Is Science Value Free?

Values and Scientific Understanding

Hugh Lacey

• Publisher: Routledge
• Year: 2005
• eBook ISBN: 9781134619740
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

Exploring the role of values in scientific inquiry, Hugh Lacey examines the nature and meaning of values, and looks at challenges to the view, posed by postmodernists, feminists, radical ecologists, Third-World advocates and religious fundamentalists, that science is value free. He also focuses on discussions of 'development', especially in Third World countries. This paperback edition includes a new preface.

1
Introduction

The idea that science is value free

The idea that the sciences are value free has long played a key role in the self-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 not interpenetrate. The one shows us to what goal we should aspire, the other, given the goal, teaches us how to attain it. So they never conflict since they never meet. There can be no more immoral science than there can be scientific morals.

(Poincaré 1920/1958:12)

Science and values only touch; they do not interpenetrate. To deny this is often perceived as to challenge that science is the pre-eminent or exemplary rational endeavor, to demean the cognitive credentials of science and to undercut its claim to produce knowledge. Lately, however, it has been much contested from an eclectic variety of viewpoints: feminism, social constructivism, pragmatism, deep ecology, fundamentalist religions, and a number of third world and indigenous people’s outlooks. Exactly what is at issue does not always emerge clearly in these contestations. The rhetoric tends to be at high volume, but the argument thin. Incommensurability seems to reign. From one viewpoint, the mounting threat of multiple irrationalities and empty voluntarism looms large; from the other, the entrenchment of ideologies.

I will attempt to sort out what is at stake in the contestation of “science is value free,” an idea that incorporates several distinct views about ways in which science and values do (ought) not interpenetrate. But those who affirm it have always recognized Poincaré’s distinction, and held that science and values touch in various ways with more or less significant effects. Too often the critics point only to aspects of the touch, but even when the focus is on alleged interpenetrations a further ambiguity arises. For “science is value free” in general hardly represents a fact. Perhaps it represents an idealization of fact.¹ It also represents a value, a goal or aspiration of scientific practices and a criterion for appraising its products and their consequences. The fact and value components cannot be separated. To the extent that “science is value free” represents a fact, or an idealization of a fact, that is because “science is value free” has been held as a value; and its being held as a value is without foundation if it is not possible for it to be increasingly manifest in fact. Thus to refute “science is value free” it is not enough to display cases where it is not manifested in fact; rather, the cognitive resources of the practices of science must be assessed for their ability and likelihood to bring about its manifestations increasingly and systematically.

In this introductory chapter I will provide an overview of the various sources of the idea that science is value free, leading to the proposal that it should be regarded as constituted by three component views: impartiality, neutrality and autonomy. Then I will outline some of the important ways in 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.

SOURCES

“Science is value free” has several sources. Its kernel is present already in the works of Galileo and Bacon. Galileo (1623/1957:270) refers to “the facts of Nature, which remains deaf and inexorable to our wishes”; and Bacon affirms, warning us to be alert to the “Idols of the mind,” the sources of error to which we are prone: “The human understanding is no dry light, but receives an infusion from the will and affections; whence proceed sciences which may be called ‘sciences as one would’” (Bacon 1620/1960: Aphorism 49).

Metaphysical/Galilean

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

Let me summarize it in contemporary dress. The world, “the facts of nature,” the spatio-temporal totality, is fully characterizable and explicable in terms of “its underlying order”—its underlying structures, processes and laws. All objects belonging to the underlying order can be fully characterized 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 the underlying order, they are part of no meaningful order, they have no natural ends, no developmental potentials, and no essential relatedness to human life and practices. Values—and objects, qua objects of value—are not represented as emergent from the underlying order of the world.

An object may come to acquire value through its relationship to human experience, practice, or social organization, but any role it plays there is played in virtue of its causal powers, of what it is qua part of the underlying order of the world, so that for explanatory purposes that it may have 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 a consequence of human causal agency. But, the view maintains, the structures, processes and laws that make up the underlying order of the world are ontologically independent of human inquiry, perception and action; they do not vary with the theoretical commitments, outlooks, interests or values of investigators. On this view, it is a “fact” that value derives from an object’s relationship to human experience, practice or social organization (that human agents generate value), and so this “fact” is explicable, in principle, in terms of underlying structure, process and law. But it does not follow from a theory that explains this “fact” that human agents themselves are objects of value.

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

Thus we arrive at one dimension of the idea of the “neutrality” of science: scientific theories have no value judgments among their logical implications. They cannot, it is said, for they contain no value categories. A second dimension is often taken to follow: that accepting a theory has no cognitive consequences at all concerning the values one holds. A third dimension is suggested too: that scientific theories are available to be applied so as to further projects linked with any values. After all, they represent “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 in fact they do not serve to inform the projects motivated by particular values, that is an entirely contingent matter. Notice that this last claim rests uneasily with another that has been heralded in the modern scientific tradition, that science serves especially well the projects of material progress; and it clashes strongly with those world-views (that “progress” intends to supplant) that consider the world to be infused with meaning or value.

