Social Science and the Moral Life
Society for the Advancement of American Philosophy
March 13, 2009
In “The Moral Philosopher and the Moral Life,” William James famously remarks, “ethical science is just like physical science, and instead of being deducible all at once from abstract principles, must simply bide its time, and be ready to revise its conclusions from day to day” (James, 1978, p. 625). The consequence of this is that, “Everywhere the ethical philosopher must wait on facts.” James’ claim is that we should apply the same epistemological rigor in our investigation of ethical claims that we apply to our investigations of any other empirical claims. It goes beyond his well-known statement that facts are “shot through” with values to insist that our values are shot-through with claims about facts.
This argument places James, and pragmatism more generally, squarely within one of the central debates in the philosophy of social science, understanding the extent to which scientific investigation should shape values. The position one takes on this matter colors ones entire understanding of the role of the social scientist. On the one hand is the social scientist as engineer, uncovering the laws of cause and effect in the social realm. On the other hand is the social scientist as social critic, using empirical inquiry to uncover not just how relationships are, but how they might be better.
This paper does not deny the obvious difference between stating “P exists” and “P ought to exist.” It does however reject some prevailing views regarding what follows from this difference. Two views are of special concern. First, the view that facts or experience haven no evidential bearing on intrinsic value judgments and hence are irrelevant to the justification of noninstrumental practical principles. Thinkers as diverse as David Hume, Max Weber, and the pragmatist Patrick Baert have alleged that the difference between facts and values is so fundamental that to appeal to one in support of another is considered a logical fallacy. This view gives us a doubled epistemology, with one set of criteria for warranting belief about facts, and another set of criteria for warranting belief about values.
The second view is that ethical principles cannot be relevant to the justification of social scientific theories. This is the converse of the naturalistic fallacy; that statements about facts cannot draw support from statements about values. Our interests in having the world be one way of another have no bearing on whether our statements about the world are actually true. This is the notion that facts are objective, values subjective. If science is to be objective, it must then concern itself exclusively with facts.
The practical consequences of this dichotomy become evident in the vacillations between hard-nosed dedication to facts and idealistic commitment to values; fatalistic conceptions of human nature and romantic notions of the pliability of human conduct; dependence on outside expertise in technical matters and anti-intellectualism in moral ones; steadfast commitment to individual responsibility and active undermining of the structures for its realization. When facts and values are held to different epistemic standards, the moral nature of empirical inquiry goes unexamined, while the material conditions that make values possible remain mysterious. Control over the direction of social policy then becomes a battle between technicians and moralizers, each showing disregard for the others’ authority.
The inability to scientifically investigate facts in conjunction with values results in lost opportunities for redirection and progress. We miss a chance to think about whether certain ends warrant the means necessary to obtain them, or how the kinds of persons we are constrain or make possible certain patterns of conduct and how these patterns of conduct influence the kinds of persons we become, or how a lay public can deliberate on matters in which they have little expertise, or how individuality can be achieved through social interaction. A thorough reevaluation of the relationship between facts and values may help reclaim the social sciences as instruments of social progress. This paper examines how bringing the social sciences into discussions of values can enhance our ability to understand three key relationships: the relationship between means and ends, human nature and conduct, and expert and lay authority.
Two figures loom powerfully the debate over the role of values in social science, Emile Durkheim and Max Weber. Durkheim argues that science can not only help us settle the “casuistic” question about how different goods relate to each other, but can provide a way of evaluating value-claims objectively. He holds that the scientific study of society can help us both identify the means which satisfy our predetermined ends and pass judgment on those ends themselves.
Max Weber strongly resists the move to place science at the center of moral inquiry. Echoing Nietzsche and foreshadowing developments in existentialism, Weber argues that scientific investigation cannot relieve us of the necessity to choose. For Weber, moral decisions are not reducible to empirical hypotheses. On his view, values are ever present in scientific inquiry, from the selection of research questions to the interpretation of results. But though our interpretation of facts in colored by the values we hold, these values are beyond the epistemic limits of empirical inquiry.
Patrick Baert (2005) offers a pragmatic account of the epistemic standards for the adjudication of facts and values, means and ends, technical and moral questions. Baert argues that pragmatism can help cure us of the Durkheimian tendency to wish for a unified science, one that can help settle both technical and moral controversy. Following Weber, he claims that a pragmatic philosophy of social science must clearly distinguish questions of facts from questions of values. The task of this paper is to look at the implications of the epistemic dimensions of the fact/value dualism. Drawing on the work of Peirce, James, and Dewey, and more recently, Hilary Putnam and Elizabeth Anderson, I argue first that this dualism is untenable, and second that it strips social science of much of its social relevance.
The first section of this paper summarizes Durkheim’s argument that the methods and epistemic standards of natural science must be maintained if there is to be a social science. The second section deals with Weber’s claim that facts provide no epistemic support for claims about values. The third section discusses various pragmatic responses to fact/value dualism, with special attention to the work of John Dewey, Patrick Baert, and Hilary Putnam, and Elizabeth Anderson.
Durkheim argues that the question concerning the relationship between facts and values, “is of the utmost importance, for on its solution depends one’s conception of the role that science, and above all the science of man, has to play” (Durkheim, 1982, p. 85).
