or absence ofroles) required of them in relation to technological objects, especially to themeans of production. As general answers they are incompatible. Yet both areneeded. Some human beings will have to be one way, some the other; andat least the second involves diminished agency.5 Feminism challenges thisbifurcation (in part, because it is often drawn along gender lines),emphasizing the universality of agency, and questioning those approachesto science—in psychology and sociology—that attempt to ground thebifurcation in inherited natural differences.

Answers to the two questions just raised highlight agency as thedistinctive human capacity shared by all human beings, but diminished inits exercise under some social conditions. Then, human beings as agentsandthe conditions under which agency is enhanced or diminishedbecome theobject of inquiry; and attempts will be made to identify strategies for anapproach (O3) that is effectively equivalent to:

O3′ The objective of a feminist approach to science is to identify(reliably) the possibilities open to human agency, the conditions forits enhancement and diminishment, and to discover means forbringing about more of the enhancing conditions.

I will make no effort to specify all the implications of following O3,6 butonly those fairly directly connected with human agency and especiallywith cognitive abilities and activities, areas of investigation proper to thebiological, psychological and social sciences. I leave it an open questionwhether a version of O3 can be followed in the physical sciences. It hasbecome commonplace to affirm that feminist insights (both critical andpositive) will be confined to the “soft” sciences. If so, so be it. But I am notso sure (cf. Tiles 1987), for one may ask: How must we think of materialthings, and what must be our characteristic stance towards them, if wewish to enhance the agency of everyone maximally? Different cultures andlocales may produce richly different answers to this question.

Constraints on theories

Following O3 involves adopting strategies that put constraints onadmissible forms of understanding of human phenomena.7 The firstconstraint is that intentional explanations have primacy in the domain ofhuman behavior. Agency is unintelligible apart from the causal efficacy ofan agent’s beliefs and desires. So, the primary categories of behavioral


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (215)

explanation will be belief, values, desire, intention and the like, rather thanposits of underlying law, process and structure. Seldom will suchexplanations be articulated with the kind of formal (deductive,mathematically articulated) organization present in materialist theories.Usually they will be articulated in the form of narratives enriched withsituation-bounded regularities (Lacey and Schwartz 1986, 1987). Thesecond constraint is that diminished agency will typically be interpreted associally produced—though, in special cases, it will be considered asphysiologically or psychologically produced—and sometimes socialaccounts are supplemented by psychoanalytic ones (Keller 1982); and (itseems to me) it will tend to be reducible to behavior that more completelyfits law-like regularities. Identifying the social/historical bounds of suchregularities will constitute an important research item. These constraints,as well as the dialectically related feminist values, draw upon the view thathuman action, since it is intentional, does not reduce to behavior, asunderstood in “scientific” psychology (psychology conducted undermaterialist strategies: Schwartz and Lacey 1982).

Strategies in which the hypotheses that can be entertained are bound toconsistency with the intentionality of human agency are incompatible withmaterialist strategies, for action cannot be represented simultaneously asintentional and lawful (Donagan 1987; Lacey 1996).8 Thus apresupposition of the feminist value complex is inconsistent with numeroushypotheses about human behavior entertained under materialist strategies,for example, those versions of materialist strategies that restrict the kindsof variables that are admissible in laws (and thus in underlying structuresand processes), and so aim to explain behavior in terms of its lawfulrelations with genes, fetal hormonal phenomena, settled brain states orenvironmental contingencies. Illustrating this Longino discusses a strategyin behavioral endocrinology that uses “the linear-hormonal model” (LHM:hormone—brain-organization—behavior),9 in which sex differences in avariety of behaviors (including performance in some mathematical tests)are lawfully attributed to differences in brain organization, which arethemselves lawfully attributed to differential roles of gonadal hormones infetal development.

Concerning research under LHM, Longino says: “Our politicalcommitments…presuppose a certain understanding of human action, sothat when faced with a conflict between these commitments and aparticular model of brain-behavior relationships we allow the politicalcommitments to guide the choice” (Longino 1990:191). “Guide the choice”to what? To rejecting out of hand any theory developed with LHM? Butthat would be an instance of fact being inferred from value, and of“wishing” (Geertz) that the world be consistent with the presuppositions ofher feminist values.10 On closer inspection, however, it is clear that thechoice Longino is referring to is to engage in research under the feminist


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (216)

strategies rather than to deploy LHM. This choice cannot properly bedismissed simply as a matter of “wishing” unless theories developed withLHM in fact manifest the cognitive values to a high degree of somerelevant domains of phenomena. Longino argues that they do not, and (ineffect) that they only appear to do so in the light of assuming that there areessential brain differences between males and females that account for agreat variety of behavioral differences. This assumption is not contained ina theory soundly accepted in accordance with impartiality, and holding it isexplicable only in terms of its being a presupposition of value complexeswhich legitimate sex differences (and male superiority) in a wide range ofsocial roles. It follows that—as the state of evidence now stands—acceptingtheories developed with LHM involves the play of a value (a “sexist”value: preserve sex differences in social roles) alongside the cognitive valuesrather than prior to their play (in the course of adopting a strategy).

The different roles of the social and cognitive values are crucial. Thereare no logical and methodological (as distinct from moral) objections toadopting the strategies of LHM on the ground of their mutually reinforcinginteractions with value complexes that legitimate widespread genderdifferences in social roles. Usually, however, those who adopt such valuecomplexes maintain that they do so in large part because “natural” genderdifferences have been established rather than that they are attemptingempirically to discover “natural” gender differences for the sake ofconfirming the presuppositions of their value complexes. No one, I suspect,adopts LHM explicitly because it can be expected to provide support for“sexist” values—hence my conclusion that, where theories developed withthe model are accepted, “sexist” values are in play alongside the cognitivevalues. Even so, it is not “bad” science per se to engage in research thatdeploys LHM (Nelson 1995, 1996), especially if it is the only or theprincipal available option in an area of investigation, as long as one onlyaccepts its theoretical products if they manifest the cognitive values highlyaccording to the highest available standards. Furthermore, the unfolding ofresearch using the model may increase the severity of testing theoriesdeveloped under any rival strategies, and this fact may provide sufficientreason to support such research as one among several approaches. While itis proper to adopt strategies whose theoretical products can be expected tobe significant for one’s moral projects, it is also important for the scientificcommunity to ensure the developed theories are subject to testing withsufficient severity.

The criticism just made of accepting theories developed with LHM islogically (but not causally) independent of the “political commitment” towhich Longino refers. Although it originated in the course of inquiry underO3, its logical and methodological force does not depend on sharing thepresuppositions and values lying behind O3. Having made the criticism,then (logically) the choice is made to follow O3, to follow a research


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (217)

approach whose strategies are dialectically linked with feminist values, andwhich runs counter to an approach in which at best theories with lowmanifestations of the cognitive values have been produced. Making thatchoice, of course, does not provide a ground for the judgment that researchdeploying LHM will not be empirically fruitful, though it does for thejudgment that the products of that research will probably not be verysignificant for feminist value complexes (Anderson 1995b). In choosing tofollow O3, one identifies human agency as the primary object of inquiry,and one is primed to query alleged limits to its possible expansion based ongender, race and other such differences. As the criticism of LHM makesclear, following O3 also contributes to “making gender visible,” bybringing to attention the play of “sexist” assumptions alongside thecognitive values in some scientific practices.

Positively, for Longino, the feminist strategy can encourage research onmodels of the brain that conflict with LHM (such as Edelman’s“selectionist model”). A model must be sufficiently complex to beconsistent with the intentionality of action—“a model [that] allows notonly for the interaction of physiological and environmental factors, butalso for the interaction of those with a continuously self-modifying, self-representational (and self-organizing) central processing system.” Moregenerally, the assumptions tend to constrain theories in the directionof “complexity, ontological heterogeneity, interaction and holism (non-reductionism).” For Longino feminist strategies involve these constraints,not because there is an alleged special “female sensibility or cognitivetemperament” that might value highly such characteristics or claim specialinsight that the world is this way, but because human agency displays thesecharacteristics, and cannot be readily recognized and may easily bediminished when these types of characteristics are not highlighted.11

Selection of data

Recall that a strategy both constrains the class of potentially acceptabletheories, and selects the kinds of empirical data deemed relevant for theappraisal of theories that meet the constraints. Recall, too (Chapter 7), thatthe act of observation does not uniquely determine the empirical data, andthat there is no set of lexical categories especially apt for reporting them.Furthermore, objects of experience are not limited to phenomenal orsubjective states; typically they include intersubjectively observable statesof affairs that involve material objects and (often) the actions andinteractions of human beings. The relevant lexicon for reporting empiricaldata depends on the objects of experience that are of interest for theinvestigation at hand. So adopting the strategies that define O3 involvesselecting certain kinds of data as especially pertinent for the investigationof agency, so that they can be brought to bear on theories that meet the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (218)

constraints. Agents act to realize their desires (shaped by their values) inthe light of their beliefs. Reflecting this, the data include reports of actionin intentional idiom. Actions are observable (Donagan 1987), and thelanguage of empirical reports of actions need not abstract from theintentionality involved. One observes (for example) that the experimenterswitched on the machine and read the meter, both actions: neither bodilymovements nor the environmental effects of bodily movements (thoughgenerally there is no action without bodily movement and its environmentaleffects), neither reflexes nor operants. Action is not behavior, as pointed outon p. 205, an object (characterizable completely in non-intentional idiom)which follows lawfully from the action of, for example, genetic,environmental and neuro-physiological variables. Actions will be observedalong with their verbal accompaniments (themselves actions), as well asnumerous contextual factors including the social interactions of which theyare part, so that where actions do not follow from a person’s authenticvalues and beliefs (when agency is diminished), it becomes possible toexplore the social factors that may be among the social causes of thatdiminishment.

LHM cannot account for action; lacking the necessary lexical resources,it cannot fit the kinds of data just described. It can only account forbehavior, its variations and its distribution across groups. Not being ableto account for action is the ground for a further criticism of LHM,different from that made earlier and one that cuts deeper. The earliercriticism simply cleared obstacles to adopting the feminist strategies. Itmaintained that theories developed in accord with LHM do not manifestthe cognitive values highly, so that the presuppositions of feminist valuecomplexes are not inconsistent with theories accepted in accordance withimpartiality, and that the belief that any of these theories are soundlyacceptable comes from (implicit) links with “sexist” values. As I pointedout, this criticism poses no logical and methodological obstacles tocontinuing to engage in research under LHM (but not exclusively), and theresults of such research can be evaluated in the light of the data and thecognitive values. Despite the alleged link with “sexist” values one mightsupport engaging in research under LHM on the grounds that its strategiesinstantiate the materialist strategies and that it is a fruitful researchprogram, the kind of research that has proved itself in many fields to bevery productive. Especially where the conditions for conducting it are inplace, we would expect this line of reasoning to be quite attractive. But itdoes not undermine the reasons proposed for choosing to adopt thefeminist strategies for investigating human cognitive powers and theirexercise. Each line of reasoning has its appeal; but neither, in actual fact, isgenerally compelling, leaving, it appears, at the present time—so far aslogical and methodological considerations are concerned—for individualinvestigators an open choice regarding which strategies to adopt, and for


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (219)

the scientific community an interest in all open options actively beingadopted by some of its members.

“Open choice of strategy” expresses, however, not the last word, butrather an interim tactic to be deployed pending further developments ofinvestigations under the various strategies. For dealing with humancognitive abilities and their exercise, feminist strategies and the strategies ofLHM (or any variant of the materialist strategies, at least as currentlydeveloped) are incompatible and produce incommensurable theories. Thelatter strategies do not possess the conceptual resources to investigateempirically the presupposition of feminist value complexes about thecentrality of human agency. Does following O3 rather than O1 (one of itsvariants, for example, LHM) express just an interest in differentpossibilities, perhaps in those of action rather than of behavior—an interestin different ‘worlds’ (Chapter 7)? If so, each could proceed unimpeded bythe other (except insofar as they might compete for resources) and, ifsuccessful, encapsulate progressively more of the possibilitiesrespectively of interest. Up to a point this is a useful way to look at therelationship between the two approaches, to consider them as dealing withdifferent ‘worlds.’

On the other hand, they compete to produce theories that apply in thesocial “world” of daily life and experience, both to explain phenomenaencountered and characterized in the course of common social practices,and to come to inform action in this realm. Given the ubiquity ofintentional idiom in the characterization of most common social practices,prima facie any approach (for example, feminist) that admits a role forintentional categories has greater explanatory power concerning humanphenomena of salience in them (for example, the phenomenon of engagingin systematic empirical inquiry itself: Lacey and Schwartz 1987) than anapproach defined by materialist strategies. For present purposes I will notdevelop this point. Rather I emphasize two others: (1) that proper testingof some theories (of domains that include characteristic human phenomenaof daily life and experience) developed following O1 requires reference to,and comparison with, results obtained using different strategies; and (2)that in order to legitimate practical applications of the products offollowing O1, one needs to show that there are no undesirableconsequences, for example, impairments of human agency (Lacey 1979) orlost positive possibilities (Chapter 8), but the research to show such thingscannot be conducted under strategies of O1 alone.

Concerning (1) consider, for example, the hypothesis (h) thatmathematical abilities in boys are greater than those in girls. This mightperhaps be thought to be established, as part of a theory utilizing LHM thatrelates mathematical ability (and its manifestations in behavior) linearly tobrain organization, on the evidence that boys perform on averagesignificantly higher than girls in an array of standardized mathematics


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (220)

tests. Longino points out that for the cited evidence to count in favor ofthis hypothesis, rather than (h′) that boys and girls are exposeddifferentially to a variety of social factors, one would also need evidencesupporting such assumptions as: the data have been analyzed so that allrelevant social factors have been controlled for; test performance is reliablyan indicator of inherent ability rather than of acquired knowledge orlearned abilities; there is one form in which mathematical ability isexpressed and that form is expressed in performance in tests of the kindgiven in the studies; and that the content of a problem has no bearing onits formal properties or on an individual’s grasp of those properties(Longino 1990:126–7). A theory containing h manifests the cognitivevalues highly, according to available standards (Chapter 3), only if itmanifests them more highly than competitors containing h′, but thosecompetitors are not developed under LHM (and closely relatedapproaches). Where proper attention is not paid to this standard, itbecomes difficult to identify undiscerned or unacknowledged roles thatvalues may be playing alongside the cognitive values. Thus the propertesting of h can be carried out only in a context where approaches otherthan O1, which permit empirical investigation of h′ (for example, O3), arealso followed.

Concerning (2) suppose it were suggested that a sensible practicalapplication of h would be, for example, to put more resources intomathematics education for boys than for girls so as not to waste resourcesteaching mathematics to those with lesser abilities. Now, apart from thequestions of the adequacy of the testing of h, of whether the intended goalof the application is desirable and of whether there might not be betterapplications, the moral legitimacy of the proposal depends on there beingno undesirable side-effects of implementing the application. The veryapplication, for example, might reinforce other social factors that “really”account for the differences of performance, and itself become one of them(Schwartz 1997) by for instance not providing the conditions that enablethe ability to develop and to be expressed (Lacey 1979). This is not an idlepossibility when one considers a child’s performance on a text to consist ofa series of actions, each responsive to a variety of the child’s desires andbeliefs. Again, following only approach O1 (any version) is inadequate tothe task of investigating such conjectured potential side-effects. When thematerialist strategies are followed exclusively important “lost possibilities”questions always remain outside of empirical purview. They limit what cancome into view, and in a way that is likely to be detrimental to thefurtherance of feminist values.

Thus, the “open choice of strategy” argument breaks down whenquestions of rigorous testing and of application are addressed in additionto those of the classes of possibilities of interest. Research on humancognitive abilities and their exercise, conducted solely under the materialist


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (221)

strategies, cannot adequately address these further questions. It needs to becomplemented minimally by research conducted under strategies thatadmit roles for intentionality and for social causation (for example, those ofO3). A similar conclusion (that it needs to be complemented by researchunder conflicting strategies) also holds for O3, but unlike the one for O1,which is often resisted, it has little practical punch. Given the currentmodes of functioning of scientific institutions, those who adopt feministstrategies cannot avoid having to “legitimate” their approach in the face ofthose who adopt materialist strategies, and thus critically testing their ownproducts against those developed under the materialist strategies (forexample, LHM). Testing across strategies, analyzing competitors in thecontext of a form of full understanding (because it wants to “uncovergender”), and critical self scrutiny are all integral parts of afeminist approach. Nevertheless, it is important to emphasize them giventhe objective of gaining theories (posits) that manifest the cognitive valuesto a high degree as assessed by the most rigorous available standards, forstrong presuppositions about agency lie behind feminist value complexes.Prima facie, approaches such as those of radical behaviorism, LHM andsociobiology have many obstacles to overcome before their interpretations(reductions or replacements) of agency and thus common socialphenomena become sufficiently detailed to compete with intentionalaccounts.13 On the other hand, detailed empirical charting ofcontemporary social phenomena will certainly display (at least in somegroups, contexts and institutions) a wide variety of law-like (or, if notstrictly law-like, statistically stable) regularities. Some of theseuncontroversially reflect deliberately adopted social conventions, but withothers it cannot be settled a priori whether (and in what cases) they aregenuinely lawful (derived from underlying law, structure and process) orhistorically variable regularities, explicable in terms of the boundaryconditions and practices of historically specific institutions that bring aboutdiminished agency (Lacey and Schwartz l986, 1987).

Hence I emphasize (though, as mentioned, it is routinely recognized byits followers) that the following of O3 needs to be complemented withresearch under the materialist strategies. Thus, following O3 could (inprinciple) have the consequence that theories-that may include hdevelopedunder a version of O1 come to be accepted of human cognitive abilities andtheir exercise. But following O1 in the customary way does not even lead toentertaining theories which deploy intentional categories. I said that thecriticism of LHM, implied in pointing out that it cannot deal with action,cuts deeper than the earlier ones designed to carve out some space forresearch under feminist strategies to proceed. It cuts deeper by movingbeyond supporting that O3 is just one legitimate option-in mutuallysupporting interaction with feminist values-among potentially many, tomaking clear that O3 (or close variant) is essential—for cognitive reasons-


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (222)

for appraising theories of human cognitive abilities and their exercise andlegitimating their practical application.

Comparing empirical adequacy across strategies

The preceding argument follows from the claim that LHM and, moregenerally, any theories (TM) generated under the materialist strategiescannot fit certain empirical data: reports of observed actions. Doesanything follow about the limitations of the theories (TF) developed underO3 from the apparently symmetrical claim that the theories it producescannot fit another kind of empirical data: reports of behavior andits relations with, for example, environmental factors, most importantlyexperimental data involving these phenomena?

This raises an important question about what constitutes the empiricaladequacy of a theory. What data should a theory be expected to fit(Chapter 10)? A theory is developed under a particular strategy. In generalit cannot be expected to fit (to put unificatory or predictive order into, asdistinct from merely to be consistent with) data generated from anincompatible strategy. An acceptable theory (of a specified domain) shouldfit the data selected by the strategy under which it was developed; asignificant theory should also apply to salient phenomena located in the“world” of daily life and experience, so that the data it fits includedescriptions (stated in its strategy’s lexicon) of phenomena in this “world.”These descriptions are not necessarily or generally those commonlydeployed within the practices in the “world” that encompass thesephenomena (cf. the discussion of the child’s swing in Chapter 7). Fromdifferent strategies we may get different and incompatible descriptions (cf.impeded free fall, damped pendulum) of the same phenomenon (theswing), all produced for the sake of explaining it and identifying itspossibilities. The most adequate of these descriptions will be the one fittedby a theory that manifests the cognitive values highly; then we havegrounds to affirm that the phenomenon as encountered in the “world” ofdaily life and experience is identical to that characterized using the lexiconof a particular strategy.

Numerous human phenomena in the “world” are described with thecategories of action and intentionality, and these same categories are centralunder feminist strategies. When I say that theories, TM, cannot account foraction, my objection is not that they cannot fit data generated in the courseof following O3, but that the descriptions that they provide ofcharacteristic phenomena of the “world,” that are routinely described inintentional idiom, lack the fine-grained detail of their intentionalcounterparts and that thus the theories they fit are weak (comparatively) inexplaining the details of the phenomena, in encompassing their possibilities,and in anticipating novel possibilities. TM perhaps account well enough for


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (223)

those actions that fall under law-like (or statistically stable) regularities;where the phenomena they apply to also include experimental phenomena,a case can be made that the theories are both acceptable and significant. Butthe strategies under which they have been developed lack the categories todefine the bounds of their application. Then, if the regular phenomena areconsidered also characteristic phenomena, it will be held that the theoryapplies in a wide-ranging way across human phenomena with only thedetails to be worked out for the more “complex” behaviors (such as thoseinvolved in the conduct of scientific research, or a conversation entered intobefore making a decision). Then, action would be understood by referenceto exemplary experimental (or mathematically modeled) phenomena (justas the motions of the swing became understood with reference to thependulum; at one and the same time an experimental artifact and amathematical idealization).

Are not TF seriously empirically inadequate since they do not fit the datagathered in these experimental investigations? Well, in line with thediscussion about which data TM should fit, they are not, since the strategiesunder which they were developed provide no (central) role forexperimental data of behavioral phenomena. TF would be seriouslychallenged only if theories (instances of TM) that fitted both theexperimental and characteristic human phenomena manifested thecognitive values highly. They will not be challenged if evidence can beoffered that the limits of application of TM are bounded and do not includecharacteristic phenomena of agency. This can best be done by proposing anexplicit account of the alleged bounds in some TF and offering socio-historical evidence that they are not gone beyond. Of course, to the extentthat such evidence has not been obtained, the more the partial groundingof TM in experiment will make them compelling, especially if the reach ofapplication of TM progressively expands as they develop. TF does not haveto fit the experimental data that support TM to be empirically adequate,but it does have to fit empirical data (reported with the use of intentionaland social/historical categories) about TM and the reach of theirapplications (see Chapter 10 for a more general analysis).

Feminist approaches and impartiality

A feminist approach to science aims to produce acceptable theories(posits), grounded in the play of the cognitive values held to high standards.It is not simply a critical perspective aiming to uncover bias in mainstreamscience, and not simply a world-view or moral vision. It does not want toreplace rationally acceptable theories with “wishings” about the way theworld is. It does criticize certain scientific claims for reflecting bias, but themore fundamental criticism is that they are false, weakly manifest thecognitive values of relevant domains of phenomena, and ignore (or not


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (224)

permit the development of) more promising approaches. Considerationsbased on the cognitive values are the most fundamental ones in thesecritiques. Pointing to the role of bias in the support of a theory indicatesthat the theory is not acceptable in accord with impartiality, and itindicates avenues to follow for its more stringent testing–test it againstcompetitors that are free from that bias. All of this makes sense because weexpect our research projects to leave a residue in the stock of knowledge. Ifwe did not expect this, we would just have the back and forth play ofbiases, with only power to settle the matter.