Epistemological and methodological/Baconian

In contrast to the Galilean, the Baconian source is epistemological and methodological. Again I summarize a contemporary version. It is through experience that we gain access to the world, which can be considered a complex repository of possibilities, of which the ones that are realized may be (increasingly) connected with our practices and our planned interventions. 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’s repository of possibilities. Considerations derived from values cannot determine what is possible. We find out what is possible only in the course of engagement with the world, through successful practices, including most importantly experimental ones. A scientific theory aims to encapsulate whatever it can of this repository of possibilities of the domains of phenomena within its compass; hence, the centrality in methodology of experiment.

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

The Baconian source of impartiality is often complemented by a view about the nature of scientific inference, or about how empirical data are related to theories so that they can serve as evidence for accepting theoretical posits, or choosing which theories to accept. The view is that scientific inference can be reconstructed in terms of accordance with certain formal rules (Chapter 3). The rules mediate between empirical data and theories in such a way that following them leads to unambiguous choices about which theories to accept, reject or deem as requiring further investigation, or at least to unambiguous assignments of degrees of confirmation to theoretical hypotheses. They provide, as it were, the means to transfer intersubjective acceptance from the available data to the theory. While there has been at times widespread agreement that scientific inference, and any rational inference, can be explicated in terms of formal rules, there has never been anything approaching unanimity about what the rules are, or even about whether they are deductive, abductive, statistical, inductive, or some combination of these. Bacon himself is usually interpreted as holding that they are inductive.

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

The success of modern science

Both neutrality and impartiality concern the content of what is posited in scientific 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 two ideas tend to fuse.² In order that there be any scientific knowledge, the Galilean idea needs to be complemented with a methodology (or procedures that can give it empirical content); and, methodologically, since objectivity cannot be had directly, intersubjectivity seems to be the best available substitute. Conversely, Baconian methodology is deployed characteristically in testing theories that meet requirements derived from the Galilean idea, although the Baconian idea itself encompasses any inquiry that is systematic and empirical (Chapters 5 and 8).

The fusion of the Galilean and Baconian ideas underlies the manifest success of modern science. Bacon promised that utility would follow from deploying his methodology. That is not what I have directly in mind. Rather it is the manifest success of modern science in increasing “the stock of knowledge.” One may identify this success primarily in terms of the discovery of objects (for example atoms, electromagnetic radiation, viruses, genes) or of the definitive entrenchment of some relatively circumscribed theories (for example the heliocentric theory of planetary motion, theories of molecular chemistry, theories of the bacterial and viral causation of diseases, theories that explain the workings of instruments). Such knowledge, of course, has been widely applied in practice: in technology, in medicine, in interpreting various phenomena of the world of daily experience; and, successful application is powerful confirming evidence in support of the knowledge. Items in the stock of knowledge have been accepted in accordance with impartiality and so their cognitive claims are compelling regardless of what values one holds. The sustained success of modern science, as it were, speaks unambiguously to the strength (but not to the certainty or unrevisability) of its cognitive claims.

A claim that is accepted in accordance with impartiality is binding regardless of the values that are held—so that the presuppositions of all practices, and the beliefs that inform all actions, should (rationally) all be made consistent with it.³ This “binding equally” should not be confused with what I have earlier called “neutrality” with its three dimensions: “consistent with all value judgments,” “no (cognitive) consequences in the realm of values,” “evenhandedly applicable regardless of values held”; nor with the stronger view (Chapter 3) that all practices and actions, regardless of the values they are intended to further, should be informed by scientific knowledge to the extent possible. Neutrality presupposes impartiality; and, when the Galilean and Baconian ideas are fused—especially if the metaethical and logical views described in the next two subsections are endorsed—it may appear that impartiality implies neutrality. But, I will argue (Chapter 4) “binding equally” is not consistent with endorsing all three strands of the idea of neutrality; and subsequent attempts to revise neutrality into a coherent thesis confront numerous difficulties.

Metaethical

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

Logical

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

Both the Humean and Baconian schemata, however, draw attention away from some other logical relations involving fact and value. From fact (especially as it is represented in scientific theory) one can infer certain matters 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 logically interact with important implications for working out the idea of neutrality in detail.

Practical and institutional

The fundamental sources of the idea of scienc...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Preface
• Acknowledgments
• 1: Introduction
• 2: Values
• 3: Cognitive values
• 4: Science as value free: provisional theses
• 5: Scientific understanding
• 6: The control of nature
• 7: Kuhn: scientific activity in different ‘worlds’
• 8: A “grassroots empowerment” approach
• 9: A feminist approach
• 10: Science as value free: revised theses
• 11: Conclusion
• Notes
• Bibliography