Posits soundly accepted in accordance with impartiality are importantfor feminist science also for the sake of informing feminist projects. Actionis more likely to be effective if based on such posits, than if based onweakly confirmed conjectures, and much more so than if based on mere“wishing” (Geertz).17 Moreover, the rational claims of such posits do notrest upon accepting the feminist value commitments. So feminist scienceaims to find out what, “free of wishing,” is the case about human agency,its material necessary conditions, the causes of its diminution, thepossibilities and means for expanding its exercise, and the barriers to itexpanding. Once again, to count as soundly accepted, posits on thesematters must display high manifestations of all the cognitive valuesmeasured against the stated standards.

It does not require commitment to feminist values to recognize thatcertain posits have (or have not) been accepted in accordance withimpartiality, just as one does not have to be committed to the modernvalues of control to recognize that numerous judgments made in materialistscience accord with impartiality. Anyone, in principle, regardless of his orher value commitments, can recognize that a theory (developed under aparticular strategy) manifests the cognitive values highly. That, I suggest, isa condition for the very possibility of critique of an approach to research Inpractice, however, since such recognition involves certain capabilities andskills, and understanding of the meanings of terms involved, and these areoften acquired only from participating in a certain kind of life or evenmoral project, many outside of the practice (be it feminist or materialist)may not always be able to recognize which judgments accord withimpartiality. I take the extent of this kind of de facto incommensurabilityto be a matter for empirical investigation, not for a priori resolution.

Impartiality remains a viable and obligatory ideal of scientific practice,regardless of the approach taken. The very idea of a feminist approach toscience does not challenge this; indeed it should insist on it. Geertz’s worryis misplaced, as long as we separate clearly the roles of social and cognitivevalues. The former influence the strategies; the latter play their role in theassessment of concrete posits. And so the role of the social value“expansion of human agency,” which links with the strategy to constraintheories in “accord with complexity, ontological heterogeneity, interaction


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (225)

and holism,” is separate from that of cognitive values; just as the role ofthe modern values of control, which have dialectical links with thematerialist strategies, is separate from the role of cognitive values. Afeminist approach to science, of course, may be criticized: one mayquestion the concrete outcomes of the research, including how wellthey reflect impartiality, or one may challenge feminist values themselvesand thus the significance of any of the concrete outcomes. Such criticismsleave the idea of a feminist approach to science sound and intact.


For my defense of the legitimacy and even indispensability of a feministapproach to science (O3) the distinction between and separation of roles ofcognitive and social values is crucial. While my account of O3 drawsheavily upon the writings of Longino, it cannot be consideredunambiguously an interpretation of her approach. In recent articles shequestions the crucial distinction between cognitive values and (other)values. She questions whether “paradigmatic constitutive values [ofscientific practices] have a solely epistemic or cognitive basis” (Longino1995:384), and later she entertains the more radical idea, “to cast doubt onthe very idea of a cognitive value” (Longino 1996:42). If her questioningcan be upheld, the defense I have mounted collapses, not only of the placeof O3, but also of impartiality in general and, with the latter, all prospectsof gaining significant knowledge.

Longino entertains a possible distinction between “constitutive” and“contextual” values of science. Constitutive values are criteria for what“counts as good or acceptable scientific judgments” (Longino 1995:353),criteria “involved in the assessment of theories, models and hypotheses,guiding their formulation, acceptance and praise, disparagement andrejection, and pursuit or abandonment” (Longino 1996:49). Contextualvalues are those that influence the process of science in any other way. Shedoes not distinguish (as I have done: Chapter 1) accepting from otherpositive stances one may adopt towards a theory, for example: provisionalentertainment, commitment to explore, application in practice; she doesnot distinguish judgments of acceptance and significance; and she does notmake explicit that theories are accepted of particular domains. Then,constitutive values include all those that serve as criteria of scientificjudgment when adopting any one of these stances, thus includingjudgments of both acceptance and significance. Constitutive values,therefore, include both cognitive values and at least those social values thatinteract with the adoption of strategies in mutually reinforcing ways. Forher, cognitive values are “characteristics enhancing the likelihood of thetruth of a theory or hypothesis” (Longino 1995:583), or characteristics


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (226)

pertaining to the degree of confirmation of a posit, or to how it stands inthe face of the empirical data that are considered relevant evidence.

According to impartiality, the cognitive values alone are constitutive ofsound scientific judgments of theory acceptance.

Values (in addition to the cognitive values) can be regarded (on myaccount) as among the constitutive values of particular approaches toresearch, since they are built into the various objectives of approaches toresearch (Oi′), and they determine what counts as a potentially significanttheory. In addition, commitment to values can properly lead to theexpectation that assessments of the manifestation of cognitive values besubmitted to tougher standards (“Rudner’s argument”: Chapter 4).Attributing these roles to values, however, does not cast doubt on the veryidea of a cognitive value; rather it presumes that there is a proper role forcognitive values distinct from that of the other values. Longino adds to thatcharacterization of a cognitive value that it is “a quality of [posits]…thatcan serve independently of context as a universally applicable criterion ofepistemic worth” (Longino 1996:42).

This addition provides the thrust to her argument. Given it the idea ofcognitive values indeed dissolves. First, what counts as a criterion of“epistemic worth” varies with the stance adopted towards a theory. Theepistemic worth of a provisionally entertained theory (for example) is notsubject to the criterion of empirical adequacy (though that of an acceptedtheory is), but it may be to that of fitting the constraints of a chosenstrategy. Second, context is often important to interpret a quality likeempirical adequacy, as illustrated in the preceding discussion of theempirical adequacy of theories developed under feminist strategies. Apartfrom the context of a strategy, to refer to a theory (of a domain) asempirically adequate is often hopelessly vague. Third, the ranking of theproposed cognitive values is contextually related to strategies, reflecting theway in which empirical adequacy is interpreted or the characteristics of thedata selected for attention by the strategies. Full rather than wide-rangingexplanatory power, the power to anticipate novel possibilities rather thanpredictive power, and capturing fine-grained details of relations rather thatquantitative precision (for example) rank higher in the context of feministstrategies; vice versa in the context of materialist strategies. It is true, then,that the proposed cognitive values cannot function as criteria of “epistemicworth” independent of context (stance adopted towards a theory orstrategy). Longino seems to suggest that this provides sufficient grounds forrejecting any clear distinction of roles between cognitive values and social,moral, etc. values, and that they should all be grouped together interactingwith one another, as it were, at the same level.

My idea of cognitive values does not require the kind of universalitypresent in Longino’s addendum, as acknowledged in the precedingthree points. It treats cognitive values as decisive criteria only for the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (227)

acceptance of theories. They exercise their decisive role only after strategieshave unfolded and prior stances have been taken towards theories. (Up tothis point values and cognitive values intermingle without creating logicalproblems.) Cognitive values characterize the relationships that must existbetween theory and data (and other theories) for a theory to be acceptable(to express some form of understanding of a domain of phenomena), andthus the criteria for choice among the various theoretical candidatesconsistent with the strategy in play While choice of strategy reflects values,and that choice has implications concerning the interpretation and rankingof (some of) the cognitive values, once it is made, theories are accepted (whenimpartiality is respected) without further mediation of values. The idea ofcognitive values needs only this kind of universality. I do not think thatLongino’s arguments undermine it.

Let us look at Longino’s arguments or rather her questions, for herconclusions are presented as not yet settled. They raise important andnovel considerations and involve two components. The first is to explorewhether an alternative list of constitutive values, which she identifies, maybe appropriate to adopt for a feminist approach to science, including formaking judgments of accepting and rejecting theories. The second is toargue that Kuhn’s list of cognitive values (accuracy, consistency, scope,simplicity, fertility) is no less grounded in (other) values than the proposedalternative list is, that only (social) values pick out one of the lists as that of“cognitive values.” I will only discuss the first component here; criticism Iwould offer of the second brings in no new considerations.

A proposed alternative list of cognitive values

What makes a theory a good or acceptable one? Within a feministapproach to science, Longino sees the following criteria being deployed:empirical adequacy novelty ontological heterogeneity, complexity andmutuality of interaction, application to current human needs, and diffusionor decentralization of power (Longino 1995:385; 1996:45 ff.). Only thefirst item is a generally recognized cognitive value. She distinguishes thelast two from the others as being “pragmatic.” The others are treated asbeing on the same level, but with functions that are contextually variable. Ithink that the role of empirical adequacy needs to be sharply distinguishedfrom the roles of the others, and that the other cognitive values mustcomplement it.

Consider novelly, “models or theories that differ in significant ways frompresently accepted theories, either by postulating different entities orprocesses, adopting different principles of explanation,incorporating alternative metaphors, or by attempting to describe andexplain phenomena that have not previously been the subject of scientificinvestigation” (Longino 1995:386). Novelty clearly is not a value


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (228)

pertaining to theory acceptance—the kind of warrant needed foracceptance is unavailable without extensive and prolonged research. Itpertains at most to other stances taken towards theories, for example,provisional entertainment, or commitment to explore. Novelty itself is opento two interpretations. On the weak one it is linked with a desire to gainscientific understanding of hitherto neglected phenomena (Longino 1995:387). Then what is important is not that the theory is novel, but that it isof these phenomena. The value in play here is a social one, “being of thesehitherto neglected phenomena” and it may properly influence the strategyadopted in investigation. Novelty then is a trivial consequence of “hithertoneglected.”

The strong interpretation refers to novel posits that are competitors withexisting ones. Novel (original) competing theories are valued under certainconditions—for example, because one wants to test a current theoryagainst more rigorous standards, or because currently available theories donot enable us to solve certain problems or do not fit strategies one wishesto adopt. Either way, the value of novelty is derivative; its value lies inbeing responsive to the considerations that underlie the desire for morerigorous testing (for example, “a suspicion of any frameworks developed inthe exclusionary context of modern European and American science”—Longino 1996:46), or in enabling hitherto unsolved problems to be dealtwith, or in furthering research under one’s favored strategies. Especiallybecause value considerations can play a role (of which one may beunaware) in one’s assessments of the acceptability of theories, it is oftenappropriate to raise the issues of higher standards of testing and ofapplicability of the theories to expanded domains of phenomena. A noveltheory may contribute to cognitively significant considerations—not quanovel theory, but qua theory that over time develops in a certain way.Novelty is not a cognitive value. Considerations pertaining to it, however,may be essential to deploying the standards with which we assess theextent of the manifestation of the cognitive values. Neither is novelty per sea value that might be added to the list of constitutive values of an approachto investigation. A value on such a list, however, might point to the needfor novel theories in certain areas, so that derivatively novelty might be animportant value in making judgments of provisional entertainment of atheory and commitment to its development.

Consider ontological heterogeneity. A plausible argument was presentedin the course of discussing Longino’s criticism of LHM that the objectiveof a feminist approach to science (O3) requires the development of theoriesthat represent phenomena of human agency as ontologically heterogene-ous. This requirement is reinforced when one attends to the more specificobjective that, I pointed out, sometimes Longino considers: “to revealgender”; for presumably the point of revealing gender is to guide projectsaiming to overcome it, and thus to expand the possibilities for the exercise


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (229)

of the capacity of agency that are currently inhibited by gender. Thisargument stresses the importance of individual differences in investigationsin the life and behavioral sciences, and the rejection of treating differencesas departures from some type which is taken as representing somethingontologically fundamental. She speaks of deriving from ontologicalheterogeneity the “rejection of theories of inferiority” (Longino 1996:47).When she does so, she is not denying the “factual” claims of such theorieson the ground of their discord with her adopted values, but pointing out thatsuch theories have not been tested against theories developed underconstraints dialectically linked with this value, so that, therefore, they havenot been appraised under appropriately high standards (cf. the precedingdiscussion of LHM). It is thus appropriate to consider ontologicalheterogeneity as among the constraints on theories required by strategiesthat are adopted in view of choosing human agency and the expansion ofthe possibilities for its exercise as the object of inquiry—and thus toconsider it as a constitutive value of the approach O3. That theories meetthis constraint derives from its mutually reinforcing links with the value ofexpanding human capacities for agency. Then, it remains for empiricalinquiry to establish whether or not theories which manifest the cognitivevalues to a high degree can be developed under it. Similar conclusionsholds for complexity and mutuality of interaction.

The other two criteria, the ones Longino calls “pragmatic” (Longino1996:48) require a different analysis. Consider applicability to currenthuman needs—“directed towards meeting the human and social needstraditionally ministered to by women” (Longino 1996:48). This can betaken as the recommendation not to apply, in practice, theories that do notinform attempts to redress current human needs, or as the recommendationto seek theories that can be applied to inform such ends. Then the value ofaddressing human needs functions as a criterion for the interest andsignificance of a theory, but not of its cognitive credentials (cf. the criticismof green revolution science in Chapter 8). It may also serve to focus ourparticular research interests within an approach to investigation defined bya certain object of inquiry, and perhaps occasion the identification of a newobject of inquiry.

Diffusion of power is a property not of theories, but of the institutionsand practices from which theories are produced. It sometimes has cognitivesignificance. In principle, the greater the range of perspectives brought tobear in the comparative assessment of theories, the higher thestandards against which the manifestation of the cognitive values can bejudged. Diffusion of power within research institutions and practices canbe expected ceteris paribus both to bring a greater diversity of participantsinto the research process (likely to increase the range of perspectivesbrought to bear, but not necessarily under all conditions), and to give


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (230)

greater authority to hitherto marginalized perspectives (cf. the discussionsof autonomy: Chapters l, 4 and 10).

Thus diffusion of power refers to an institutional and practical value,deriving from its service to the rigorous assessment of theories in the lightof the cognitive values, that is, to impartiality. One dimension of this isthat of creating conditions in which contributions to knowledge fromlocal, traditional and indigenous sources may not be ignored (Chapter 8).It also ceteris paribus serves the interest of applicability to current humanneeds.

Now consider empirical adequacy. As discussed on p. 212–14, itsinterpretation can vary with the strategies under which inquiry isconducted. Empirical adequacy is the value that a theory “fit” the relevantclass of available empirical data, where the adopted strategy determineswhat counts as “relevant.” Once a class of data has been picked out asuniquely relevant, it can bear evidentially only upon posits of certain types.But I have maintained that adequate testing of posits requires meetingstandards that involve comparative assessment. This is especially pertinentwhen dealing with posits which—if considered incorporated into the stockof knowledge—challenge the presuppositions of moral projects (forexample, feminism). Here empirical adequacy must function in concertwith the other cognitive values. Consider TM (theories developed undervariants of the materialist strategies). They were criticized not for beingempirically inadequate in the context of the materialist strategies, but forlacking (comparatively) explanatory, unificatory and anticipatory(predictive) power among the realm of phenomena that are described in thecourse of daily life and experience with action and intentional terms, andfor lacking the resources to shape interpretive accounts of the successes,failures and bounds of application of competing theories. This criticismcannot be sustained unless empirical adequacy is complemented withseveral other cognitive values, so that Longino’s own criticisms (as I havedeveloped them) depend on commitment to something like a standard listof cognitive values.

Longino maintains that a cognitive value, such as empirical adequacy,does not have “a solely epistemic or cognitive basis.” I concur: sinceadopting a strategy is partly rationalized in view of its mutually reinforcinginteractions with certain social values, values contribute to some extent tothe interpretation of empirical adequacy that one brings to bear on one’shypotheses. She goes on to say: “Empirical adequacy is valued for,among other things, its power when guiding inquiry to reveal both genderin the phenomena and gender bias in the accounting of them” (Longino1996: 45). I take this to mean that empirical inquiry conducted underfeminist strategies, properly responsive to the cognitive value of empiricaladequacy (together with, I add, a few other cognitive values) has the powerto reveal gender and, therefore, it is able to inform feminist projects, and


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (231)

that the inquiry gains value from that fact. The products of this approachto inquiry, but not generally those of O1, can be adopted to informpractices of the feminist moral project. This gives them a value, lacking forthe products of variants of the approach O1, provided that they come toinform practices effectively. Effectiveness in application presupposes thatthe products manifest highly the cognitive values.

I mentioned on p. 212 that it is possible (but not likely) that, within afeminist approach to science, some theories expressing LHM might becomesoundly accepted. If they were, they would be no less valuable thantheories that reveal gender, if soundly accepted, are. Their coming to beaccepted would show that certain presuppositions of the feminist moralproject could not be sustained. That would no doubt be a disappointment,but it would not be a ground to reject the outcomes of the inquiry, and toattempt to act contrary to them would be folly. The point, of course,generalizes: where feminist inquiry successfully reveals gender, to attemptto act contrary to the finding threatens to entrench what has been revealed.The role of the cognitive values concerns—once strategies have beenadopted—only how the logical gap between data and posits is to be closed.This requires a clear distinction of role between cognitive values and othervalues.

I do not think that Longino’s arguments undermine the distinction ofroles between cognitive and social values (see Chapter 10). Equally, myarguments do not undermine, rather they reinforce, other conclusions shehas reached: that scientific knowledge emerges from a process in whichvalues are pervasive; that theories have desiderata linked with significanceover and above cognitive ones; that values (since they influence thestrategies adopted) and therefore social conditions partly (largely?) explainwhich theories come to be produced and accepted in the scientificcommunity (Chapter 10); and that social values can be manifested intheories alongside the cognitive values (contributing to their significancefor certain value complexes, but not their acceptability), since theapplicability of some (kinds of) theories may serve particular valuecomplexes especially well (Chapters 4 and 10). The crucial distinction ofroles of the cognitive and (other) values can be reconciled with thepervasive role of values in the processes of science. It is not merely a formalreconciliation. The very intelligibility and rationale of gaining andattempting to apply systematic empirical understanding (science) dependson it.

The details of the reconciliation will be developed in the next chapter inwhich I attempt to articulate what can and cannot be defended of the threecomponents of the idea that science is value free.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (232)

10Science as value free: revised theses

I introduced theses of impartiality, neutrality and autonomy (inChapter 4); but they concerned only inquiry conducted predominantlyunder the materialist strategies. It is now clear that this involves a seriousoversimplification. At least in principle, and to some extent in practice, theobjective of science—to gain understanding of phenomena—may (must?)be pursued with a number of different approaches, each deploying its ownparticular strategies. Where does this leave the view that the sciences are,or ought to be, value free? In view of the argument that no theories canbecome accepted outside of research conducted under strategies which areadopted in part because of their mutually reinforcing interactions withparticular value complexes, does the idea of science as value free retain anysense? Or should it be dismissed simply as the allure of a way of thinkingthat disguises the role of values (the modern values of control) in thepredominant approaches to modern science?

To address these questions, I will develop revised versions of impartialityand neutrality that leave open (in principle) a multiplicity of approaches toscientific inquiry. My answer will be impartiality can and ought to besustained as a viable and important thesis; that neutrality, despiteambiguities that make it difficult to sustain in a clear-cut way, can beinterpreted so as to be defensible; but that only fragments of autonomy canbe sustained.


I proposed (Chapter 4) the following thesis of impartiality (in abbreviatedform):

I′ 1 The cognitive values are distinct from other values.

2 T is accepted of D if, and only if, T (of D) manifests the cognitivevalues highly in relation to E—where T and E fit the materialiststrategies.

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (233)

3 T is rejected of D, if and only if, T′ is accepted of D, where T and T′are inconsistent.


4 Values are not among the grounds for accepting and rejectingtheories; and

5 Fitting the materialist strategies is the only antecedent require menton T and E.

I′ builds in the materialist strategies; in it T and E represent theories anddata (drawn from some domain of phenomena, D) that are constrained andselected respectively by the materialist strategies. Thus, by definition,judgments made under alternative strategies cannot be made in accordancewith impartiality. This is unsatisfactory, for those who adopt alternativestrategies aspire to making judgments of which (4) holds. They, too, do notwant their judgments clouded by extraneous considerations; they want toseparate the genuine from merely logical possibilities, to test their theoriesagainst the best materialist theories and to have their practical projectsinformed by sound empirical knowledge. We need a statement ofimpartiality that may apply to research conducted under any strategies.

I′ faces the further difficulty that (4) and (5) cannot be held togetherconsistently, since (5) involves that the modern values of control are amongthe grounds for rejecting a whole class of theories, those that do not fit thematerialist constraints, thus contradicting (4). I say “rejecting,” rather than“not investigating,” for a policy of not investigating is tantamount torejecting. Then, only if the applicability of I′ were clearly restricted totheories of domains whose possibilities are exhausted by their materialpossibilities, could (2)/(3) be sustained as a value of scientific researchpractices. (I am assuming that (1) has been satisfactorily established.) Onlyif all the possibilities of phenomena were material possibilities, would itmake sense to restrict the interpretation of impartiality in a way thatpresupposes the deployment of the materialist strategies. To put such acondition upon all scientific practice, prior to the outcome of inquiry overthe long haul (without a sound argument for materialist metaphysics)would be an instance of “wishing” (Geertz: Chapter 9): treating the world,insofar as it comes into contact with our practices, as an object ofpossible control, as we may “wish” it to be, but which we have no soundground to affirm that it is.

Could these problems be avoided by revising I′ so that acceptance oftheories becomes relativized to strategies, as in the following (using the sameabbreviated form as on p. 224–5 with I′):


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (234)

I″ 1 The cognitive values are distinct from other values, and they maybe manifested in theories developed under a variety of differentstrategies.

2 T is accepted of D under S if, and only if, T (of D), in relation toE, manifests the cognitive values highly, and to a higher degreethan any rival theory does—where T and E fit S.

3 T is rejected of D if, and only if, T′ is accepted of D, where T andT′ are inconsistent.


4 Values are not among the grounds for accepting and rejectingtheories.

Values may play a role in the choice of which strategy to adopt, but this isconsistent with I″, which may be seen as a simple generalization of I′. Anytheory accepted in accordance with I′ will also be accepted in accordancewith I″—except that now it will be accepted under the materialiststrategies, and so accepted explicitly of domains of material possibilities ofthings. Theories, that do not fit the constraints of the materialist strategies,are now not rejected because they violate these constraints, but onlyrendered insignificant for not pertaining to the valued class of possibilities.Relativized acceptance need not lead to a problematic relativism of rationaljudgment. Different judgments reflect different foci of interest, differentvalued classes of possibilities, but that a theory is accepted under S (of D)depends only on its manifesting the cognitive values to a high degree inrelation to the selected data—a judgment that can arguably be made (inprinciple) by everyone regardless of whether or not they actively conductresearch under S and adopt the values linked with S. “T is accepted of D”is relative to S; but “T is accepted of D under S” is not relative to anything.

The relativizing of acceptance in I″ is consistent with a number of thingsthat I have emphasized: the separation of the levels of strategies andtheories, a role of values in the adoption of strategies which does notthreaten that sound theory choices are responsive only to the data and thecognitive values, the claim that there is no way to pursue thegeneral objective of science, to gain understanding of phenomena (O)—especially its component: to encapsulate the possibilities of things and theways to realize some of them—except under the strategies of a particularapproach (Oi: Chapter 5). On the other hand, it is marred by a seriousambiguity. Does “rival theory” designate a theory that fits the constraintsof S, or one developed under any strategies? If the former, then “acceptingT of (D)” would be, in effect, “accepting that T (of D) is the best theory (ofD) developed under S”; and “rejecting T”; would be “accepting a better


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (235)

theory (inconsistent with T) developed under S”. This would avoidrejecting rivals because of their lack of fit with S and thus in part forreasons that include values. If the latter, there would be no need to resortto relativizing acceptance and rejection of theories. If we can compare thedegrees of manifestation of the cognitive values in theories developed underdifferent strategies, why would we be content to compare them only intheories developed under the same strategies? Why keep the theoreticalproducts of different approaches separate, without critical interaction withone another, as if they are not all framed by O? There is a reason, but itdoes not provide the whole story. Unless a strategy unfolds of its ownimpetus, it is not likely to develop theories that manifest the cognitivevalues to a high degree; fruitful comparison of theories across strategiesrequires sufficiently developed theories (in relevant ways of the samedomains). At least sometimes, there are two moments to the acceptance oftheories: acceptance that T is the best theory of D under S, and thenunqualified acceptance of T (of D).

Sometimes we do compare the cognitive merits of theories developedunder different strategies, and judge theories developed under one strategyto manifest the cognitive values of comparable domains of phenomenamore highly than those developed under another (incompatible) strategy.1

When we do this the cognitive values “encapsulation of the possibilities ofa domain of phenomena,” “explanatory power” and “capability to definethe limits of application of a theory” are likely to be most salient. Since aparticular approach is linked (in mutually reinforcing interactions) withparticular values, the manifestation of “encapsulate the possibilities…”prima facie may be expected to be restricted. Attention to it alerts us not toidentify readily the possibilities afforded by the phenomena of a domainwith those possibilities of the phenomena that are of interest in light ofthese values; to attempt to define the limits of applicability of theoriesdeveloped under a particular strategy; and to raise the question of whetherthe lexicon of the strategy being followed has the resources to characterizeall the humanly and environmentally relevant effects of application oftheories developed under the strategy (Chapter 8, and the later sections ofChapter 7).

It is always pertinent, in principle, to ask: concerning specified domainsof phenomena that we are investigating, could we adopt alternativestrategies that would enable us to gain theories that would manifest thecognitive values (especially the three just listed) more highly? If the answeris “yes,” then ceteris paribus that is sufficient reason to adopt thealternative strategies for the investigation of those phenomena, regardless ofthe (social) values linked with the initial strategy, and it might also providethe occasion to reassess commitment to these values. In this way thecognitive values may also play a role in considerations of the choice ofstrategies for investigating certain domains of phenomena, but that does not


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (236)

annul the mutually reinforcing interaction of the strategies with a set ofvalues. Links with certain values and not others, thus, may be conducivefor gaining theories that manifest the cognitive values most highly ofspecific domains -as illustrated in the ways that the modern values ofcontrol are conducive for gaining theories that encapsulate the materialpossibilities of things and that represent fully the possibilities of certaindomains in which relevant human influence is absent, feminist values forgaining theories about human cognitive abilities (Chapter 9), and thevalues of authentic development for gaining access to possibilities (inimpoverished countries) of agricultural practices that would serve bothsocial justice and environmental integrity (Chapter 8).

We need a formulation of impartiality that, unlike I′, does not build inthe adoption of a particular strategy and that, unlike I″, permits impartialjudgments in theory choices made across strategies. To this end we mighttry replacing I″(2) (again maintaining the abbreviated form) with (2a) andmaking the obvious related adjustment to (3).

(2a) is accepted of D if and only if T, in relation to E, manifests thecognitive values highly—where T is constrained by S, and some of itsrivals are constrained by S1, S2,…(S1 ≠ S), S), and E includes the dataselected under all of S, S1, S2…; and making the obvious relatedadjustment to (3).

But E cannot include all these data. Data are sought out and recordedunder a strategy, from which the lexicon of observational reports isobtained. Generally, competing strategies are incompatible and (to someextent) their lexicons are incommensurable. Recall that where anAristotelian scientist (A) observes an instance of inhibited free fall, aGalilean (G) observes a damped pendulum (Chapter 7). G’s theories do nothave to “fit” A’s data, and vice versa. In G’s theories there are no inhibitedfree falls, and what A describes as an inhibited free fall is considered to bein fact a damped pendulum. A theory must—so to speak—only fitdata that express facts of phenomena of D. Methodologically this meansthat, while a strategy remains actively in contention, theories developedunder it are required only to fit data selected under it. This raises questionsabout D, the domain of phenomena of which theories are in competition.We grasp phenomena in scientific ‘worlds’ and/or social “worlds”(Chapter 7), with lexicons developed either in scientific practices or thesocial practices of daily life and experience. If the items of D arecharacterized solely with scientific lexicons, then generally the question ofcomparing theories across strategies will not arise, for a phenomenoncharacterized with the lexicon of S1 will be part of a different ‘world’ fromthat of one characterized with the lexicon of S2; theories developed underthe different strategies will not compete of domains in the same ‘world.’


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (237)

There can be competition only when there is a shared characterization ofthe items of D—in a lexicon of social practices of daily life and experience.G and A competed of domains that included such phenomena as children’sswings, flights of arrows and planetary movements. Each proposedapplications to phenomena of this social “world,” of which both theirscientific ‘worlds’ were “sub-worlds.” The categories of this “world”include those of both ‘worlds’; indeed scientific activity in general isunintelligible when abstracted from the “world” in which it gains thenecessary conditions to be pursued, and the contexts in which to be appliedand in which its theories gain significance. Thus, with these categories itbecomes possible to formulate that a phenomenon, f, characterized withthe categories of a common social practice (for example, playing on aswing) is identical to f, characterized with the categories of A, or to f,characterized with those of G. Let D, a domain of the “world,” berepresented by {f1, f2,…}. When A is applied to D, it represents it as D={f1,f2,…}; and G as D={f1, f2,…}. Let E be the class of data formulated with thelexicon used in D; and E with the one used in D. The items of E and E arelargely (not entirely) incompatible. G is accepted of D because it manifeststhe cognitive values, in relation to E, more highly than A does in relation tothe items of E (Chapter 7). It is accepted of D because it is accepted of D.This judgment does not require that either theory be expected to fit dataother than those selected in accordance with the strategy under which itdeveloped.2 Generally the question of comparison of theories acrossstrategies arises only in the context of the application of theories todomains of phenomena in the realm of daily life and experience. It is oftenof great importance when facing practical (technological) applications oftheories, especially questions about their legitimation.3

I now offer the following statement of impartiality:

I 1 The cognitive values are distinct and distinguishable from othervalues, and they may be manifested in theories developed under avariety of different strategies.

2 T is accepted of D under S if, and only if, T is accepted of D undera strategy S; and so, in relation to E, manifests the cognitive valueshighly according to the most rigorous available standards; and to ahigher degree than any rival theory manifests them in relation tothe data appropriate in the light of the strategy under which itdeveloped—where T meets the constraints of and the items of Ehave been selected in accordance with S, and some of the rivals are(were) developed and appraised under different strategies.

3 T is rejected of D if, and only if a rival theory (T′) is accepted of D,and T and T′ are inconsistent, regardless of the strategies underwhich T′ developed.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (238)


4 Values and assessments of a theory’s significance are not among thegrounds for accepting and rejecting theories.

I articulates nicely the value of impartiality.4 It is well expressed in thepractices of systematic empirical (scientific) inquiry to the extent thattheories accepted in them are accepted in accord with I, and that they areconducted so as to increase the number, proportion and variety of theoriesaccepted in accord with I. Adopting I requires attentiveness to theconditions under which its expression can be furthered. It is also a value ofscientific practices, I propose, that all theories that inform a practical(technological) application and the (factual) presuppositions of thelegitimacy of the applications be accepted in accordance with I.5 Judgmentsmade in accord with I will be preceded by judgments made in accordancewith I″. In the case that D is not constituted by phenomena in the realm ofdaily life and experience, I and I″ (or I′) tend to collapse together, althoughif D can be maintained to be open in principle to investigation under adifferent strategy, accord with I will not be reached prior to comparativetesting against theories developed under the different strategies. Then,before a proper comparison can be made, theories under the respectivestrategies must be appropriately developed.6

Impartiality is a demanding thesis and its manifestations are not casuallyupheld. There are, however, many theories (of specified domains) that havebeen accepted in accordance with I, and that also are significant for manyvalue complexes. Indeed, exemplary cases have been developedand consolidated under the materialist strategies (though I is not always a“fact” of theories actually accepted under these strategies). Within manyspaces, whose kinds of boundary conditions we can broadly specify,phenomena have been soundly grasped—as evaluated by the cognitivevalues—in terms of being generated from underlying law, process andstructure, and their possibilities are exhausted by their materialpossibilities. The world is this way; we have not taken it to be this waybecause of our desire to control it. In the long run, part of the reason tocontinue following the materialist strategies is the considerable and on-going empirical success gained from following them, the success of comingto accept theories (of certain domains) in accordance with I.

Separate roles for cognitive and other values

When one adopts the materialist or any other strategies, in effect one laysout in the most general terms the kinds of phenomena and possibilities to beinvestigated. There is nothing logically improper about social valuesstrongly influencing this choice. Then, the acceptability of theories


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (239)

constructed under the strategies is judged in the light of the data and thecognitive values (where having been developed under a particular strategyis not a cognitive value: Chapter 5). The important thing is to keep theroles of the social and the cognitive values separate. Their different rolesreflect different (logical) moments connected with making theory choices.At one moment, when we ask: “What characteristics must theories have tobe provisionally considered?”, strategies play the key role. They serve toeliminate theories that do not fit their constraints. They function (logically,not necessarily temporally) first. In principle, with respect to a givendomain of phenomena, an array of incompatible theories will fit theconstraints; the play of the strategies is insufficient to determine whichtheory to accept. Then (logically) at the second moment, when we ask:“Which (if any) of the theories, with these characteristics, is to beaccepted?”, one of the theories from the array may be accepted. Here, theplay of the cognitive values, in the light of the empirical data and otheraccepted theories that are available, is (according to I) decisive.7

At the moment of concrete theory choice only the cognitive valuesproperly play a role. If, given the data available in the current state ofinvestigation, the cognitive values do not suffice to make a clear choice, notheory may be considered soundly accepted and the matter must remainopen to further investigation. If, in actual fact, a theory is accepted underthese conditions, then a value has played a role (improperly) alongside thecognitive values in making the judgment of acceptance. This does notpreclude values playing a proper role, however, not alongside thecognitive values, but at the moment when the strategies function,interacting dialectically with them. Strategies are adopted because ofinterest, typically derived from values, in the possibilities that may beencapsulated in theories constructed and consolidated under them.Whereas strategies lay out the general features of the possibilities ofinterest, a properly accepted theory encapsulates what the genuinepossibilities are.

One may adopt strategies, then, because of their relationship to values,because they may enable us to encapsulate possibilities that may informone’s moral and social projects. That does not imply that one eliminatesfrom consideration theories that do not fit the strategies because onebelieves that they are false; rather one may do so because they do notprovide a means to identify the possibilities of interest. One tends to adoptstrategies under which it can be expected that significant theories will bedeveloped. Now, a significant theory is also a soundly accepted one; but atheory, that is soundly accepted of certain domains of phenomena and thatreliably encapsulates their possibilities, may have no relevance for one’spractical projects. (Dismissing a soundly accepted theory as insignificantdoes not imply contradicting that theory.) Adopting a particular strategydoes not (and cannot) commit one to accept any theory; rather it frames


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (240)

the quest to construct and consolidate theories of certain kinds, but itprovides no guarantee that the quest will be successful. Not all strategiesare like the materialist strategy, such that the world lends itself to revealingcertain of its possibilities (many of which can be realized through ourinteractions with it) in the course of research under them. Persistent failureto develop under given strategies theories, which manifest the cognitivevalues to a high degree, is ceteris paribus a decisive ground to abandonthose strategies. Thus, the adoption of strategies is not only linkeddialectically with values, but also is under long-term empirical constraint.

A theory is properly accepted (rationally believed to encapsulate thepossibilities) of a domain only if it manifests the cognitive values to a highdegree according to the highest standards for assessing the degree of theirmanifestation in theories. The values, that make these possibilitiesinteresting and that may motivate the provisional entertainment of theoriesthat fit the strategies that are dialectically linked with them, play no(proper) role in judgments of acceptance. They do (ought) not functionalongside the cognitive values. The distinction and separation of momentsis methodologically and logically essential. The answer to “On whatgrounds is T (of D) accepted?” properly includes no appeal to values, butthe answer to “Why does the scientific community investigate theories thatare in line with certain strategies?” will typically do so; appropriately so.Although with the passage of time that will not be sufficient by itself,and will need to be bolstered with evidence that following the strategiesleads to success in consolidating theories that manifest the cognitive valuesto a high degree.

The answer to “How did it come about that T (of D) became accepted?”will also involve reference to values—those which sustained the researchactivity and motivated following the strategies. Values play a causal role inthe development and consolidation of theories. That is consistent with I,for I is a thesis about the sound acceptance of theories, not one about howto explain that a theory came to be accepted. Pointing to the grounds forsoundly accepting T (of D) is not the same thing as explaining why T cameto be soundly accepted. The latter requires also reference to the process ofinvestigation: Why, for example, was T provisionally entertained in thefirst place? Why did certain investigators commit themselves to itsdevelopment? Why was D considered of interest? Theories which satisfy Iare the residue of a value-implicated process. Thus explaining that T hascome to be accepted will involve reference to values, but this is irrelevant towhether T is accepted in accordance with I or not. The question: “Why is Taccepted?” is ambiguous between “On what grounds is T accepted?” and“How has it come about that T is accepted?” Some social constructionistshave assumed that good answers to the latter annul or need not pay heed toanswers to the former, and thus have erroneously concluded that the role


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (241)

of values (and the power that may support their role) in good answers tothe latter shows that impartiality cannot be realized.

Regardless of what values pertain to the causal history in which T becomesaccepted, I may be satisfied, but that causal history may also bring it aboutthat T be an object of value only within certain value complexes, especiallythose that interact in mutually reinforcing ways with the strategies underwhich T developed. Accepting T does not imply its general significanceacross value complexes. At the same time attending to the causal historywithin which it becomes accepted is not irrelevant to judgments aboutwhether T has been accepted in accordance with I, for that history canalert us to whether T has been tested against an appropriate array ofcompetitors, including (in principle) competitors developed undercompeting strategies. More generally, it provides the evidence pertinent tojudging if the degree of manifestation of the cognitive values in T has beenestimated according to the most rigorous available standards. Failing toattend to the causal history may hide the covert role that values mayactually be playing alongside the cognitive values when appropriatecompetitors are not entertained.9 In many of these situations, a mistakeninference is covertly made from accordance (discordance) with I′ toaccordance (discordance) with I,

The judgment that T (of D) is accepted in accordance with I is logicallyindependent of the values that are held. Thus, when well made, it isbinding generally regardless of the values held, regardless of the possibilitythat the adoption of certain values may be a necessary condition forengaging in the research activities (or even for having the skills andcapabilities to do so) that enable such a judgment to be made, andregardless of the possibility that the theory may be significant only wherethese values are adopted. Holding different values, then, is not a groundfor denying the cognitive value of such a judgment, or for acting on the basisof beliefs that are inconsistent with it. But holding them may void thejudgment of the significance of the theory, and provide a reason to engagein investigation exploring different classes of possibilities by adoptingdifferent strategies.

Rejection of theories

So far in this section I have focused on the acceptance of theories. When weturn to their rejection additional matters arise. If T has been soundlyaccepted of D, then, in accordance with I, theories that are inconsistentwith T will properly be rejected. In such cases, rejection follows from thedata and deployment of the cognitive values. Adopting strategies involvesanother kind of “rejection.” Not only does it underlie choosing those(embryonic) theories that are to be provisionally entertained and adheredto for the sake of their development, but also characterizing those that are


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (242)

simply not to be considered, those that are to be “rejected” out of hand or“side-lined.” What is the status of the side-lining of theories when, forexample, under the materialist strategy theories that deploy teleological,intentional and sensory categories are side-lined?

Several cases need to be distinguished. In the first, the side-lining reflectssimply a lack of interest in the domains of phenomena and classes ofpossibilities that such theories might illuminate, and thus the judgment thatthese theories will be insignificant. Per se it has no cognitive repercussions.It may come to have some. Suppose that the side-lining of a type oftheories is done by a group of scientists who adopt S, but that in thescientific community as a whole other groups adopt different strategies, sothat we would expect that the appropriate array of theories would bedeveloped to test the accord with I of T’s eventual acceptance. Then,sidelining would be by a group for the sake of the serious unfolding of S,and it would be accompanied by the active tolerance of groups adoptingdifferent strategies, and leading towards an eventual critical interactionwith the theories produced under the other strategies. If the whole scientificcommunity adopts S, then side-lining amounts to rejection that does notaccord with I.

In the second case, side-lining does have a clear cognitive dimension andis equivalent to rejection. In I, a theory is said to be accepted of aparticular domain, or domains, of phenomena. A theory, accepted underthe materialist strategies in accordance with I, will be accepted of a widerange of such domains. It is reasonable to affirm that the success of scienceconducted under the materialist strategies has amply demonstrated that,concerning spaces with certain kinds of boundary conditions, phenomenacan be well grasped under these strategies, legitimating the rejection oftheories that do not fit the constraints of the strategies. This is a kind ofmeta-induction. It supports following the materialist strategies exclusivelywhen dealing with phenomena in such spaces, and they may be the vastmajority of phenomena that gain the attention of physicists in the normalcourse of events. I do not know how to define precisely “certain kinds ofboundary conditions,”10 but the meta-induction definitely does not extendto spaces in which human agency is a relevant causal factor, for there thesuperiority of materialist to intentional explanations has not beendemonstrated (Chapters 6 and 9).

The third case concerns spaces where human agency is a relevant causalfactor. Of course, one may have no interest in investigating phenomena inthese spaces, so that then side-lining theories that deploy non-materialistcategories amounts to simply opting out of research on them. If one doesinvestigate them, however, there is no cognitive ground to reject non-materialist theories out of hand. The only grounds would be rooted invalues. In Chapter 1, I discussed the long tradition in which science hasbeen identified with that mode of empirical inquiry that follows the


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (243)

materialist strategies. If, in the name of doing “science,” the scientificcommunity attempts to understand the phenomena of agency throughfollowing the materialist strategies exclusively, then de facto values will beamong the grounds for the theories that come to be accepted in thecommunity, specifically those values, commitment to which partly led to theout-of-hand rejection of non-materialist alternatives, and thus to the failureto test one’s theories against such alternatives.” Then, any degree to whichthe cognitive values become manifested in these theories will not bemeasured against the highest available standards—an ironic twist to thetradition that insists on science as value free.

The term “modern science,” I pointed out (Chapter 5), typicallyconnotes both systematic empirical inquiry and inquiry conducted under thematerialist strategies. The idea of science as value free has been articulatedas part of the self-understanding of the tradition of modern science,and both connotations permeated the sources of the idea. And it is truethat, under the materialist strategies, we have had great success in comingto accept, in accordance with I, theories of a wide range of spaces.Nevertheless, for an important range of spaces—where human agency is arelevant causal factor—inquiry, conducted exclusively under the materialiststrategies, cannot produce theories that are accepted in accordance with I.To be accepted in this way the theories of the phenomena in these spacesmust be tested in comparison with theories (posits) developed under otherstrategies, and some posits (for example, “No lost possibilities”:Chapter 8) are not amenable to investigation under the materialiststrategies. Where inquiry is limited to the compass of the materialiststrategies posits like these, that may function in the legitimation ofapplications of theories, function as if a priori, even though they are opento empirical inquiry (Lacey 1997c). I is a value of the practices ofsystematic empirical inquiry. That is why I have stipulated that “science”be identified with “systematic empirical inquiry” and not with “modernscience” with its dual connotations. I push my stipulation becauseimpartiality, an ideal carried by the tradition of “modern science,” can berealized generally only within the widened notion of “science,” whichadmits of systematic empirical inquiry under a variety of strategies, amongwhich the materialist strategies perhaps have pride of place.


As I have emphasized, it is important to keep the roles of the social and thecognitive values separate, and that of the social values confined to itsproper moment. Nevertheless, if social values have a role at all then itfollows that theories, even when accepted in accordance with I, may not beneutral, at least in the sense of “evenhandedness in application,” of beingsignificant for any viable value complex, N′ (3) (Chapter 4). They may not


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (244)

be generally significant because the possibilities encapsulated in a theorymay be of little interest for practices which express values other than thoselinked with the strategies that generated the theory, because any practicalapplication of the theories may require conditions and have consequencesthat undermine the embodiment of the preferred values; as the possibilitiesof increasing crop yields that also involve the seed becoming a commodityhave little interest where prevailing social values emphasize enhancing localwell being, agency and community (Chapter 8).

I introduced provisionally (Chapter 4) the following thesis of neutrality(in abbreviated form):

N′ 1 Accepting T in accordance with I′ implies no value commitments.2 Accepting T neither undermines nor supports holding any one of

the range of viable value complexes; and3 If T is accepted in accordance with I′, it is significant to some

extent for any viable complex.

The discussion of the seed (Chapter 8) provides an example of thebreakdown of N′(3). It shows that theories, accepted in accordance with I′,need not be significant (even in principle) across all viable value complexes.Can N′ be revised to meet this counter-example? The discussion of the seedalso points to the need for a statement of neutrality that applies where I,rather than just I′, is applicable. In addition it provides a hint worthexploring about how neutrality might be revised, when it points out that thesame genetic theory that informs the research on hybrid seeds that are thebasis of the green revolution is also compatible with comparable increasedcrop yields being obtained from appropriately selected pure varieties ofseeds.

Perhaps neutrality should be restricted to “fundamental” theories—andthen affirm that fundamental theories apply equally both for valuecomplexes that highlight the modern values of control and for those incontinuity with local traditions. Fundamental genetic theory, for example,does apply for both, but is it equally significant? It seems clear that it isnot, since it is more readily technologically applied, and the material andsocial conditions required for its development are more readily available(without undermining the value complex adhered to), where the modernvalues of control are adhered to. Significance is a matter of degree, and thismust be built into any defensible version of neutrality. I will do this, but Iresist formulating a thesis of neutrality pertaining to “fundamental”theories since I do not recognize a sharp separation of “fundamental” and“applied” science.

Instead, I suggest substituting in place of N′(3) the following, which doesnot require all theories accepted under the materialist strategies to be


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (245)

significant for all viable value complexes, but only that for any viable valuecomplex there are some accepted theories that are significant for it:

3′ For any viable value complex, there are, in principle, some theoriesaccepted in accord with I′ that are significant to some extent; and(perhaps) the more wide-ranging (“fundamental?”) a theory, themore it is significant to some extent across a greater range of viablevalue complexes.

Replacing (3) by (3′) meets the objection that was raised but, like (3), itdoes not satisfy those (see Chapter 4) who think that N′ does not capture astrong enough sense of “evenhandedness.” It is compatible with vastdifferences in the extent of significance for different value complexes (andeven, for some, when all things are considered an overall negative value).The significance of theories accepted, in accord with I′ or I, under thematerialist strategies is assuredly greater in those value complexes thatcontain the modern values of control (Chapter 6). The mutually reinforcinginteraction between the materialist strategies and the modern values ofcontrol sets a bound to the scope of the “evenhandedness,” or comparablesignificance, of these theories. I remain unable to move towards a strongerreflection of theories having comparable significance without falling intoparadox (Chapter 4). (3′) recognizes that the significance of theories cantranscend the context of their origin and development, and (with itsaddendum) it acknowledges that there is something about properlyconducted “fundamental” science that in general carries a measure ofneutrality with it. That is not a trivial matter. It also reflects properly thatit is difficult to generalize about neutrality. methodological considerationsand judgments made in choosing theories (including that a theory isaccepted in accordance with I) alone cannot settle what is the range of valuecomplexes across which a theory has significance and to what extent.Empirical investigation, involving social factors, is needed case by case.

Modifying N′ so that (3) is replaced by (3′) changes the focus ofneutrality -its key third component now points to a (desired) feature of theresearch practices engaged in under the materialist strategies rather than toa feature of each accepted theory itself. In order to make this explicit, Irevise N′ as follows:12

N″ Scientific practices conducted under the materialist strategiesgenerate theories (accepted in accordance with I) such that:1 accepting these theories implies no value commitments;2 accepting them neither undermines nor supports holding any oneof the range of viable value complexes; and3 ′for any viable value complex, there are (in principle) someaccepted theories that are significant to some extent.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (246)

Given that in all cultures some practices of control are valued (howeversubordinated they may be to other social values, Chapter 6) and thatnumerous phenomena (regarding which relevant human causal influence islacking, Chapter 6) are shared in the daily life and experience of allcultures, we would expect that N″(3′) would be highly manifested—andit is; so also is it that only where the modern values of control are part of avalue complex are theories accepted under the materialist strategiesroutinely significant to a high degree.


Shifting the focus from individual theories to scientific practices is fruitful.Part of the appeal of the initial idea of neutrality is that science can serve,more or less evenhandedly, the interests of all (viable) value complexes.The shift keeps that idea alive. But, if scientific practices are limited tothose conducted under the materialist strategies, the “evenhandedness”expressed in N″ does not amount to comparable significance across viablevalue complexes, for overwhelmingly theories are of greater significancefor those complexes that include the modern values of control. Can weidentify such a sense of “evenhandedness” if we do not limit scientificpractices in this way? And, in doing so, can we also keep alive the hope thatwhere the interests of value complexes clash, aspects of the conflict that areimplicated in posits about the world—about nature and human nature,about what is and is not, about what is and is not possible—can inprinciple be settled decisively (at least in the long run) through investigationthat is systematic and empirical, and aiming to produce theories (posits)that accord with impartiality? We find hints again in the discussion of “nolost possibilities” (Chapter 8).

Recapitulating: in view of its adverse social and ecological effects, thesignificance of the scientific theories that inform the green revolution islargely restricted to the interests of the market and of other values closelylinked with the modern values of control—and does not extend, forexample, to those viable value complexes that highlight local well-being,agency and community. It was objected that these latter value complexesare not viable because their presuppositions contradict the posit: “Outsideof green revolution (and successor) technologies there are no possiblemechanisms to increase crop yields commensurate with the need to meetthe basic food needs of the poor peoples of the world—hence adoptingthese technologies, and implementing the necessary social reorganizations,involves no lost possibilities relevant to addressing the need to providefood.” Clearly, if “no lost possibilities” were a posit accepted inaccordance with I, any value complex whose presuppositions contradictedit would not be viable. But research, conducted entirely under thematerialist strategies, could not establish it as such a posit, for that research


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (247)

does not investigate developments that might be made under differentstrategies, for example, strategies that involve a dialectical interplay oftraditional knowledge and research under the materialist strategies.

It is difficult to engage in empirical investigation of “no lost possibilities”because it necessitates the comparison of results gained from incompatiblestrategies, because relevant alternative strategies are not developed toanything like the extent to which the materialist strategies are, and becausethe conflicting strategies are respectively linked in mutually reinforcingrelations with social values and socio-economic projects with littlecontemporary access to power (Lacey 1997c). Thus, gaining the materialand social conditions for conducting the relevant research may be virtuallyimpossible because power denies their provision. Put simply, the materialand social conditions required for the relevant research are incompatiblewith the interests of the market. Their absence, thus, may reflect not theinherent inability of alternative forms of research to develop but thehegemony of the power of the market and its related forces. Whatever theexplanation for the actual underdevelopment of research under alternativestrategies may be, without it “no lost possibilities” cannot be accepted inaccordance with I; so that where this posit is accepted with virtualcertitude one rightly suspects that it is accepted because it is apresupposition of value complexes containing the modern values ofcontrol, or that it is functioning as a disguise for the “realities” of power.Without it, furthermore, research cannot be expected to generate theories ofmuch general significance for those who adhere to value complexes inwhich the value of control is subordinate.13

Part of the appeal of the initial idea of neutrality is that judgments aboutposits like “no lost possibilities,” which are at the heart of socialcontroversies, could be grounded in the outcomes of systematic empiricalinquiry—and that in this way “evenhandedness” (“comparablesignificance”) would be obtained across the widest range of viable valuecomplexes. It is now clear that research conducted solely under thematerialist strategies cannot adequately reflect “evenhandedness,” and takeus to a thesis of neutrality more robust than N″. Perhaps it can be reflectedmore robustly, while not rejecting what is sound in N″ when scientific(systematic empirical) inquiry is taken (in principle) to include practicesconducted under a variety of strategies.

New statement of neutrality

In the light of these considerations, I offer the following formulation ofneutrality:14


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (248)

N Practices of scientific (systematic empirical) inquiry conducted undera variety of strategies generate theories, that are accepted in accordwith I, such that:

l accepting these theories implies novalue commitments;

2 accepting them neither undermines nor supports holding any one ofthe range of viable value complexes; and

3 in principle, for any value complex that remains viable as the stock oftheories (accepted in accordance with I) expands in the course ofresearch that puts its presuppositions to empirical test:(a) there are some accepted theories, developed under materialiststrategies, that are significant to some extent; and(b) there are some accepted theories, some of which may be developedunder non-materialist strategies, that are highly significant.

N incorporates what is sound in N″, and it does—recognizing that controlhas some place in all value complexes—grant a unique kind of recognitionto the materialist strategies in (3a). It keeps contact with the initial idea ofneutrality, including, for example, by retaining item (1), which it effectivelyshares with N″. But N also goes beyond the initial idea and removes anyparadox that arises from (covertly) interpreting it in a way that treatsparticular value-linked strategies (materialist) as the source of neutrality. Italso introduces a measure of interpretive complexity; neutrality gainshigher manifestation in the context of inquiries conducted under multiplestrategies, enough strategies to enable the presuppositions of currentlyviable and adhered to value complexes to be submitted (in some measure)to empirical investigation. The gain is a coherent idea and a value that isworth aspiring to manifest more fully, even though it may have littlecurrent manifestation and its prospects for fuller short-term manifestationmay be dim in the light of current economic “realities” (Lacey 1997c).

I have maintained that human flourishing, in all of its dimensions andvariations, and for as many people as possible, is my fundamental point ofreference (cf. Kitcher 1993:391; Dupré 1993:244, 264). How adequatelythis value can coexist in the same value complexes as the modern values ofcontrol is a matter open to on-going investigation and controversy aboutthe appropriate forms of empirical inquiry in which to conduct it. There isnot a unique value complex to which adherence is required for living aflourishing life; equally not any fancied and articulated value complex cannourish flourishing, but only those whose presuppositions fit with what ispermitted by nature and human nature. I am trying to explicate an idea thatscientific (systematic empirical) practices be neutral across all (actuallyadhered to) value complexes whose presuppositions fit with what is


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (249)

permitted by nature and human nature. The best criterion we have of thisis that the presuppositions be consistent with what has been accepted inaccordance with I-hence the role of viable value complexes in N.

As I have defined it, “viable” is an historical variable: because the realmof genuine human possibilities changes with historically variable socialstructures, and because further investigation of the presuppositions of acurrently viable value complex may render it non-viable. Thus, if for aparticular (currently) viable value complex there are no accepted theoriesthat are highly significant, this may reflect only that subsequent researchwould render that complex non-viable—though obviously this could not beknown prior to conducting the research It is hardly an undesirablelimitation of the scope of neutrality that it not extend across those valuecomplexes whose current viability derives only from the failure to carry outthe research which, if conducted, would produce posits accepted inaccordance with I that are inconsistent with their presuppositions. On theother hand, it is undesirable that it not extend across value complexes thatare viable and currently adhered to by some groups. However, where thevalues of the complex are not highly manifested because of the play ofpower, the general idea of neutrality resists the identification of the“viable” with what the “realities” of power make the “socially(historically) sustainable” (Chapter 4) or with the dominant tendencies ofthe moment. Neutrality should extend across those value complexes thatare not only viable and currently adhered to by some groups, but also remain(or would remain) viable as the stock of theories (posits) accepted inaccordance with I increases (or would increase) in response to researchaiming to test the presuppositions of the complexes empirically and thenceto assess the consistency of the presuppositions with the expanded stock oftheories accepted in accordance with I—hence the long qualificationintroduced in N(3).

N thus would become more highly manifested if research were to beconducted so as to expand the stock of theories accepted in accordancewith I in a way that submits the presuppositions of all currently adhered tovalue complexes to empirical testing. Clearly any theories (posits) thatbecome accepted in the course of research that tests, and comes to support,the continued viability of a particular value complex are likely to besignificant for the complex. As illustrated with “no lost possibilities,” suchresearch needs to be conducted under a multiplicity of strategies. For N tobecome highly manifested enough strategies would have to be in play toprovide an opportunity for the adherents of all (many) viable valuecomplexes to attempt to consolidate significant theories and for theirpresuppositions to be put to some measure of empirical test. Conducting therelevant research requires material and social conditions, which may onlybe available where specific social values are highly manifested, such as thevalue of nurturing a wide variety of ways of leading flourishing lives that


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (250)

are open to as many as possible. Those conditions may not bereadily available where the modern values of control are well entrenched.Then N would not be endorsed as a value, though N″ might be. Speakingloosely, N itself is not “neutral” (cf. Harding 1998). If scientific practicesare valued in general across value complexes, it is not because N representsa value in all of them.

Certainly N does not represent a “fact” of current scientific practice, asillustrated in the case (and cases of this kind can readily be multiplied) ofthe science that informs the green revolution. Where scientific practices areconfined to those carried out under the materialist strategies, N cannot behighly manifested, for their products are especially significant for valuecomplexes that contain the modern values of control. Given currenttrajectories it would be rash to anticipate the fuller manifestation of N.Indeed, it does not represent a value endorsed in much contemporaryscientific practice, although I think that arguments, for example, thatfeminist perspectives serve not to bring values illicitly into science, but todetect and challenge “bias” already present in scientific practices andproducts, indicate endorsem*nt of something like N, as well as of I(Chapter 9).

Neutrality; the tension between N and N″

Perhaps it is N″ that articulates best the value of neutrality widely endorsedin the tradition of modern science—or perhaps only its first two items?More likely, I think, neutrality is actually taken to be manifestedsufficiently (by its proponents) as long as more than one sociallysustainable value complex (with endless individual variants), for whichscientific theories are highly significant, is in play in the political arena,where each of them contains the modern values of control; and collectivelytheir scope of manifestation continues to expand to include more and morepeoples from “underdeveloped” societies. Neutrality, so understood, isframed by “progress” or “development”, by the modern values of control.The essential link with these values is not taken as a sign of the absence ofneutrality (as such a link with any other values would be) only because it isassumed that the modern values of control have become universal values.With the unfolding of modernity, numerous pre-modern value complexeshave been rendered historically unsustainable. Some have been shown to benon-viable; others have simply succumbed to superior power or the effectsof invasion. It does not violate the claim of “evenhandedness” thatscientific developments have revealed some value complexes to benonviable. It does violate the claim that scientific theories have served asinstruments to inform practices of the powerful that have destroyed theconditions for sustainability of some viable value complexes,especially when—as in “no lost possibilities”—scientific practices provide


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (251)

no investigations to counter ideological claims affirming the non-viability ofa value complex.

N″ is too weak to sustain “evenhandedness” (“comparablesignificance”). One may argue against the value of “evenhandedness,”maintaining that the value of scientific inquiry derives, for example, fromthe links with the modern values of control or from individualist accountsof human nature. I do not rule this out of hand. My point here is that suchlinks with social values should not be disguised under the cloak ofneutrality and that, where the argument draws upon the claim that themodern values of control ceteris paribus lead to enhancements of humanflourishing, it is based on a posit that has not been accepted in accordancewith I (and could not be as long as the conduct of research is confined tothe materialist strategies).15 If one concludes that N does not represent avalue desirably expressed in modern scientific practices and that theneutrality articulated in the tradition of modern science just boils down toN″, then it would follow that modern science is not seriously“evenhanded.” That, however, would carry a sting only among those whoseriously entertain value complexes in tension with the modern values ofcontrol.

While I do not doubt that most members of the scientific communitywould endorse an articulation like N″ rather than N, I suspect that manyof them would be reluctant to give up aspiring for a more robust“evenhandedness” altogether. In the scientific community, as in otherinstitutions (Chapter 2), there may be contestation about the values to beexpressed in its practices with different articulations coming to the fore atdifferent times as unavoidable tensions are played out in different ways. Themodern scientific tradition has developed in mutually reinforcinginteraction with the modern values of control; at the same time it hasinsisted that power (whether exercised with overt force or covertly throughthe normal functioning of dominant social institutions) should not definethe scope of scientific inquiry, but that in principle none of the possibilitiesof nature and human nature lies beyond the scope of scientificinvestigation. There is tension here, intensified when considering whether“science” should be defined by inquiry conducted under the materialiststrategies or also include any systematic empirical inquiry (Chapter 5).Further tensions can be seen when science is linked with “progress”:science serves the fuller manifestation of the modern values of control, butit also stands opposed to dogma and ideology wherever they are to befound, not just in pre-Enlightenment times and societies. In informing theprojects of “progress,” science promises soundly grounded knowledge notonly about means to ends and about what is technologically possible, butalso about the consequences and side-effects of technologicalimplementations and alternatives to them—that is part of“evenhandedness.” Thus, for example, in the case of the seed,


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (252)

“evenhandedness” is ignored when scientific applications further theinterests of agribusiness in a context where “no lost possibilities” (since notgrounded in empirical investigation) functions in an ideological role. Ipropose N as an expression of neutrality aiming to capture this alternativetendency carried by the tradition of modern science.

I endorse N as a value of the practices of science. Endorsing N goes handin hand with endorsing I. Clearly N presupposes I; if theories are notaccepted in accordance with I the question of neutrality does not arise. Inaddition, ceteris paribus the degree of expression of I in scientific practicesis increased as they express N fully. Research expressing N provides therange of theories needed for testing the manifestation of the cognitivevalues in theories against the highest standards, and it ensures that agreater variety of theories. can be tested in such a way that they canbecome accepted (or rejected) in accord with I. The high expression of Nis, thus, a condition for I to be expressed widely throughout the range ofsystematic empirical inquiries. While I endorse N, I am well aware of theenormous practical difficulties confronting its fuller expression, for thatwould require fundamentally reconceiving and reorganizing the institutionsof science. I endorse it because human flourishing, in as many dimensionsand forms, and for as many people as possible, remains my point ofreference. Human flourishing is enhanced when actions can be informed bysound empirical knowledge (wherever possible by posits accepted inaccordance with I), so that it would be well served if there were available avariegated body of knowledge which provided, for each viable valuecomplex, significant posits that were able to inform activity within any oneof them—with the horizon that all adhered to viable value complexesmight become socially sustainable.

Neutrality is not “neutral”

So my endorsem*nt of N rests ultimately on a value, moreover one intension with the modern values of control—N itself is not “neutral”; itdoes not command universal adherence. It articulates a kind of neutralityof scientific practice which cuts across value complexes including thosethat contain the modern values of control, since the inquiries it refers toincorporate those conducted under the materialist strategies, and so theyproduce theories that are significant where the modern values of controlare manifested. But where multiple strategies are in play, the conditionsthat sustain inquiry under them would be likely to curtail the amount ofresearch conducted under the materialist strategies. Such research would beappropriate and sufficient where control is a subordinate value,but generally not where the modern values of control are adhered to. Theinterests of the modern values of control are not generally well (pre-


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (253)

eminently) served by scientific practices that aspire to the fullermanifestation of N.

The character of the neutrality of scientific practice cannot be settled bycognitive and methodological considerations, but ultimately by one’sfundamental values. These values, in turn, are items in value complexeswhich also have presuppositions. A presupposition of the values connectedwith both the grassroots empowerment and feminist value complexes(Chapters 8 and 9) is that a variety of distinctive modes of humanflourishing are possible and that they could be realized without causing thekind of human, social and ecological devastation that has come in the wakeof modernizing “development.” Value complexes that include the modernvalues of control have been articulated with various presuppositions, forexample, that human flourishing is uniquely enhanced within them, andthat there are no possibilities (other than massively destructive ones) for theforeseeable future outside of social forms in which they are manifested(Lacey 1997c).

Clearly none of these presuppositions (or their negations) at present canbe accepted in accordance with I. They are not included in the currentstock of soundly accepted theories and posits; they express anticipations,expectations, hopes, faith and the like. They also are about the possibilitiesof nature and human nature, so that, in principle, empirical investigationcan bear on them in the long run, provided that it is conducted undermultiple strategies. Such investigation, whose practices would aim toexpress N, is appropriate not only if we desire to explore the range of waysin which flourishing lives can be lived, but also if we desire to hold allclaims of the possibilities of nature and human nature—especially thosepertinent to value complexes of contemporary social and politicalsignificance, and to the controversies among them—to consistency with thelong-term outcomes of empirical investigation. Where scientific practicesaim only to express (variants of) N″, such presuppositions will remain inprinciple outside of long-term empirical purview; and the argument for thepresuppositions of the modern values of control will remain derivative toadhering to these values. Ironically, then, I appeal to neutrality in critiqueof the mainstream tendencies of modern science.

I have argued that N is worth endorsing, but also indicated the greatdifficulties confronting its fuller expression. In matters pertaining to values,moving from the outcomes of argument to action may be undermined bythe unavailability of the material and social conditions necessary for theaction to proceed. It is difficult to figure out what to do when one endorsesN, though we find anticipatory pointers in Chapters 8 and 9. Thesituation is different with N″, for there are institutions well entrenched incontemporary society in which research conducted under the materialiststrategies has a well established tradition, and in which the modern valuesof control are highly manifested and embodied. Thus the path to


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (254)

equilibrium through “adjustment” leads readily to engaging in researchunder the materialist strategies; whereas endorsing N arises in the paths of“creative marginality” and “transformation from below” (Chapter 2). Inthis way, the widespread adherence to N″ can be explained, even if—as Ithink—the argument favors adherence to N.

A further difficulty also arises: how to identify clearly what is to count as“science,” “scientific” institutions and practitioners of “science” when thecompass of what is to be included under “science” is expanded. Again, it isnot the use of the word “science” that is in dispute; I want to includeresearch conducted under the materialist strategies as one (very important)instance of systematic empirical inquiry, other instances of which includethat conducted under the grassroots empowerment and feminist strategies(Chapters 8 and 9). The collected body of systematic empirical inquiries(carried out under a variety of strategies) and their theoretical products arenot an integrated, unified whole, and many of these inquiries do not (inactual fact) even interact with one another. Regardless of the strategyunder which a theory (posit) is generated and consolidated, what isimportant (for present purposes) is whether or not it is or (in principle) couldbe acceptable of some domain of phenomena in accordance with I.

Endorsing N leaves it open that particular investigators (and groups ofthem) work exclusively under particular, favored strategies, and do sobecause of the strategies’ mutually reinforcing interactions with theirparticular, adhered to values, provided that they do not (cognitively)delegitimize research under all other strategies; and remain open, inprinciple, to rethinking their position in the light of the outcomes of thealternative research projects that they would tolerate being carried outwithin the scientific community. Endorsing N does not free us fromcontroversy, including that of the most intractable kind, though it maypermit people to enter into controversy entertaining more modest and lessself-interested claims.


So, is science neutral?Is science—either that practicc conducted largely exclusively under the

materialist strategies, or the collected body of systematic empirical inquiries—a highly rated object of value for all value complexes?–No.

Does accepting theories, in accordance with I, rationally commit one toparticular value judgments?—I have not challenged this.17

Can theories, accepted in accordance with I, cognitively reinforce orundermine the values one holds?—Yes.

Are theories, soundly accepted in accordance with I, generally significant for all viable value complexes?—No.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (255)

Is the neutrality of scientific practice well manifested in actual fact?—N″is; N is not; and no version of neutrality that includes a robust version of“evenhandedness” is.

Is neutrality a value to be adhered to in on-going scientific practice?—Ifthe modern values of control are adopted, N″ (but not N insofar as it goeswell beyond N″) tends to be a highly rated value; otherwise, it issubordinated to some other values. If the value of nurturing the many andvarious modes of human flourishing is adopted, N is a highly rated valueand efforts to develop multi-strategy alternatives to the materialiststrategies will be endorsed. Either way, taking neutrality to be a highlyrated value springs not from cognitive and methodological considerationsalone, but also from commitment to a particular social value.


Autonomy was proposed (Chapter 4) as a view about features of scientificpractices, and the processes of their conduct, desired for the sake offurthering the manifestation of impartiality and neutrality. I offeredprovisionally the following formulation (abbreviated version):

A′ 1 Scientific practices aim to gain theories that are accepted inaccordance with I′ and whose acceptance accords with N′.

2 They are conducted without “outside interference” by thescientific community which: a) defines its own problems, etc,; b)has unique authority with respect to matters of method, etc.; c)determines who is admitted into the scientific community, andwhat counts as competence and excellence; d) shapes scientificeducation and scientific institution; e) forms its members in thepractice of the “scientific ethos”; and f) exercises its responsibilityto the public fully by acting in accord with items a)-e).

3 The scientific community conducts its investigations in self-governed institutions which are free from “outside interference”,but provided with sufficient resources in order to conduct itsinvestigations efficiently.

As with I′ and N′, it is presupposed in A′ that scientific inquiry is conductedvirtually exclusively under the materialist strategies; so, minimally, it mustnow be revised in order to fit with I and N. Revising A′ turns out to be anelusive quest, leaving autonomy at best a distant and tension-filled ideal.

Autonomy is subordinate to impartiality and neutrality. Thus, asatisfactory thesis of autonomy will include (leaving open for now what theother items might be):


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (256)

A 1 The practices of scientific (systematic empirical) inquiry aim tobring about in themselves fuller expressions of I and N.............

A(1) presupposes that the fundamental objective of scientific (systematicempirical) inquiry is to gain understanding of phenomena (O) and thisincludes the encapsulation of their possibilities. O, however, cannot bepursued directly but only via following particular approaches (Oi), aredefined by the adoption of particular strategies (Chapter 5), and whichenable the encapsulation of possibilities of particular kinds. Adopting thematerialist strategies represents following one particular approach (O1),one among (potentially) many, the one that enables the encapsulation ofthe material possibilities of things. It is adopted almost exclusively becauseof its mutually reinforcing interactions with the modern values of control(Chapter 6). Then, we see that A′(2) (and (3)) cannot be sustained, for theconfining of research to that conducted under the materialist strategies isintelligible only in terms of the connection with an “outside interference,”the interests of the modern values of control—an “outside interference”already, as it were, built into A′(1).

A′ is not responsive to the need to choose an approach to follow, and soit does not recognize the place where values can play their proper andessential role in scientific practices. “Outside interferences”—such as values—must play a role in the process of scientific inquiry at the (logical)moment when strategies are adopted. A′ can be rendered intelligible byinterpreting “outside interference” as “outside interference other than thatimplicit in the mutually reinforcing interaction between the materialiststrategies and the modern values of control.” Under this interpretation onemight hold that A′ is actually reasonably well expressed in the practices ofmodern science (despite notorious departures from it), especially ifallowance is made for the “compromises” discussed in Chapter 4—somuch so that its widespread endorsem*nt among the members of thescientific community and scientific institutions may be an important causalfactor in the inadequate expressions of I and (especially of) N incontemporary scientific practices. But, no matter how much it may beactually expressed in scientific practices, A′ makes no general claim tobeing adhered to, for it applies only to practices that generally can furtherthe manifestation of N″ but not N. Endorsing A′ (with obvious minormodifications) makes sense, as the expression of a value of scientificpractice, where N″ is endorsed, thus where the modern values of controlare adhered to—but not necessarily among the general public, and certainlynot among those (Chapter 8) who are attempting to develop systematicempirical (scientific) approaches that are not linked with the modern valuesof control.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (257)

Not to endorse A′ is not the same thing as to reject it outright. Myobjection is that it rests on misidentifying O with O1 But O1 does expressthe objective of an important approach to science; so while there is not ageneral interest in its pursuit or in the unqualified endorsing of A′, it isimportant that the results obtained under it accord with I. When O1 issubordinated to the aim expressed in A(1), how can we ensure that itstheoretical products accord with I? How should scientific practices beconducted so that values play their role at the admissible moment, withoutopening the door to any “outside factors” (dogma, ideology, government,corporations, popular opinion) playing any role that might be chosen forthem by those with the power to do so? I propose that O1 should bepursued according to a thesis of localized autonomy: within certainboundsdefined, I suggest, not exclusively by self-governing scientificinstitutions, but in the course of inclusive democratic deliberation so as tobe responsive to N-the items of A′(2) should be endorsed, with virtueslisted in the “scientific ethos” multiplied to include “openness to theresults, accepted in accordance with I, of research conducted under otherstrategies,” “sensitivity to the distinction of levels—of strategy and theory -and to the legitimate role of values being confined to the former level,” and“interest in discerning and submitting to empirical investigation, wherepossible, any posits that are key to important current value disputes.”19 I willnot explore this notion of “localized autonomy” further.

In the context of O, and the recognition of multiple approaches tofurthering it (each linked with particular values), it is difficult to identifyitems that might be added to A(1) in order to state a generally defensiblethesis of autonomy. The interest in the fuller manifestation of I and N,however, remains. How must scientific research be conducted, and in whatkinds of communities, in order to gain more theories (posits) accepted inaccordance with I in the course of inquiries that increasingly express N? Inview of the preceding discussion of neutrality, I suggest, pending theirfurther testing in relevant empirical investigation, that at least the followingconditions need to be satisfied (cf. Longino 1990:76). First, researchis conducted under a range of strategies, which are sufficiently varied topermit that, for each viable value complex currently adhered to, someresearch is conducted under a strategy in mutually reinforcing interactionwith it. Then research is being conducted so that (in principle) if it issuccessful, for each viable value complex, there will be produced significanttheories, and in the course of its conduct, the presuppositions of all thesevalue complexes, if at all possible, will be put to some measure of empiricalinvestigation. In order to further the manifestation of N, values are not keptout of roles in research practices (or their role disguised); ratherappropriate roles are provided for a variety of value complexes.

Second, the distinction of levels (of strategies and theories) is strictlyadhered to, with the role of values limited to that of providing support for


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (258)

the adoption of particular strategies-delineating the general features of thepossibilities of interest to be investigated and framing the context in whichtheories consolidated under the strategies are likely to be significant -andnot permitted to intermix with the cognitive values in making judgments oftheory acceptance and rejection. So that the distinction of levels (strategyand theory) can be respected, the practices of each community, engaged insystematic empirical inquiry, are granted a version of localized autonomy,the details of which still need to be worked out, parallel to that introducedfor communities pursuing O1.

Third, the research is conducted by a variety of interacting communitiescommitted to systematic empirical inquiry. While a particular communitymay (and typically will) conduct research principally of a particularapproach, there is sufficient interaction (and the institutional structures toensure its being sustained)-framed by the objective O-that the theoriesaccepted in accordance with I of each (when relevant) can be broughtcritically into the deliberations of the others. Then the standards of testingof all will be raised, and posits of importance to some value complexes willnot be dropped from consideration prior to the appropriate investigationof them being carried out. Fourth, the items of the “scientific ethos” areexpanded to include those listed on p. 250 in connection with thediscussion of the localized autonomy of research under the materialiststrategies.

Fifth, the matter of who (and the qualifications of whom) is regarded asa member of a “research” community admits of considerable flexibility,and is open to ongoing contestation. It may lack ready definition (in termsof, for example, qualifications), especially (Chapter 8) in view of therebeing approaches attempting to develop a dialectic between traditionalknowledge practices and research conducted under the materialiststrategies. Then, a “research” community may not be separable from acommunity engaged in a particular set of practices (forexample, agricultural)—systematic empirical inquiry may be a part of amore general, on-going practice.

Finally, in part elaborating some of the earlier conditions, thecommunity of communities engaged in systematic empirical inquiry needsto manifest “diffusion of power” (Longino: see Chapter 9). Diffusion ofpower, in principle, functions as a counter to the impact of values onscientific production present because science is produced mainly ininstitutions, such as universities and research institutes, which are anintegral part of the dominant socio-economic system. These institutionsoften reinforce certain values, perhaps misidentifying them with items ofthe “scientific ethos,” such as the primacy of the intellect, individualism,competivism, and virtually exclusive commitment to one’s scientificdiscipline. They require of the scientist an intensive training, and thencertain priorities of time and commitment. In these institutions, being a


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (259)

scientist is a “full-time” commitment, and family, social and otherinvolvements must adapt to it; otherwise one will fall behind in thescientific competition. It involves a way of life that makes it difficult tocoexist with lives that express certain other values, for example, thoseexpressed through participation in programs for social transformation(except those of modernizing: Chapter 8). Moreover, it is a way of life thattends to be granted a sort of “epistemic privilege” (cf. Longino 1993), aprivileged place for making authoritative judgments about what is andwhat is not possible. So widely is it accepted that modern scientific theories,developed under the materialist strategies, provide the best encapsulationsof the possible; a privileged place which generates asymmetric relations,which are conducive to relations of domination between those whosepractices are informed by the scientists’ knowledge, and those who areexpected to adapt their lives to conform to the imperatives of scientificknowledge. If one does not adopt such values, one will not be able toparticipate in the experimental and theoretical practices of modern science,and so one will not gain the experiences necessary to be an “expert” in thedomain of scientific knowledge. But, if one does adopt them, one will mostlikely not be able to recognize the practices from which alternative formsof social transformation might develop. Adopting the values, then, willenable one to gain access to more and more material possibilities thatmight be realized through technological practices, but it may leave one invirtual ignorance of novel social possibilities (and the material possibilitiesthat they might introduce) that actually existing reality may permit.

In the light of the dialectical connections that I have referred to, I wish tosuggest—as an item for further discussion—that there is no way of life andthere are no institutions, which enable one to explore both the fullest rangeof modern technological possibilities and the possibilities of socialtransformation that would adequately serve the poor majorities of theimpoverished countries and the interest of expanding the capabilities ofhuman agency in general.

Adopting the social value of diffusion of power, therefore, has far-reaching implications. I have defended it in terms of its potential to servecognitive ends: to produce higher standards for estimating the degree ofmanifestation of the cognitive values, and to gain access to possibilities thatcannot be encapsulated under the materialist strategies (Chapters 8 and 9).As a social value it can also be defended in terms of its contribution tosocial justice. I think it important not to dissociate the two defenses. Whenthey become dissociated and the one from social justice prioritized,furthering diffusion of power can also lead to a lowering of cognitivestandards and the failure to gain the kind of knowledge that is needed toinform the projects of social transformation (Chapters 4 and 9).

I am unable to bring these conditions together into a sharp thesis thatmight replace A′, that could be endorsed on similar terms to N. There are


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (260)

several reasons for this. In the first place, the set of approaches that mightbe adopted under the general umbrella of O is too ill-defined (thoughcertain of its members can be readily identified: Chapters 8 and 9), as alsois the set of research communities to be considered included in the body ofinteracting research communities. Second, the value of research meetingconditions like these is contested by those who adhere to the modernvalues of control who aspire only to manifest N″ and not N so that, forlack of social and material conditions and the presence sometimes ofoutright opposition backed by power, numerous potential approaches toscience are not entertained or sometimes even conceived. This renders iteven more difficult to hope to identify concretely a body of communitieswhose research, collectively and interactively, might be sufficient to ensurethe fuller expression of N. Third, endorsing N and seeking ways to enhanceits expression, are likely to lead one to challenge the practices of any actualcommunity (of communities) engaged in scientific research for lackingsignificance for certain specified value complexes; and one may need toback this challenge with concerted political action to bring thecommunity’s practices more into accord with N.

In such a situation, autonomy—other than pragmatically anddemocratically worked out “localized autonomy”—has little meaningful,short-term relevance. We can only conjecture about what a community ofcommunities of research would be like for which we might desire toendorse a form of autonomy—that is, a community sufficiently specifiedand institutionalized, whose on-going conduct sufficiently reflected the sixconditions listed on pp. 251–2. For the long run, however, suchconjecturing is not idle for, given commitment to the multiplicity andvariability of forms of human flourishing, N is worthy of much fullerexpression. If eventually there are created institutions that serve effectivelyto further it, then a version of autonomy that is consistent with the sixconditions may be able to be stated and defended. But that remainsconjecture. I add the further conjecture that it would further the traditionalgoals of the university if it committed itself to the furtherance of N.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (261)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (262)


Is science value free? Do the practices of scientific inquiry express thevalues of impartiality, neutrality, and autonomy? Impartiality: that atheory is accepted (of a domain) if, and only if, in relation to theappropriate empirical data, it manifests the cognitive values to a highdegree according to the most rigorous available standards. Neutrality: that,for any viable value complex, there are theories (accepted in accordancewith impartiality) which may be applied so as to further significantly themanifestation of the values that constitute it. Autonomy: that scientificpractices proceed, and scientific communities and the institutions thatsupport them are constituted, for the sake of furthering manifestations ofimpartiality and neutrality without direction (“interference”) from factorsother than the data, the cognitive values and the ingenuity of thepractitioners, in particular without direction from (non-cognitive) values.

Impartiality and neutrality have presuppositions: impartiality thatcognitive values are distinct and distinguishable from other kinds of values;neutrality that accepting a theory is logically consistent with making anyvalue judgments, and that accepted theories leave open a range of viablevalue complexes. I have not questioned these presuppositions. I havedefended endorsing the three values against common criticisms that theyare not always actually well manifested: not all theories actually acceptedin the scientific community accord with impartiality; most actuallyaccepted theories (contrary to the aspiration of neutrality) serve to furtherthe manifestation of some value complexes (those containing the modernvalues of control) much more than that of others; and the claim ofautonomy must contend with the numerous compromises into which thescientific community enters with the values of those institutions whichprovide for its material and social conditions. For even when the threevalues are not manifested highly in actual fact, they may still be expressedin the practices of scientific inquiry. Then, the key question becomes: Arethe practices of scientific inquiry so constituted that (whenproperly conducted) there is progressive movement in the direction of thehigher manifestation of impartiality, neutrality, and autonomy? If so (forall three values), then “science is value free” would be well grounded.

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (263)

The fundamental objective of scientific inquiry is to gain understandingof phenomena: to chart and explain their workings and features, toencapsulate the possibilities (including novel ones) that they allow, and todiscover how some of the hitherto unrealized ones may become realized.The cognitive values designate those characteristics of theories (and theirrelationships with empirical data and other theories) which indicate thatthey serve to fulfil this fundamental objective with respect to some domainsof phenomena. The objective of gaining understanding, however, providesby itself no direction to scientific inquiry. In order to pursue it, it isnecessary to follow a particular approach to inquiry, where an approachhas the objective to investigate possibilities of a specified kind. Following aparticular approach involves adopting strategies which constrain the kindsof theories entertained and select the kinds of empirical data which thosetheories should fit. A strategy gives direction to research; and the adoptionof a strategy gains support, from which the grounds to adopt it cannot beseparated, from its mutually reinforcing interactions with (moral andsocial) values. Sometimes a strategy will be adopted deliberately andconsciously because of its relations with values; though, more typically,certain strategies frame practices of inquiry that are solidly entrenched (sothat adopting them seems not to require rational justification) because thevalues with which they bear reinforcing relations are deeply manifested andembodied in prevailing social institutions—as the materialist strategies areentrenched in view of their mutually reinforcing relations with the modernvalues of control.

Since engaging in scientific inquiry (which I construe as any form ofsystematic empirical inquiry) requires adopting a strategy, and adopting aparticular strategy becomes intelligible only in view of its mutuallyreinforcing relationships with values, autonomy cannot be sustained.Values pervade, and must pervade, scientific practices and (in significantpart) account for the direction of inquiry and for the kinds of possibilitiesattempted to be encapsulated in theories.

The link with values becomes further apparent when it is observed thattheories that become accepted in the course of inquiry conducted undergiven strategies generally are applicable in ways that serve to furtherespecially the values that have reinforcing links with the strategies.Theories accepted under the materialist strategies, for example, onapplication serve especially well interests framed by the modern values ofcontrol. Since the materialist strategies are adopted virtually exclusively inmodern scientific practices, this indicates that these practices do notactually serve the fuller manifestation of neutrality well. Does this meanthat neutrality cannot be sustained? I have suggested that it does not, thatneutrality can be sustained as a value of scientific practices provided thatthere are in play within scientific communities (the body of communities inwhich systematic empirical inquiry is conducted) a multiplicity of strategies,


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (264)

among which the materialist strategies may have an esteemed place—as inthe formulation, N (Chapter 10). There is good reason, I have alsosuggested, to endorse N; but then the ground of neutrality is not in amethodology based in strategies which deploy a lexicon that is devoid ofvalue terms, but in the value of the social sustainability of a variety of formsof human flourishing (not limited to those that are implicated in themodern values of control). On the other hand, N can become manifestedmore fully only if there is significant change in the conduct of scientificpractices (away from the virtually exclusive adoption of the materialiststrategies) and the character of their supporting institutions, and—in lightof the mutually supporting relationships of strategies with social values—this can only happen if there is change in the values that are predominantlymanifested and embodied in social institutions.

N would be endorsed, I suspect, only among those who adopt strategiesincompatible with the exclusive role of the materialist strategies, and whoquestion the modern values of control on the ground that theirmanifestations and embodiments threaten the social sustainability of valuedforms of human flourishing. It is not endorsed in the mainstream ofmodern scientific inquiry. Moreover, practices in which the materialiststrategies are adopted almost exclusively do not and cannot come toexpress N more fully. They are linked too closely with “progress,” withfurthering the modern values of control. But they do retain a residue ofneutrality; accepted theories may be applied to further significantly theinterests of all value complexes that contain the modern values of control.For those who think that the trajectory of the future has been set and that,in fact, there are no significant realizable possibilities for the future outsideof value complexes that contain the modern values of control, this residueis enough of neutrality; for them N would represent either idle rhetoric orillusory aspiration. Whether one endorses N or the residue of neutrality,then, depends on the values one adheres to and on one’s anticipations offuture possibilities; the matter depends not on issues of scientificmethodology, narrowly conceived, but on substantive value controversyand disagreements open to investigation in the social sciences.

What of impartiality? That values are implicated in the adoption of thestrategies under which theories are generated, developed and consolidatedpermits (in principle) that theories can be accepted (of various domains ofphenomena) solely in the light of the empirical data, otheraccepted theories and the play of the cognitive values. Numerous theories,developed under the materialist strategies, have been accepted inaccordance with impartiality—though actual departures from it are notuncommon, especially when certain kinds of theories are rejected, notbecause of their cognitive failures, but because they do not fit thematerialist strategies. The practices of scientific (systematic empirical)inquiry are constituted so as to permit fuller manifestations of impartiality


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (265)

in the making of theory choices. Thus, impartiality remains a defensibleand, I believe, obligatory value of scientific practices. It is a value rooted inthe very objective of gaining understanding, a requirement of being able toseparate the genuinely possible from the merely conceived or desired to bepossible.

In various ways understanding informs our activities and socialpractices; it informs means to ends, the possibility of desired ends, and thepresuppositions that lie behind the legitimation of our projects. The valueof the social sustainability of a wide variety of forms of human flourishingsupports that the fuller manifestation of impartiality is a value. Impartialityis manifested more fully when (1) increasingly the theories that are actuallyaccepted in the course of scientific inquiry accord with it, and (2) the rangeof phenomena of which theories are accepted in accordance with itcontinues to expand—moving towards the ideal in which our activities(both their concrete conduct and their legitimation) are informed whereverpossible by understanding that accords with it. Conducting inquiry almostexclusively under the materialist strategies has produced a remarkablenumber of theories accepted of numerous and various domains ofphenomena (the range of which continues to expand). It does not permit,however, moving towards the ideal, for it lacks the means to investigatewhole classes of possibilities, those available to phenomena when we do notabstract them from their human and social contexts, for example, toinvestigate the question of whether there are significant realizablepossibilities for the future outside of those that serve the interests of valuecomplexes containing the modern values of control. Manifestations ofimpartiality can only move towards this ideal when inquiry is conductedunder a variety of strategies. Only when diverse values lead to the adoptionof a variety of competing strategies—rather than when it is attempted “tokeep values out of science,” but when in fact the attempts become adisguise for science being conducted under one kind of strategy and thusunder the influence of one set (no matter how “universally endorsed”) ofparticular values—are the conditions in place for the kind of manifestationof impartiality that could approach the ideal. When, and only when, theseconditions are in place, can we also expect that fuller manifestations ofimpartiality will be accompanied by fuller manifestations of neutrality (N).

Is science value free? It is not, since autonomy cannot be sustained. But itis not illuminating to sum this up with the slogan: “Science is not valuefree.” I have been more concerned to articulate the positive ways in whichvalues may (legitimately) play a role in science, and with how these arecompatible at least with impartiality, a core component of “science is valuefree”; and to make clear that their legitimate role actually requirescommitment to impartiality, for the adoption of a strategy—though itsrationale comes from relations with values—is not justified in the long run


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (266)

unless it leads to the development of theories that become accepted inaccordance with impartiality.

Values, I have said, pervade and must pervade the practices of scientific(systematic empirical) inquiry, the practices whose objective is gainingunderstanding of phenomena. That is because scientific inquiry isconducted under strategies, the adoption of which is explicable in terms oftheir mutually reinforcing interaction with particular values. Values have aproper role to play at the moment of the adoption of strategies. But theirlegitimate role does not extend to playing alongside the cognitive values inmaking theory choices.

The separation of moments—adopting strategies, accepting theoriesisessential, with values playing their role only at the moment of adoptingstrategies. It permits the retention of impartiality; and makes it possible toanticipate that conducting research under a variety of strategies might leadto fuller manifestations of both impartiality and neutrality (N). It alsohelps to explain why inquiry conducted almost exclusively under thematerialist strategies lacks neutrality (N) and manifests at most its residue.It also allows us to identify mechanisms (when values, consciously or not,come to function alongside the cognitive values) that explain howimpartiality can actually be departed from in scientific practice. Theseparation of moments underlies a synthesis of the genuine achievements ofscientific inquiry conducted under the materialist strategies, and what issound in the many contemporary forms of criticisms of science; withouthaving to limit scientific practices to those of the former (with its oftenimplicit commitment to materialist metaphysics) or to embrace the radicalrelativism often associated with the latter

The account enables us to pick out what is and what is not defensible in“science is value free”; (partly) to explain why neutrality (N), althoughdefensible, is not widely endorsed, why impartiality is not always wellmanifested in actual fact, and why autonomy seems defensible to thosewho do not accept that there are questions about choice of strategy tobe addressed; and to characterize in broad terms how inquiry must beconducted if there is to be fuller manifestation of neutrality and of (anideal of) impartiality. It also provides a setting in which value-basedcriticisms of current scientific practice can be made. The bulk of modernscientific inquiry is conducted under the materialist strategies, and theseinteract in mutually reinforcing ways with the modern values of control.How would the practices of scientific inquiry have to be re-constituted sothat its strategies come to interact in similar reinforcing ways with thevalue of the social sustainability of the widest possible variety of forms ofhuman flourishing? The discussion of the “grassroots empowerment” and“feminist” strategies provides some initial hints. With the separation ofmoments in my account, this becomes an intelligible question. It is also oneof the urgent moral questions of our day.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (267)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (268)


1Introduction: the idea that science is value free

1 As commonly held among logical empiricists (Feigl 1961:15) and criticalrationalists (Popper 1972:29).

2 Descartes and Newton played the key roles in consolidating the fusion of thetwo ideas in the tradition of modern science.

3 Implementing this proposal involves presuppositions about authoritativetestimony (Coady 1992) and the availability of appropriate channels ofcommunication. Deepening it requires consideration of “Rudner’s argument”

4 (Chapter 4). See Cupani (1998) for a defense, and Hull (1998) for achallenge.

5 Arguments of this kind are often part of “feminist empiricism” (Anderson1995a; Nelson l996).

6 My usage differs from that of, for example, van Fraassen (1980).7 Almost all of them have been elaborated by Weber (1949). He also maintains

that commitment to “science is value free” does not imply moralindifference, but is a presupposition of sound moral thinking and action.

8 Roughly, the fusion of the Galilean and Baconian ideas under discussionsrests upon yet another Baconian idea, the valuing of “utility.”

9 “Strategy” serves many functions similar to Kuhn’s (1970) “paradigm,”“disciplinary matrix” and “structured lexicon”—and, to some extent,Lakatos’ (1970) “research programme.” I do not use Kuhn’s notion, becauseof its changes over the years, to avoid entering into questions of Kuhnianinterpretation, to keep a distance from some of the connotations associatedwith his terminology, and so that I assume full responsibility for theimplications of my use of the notion. Nevertheless, my indebtedness to Kuhnis great; “strategy” is certainly an intellectual descendent of “paradigm.”Note also that there are other dimensions of strategies (and paradigms) that Iwill not develop in this book, e.g., concerning the provision of guidance forthe deployment of the auxiliary hypotheses that mediate between theory anddata.

10 Holist epistemologies, that do not readily permit differentiation of levels,often underlie criticisms of “science is value free.”

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (269)


1 This chapter is a shortened version, with some changes, of parts of an articleI co-authored with Barry Schwartz (Lacey and Schwartz 1996). In it weanalyzed how values can be considered simultaneously to be objects ofcognitive and moral evaluation, and objects of social and psychologicalexplanation. Further details on many points raised in the text can be found inthe article.

2 On the plurality of values and the variety of appropriate relations withobjects of value, see Anderson (1993).

3 In this section, I draw freely upon Nerlich (1989) and Taylor (1985).4 Values discourse becomes deformed when “values” become separated from

their other modes, especially when holding values becomes identified solelywith their articulation (Lacey and Schwartz 1996).

5 We also say that “X values v,” where v is “an object of value” for X. Usually“X values v” fills in for an expression like “X values that ø be characterizedby a certain type of relationship with v.” Each object of value is associatedwith specific relationships (Anderson 1993).

6 Graham Nerlich (correspondence) has strongly questioned my tying ofevaluation to future orientation.

7 Where one discerns, in interpretative inquiry, that a value is manifested fairlyconstantly in a person’s behavior, but the person displays no reflectivenessabout its desire and belief dimensions, and displays no particular commitmentto narrow the gap, I will refer to the person as merely having, rather thanholding the value.

8 “Holds” as distinct from simply “has” a value (see previous note).9 McMullin (1983) and Nagel (1961) refer to all three kinds of these

judgments as “value judgments.” McMullin calls the first two kinds ofjudgments “valuing,” and the third “evaluating.” Nagel (1961) refers to themrespectively as judgments of “appraising” and “characterizing.”

3Cognitive values

1 Some of the beliefs that inform one’s actions may remain unevaluated and notbecome objects of deliberation (or even recognition) when they derivecausally from: mechanisms for the inculcation of ideology, the absence ofsocial institutions in which alternative views are articulated and criticallydiscussed, prejudice, habit, brainwashing, indoctrination, desire to conform,expediency, opportunism, unreliable authorities, threat, ignorance ofalternatives, exigencies and limitations of experience, unsound argumentativemoves, cognitive errors, behaviorist reinforcement, projections of classinterests or personal desires, and simple unreflectiveness or other personalvices.

2 We do not always have means available to resolve the contradiction. There isa manner of speaking in which people say “p is true for me, but apparently


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (270)

not for you.” I take this to mean something like: “p is sufficiently wellsupported to warrant informing activities connected with my attempts tomanifest more fully my most fundamental values; but apparently not so withyours.” It may be that most of the beliefs that inform our activity are of thistype—then we may be mired irremediably, in “de facto relativism.” Even ifthis is our predicament, it does not provide a reason not to be responsive toargument and evidence.

3 At times the interpretative activities of identifying the cognitive valuesmanifest in beliefs and the values manifest in behavior will be intertwined.When considering certain types of beliefs, their having a specified causalorigin—for example, origin in appropriate experiences—can be a highly ratedcognitive value. This point is heightened when we take note that most of ourbeliefs rest upon the testimony of others (Coady 1992), so that there will becognitive values of the type “testified to by a reliable authority.”Furthermore, reliable observational reports, especially of certain socialphenomena, may require observers who manifest particular values, since thephenomena may be opaque to those who lack a certain kind of moralsensibility (Lacey 1997c).

4 Or, at least, for each kind of object of investigation, there should be a correctset of cognitive values to frame the investigation. The possibility that, tosome extent, cognitive values may be “relative to” the object of investigationwill not be explored in this book.

5 The controversy over creationism partly concerns the list of cognitive valuesto be embodied in public educational institutions.

6 McMullin (1983) points out that, ironically, the attempts of the positivist-influenced traditions to reduce cognitive value to the outcomes of rule-governed operations all covertly traded on subtle cognitive value judgments.The application of inductive rules presupposed “judgments”—which are notrule-governed—about “curve-fitting, extrapolation and estimates ofrelevance”; and in order to characterize “basic statements,” Popper (1959)appealed to conventions whose deployment involved not rules, but such(cognitive) value judgments as “easy to test” and “likely to gain agreement.”Finally, Carnap (1956) based the choice of theoretical language(sensationalist, physicalist, etc.) on such values (cognitive, pragmatic?) as“efficiency, fruitfulness and simplicity.”

7 The way of looking at things can be traced back to Kuhn (1977) withanticipations in the “Postscript” to Kuhn (1970); it has been most richlydeveloped by McMullin (1983, 1993, 1996): see Lacey (1997b).

8 Sometimes, cognitive values are called “epistemic values” or “epistemicvirtues” (McMullin 1983, 1996), or “epistemic utilities” (Hempel 1981). I amcontrasting cognitive values with relations (between theories and data)which, like proof in mathematics, obtain only if the relata are connected in away that can be established by a set of formal rules. In another sense of“rule” (for example, Eldridge 1997), rules are followed whenever one acceptsa theory, for example, the rule: accept only theories that manifest thecognitive values highly; and scientific practices are structured by various“rules”: orderly, somewhat codified, and normative constraints.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (271)

9 The list simply draws together items that have been proposed in the literature:especially Kuhn (1977) and McMullin (1983; 1993; 1996); also Ellis (1990),Laudan (1984), Longino (1990), MacIntyre (1977), Newton-Smith (1981)and Putnam (1981; 1987; 1990).

10 Van Fraassen (1980) defines empirical adequacy, in the context of thesemantic model analysis of theories, as: T has sub-models that are isomorphicwith sets of observable phenomena of relevant domains. These sets cannot beidentified with sets of items of E, since E consists of actual data. However, itseems to me, we have reason to believe that T is empirically adequate in VanFraassen’s sense if, and only if, (l)-(4) are satisfied.

11 For queries about the last clause see Brusch (1989). 12 Or non-triviality: not all statements well formed with the categories of the

theory are entailed within the theory (da Costa and Bueno 1998).13 Salmon (1966) regards “analogy” and some other items cited here as factors

that are relevant to the assessment of “prior probabilities” for purposes ofBayesian calculations.

14 See Lakatos (1970) on various senses of “ad hoc.”15 I am not aware of the issue of standards having been directly addressed in the

literature on cognitive values. I offer here a preliminary discussion that needselaboration and critical interaction.

16 In Lacey (1979) I also argue that the data considered by radical behaviorismdo not meet the representativeness standard.

17 For example, empirical investigation leads researchers in medical science toregard double-blind testing as more reliable than other methods (Anderson1995a; Laudan 1984).

18 This standard functions in close concert with relevant to the criticalcanfrontation of competing theories (above), reflecting Feyerabend’s (1965,1975) argument that theoretical conflict can be an instrument for increasingempirical adequacy. In practice, it can be very difficult to interpret because ofthe well known thesis of the underdetermination of theories by data. Notealso that a certain explanatory ideal (e.g., mechanism, holism, organicism,intentionality) may function as a standard, though often it may be difficult toseparate its role from that of strategies (introduced in Chapter 4; see alsoChapter 5).

4Science as value free: provisional theses

1 The notions of “accepting,” “applying” and “significance of” a theory wereintroduced in Chapter 1, as was the terminological convention to distinguish“cognitive values” from (other kinds of) “values.”

2 Elsewhere (Lacey 1997b, 1999a, 1999b) I have called them “materialistconstraint/selection strategies” in order to use a terminology which alwaysputs the two roles of strategies at the center of attention.

3 The Galilean idea (Chapter 1) is metaphysical; adopting the materialiststrategies follows from accepting this idea (materialist metaphysics). Myterminology, “materialist strategies” derives from this (see McMullin 1999


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (272)

for an objection). Without denying the historical importance of themetaphysical idea, adopting the materialist strategies can be de-linked fromit. In this chapter, I treat the materialist strategies as a commonplace ofscientific practice. Then (Chapters 5–7) I argue that the ground of theiralmost exclusive use lies in links with values, rather than metaphysics.

4 For further clarification see Lacey (1999a). Some fields of science do notdisplay the same kind of commitment to use of the materialist strategies. Wemay discern a spectrum of cases, ranging from the thoroughly materialistic tothe “historical” (evolution, geology), to the naturalistic (natural history,taxonomy), to ecology, to the human and social sciences—so that in what isgenerally identified as “modern science,” a range of strategies are in play.

5 Some comments: First, if one were to accept a rule-governed account of thecriteria of cognitive value, item (1) of I′ could be dropped, and (2) would bereplaced by something like: “T is accepted of D if, and only if, the degree ofits cognitive value, as calculated through application of the rules, issufficiently high….” Second, agreeement in the scientific community does notsuffice for the acceptance of T, according to I′. But, given that the standardsinclude having adequately responded to criticisms, a significant amount ofa*greement is a necessary condition for sound acceptance. Third, values otherthan cognitive values, may be manifested too in T, alongside the cognitivevalues, but this is irrelevant to soundly accepting T. Insisting that T manifestthe cognitive values according to the highest available standards is a meansfor ensuring that such values do not play a (covert) role in T becomingaccepted.

6 (R) is clearly a condition upon the legitimacy of any practical application ofT. Many commentators have held that it is only that.

7 Questioning, of course, often does come to an end, especially in the naturalsciences where it is uncontroversial that (R) holds widely.

8 See the discussion of “no lost possibilities” (Chapter 8; Lacey 1997c).9 The notion of scientific knowledge as a public good is under strong challenge

at the present time, especially in the light of the intertwining of biotechnologyresearch with business interests that has been reinforced by extensions ofintellectual property rights to cover, e.g., genes. Neutrality may well berapidly disappearing as part of the self-understanding of the scientificcommunity. So, too, may the link between the materialist strategies and themodern values of control (Chapter 6)—leading to scientific researchstrategies progressively becoming linked piecemeal with a variety ofcompeting economic interests.

10 There is another view (which I will not discuss in this book)—call it “novalue implicatures”—that might be included here: Accepting a theory inaccordance with I′ has no implicatures (“pragmatic implications”—Root1993, drawing upon Grice) in the realm of values.

11 Since there may be theories accepted of domains with only the most remoteconnection with daily life and experience and practical activities, to thestatement of “evenhandedness” should be added the qualification; “…theories, accepted in accordance with I′ if they are applicable at all….” I willassume this qualification to be in place here and in all subsequent revisions of“evenhandedness”.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (273)

12 This argument is elaborated and clarified in Lacey (1997c).13 “Consistent with all value judgments” is presupposed in the formulation of

the other components of neutrality It should be understood as containing aqualification. Earlier (Chapter 3), I suggested that it accords with an informalideal of rationality that if one holds the belief that p then one negativelyvalues any of one’s actions that might be informed by ¬p; they are irrational.I interpret N′(l) (and all its successors) to be qualified by “with the exception:X’s acceptance of T (of D) commits X rationally to valuing negatively anyapplication X might make of negations of T (of D).” This qualification doesnot threaten “evenhandedness,” for minimal rational commitments of thistype are endorsed in all value complexes.

14 Another set of questions arises here too, for the issues of autonomy andresponsibility of scientists are intimately connected with how scientistsexercise their authority. Under what conditions “should” scientific claims beheld by all, including non-scientists, on pain of the charge of “irrationality?”How should science interact with popular beliefs, outlooks and worldviews?How does science illuminate the realm of daily life and experience? Why, andwithin what bounds, should non-scientists accept the authority of scientists?Presumably, scientists, qua scientists, have at most cognitive and not moralauthority. What happens when there is tension between them? In what waysdoes the authority of scientists, in actual practice, contribute to theentrenchment of ideology? It is not uncommon for scientists, claiming to bespeaking authoritatively, to affirm that science either presupposes or supportsa metaphysical position, e.g., materialist metaphysics. In doing so, the claimgoes beyond the evidence. Some forms of “anti-science” critique hold: that isthe way science is; it always makes claims implicated in values!

15 Scientific institutions can undermine their own claim to autonomy in anumber of other ways: for example, if they cease to be sources ofauthoritative testimony regarding what has been accepted in accordance withimpartiality and neutrality—by tolerating or not investigating charges offraudulent results, by only revealing results that serve special interests, or byusing their authority so as to portray ideology as scientific knowledge. Wheresuch things happen it may become appropriate to extend the areas ofcompromise to include also items (b) and (d) of A′(2). Finally, where theinstitutions in which scientific research is actually conducted turn out not tobe genuinely self-governed (think of the research units of the tobaccoindustry), they can make no claim to be granted autonomy—freedom frompublic interference—on grounds connected with impartiality and neutrality.

5Scientific understanding

1 Formal consistency with theories soundly accepted under the materialiststrategies (or any other strategies) is a cognitive value.

2 See, for example, Laudan’s (1984) analysis of the gradual rejection of the(supposed) cognitive value articulated in Newton’s “Hypotheses non fingo.”


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (274)

3 Shiva (1988) and Anderson (1995a) make the same point without theparadoxical rhetoric.

4 Theories accepted under the materialist strategies are unrivalled in havingdemonstrated high manifestations of the cognitive values. It does not followrationally that we should always use soundly accepted theories of this kind toinform our actions (Chapter 3). If one aims for such objectives as thestrengthening of one’s cultural heritage, the empowerment of oppressedpeoples, the overcoming of poverty, the enriching of relations among people,the strengthening of democracy, and the creation of ways of life that do notdevastate the environment, then modern science (conducted under thematerialist strategies) has (can have) little to contribute, except in asubordinate way regarding some of the technical projects that will beinvolved in acting for those objectives (Chapter 8).

5 Laudan (1984) has proposed that the identification and adoption of cognitivevalues can be rationally reconstructed in terms of a “reticulated model,”which (paraphrased into my terminology) involves two-way interactionsbetween each pair of the triad: {theories, scientific practices, cognitivevalues}. For example, cognitive values “justify” scientific practices, andscientific practices “exhibit the realizability” of the cognitive values; andtheories (chosen in the course of scientific practices) and cognitive values“must harmonize.” My account, in requiring that cognitive values be criteriaof theory choice actually used in scientific practices, and that they be possiblymanifested, incorporates Laudan’s considerations. His account, however,does not distinguish between strategies and cognitive values, and does notinvolve significantly my fourth kind of consideration. He seems to hold thatthe objective of science is simply to gain theories that manifest highly thecognitive values that are currently adopted in scientific practice.

6 The argument deployed here has the same structure as that involved inReichenbach’s “pragmatic vindication of induction” (Salmon 1966); it is apriori. On the other hand, for example, the claim: “if we are to obtainempirically adequate theories in medical research meeting the standard ofreliability, we must use double-blind testing procedures,” is an empirical one.In general, procedures that are followed in order to bring about the highmanifestation of cognitive values in theories need empirical vindication(Laudan 1984).

7 But see Chapter 4, note 4.8 Made by John Clendinnen (in discussion). See also Chapter 7, note 18.9 See Lewontin (1993) on the irreducible interaction between the biological

and environmental causes of diseases.10 And of the conditions provided for the research (Rouse 1987).11 O′ is adapted from Ellis’ statement of the realist’s aim of science (borrowed

from van Fraassen 1980), and O′ from his own version of the aim (Ellis 1985).(Recall—Preface—I will not discuss empiricist versions of the objective ofscience.) Hempel (1983a) proposes that the objective of science may be putas: “seeking to formulate an increasingly comprehensive, systematicallyorganized worldview that is explanatory and predictive.” Acknowledging thatthis is very vague, he goes on to suggest that adopting a suitable set ofcognitive values may be regarded as “attempts to articulate this concept


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (275)

somewhat more fully and explicitly.” Then, the objective of science becomesto gain products that more fully manifest the cognitive values—so that theydefine the objective of science rather than gain justification from it (cf. note5).

12 My account of multiple approaches has much in common with Tiles’ (1985,1987) view that scientific research unfolds in the context of agreementsabout empirical classificatory schemes and metaphysically derived constraints(which carry with them an explanatory ideal) on theories, and with Dupré’s(1993) account of the disunity of science.

13 Another argument for adopting the materialist strategies exclusively goes:Physical science developed when it rejected strategies which deployteleological categories, then biological science did the same, so we wouldexpect that the same would happen in the human and social sciences.Another is sometimes used among those who adopt Bayesian approaches:The success of the materialist strategies confers high “prior probabilities”upon theories that are constrained in accord with these strategies. All sucharguments involve what I call “the overgeneralization of meta-inductions”(Chapter 10).

14 Where I use “underlying structure, process and law,” McMullin (1996,1999) uses “causal structure of the natural world” (Lacey 1997b, 1999b).

15 It also follows that the problem with (for example) Lysenkoism was not inconception, but in implementation; its strategies were retained and practicallyadopted even though they failed (perhaps after an initial short-runappearance to the contrary) to generate theories that manifested the cognitivevalues highly.

6The control of nature

1 These remarks raise important questions about who are the agents thatexercise control and the social relations under which they do so, and theextent to which these relations include relations of control among humanbeings, as well as material objects (Lacey 1990; Leiss 1972).

2 Elsewhere I compared it to the relationship that Weber spoke of as an“elective infinity” between Protestantism and capitalism (Lacey 1999a,1999b).

3 Shiva (1988, 1993) maintains that the predominance of research conductedunder the materialist strategies derives from the hegemony of capitalism, towhich she holds the modern values of control to be subordinate. I discuss thisin Lacey (1999b).

4 Why the modern values of control have become, and remain, ascendant is aquestion of major importance whose exploration lies beyond the scope of thisbook.

5 All unaccompanied page references are to Taylor (1982).6 The argument presented here, though it evolved from freely interpreting

Taylor’s argument, does not pretend to be faithful to his in its details,


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (276)

terminology and objectives. (See Lacey (1986) for direct analysis anddiscussion of his argument.)

7 For a generalization of the argument in play here see Lacey (1997c).8 The displacement argument does not imply that there are no losses that

accompany the gains of the displacement, for example, as Taylor points out,a sense of attunement with nature.

9 When Taylor returns to his argument (Taylor 1995), he seems to limit hiscase to the displacement argument.

10 Unless one were committed to materialist metaphysics as Taylor is not(Taylor 1970, 1985; Lacey 1990).

11 I have borrowed this broadly neo-Kantian point from Kuhn (Chapter 7). It isnot completely decisive. Some of the terms of materialist lexicons may beabsolute, but nothing about scientific practices assures us that this is so.Kuhn’s point is compatible with the truth of scientific realism (Sankey 1997),but the practices of science themselves seem more amenable directly tocarefully crafted constructivist accounts of their object of inquiry.

12 Of course we do share the world—the natural/historical world of which weare a part, which includes a “world” of social practices and objects(Chapter 7); that is why the displacement argument is sound.

13 Taylor also elaborates his argument using the notions of “disinterestedrepresentation” (Taylor 1995:11) and “disengaged perspective” (89–90). Myargument, that they do not serve to separate adoption of the materialiststrategies from the link with the modern values of control, will be quitefamiliar by now.

7Kuhn: scientific activity in different ‘worlds’

1 See Chapter 1, note 9, on the relationship of my “strategy” to Kuhn’s“paradigm.” Throughout this chapter, since my concern is withKuhnoriginated ideas and not with Kuhnian scholarship, I will refer to manyideas expressed in my terminology as Kuhn’s.

2 For the logic in play here and its limitations see MacIntyre (1988, last fewchapters).

3 The developments and changes of Kuhn’s position, and the wide range ofinterpretations of his views, are discussed critically in Hoyningen-Huene(1993) and Sankey (1994).

4 While the transition of the sixteenth and seventeenth centuries, “the scientificrevolution,” has been the most dramatic transformation of the kind thatKuhn discusses, he thinks that less dramatic instances are relatively common;since, for Kuhn, there are as many scientific ‘worlds’—or ‘sub-worlds’—asthere are fields of inquiry.

5 Sometimes A will refer to Aristotelian-type theories, sometimes toAristotelian scientists; similarly G. In context no ambiguity will arise.

6 The notions of “acceptance,” “significance” and “application” of a theorywere introduced in Chapter 1.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (277)

7 No doubt it also derives (in part) from the character of general humanperceptual mechanisms.

8 Conversely, we have lost much of the observational skill of A, and ofnumerous “pre-modern” forms of local knowledge and the practical skillsassociated with them.

9 It is part of the meanings of teleology and lawfulness (accord withdifferential equations) that they exclude each other. In G’s explanations, whathappens when a process is under way is explained by the initial conditions ofthe process (and the boundary conditions of the space) and the laws, not by atendency to reach a final state.

10 See the “displacement” argument in Chapter 6.11 The specifics of the planetary theories of A and G that I cite are all well

documented in standard histories of the subject (e.g., Kuhn 1957).12 A did not reject his general theory as a theory of cosmologically significant

phenomena when he accepted Ptolemaic astronomy in place of Eudoxan. Inthis case, other things were not equal: there could be an instrumentalistaccommodation, because Ptolemaic theory was not part of a morecomprehensive theory that was incompatible with A with respect to a range ofsubdomains, it was only more empirically adequate than Eudoxan theorywith respect to planetary movements and appearances, and Ptolemaic theorydid not manifest the other cognitive values very highly with respect to thesephenomena.

13 A queried any move from artifact (including experiment) to nature. Theunification, brought about by Newton, of certain experimental and naturalphenomena served to undermine A’s query, and to legitimate the readymoves made from experiment to nature in modern science. While A’s querycannot be sustained generally, it is always potentially relevant. It is alwaysappropriate to ask whether the natural space is sufficiently like theexperimental one for extrapolation to be made without further investigationof the natural space (Lacey 1984; Schwartz and Lacey 1982: Chapters 2, 9).

14 Later in the scientific tradition, especially when technological applicationbecomes of major importance, a scientific ‘world’ and a social “world”might mutually support each other’s development.

15 John Clendinnen (in discussion) has challenged that, with remarks like this, Imischaracterize the nature of modern (Galilean) science. See “objections” tomy account of full understanding (Chapter 5).

8A “grassroots empowerment” approach

1 In place of the term “authentic,” which I borrow from Latin Americandiscussions, others have used “alternative,” “appropriate,” “integral,” and“popular” (Lacey 1994). (“Sustainable development” is ambiguous and opento be co-opted by versions of both modernizing and authentic development:several articles in Sachs 1993). No one of these terms has gained generalcurrency and all of them have disadvantages (Fabián 1991; Escobar1995; Goulet 1988). This section represents my attempt to offer a summary


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (278)

synthesis of thinking that is characteristic of the popular movement in LatinAmerica (see p. 185) on development and its processes. In part, I draw frominsights taken from the “anti-development” (e.g., Escobar 1995), “liberationtheology” (e.g., the essays of Ellacuría, in Hassett and Lacey 1991) and “newsocial movement” (e.g., Escobar and Alvarez 1992) literatures.

2 They provide the model for my fifth path toward equilibrium,“transformation from below” (Chapter 2), and they often are called “newsocial movements” in sociological writings.

3 The language of “human rights”—the full list: social/economic/cultural as wellas civil/political—is widely used in the articulations of authentic development,often under an innovative interpretation. Human rights can be taken torepresent what human beings can become; they constitute a check-point onthe legitimacy of institutions that shape people’s lives. Looked at this way,human rights represent aspirations, whose realizability is asserted in themodern array of rights’ documents; appropriate institutions are being soughtfor their protection and provision, and for their absence current institutionscan be held accountable (Lacey 1991b; several articles in Hassett and Lacey1991; Faria l994).

4 This does not imply that theory has no role in informing the struggle,especially concerning the structural obstacles to its progress and how theinteraction of human nature with historical conditions shapes the range ofpossibilities open to serious exploration (Lacey 1997c).

5 My usage of “appropriate technology” may be narrower than customaryones (Fisher 1987).

6 This subsection summarizes parts of Shiva (1991). For documentation andfurther details see the writings of Shiva listed in the Bibliography; also Tilesand Oberdiek (1995: Chapter 6).

7 See also Shiva (1997) and, for some critical remarks, Lacey (1999b). Notealso that, in addition to the social and ecological critique of the greenrevolution, a methodological critique of its research practices can also bemounted (Lacey 1998).

8 Cf. Chomsky (1993:114–17) on the connection between imperialism and thedestruction of forms of knowledge (also Harding 1998).

9 Implementing the green revolution itself, as distinct from its legitimation inthe face of criticisms, does not presuppose “no lost possibilities”; alternativeproposals presuppose that there are lost possibilities. The former presupposesonly that the possibilities uncovered by green revolution research can berealized in practice. Of course something is lost: local traditions areundermined and community values can no longer be aspired to. However—the objection maintainsthe cause of this is not the implementations of thegreen revolution (and their social requirements), but the inability oftraditional forms and their (possible) contemporary transformations to meetthe basic needs of the poor.

10 For example, Shiva (1991), several articles in Suárez (1990) and in Delgadoet al. (1990), Ambrose (1983), Kloppenburg (1988), Rist and San Martin(1991), Delgado (1992).


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (279)

11 See especially Shiva (1988, 1991); also Levins and Lewontin (1985), Rist(1992), Brusch and Stabinsky (1996); several articles in Delgado et al.(1990), Marglin and Marglin (1990) and Nandy (1988).

12 Quotes from Shiva (1991:26, 29, 97 respectively). 13 Cf. the interactionist account of the causation of tuberculosis in Lewontin

(1993).14 For a discussion of the ethical issues raised here, see Lacey (1998).

9A feminist approach

1 I also draw upon Longino (1987, 1992, 1993, 1995, 1996), Anderson(1995a, 1995b), and discussions with Lynn Hankinson Nelson.

2 Quotes from Longino 1990:189, 150, 171, 190 respectively.3 Cf. the “paths to equilibrium” (Chapter 2), especially the fourth (“the quest

for power”) and fifth (“transformation from below”), and the highlighting ofenhancing grassroots agency among the values of authentic development(Chapter 8).

4 Instead, at best they asked: “What material benefits flow from successfullyadopting the stance of control?”

5 Cf. the complementary paths towards equilibrium, “adjustment” and“resignation” (Chapter 2).

6 Longino’s (and Anderson’s) version of feminism is very inclusive, and it canbe endorsed from a wide variety of value complexes including those linkedwith authentic development. (If “agency” is complemented with “solidarity,”a rich synthesis of O2 and O3 can be developed.) I take this kind ofinclusiveness to be a virtue. The label “feminist” is appropriate because thevalues drawn upon, though defensible generally as “liberatory,” have beenparticularly highlighted in recent feminist discussions. Furthermore, althoughO3 concerns in principle all the conditions that diminish agency, itsdevelopments to date have put special (though not exclusive: e.g., Harding1998) emphasis on those that diminish women’s agency.

7 Where I use “strategy,” Longino uses “global assumption.” The differentuses may reflect more than terminological preferences.

8 Anderson (in discussion) questions that the intentionality of action isinconsistent with its lawfulness by citing rational choice theory in economics.This is an important question and deserves more detailed attention than I canprovide here. Briefly, my response is that rational choice theory remains largelyan ideal type, explaining little of the interesting detail of action; and that,where it does succeed in representing action (behavior) lawfully it is incontexts in which I describe the agency as diminished, in which (for example)structural conditions constrain the available options so narrowly that theagent’s adopted values are not able to be manifested to any significant degree.Then the intentional terms “preference,” “belief,” etc. could readily bereplaced by explicitly non-intentional terms (as “purpose” is replaced by“reinforcement” in many behaviorist theories) without loss of explanatoryand predictive power. On the other hand, I agree with Anderson that action,


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (280)

insofar as it is intentional, does have law-like features. That diminishedagency can be interpreted as socially produced reflects the fact that agency hasstructural enabling conditions. It also has physiological enabling conditions.Physiological and social structures provide both enabling conditions andpotential constraints on agency—so that in some respects action is necessarilyimplicated in law-like regularities.

9 Longino (1990: Chapters 6, 7). I report Longino’s discussion of LHM forpurposes of illustration only. (I make no effort to evaluate it.)

10 Haark (1996) seems to interpret Longino in this way (cf. Anderson 1995b). 11 Reference to and quotes from Longino 1990:143–61, 185, 187 respectively12 To avoid repetition of what by now will be familiar themes I only sketch an

argument for these two points.13 Addressing arguments that intentionality is characteristic of only one stance

(adopted for pragmatic ends) among others (design and physicalist) thatmight be adopted towards human beings (e.g., Dennett 1987) lies beyond thescope of this work.

14 From the perspective of those following O3, this consequence is, of course,highly unlikely; and, if sufficiently general in range of application, alsoparadoxical (Donagan 1987; Lacey 1996).

15 As, for example, with theories of radical behaviorism and cognitivepsychology

16 For an instance of the relevant kind of argument see Schwartz, Schuldenfreiand Lacey (1978) and Schwartz and Lacey (1982).

17 If action based upon such posits is successful, that will provide furtherconfirming evidence of the posits—just as successful technologicalapplications provide further evidence supporting the theories that informthem. Moreover, without action committed to implementing some of therelevant possibilities (which may require considerable social change) it isimpossible to explore them, for there will be no empirical basis upon whichto base hypotheses about them. It is in this sense that appraisals of theoriesdeveloped within the feminist approach need to pay heed to the outcomes ofpolitical action.

18 Such investigation might lead to qualified support for feminist “standpointepistemologies,” discussed (for example) in Harding (1996).

191 will not discuss the epistemological framework in which Longino presentsher argument.

10Science as value free: revised theses

1 For example, the criticism of the “linear hormonal model” in Chapter 9; seealso Chapter 7.

2 The account here generalizes an argument presented in Chapter 9; it isinfluenced by MacIntyre (1977).

3 For example, “no lost possibilities,” Chapter 8.4 The conclusions drawn about Rudner’s condition (Chapter 4) apply

unchanged in connection with /.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (281)

5 It is a value, not necessarily an over-riding one. In context, the urgencies ofthe moment or the necessity to act may make it subordinate to other socialvalues, and the costs of undertaking the relevant inquiry may not warrantlosses that might be incurred in other valued activities. Where it issubordinated, I suggest, the tradition of modern science has alwaysarticulated (but not always acted upon): clearly separate accepted theoriesfrom ideology; do not subordinate understanding to the interests of power—so that if one “must” act in a context where one’s action is informed by, and/or its legitimation presupposes, less than soundly accepted theories (posits), beaware that this is so and monitor outcomes carefully for possible undesirable,unintended side-effects.

6 This is one reason why it is difficult to reach accord with / in psychology andthe human sciences in general (Lacey 1980, 1997c).

7 Brian Ellis pointed out to me (in discussion) that sometimes the cognitivevalues operate prior to any considerations of strategies; that their play aloneis sufficient to exclude from consideration certain hypotheses, e.g., that theuniverse came into being five minutes ago. (See also Ellis 1990.)

8 The famous cases of Lysenko and the persecution of Galileo have come tosymbolize what happens when theories are accepted in violation of I; theyturned into tragedies when power was also used to uphold these theories.

9 Examples may be found in Lewontin 1993; Longino 1990; Nelson andNelson 1995.

10 Elsewhere, borrowing from Bhaskar (1975), I have referred to them as theconditions that define “closed spaces,” spaces that in virtue of theirstructures ensure that, to a reasonable approximation, only a small number ofspecified forces and influences are acting (Lacey 1984, 1990; Lacey andSchwartz 1986, 1987). Following criticism from Marcos Barbosa de Oliveira,I now think this is too narrow a specification.

11 The mechanism involved here—call it “the overgeneralization of meta-inductions”—is a common partial source of the de facto violation ofimpartiality, for example, in the examples discussed by Longino (1990) andNelson and Nelson (1995), especially those about alleged gender and racialdifferences in human abilities and capacities.

12 The statements of N″(1), (2) are different from those of N′(1), (2) only inminor matters of wording, abbreviation and adaptation to the context of I. Iwill leave open whether the addendum, included in (3′) in the text on p. 237,should be included definitively.

13 With it, it should be kept in mind, “no lost possibilities” might be established,and then these value complexes would be rendered unviable. It is an oddnotion of “neutrality” that denies potentially significant options “an(empirical) run for the money,” and which leaves it for the significance to“trickle down” to where the modern values of control are not highlymanifested. As in economic matters, “trickle down” is a pale shadow of“evenhandedness.”

14 I take “viable value complex” to be redefined here as “value complex whosepresuppositions are consistent with all theories accepted in accordance withI.”


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (282)

15 Accepting posits like these, without accord with I, and acting on them is tocourt disaster (especially for the poor) and exacerbation of the contemporarysocial and ecological crises (cf. Chapter 8).

16 My argument for endorsing N supports the exploration of multiplestrategies, but—since it is based on considerations of human flourishing—itdoes not support the moral legitimacy of research conducted under anystrategies whatsoever.

17 will not subject item (1) (shared by N and N″) to scrutiny in this book.Recall that the materialist strategies ensure that it is true for theoriesdeveloped under them by deploying lexicons that contain no value terms.That argument no longer obtains when multiple strategies are in play,especially in the social sciences whose descriptive, explanatory andanticipatory lexicons may include value-laden terms. Elsewhere (Lacey1997c) I explore some arguments against (1) in detail.

18 In the case of N, I would urge, the social value involved makes a compellingclaim on all of us—but the argument for that cannot be developed here.

19 It is not important that every individual investigator possess all these virtues,but that there be a healthy spread of them throughout the scientificcommunity.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (283)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (284)


Ambrose, R. (1983) “Agricultural research and breaking the cycle of dependency,”in Science for the People 15(6): 6–31.

Anderson, E.S. (1995a) “Feminist epistemology: an interpretation and defense,” inHypatia 10:50–84.

——(1995b) “Knowledge, human interests, and objectivity in feministepistemology,” in Philosophical Topics 23:27–58.

——(1995a) “Feminist epistemology: an interpretation and defense,” in Hypatia 10:50–84.

——(1993) Values in Ethics and Economics, Cambridge, MA: Harvard UniversityPress.

Bacon, F. (1620/1960) The New Organon and Related Writings, Indianapolis:BobbsMerrill.

Bernstein, R.J. (1983) Beyond Objectivism and Relativism: science, hermeneuticsand praxis, Philadelphia: University of Pennsylvania Press.

Bhaskar, R. (1986) Scientific Realism and Human Emancipation, London: Verso.——(1979) The Possibility of Naturalism: a philosophical critique of the

contemporary human sciences, Atlantic Highlands, NJ: Humanities Press.——(1975) A Realist Theory of Science, Atlantic Highlands, NJ: Humanities Press.Bronowski, J. (1961) Science and Human Values, London: Hutchinson.Brush, S.B. (1989) “Prediction and theory evaluation: the case of light bending,” in

Science 246:1126–9.Brush, S.B. and Stabinsky, D. (1996) Valuing Local Knowledge: indigenous people

and intellectual property rights, Washington: Island Press.Bunge, M. (1980) Ciencia y Desarrollo, Buenos Aires: Siglo Veinte.Campbell, N.R. (1957) Foundations of Science, New York: Dover.Carnap, R. (1967) The Logical Structure of the World and Pseudoproblems in

Philosophy, Berkeley: University of California Press.——(1956) “Empiricism, semantics and ontology,” in R.Carnap Meaning and

Necessity, Chicago: University of Chicago Press.Chomsky, N. (1993) Year 501: the conquest continues, Boston: South End Press.Coady, C.A.J. (1992) Testimony: a philosophical study, Oxford: Clarendon Press.Cupani, A. (1998) “A propósito do ‘ethos’ da ciência,” in Episteme (Brazil) 3, 16–

38.da Costa, N. and Bueno, O. (1998) “Paraconsistência: esboço de uma interpreta-

ção,” in N.da Costa, J.-Y Béziau, and O.Bueno Elementos de TeoriaParaconsistentes de Conjuntos, Campinas (Brazil): Coleção CLE.

Delgado, F. (1992) La Agroecologia en las Estrategias del Desarrollo Rural, Cusco:Centro de estudios regionales andines “Bartolomé de las Casas”.

——et al. (1990) Agroecologia y Saber Andino, Cochabamba: AGRUCO-PRATEC.

Lacey_1999_Is Science Value Free.pdf - [PDF Document] (285)

Dennett, D.C. (1987) The Intentional Stance, Cambridge, MA: MIT Press.Donagan, A. (1987) Choice: the essential element in human action, London:

Routledge & Kegan Paul.Dupré, J. (1993) The Disorder of Things: metaphysical foundations of the disunity

of science, Cambridge, MA: Harvard University Press.Eldridge, R. (1997) Leading a Human Life: Wittgenstein, intentionality and

romanticism, Chicago: University of Chicago Press.Ellis, B.D. (1990) Truth and Objectivity, Oxford: Basil Blackwell.——(1985) “What science aims to do,” in P.M.Churchland and C.A.Hooker (eds)

Images of Science, Chicago: University of Chicago Press.Escobar, A. (1995) Encountering Development: the making and unmaking of the

Third World, Princeton: Princeton University Press.Escobar, A. and Alvarez, S.E. (eds) (1992) The Making of Social Movements in

Latin America: identity, strategy and democracy, Boulder: Westview Press.Fabián, R.R. (1991) “Comentários críticos a las teorias de desarrollo predominan-

tes: hacia otro concepto del desarrollo de las comunidades,” Realidad 4:713–61.

Faria, J.E. (ed.) (1994) Direitos Humanos, Direitos Socias e Justiça, São Paulo:Malheiros Editores.

Feigl, H. (1961) “Philosophical tangents of science,” in H.Feigl and G.Maxwell(eds) Current Issues in the Philosophy of Science, New York: Holt, Rinehartand Winston.

Feyerabend, P. (1989) “Realism and the historicity of knowledge,” in The Journalof Philosophy 86:393–406.

——(1979) Science in a Free Society, London: New Left Books.——(1975) Against Method, London: New Left Books.——(1965) “Problems of empiricism,” in R.S.Colodny (ed.) Beyond the Edge of

Certainty, Englewood Cliffs, NJ: Prentice Hall.Fisher, F.G. (1987) “Ways of knowing and the ecology of change: understanding

non-technical barriers to appropriate technology,” UNESCO Network forAppropriate Technology Seminar, The University of Melbourne.

Galilei, Galileo (1623/1957) The Assayer, excerpts in S.Drake (trans.) Discoveriesand Opinions of Galileo, Darden City, NJ: Doubleday.

Goulet, D. (1988) “Tasks and methods in development ethics,” in Cross Currents38: 146–53.

Haark, S. (1996) “Science as social?—yes and no,” in Nelson and Nelson (1996).Hacking, I. (1993) “Working in a new world: the taxonomic solution,” in P.

Horwich (ed.) World Changes: Thomas Kuhn and the nature of science,Cambridge, MA: MIT Press.

——(1983) Representing and Intervening, Cambridge: Cambridge University Press.Harding, S. (1998) Is Science Multicultural? postcolonials, feminisms, and

epistemologies, Bloomington, IN: Indiana University Press.——(1996) “Rethinking standpoint epistemology: what is ‘strong objectivity’?” in

E.F.Keller and H.E.Longino (eds) Feminism and Science, Oxford: OxfordUniversity Press.

——(ed.) (1993) The “Racial” Economy of Science: toward a democratic future,Bloomington, IN: Indiana University Press.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (286)

Hassett, J. and Lacey, H. (eds) (1991) Towards a Society that Serves Its People: theintellectual contribution of El Salvador’s murdered Jesuits, Georgetown:Georgetown University Press.

Hempel, C.G. (1983a) “Values and objectivity in science,” in R.S.Cohen and L.Laudan (eds) Physics, Philosophy and Psychoanalysis, Dordrecht: Reidel.

——(1983b) “Kuhn and Salmon on rationality and theory choice,” in The Journalof Philosophy 80:570–2.

——(1981) “Turns in the evolution of the problem of induction,” in Syntheses 46:389–404.

——(1965) “Science and human values,” in Hempel Aspects of ScientificExplanation, New York: The Free Press.

Hesse, M. (1977) “Theory and value in the social sciences,” in C.Hookway and P.Pettit (eds) Action and Interpretation, Cambridge: Cambridge University Press.

Hoyningen-Huene, P. (1993) Reconstructing Scientific Revolutions: ThomasS.Kuhn’s philosophy of science, Chicago: University of Chicago Press.

Hull, D.L. (1988) Science as Process: an evolutionary account of the social andconceptual development of science, Chicago: University of Chicago Press.

Hume, D. (1739/1968) A Treatise of Human Nature, Oxford: Clarendon Press.Joseph, G. (1980) “The many sciences and the one world,” in The Journal of

Philosophy 77:773–91.Keller, E.F. (1982) “Feminism and science,” in Signs: Journal of Women in Culture

and Society 1:589–602.Kenney, M. (1986) Biotechnology: the university—industrial complex, New

Haven: Yale University Press.Kitcher, P. (1993) The Advancement of Science: science without legend, objectivity

without illusions, New York: Oxford University Press.Kloppenburg, J.R. (1988) First the Seed: the political economy of plant

biotechnology, 1492– 2000, Cambridge, Cambridge University Press.Koyre, A. (1957) From the Closed World to the Infinite Universe, New York:

Harper Torchbooks.Kuhn, T.S. (1993) “Afterwords” in P.Horwich (ed.) World Changes: Thomas Kuhn

and the nature of science, Cambridge, MA: MIT Press.——(1977) “Objectivity, value judgment and theory choice,” in Kuhn The Essential

Tension, Chicago: University of Chicago Press.—— (1970) The Structure of Scientific Revolutions, 2nd edn, Chicago: University of

Chicago Press.——(1957) The Copernican Revolution: planetary astronomy in the development of

Western thought, Cambridge: Harvard University Press.Lacey, H. (1999a) “Scientific understanding and the control of nature,” in Science

and Education 8: (13–35).——(1999b) “On cognitive and social values: a reply to my critics,” in Science and

Education 8: (89–103).——(1998) “The dialectic of science and advanced technology: an alternative?” in

Democracy and Nature 4:34–53.——(1997a) “Ciência e. valores,” in Manuscrito 20:9–36.——(1997b) “The constitutive values of science,” in Princípia 1:3–40.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (287)

——(1997c) “Neutrality in the social sciences: on Bhaskar’s argument for anessential emancipatory impulse in the social sciences,” in Journal for theTheory of Social Behavior 27:213–41.

——(1996) “Behaviorisms: theoretical and teleological,” in Behavior andPhilosophy 23:61–78.

——(1995) “The legacy of El Salvador’s murdered Jesuits,” in Journal for Peaceand Justice Studies 6:113–26.

——(1994) “¿Qué tipo de ciencia le sirve al desarrollo auténtico?” Realidad 39:369–82.

——(1991a) “Understanding conflicts between North and South,” in M.Dascal(ed.) Cultural Relativism and Philosophy: North and Latin Americanperspectives, Leiden: E.J.Brill.

——(1991b) “Liberation theology and human rights,” in Proceeding from the 38thAnnual Meeting of the Rocky Mountains Council of Latin American Studies,101–10.

——(1990) “Interpretation and theory in the natural and human sciences,” inJournal for the Theory of Social Behavior 20:197–212.

——(1986) “The rationality of science,” in J.Margolis, M.Krausz and R.A. Burian(eds) Rationality, Relativism and the Human Sciences, Dordrecht: KluwerAcademic Press.

——(1985) “On liberation,” in Cross Currents 35:219–41.——(1984) “Constraints upon acceptable theories: Chomsky’s analogy between

language acquisition and theory formation,” in Pacific Philosophical Quarterly65: 16–78.

——(1980) “Psychological conflict and human nature: the case of behaviorism andcognition,” in Journal for the The otyqfSocialJlehavwr 10:131–55.

——(1979) “Skinner on the prediction and control of behavior,” Theory andDecision 10:353–85.

——(1974) “The scientific study of linguistic behavior: a perspective on theSkinner-Chomsky controversy,” in Journal for the Theory of Social Behavior 4:17– 51.

Lacey, H. and Schwartz, B. (1996) “The formation and transformation of values,”in W.O’Donohue and R.Kitchener (eds.) The Philosophy of Psychology,London: Sage.

——(1987) “The explanatory power of radical behaviorism,” in S.Modgil and C.Modgil (eds) B.F.Skinner: Consensus and controversy, London: Falmer Press.

——(1986) “Behaviorism, intentionality and socio-historical structure” inBehaviorism 14:193–210.

Lakatos, I. (1970) “Falsification and the methodology of scientific researchprogrammes,” in I.Lakatos and A.Musgrave (eds) Criticisms and the Growthof Knowledge, Cambridge: Cambridge University Press.

Laudan, L. (1984) Science and Values: the aims of science and their role inscientific debate, Berkeley: University of California Press.

Leiss, W. (1972) The Domination of Nature, Boston: Beacon Press.Levins, R. and Lewontin, R. (1985) The Dialectical Biologist, Cambridge: Harvard

University Press.Lewontin, R. (1993) Biology as Ideology, New York: Harper Perennial.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (288)

Lloyd, E.A. (1996) “Science and anti-science,” in L.H.Nelson and J.Nelson (eds)(1996) Feminism, Science and the Philosophy of Science, Dordrecht: KluwerAcademic Press.

Longino, H.E. (1996) “Cognitive and non-cognitive values in science: rethinkingthe dichotomy,” in L.H.Nelson and J.Nelson (eds) Feminism, Science and thePhilosophy of Science, Dordrecht: Kluwer Academic Press.

——(1995) “Gender, politics, and the theoretical virtues,” in Synthese 104:383–97.——(1993) “Subjects, power and knowledge: description and prescription in

feminist philosophies of science,” in L.Alcoff and E.Potter (eds) Feministepistemologies, New York: Routledge.

——(1992) “Essential tensions—phase two: feminist, philosophical, and socialstudies of science,” in E.McMullin (ed.) The Social Dimensions of Science,Notre Dame: University of Notre Dame Press.

——(1990) Science as Social Knowledge, Princeton: Princeton University Press.——(1987) “Can there be a feminist science?” in Hypatia 2:51–64.MacIntyre, A. (1988) Whose Justice? Whose Rationality? Notre Dame: University

of Notre Dame Press.——(1981) After Virtue, Notre Dame: University of Notre Dame Press.——(1977) “Epistemological crises, dramatic narrative and the philosophy of

science,” in The Monist 60:453–72.Marglin, F.A. and Marglin, S.A. (eds) (1990) Dominating Knowledge:

development, culture and resistance, Oxford: Clarinden Press.Margolis, J. (1995) Historied Thought, Constructed World: a conceptual primer

for the turn of the millennium, Berkeley: University of California Press.Maxwell, N. (1984) From Knowledge to Wisdom: a revolution in the aims and

methods of science, Oxford: Blackwell.McDowell, J. (1994) Mind and World, Cambridge, MA: Harvard University Press.McMullin, E. (1999) “Materialist categories?” in Science and Education 8: (37–44).——(1998) “A case for scientific realism,” in J.A.Kourany (ed.) Scientific

Knowledge: basic issues in the philosophy of science, 2nd edn, Belmont, CA:Wadsworth.

——(1996) “Values in science,” in W.Newton-Smith (ed.) A Companion to thePhilosophy of Science, Cambridge: Blackwell.

——(1993) “Rationality and paradigm change in science,” in P.Horwich (ed.)World Changes: Thomas Kuhn and the nature of science, Cambridge, MA:MIT Press.

——(1983) “Values in science,” in P.D.Asquith and T.Nickles (eds) PSA 1982, vol.2, East Lansing, MI: Philosophy of Science Association.

Merton, R. (1957) Social Theory and Social Structure, Glencoe: Free Press.Midgley, M. (1979) Beast and Man: the roots of human nature, London:

Routledge.Murdoch, I. (1992) Metaphysics as a Guide to Morals, London: Penguin.Nagel, E. (1961) The Structure of Science, New York: Harcourt, Brace and World.Nandy, A. (ed.) (1988) Science, Hegemony and Violence, Delhi: Oxford University

Press.Nelson, L.H. (1996) “Empiricism without dogmas,” in L.H.Nelson and J.Nelson

(eds) Feminism, Science and the Philosophy of Science, Dordrecht: KluwerAcademic Press.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (289)

——(1995) “A feminist naturalized philosophy of science” in Synthese 104:399–421.

Nelson, L.H. and Nelson, J. (1995) “Feminist values and cognitive virtues,” in D.Hull, M.Forbes and R.M.Burian (eds) PSA 1994: proceedings of the 1994biennial meeting of the Philosophy of Science Association, vol. 2, EastLansing, MI: Philosophy of Science Association.

Nerlich, G. (1989) Values and Valuing: speculations on the ethical life of persons,Oxford: Oxford University Press.

Newton-Smith, W.H. (1981) The Rationality of Science, London: Routledge &Kegan Paul.

Poincaré, H. (1920/1958) The Value of Science, New York: Dover.Popper, K.R. (1959) The Logic of Scientific Discovery, New York: Harper.——(1972) Objective Knowledge, London: Oxford University Press.Putnam, H. (1990) Realism with a Human Face, Cambridge, MA: Harvard

University Press.——(1987) The Many Faces of Realism, La Salle, IL: Open Court.——(1981) Reason, Truth and History, Cambridge: Cambridge University Press.——(1978) Meaning and the Moral Sciences, London: Routledge & Kegan Paul.Rescher, N. (1965) “The ethical dimension of scientific research,” in R.G.Colodny

(ed.) Beyond the Edge of Certainty, Englewood Cliffs, NJ: Prentice-Hall.Rist, S. (1992) Desarrollo y Participacion: experiencias con la revalorizacion del

conocimiento campesino en Bolivia, Cochabamba: AGRUCO.Rist, S. and San Martin, J. (1991) Agroecologia y Saber Campesino en la

Conservación de Suelos, Cochabamba: AGRUCO.Root, M. (1993) Philosophy of Social Science, Oxford: Blackwell.Rouse, J. (1987) Knowledge and Power: towards a political philosophy of science,

Ithaca: Cornell University Press.Rudner, R. (1953) “The scientist qua scientist makes value judgments,” in

Philosophy of Science 20:1–6.Sachs, W. (1993) Global Ecology: a new arena of political conflict, London: Zed

Books.Salmon, W.C. (1983) “Carl G.Hempel on the rationality of science,” in The Journal

of Philosophy 80:555–62.——(1966) The Foundations of Scientific Inference, Pittsburgh: University of

Pittsburgh Press.Sankey, H. (1997) Rationality, Relativism and Incommensurability, Aldershot:

Avebury.——(1994) The Incommensurability Thesis, Aldershot: AveburySchwartz, B. (1997) “Psychology, idea technology, and ideology,” in Psychological

Science 8:21–7.——(1986) The Battle for Human Nature: science, morality and modern life, New

York: Norton.Schwartz, B. and Lacey, H. (1982) Behaviorism, Science and Human Nature, New

York: Norton.Schwartz, B., Schuldenfrei, R. and Lacey, H. (1978) “Operant psychology as

factory psychology,” in Behaviorism 6:29–54.Scriven, M. (1991) Evaluation Thesaurus, 4th edn, Newbury Park, CA: Sage.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (290)

——(1974) “The exact role of value judgments in science,” in R.S.Cohen and K.Schaffner (eds) Proceedings of the 1972 Biennial Meeting of the Philosophy ofScience Association, Dordrecht: Reidel.

Shiva, V. (1997) Biopiracy: the plunder of nature and knowledge, Boston: SouthEnd Press.

——(1993) Monocultures of the Mind: perspectives on biodiversity andbiotechnology, London: Zed.

——(1991) The Violence of the Green Revolution, London: Zed.——(1989) Staying Alive: women, ecology and development, London: Zed.——(1988) “Reductionist science as epistemological violence,” in A.Nandy (ed.)

Science, Hegemony and Violence, Delhi: Oxford University Press.Skinner, B.F. (1972) “Are theories of learning necessary?” in Skinner Cumulative

Record, 3rd edn, New York: Appleton-Century-Crofts.Solomon, M. (1994) “Social empiricism” in Noûs 28:325–43.——(1992) “Scientific rationality and human reasoning,” in Philosophy of Science

59:439–55,Suárez, B. (ed.) (1990) ¿Biotechnologia para el Progreso Agricola de México?,

Mexico City: Centro de Ecodesarrollo.Taylor, C. (1995) Philosophical Arguments, Cambridge: Harvard University Press.——(1985) Human Agency and Language: philosophical papers, vol. 1, Cambridge:

Cambridge University Press.——(1982) “Rationality,” in M.Hollis and S.Lukes (eds) Rationality and

Relativism, Cambridge, MA: MIT Press.——(1981) “Understanding in the Geisteswissenschaften,” in S.H.Holtzmann and

C.M. Leich (eds) Wittgenstein: To follow a rule, Cambridge: CambridgeUniversity Press.

——(1980) “Understanding in human science,” in The Review of Metaphysics 34:24–38; 47–55.

——(1971) “Interpretations and the sciences of man,” in The Review ofMetaphysics 25:3–51.

——(1970) “The explanation of purposive behavior,” in R.Berger and F.Cioffi (eds)Explanatian in the behavioral sciences, Cambridge: Cambridge UniversityPress.

Tiles, M. (1987) “A science of Mars or Venus” in Philosophy 62:293–306.——(1986) “Mathesis and the masculine birth of time,” in International Studies in

the Philosophy of Science 1:16–35.——(1985) “Correcting concepts,” in Ratio 27:19–35.Tiles, M. and Oberdiek, H. (1995) Living in a Technological Culture: human tools

and human values, London: Routledge.van Fraassen, B.C. (1980) The Scientific Image, Oxford: Clarendon Press.Weber, M. (1949) The Methodology of the Social Sciences, New York: The Free

Press.White Jr, L. (1968) Machina ex Deo, Cambridge, MA: MIT Press.


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (291)


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (292)


A see AristotleA see autonomyadjustment, path of 32–4, 38, 240 266agency 182, 187, 192, 198–5, 203–10,

208, 210, 214–20, 229–6, 247agent see agencyagroecology 190–8approaches to scientific investigation

18–2, 57–9, 87, 100, 102–4, 107–109, 127, 149, 171, 177, 195–199,203, 207, 211, 213, 215, 218, 220–7,242–53, 249;see feminist approaches, grassrootsempowerment approaches,materialist approaches

Anderson, E.S. 15, 89, 198, 203, 253,258, 260, 266, 267

applied research see fundamentalresearch

Aristotle (A) 52, 75–7, 145–176, 222–29, 263

Aristotelian see Aristotleattunement with nature 76, 111, 130,

132, 142, 162–5, 174, 262autonomy 1, 8–9, 11, 16, 18, 20, 31,

81–7, 87, 165, 172, 196–1, 216, 218,241–54, 247–60, 259, 260;localized 243–1, 247

belief(s) 12, 23, 26–8, 41–8, 65, 149,153, 156, 201, 204, 228, 255;modes of beliefs 48–49

Bacon, F. 1, 3–6, 18–2, 60–2, 67, 102,104–7, 115–7, 178, 199, 253;Baconian utility 60–2, 107–9

Barbosa de Oliveira, M. 268Bernstein, R.J. 52, 56Bhaskar, R. 6, 58, 101, 268biotechnology 83, 186, 188–2, 192,

194–8, 259

Clendinnen, J. 261, 263Coady, C.AJ. 63, 253, 257cognitive value 8, 14–15, 18–18, 54,

60, 65, 68, 69, 89, 229;criteria of 54–6, 87, 91–2

cognitive values 11, 15–16, 18–2, 26,43–65, 65, 68, 69, 70–4, 80, 87, 89–94, 96, 97–109, 128–1, 136–9, 140,144, 149, 153, 163, 166–79, 177–1,192, 195, 196, 202–7, 205–14, 212–22, 218, 221–35, 238, 244, 246, 247–59, 257, 260, 261;vs. rules 4, 18, 54–7, 93, 257, 258;modes of 49–53, 89–90;see standards for estimating thedegree of manifestation of

complex of values 29, 31, 38, 38–39,43–7, 63–5, 69, 72–8, 79, 104, 111,113, 125, 176–0, 188, 190, 195, 202,218, 224, 227, 136–9, 235–7, 244,268;see also viable value complex

comprehensiveness 130, 134, 136–46constraint/selection strategies 259;

see strategiescontrol of natural (material) objects 19–

1, 60–2, 76, 104, 106, 109–47, 162,164, 171, 174, 185, 199–4, 232, 235,266;


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (293)

possibilities of 76, 114, 117, 125,134;objects of 117–28, 219;requirements for 117–9, 133, 138;see also modern values of control

creative marginality, path of 34–7, 240

Descartes, R. 52, 59, 115, 253Donagan, A. 45, 201, 204, 267development 79, 178, 179–89, 190–4,

237, 266;authentic 180, 181–89, 193–99,222, 266;modernizing 180, 182, 184, 189–3,192, 195, 239, 264

displacement 130, 132–38, 143, 262,263

ecology 95, 97–102, 111, 131–4, 136–38, 178, 187, 190, 194, 268;see also agroecology

Ellis, B.D. 59–1, 93, 257, 261, 268empirical adequacy 57–59, 61–4, 92,

94, 108, 161, 167, 174–8, 208–4,213, 216–2

encapsulations of possibilities 3, 59, 64,76, 98, 101–3, 104, 107, 114, 117,127, 156, 176, 182, 188, 190, 193,199, 205, 221, 226, 242, 245

evenhandedness 5, 20, 74, 77, 80–2,230, 232–40, 237–5, 241, 259, 268

explanatory power, wide-ranging andfull see understanding, wide-rangingand full

feminist approaches 19–2, 103, 144,178, 196–23, 237, 239–7, 253

feminist strategies see feministapproaches

feminist values see feminist approachesFeyerabend, P. 59, 104, 150, 258fruitfulness 165–79, 205fundamental research 100, 16, 123–6,

128, 232;vs. applied research 100, 119, 125,184, 194, 231

fundamental theory 115, 193, 231

G see GalileoGalilean see GalileoGalileo 1, 3–6, 9–10, 14, 18–2, 67, 75–

8, 94, 100, 101, 144, 145–79, 17, 18,222–, 253, 263, 268

Geertz, C. 196–1, 202, 211, 219grassroots empowerment approaches

19–2, 103, 17, 18, 18, 239–, 253grassroots values see grassroots

empowerment approaches anddevelopment, authentic

green revolution 100, 186–7, 216, 231,233, 237

Hacking, I. 62, 147, 158, 165Harding, S. 74, 85, 23, 26, 26, 267Hempel, C.G. 3, 6, 56, 25, 261Hoyningen-Huene, P. 148, 262human nature 39–54, 76–8, 109, 18,

21, 23, 23, 23, 264human flourishing (fulfillment, well-

being) 16, 18, 25–7, 45, 104, 111–15, 132, 144, 162, 17, 17, 17, 235–,23, 24, 250–6, 268

I′, I″ see impartiality, provisional thesesof

I see impartiality, revised thesesimpartiality 1, 3–6, 8–12, 15, 17–20,

21, 41, 42, 56, 65, 65–73, 76, 77, 79,81, 86, 87, 94, 107, 128, 145, 172,17, 18, 18, 18, 210–1, 19, 20, 218–3,21, 23, 241–5, 247–6, 260;provisional theses of 69, 20, 220,222;revised thesis of 224

insignificance see significanceintellectual property rights 18, 18, 259intentional explanation 19, 23, 29, 45–

6, 109, 125, 126, 18, 18, 206–1, 209–24, 228–5

intentional understanding seeintentional explanation

intentionality see intentionalexplanation

Kloppenburg, J. 18, 18, 264


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (294)

Kuhn, T.S. 8, 52, 59, 91, 92, 105, 144,145–79, 19, 25, 25, 25, 263

Laudan, L. 52, 92, 93, 25, 25, 25, 261Lewontin, R. 10, 85, 18, 25, 26, 268linear-hormonal model (LHM) 203–8,

205–1, 214, 20, 267Longino, H.E. 11, 85, 198–2, 244–, 25,

266, 267, 271

MacIntyre, A. 2, 52, 59, 84, 87, 167,25, 25, 267

material possibilities 19, 98–100, 104,107, 119, 125–8, 135, 137–40, 141–5, 171, 172, 219, 222, 224, 242, 245–3

material world, the 128–47materialist approach 102materialist metaphysics 60, 80, 101,

104, 125–30, 139, 141, 144, 170, 20,24, 260

materialist strategies 19, 67–9, 69, 74,79–2, 87, 90–2, 94, 96, 98, 100–108,114–30, 129–47, 176, 177, 186–9,192–9, 195, 18, 18, 18, 205–1, 19,20, 218–, 224–5, 247–6, 25, 25, 25,268

materialist understanding see materialiststrategies

McMullin, E. 52, 56–59, 91–3, 127,144, 25, 25, 25, 261

modern values of control 19, 80, 87,102, 109, 114–47, 171, 172, 175–9,176–, 17, 17, 188–, 195, 18, 18,199–, 19, 218–, 20, 231–5, 247–6,259

moral values 14–15, 17, 19, 90, 249

N′, N″ see neutrality, provisional thesesN see neutrality, revised thesisNagel, E. 16, 75, 255Nelson, L.H. 59, 18, 25, 26, 268Nerlich, G. 255neutrality 1–6, 8–11, 17, 18, 20, 21, 39,

41, 64, 73–81, 86, 87, 105, 128, 138,172, 16, 17, 17, 18, 20, 231–48,242– 4, 247–60, 260;

provisional theses of (N, N) 76, 231–8;revised theses of (N) 234–1

Newton, I.53, 60, 253, 260Newton-Smith, W. 59, 60, 257, 263

Oberdiek, H. 185, 264

paths towards equilibrium 32, 240,266;see also adjustment, creativemarginality, quest for power,resignation, transformation frombelow

personal values 23–7, 29, 31, 38, 41,61–64, 83–5

Poincaré, H. xi, 16, 72Popper, K.R. 52, 55, 62, 253popular organizations 37, 172,

185ndash;6posits see theoriespossibilities 3, 6–8, 19–2, 39–4, 47–9,

59, 61, 76–8, 94, 95, 99–107, 111,113, 125–7, 127, 131–6, 137, 141–4,148, 158, 162–6, 171, 172–7, 177–1,18, 18, 207, 209, 219–222, 225–3,21, 236, 237, 239, 244–3, 249, 266,267;encapsulation of see encapsulationof possibilities;lost possibilities 99, 100, 189, 190–7, 206–11, 230, 233–4, 236–3, 238,259, 264, 267, 268;unrealized possibilities 42, 47, 101,102, 103, 131;see also material possibilities;control, possibilities of

progress 3, 74, 81, 179, 237–4, 250Putnam, H. 6, 15, 16, 39, 45, 51, 52,

104–6, 257

quest for power, path of 36–8, 266

rationality 5, 16, 45–8, 49, 112, 114,130, 136, 149, 165, 175, 179;informal ideal of 47

resignation, path of 33–5, 38, 266


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (295)

responsibilities of scientists 82, 85–7,259

Rouse, J. 148, 158, 188, 261Rudner, R. 16, 64–6, 70–4, 213, 253,

267rules see cognitive values vs. rules

Salmon, W.C. 56, 261Sankey, H. 56, 148, 260Schwartz, B. 16, 33, 45, 125, 201, 206–

xi, 255, 263, 267, 268science see approaches to scientific

investigation, science as value free,systematic empirical inquiry

science as value free xi–1, 7–8, 11–12,15–18, 20, 54, 57, 65, 86, 87, 128,197, 218–4, 229, 247–60, 253;see also autonomy, impartiality,neutrality

science, objective of 18–19, 93–4, 101–3, 105, 123, 156, 172, 176–81, 195,218, 220–7, 242, 244, 247, 260, 261

science, success of 4–5, 7, 13, 126scientific ethos 9–11, 16, 63, 81–3, 147,

149–3scientific theories see theoriesscientific understanding see

understandingScriven, M. 3, 15–17, 54seeds 95, 98–100, 118, 178, 186–187,

231, 238;as commodities 188, 195, 230

Shiva, V. 188–7, 259, 264, 266significance 14–16, 47, 48, 49, 53, 54,

60, 62, 65, 69, 74, 78, 81, 120–2,128, 130, 132–37, 138, 162, 166,171–5, 176, 184, 188–3, 192, 197,203, 209, 212, 215, 217, 220, 223–43, 241, 243–51, 246, 258

social structures 33, 34, 35, 36, 37, 106–9, 175, 183, 236

social values 14–15, 17, 19, 23, 27–29,41, 60, 75, 78, 80, 82, 90, 92, 100,106, 108, 109, 112–4, 119, 124, 128,134, 144–8, 168–71, 172, 175, 178,186, 195, 199–4, 202, 212–22, 222,

225, 230, 232, 234, 236, 237, 246,249, 267, 268

standards for estimating the degree ofmanifestaion of cognitive values 61–5, 71–4, 166, 168, 175, 203, 207–15,213, 195–21, 227, 229, 238, 244,246, 258

strategies 19–2, 60, 103, 107, 128, 139,142–5, 148–1, 152–5, 161, 165–79,176, 178, 184, 192, 195–197, 201,203, 205–10, 209, 213–18, 216–2,218–6, 222, 224, 227, 230, 233–2,240, 242–1, 249–59, 253, 258, 260,262, 266;see also feminist strategies,grassroots empowerment strategies,materialist strategies

systematic empirical inquiry 5, 18–1,100–100, 102, 108, 133, 144, 172,178, 193, 196, 206, 217, 224, 229–6,233–1, 237–5, 240, 242–2, 247, 250

systematic empirical understanding seesystematic empirical inquiry

Taylor, C. 23, 59, 63, 98, 109, 129–47,158, 257, 263

Tiles, M. 59–1, 100, 105, 185, 201,261, 266

theories, acceptance of 2–4, 8, 10, 12–13, 15, 20, 60–2, 64–7, 69–1, 72, 74,77–9, 81, 87, 115, 121, 128, 129–46,149, 161, 166, 177, 188, 190, 195,197, 205, 209, 212–19, 216, 220–7,224–43, 238–8, 243–2, 247, 252,258, 259, 260, 267;adherence to 12–13, 84, 213, 229;application of 2, 12–14, 16, 60–2,65, 70, 74, 77–79, 120, 123, 140,161–7, 168–7, 193, 206–13, 212,215–1, 223–30, 247, 259, choice of11, 15, 18, 18, 83–6, 91, 92–3, 106,145, 153–6, 166, 176, 213, 225,231, 251;consolidation of see acceptance of;provisional entertainment of 12–13,60, 63, 67, 69, 84, 212, 214, 225–4;significance of see significance;


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (296)

see also fundamental theorytraditional local knowledge 19, 53, 100,

103, 118, 133, 144, 185–90, 191–9,233, 245

truth 39, 43, 45, 47, 49, 51, 53, 60transformation from below, path of 37–

8, 240, 264, 266

understanding 18, 20, 87, 94, 95–100,127–30, 130;forms of 96–100, 128–46, 179, 188,195, 200, 202, 214;full 58, 96, 99–100, 135, 137–40,143, 163–6, 170–3, 178, 184–9,193–7, 207, 213, 263;materialist see materialist strategies;wide-ranging 58, 61, 96, 98, 99,118–20, 137–40, 143, 170, 184,213, 231, 251–9

value, object of 2, 21, 73, 106, 113,124, 125, 176, 227, 240

value free see science as value freevalue complex see complex of valuesvalue judgment 2, 5–6, 15, 16, 38–4,

55, 62, 70–3, 74, 76–9, 241, 247,255

values 2, 5, 7, 9–12, 14–20, 21–42, 43,63, 64, 69, 74 74, 81, 99, 101, 127,135, 140–5, 163, 170, 171–8, 178,185, 188, 190, 193, 196, 201–6, 207,218, 226–35, 237, 238–6, 241–50,245, 247, 250–59;adopting 38, 43, 77;articulated 24, 27, 30, 31, 38, 267;degree of manifestation of 39;embodied 18, 25, 38, 39, 40, 108,179, 181, 195, 230, 239;expressed 18, 24, 27, 224, 237–6,242, 246–4;kinds of 26;manifested 10, 17–18, 24, 27, 30,31, 41, 79, 108, 127, 182, 199, 235–6, 251;modes of 24–7, 27–29;personalized 28, 41;traditional 74–7;

woven into 24, 27, 32, 37, 39, 40,43, 48, 128;holding vs. having 23, 26, 34, 38,255;see also cognitive values, complex ofvalues, modern values of control,moral values, personal values, socialvalues

viable value complex 77–81, 189–5,231, 233–43, 238, 241–51, 268;vs. historically and sociallysustainable value complex 79, 237–4, 238

violence 34, 37, 38, 181, 193

world, the see ‘world’, “world”;material world, the

‘world’ 145–79, 206–11, 223, 263“worlds” 14, 147–54, 155, 160–7, 171–

5, 175–9, 206, 209, 223, 263


Lacey_1999_Is Science Value Free.pdf - [PDF Document] (2024)
Top Articles