Anticipating post-positivists generations beyond him, Durkheim argues that “every means is an end” (1982, p. 86). Since there are typically multiple avenues leading to any given end, to describe any one as “best” is to make a value judgment. Should we treasure speed or safety? Less risk or more reward? Maximal enjoyment or minimal cost? The operationalization of terms such as “costs” and “benefits” is only possible once value judgments have been made. Thus, Durkheim argues that if sociology, “cannot guide us in the determination of our highest ends, it is no less powerless to determine those secondary and subordinate ends we call means” (Durkheim, 1982, p. 86).
Durkheim argues that the only way to avoid these problems is if the value of various social arrangements can be objectively observed through scientific study. But before the sociologist can confront the epistemological question of when she or he can claim that something is good, she or he must first confront the most basic question of ethical theory: what is good? If we can state specifically what the good is, then we may create a method by which to tell if a given social arrangement corresponds with it. Further, we can articulate an epistemology which can guide us in deciding when our claims that something is good are justified.
Durkheim argues that the definition of good cannot be posited a priori, but must be discoverable through an analysis of actual conditions. On his view, such analysis reveals that for, “societies, as for individuals, health is good and desirable; sickness, on the other hand, is bad and must be avoided” (Durkheim, 1982, p. 86). With the definition of goodness established, the sociologist now needs, “an objective criterion, inherent in the facts themselves, to allow us to distinguish scientifically health from sickness” (Durkheim, 1982, p. 86). That is, she must find some observable marker which clearly indicates when a social phenomenon is healthy rather than sick.
Durkheim claims that generality is the objective criterion by which health can be distinguished from sickness. To determine whether a phenomenon is good for a given social group at a certain time, the sociologist examines whether that thing is common to the entire species to which that social group belongs. For instance, if the practice of punishing criminals is common among all social groups at a certain stage of their development, then we may safely determine that the practice is a sign of health. Just as the average child during certain stages takes on qualities which though unpleasant, indicate normal development, societies do the same. And just as physiology concerns itself with the average child, and calls pathological all that deviates from the average, the sociologist seeks to discover what the average is for a given social species, and then uncover the pathological deviations from this norm.
For Durkheim, facts and values are on equal epistemological footing. To make a substantive evaluation of the worth of a phenomenon is make a claim about a fact about it. The process for the verification of factual and normative claims is the same; it is a process of examining what properties are common to the species and then seeing the extent to which a given specimen differs. According to Durkheim, this allows, “science…to throw light on practical matters while remaining true to its own method” (Durkheim, 1982, p. 86). Unlike positivists, Durkheim believes that normative claims are indeed verifiable. Indeed, he claims that if sociology cannot help verify claims about what we ought to do, then it has no justification for its existence. In contrast to idealists, he believes that this verification occurs only through systematic empirical inquiry. Moral science, as he refers to it, is an inductive rather than deductive discipline.
In his 1913-1914 lectures on pragmatism, Durkheim states that certain elements of his thought are compatible with the general thrust of pragmatism. Though he strenuously objects to what he perceived as an overly constructivist account of truth and reality, he largely agrees with William James’ famous statement that the moral philosopher must everywhere “wait on facts” before she issues her final judgment (Durkheim, 1983, p. 54). Durkheim shares with James the basic commitment that values exist only in a specific social context. As James observes, it is equally true that a world without people is a world value, and a world without value is possible only in a world without people (James, 1978).
Yet more important for the purpose of this paper is Durkheim’s insistence that facts and values, means and ends, are not ontologically distinct classes. Dewey makes a similar point in his claim that the difference between means and ends is better thought of as a difference between nearer and remoter ends (Dewey, 1988). If values are not different in kind from facts, this lends plausibility to Durkheim’s claim that judgments about both should be subject to the same epistemic standards. This provides some defense against accusations that he is guilty of the naturalistic fallacy—that is, that he is attempting to ground a claim about the way the world ought to be in a claim about the way the world is. For Durkheim, the statement that a given society ought to do is only intelligible as a statement that this society is constituted such that this action promotes its health. Just as it is a matter of fact that peanuts are toxic to certain humans, some social practices are toxic to specific societies. The sociologist is not merely a cultural engineer, she is a cultural doctor; she must recognize the appearance of health, sickness, their respective causes, and their remediation. Her claims about what signifies health are on epistemic parity with her claims about its causes.
Patrick Baert (2005), in his discussion of Weber’s relationship to pragmatism, points to two central aspects of his understanding of the role of values in social inquiry. The first is Weber’s claim that sociology does not emerge ex nihilo; social inquiry arises due to concrete needs and values, and the basic questions which form its substance are shaped by these. Weber’s second claim is that though social inquiry may find its genesis in values, it does not itself terminate in value judgments. That is, sociology cannot inform us as to which modes of social organization are best in a given context.
Thus, in a limited sense, the sociologist is able to maintain a position of value-neutrality. The epistemic standards for an inquiry into factual matters are distinct from those involved in making value-judgments. Weber’s reliance on this distinction is clear in his use of ideal types; while the construction of an ideal type depends on the researcher’s interest, whether or not a given phenomenon is accurately described by an ideal type is an objective matter of fact. Whether it would be better or worse for the phenomenon to correspond to the ideal type is a subjective question and unfit for scientific inquiry. Weber insists that ideal types be treated as the researcher’s logical constructs, rather than accounts of reality or as claims concerning value.
For Weber, sociology is an exclusively technical discipline. To repeat the metaphor used above, the sociologist is an engineer rather than a doctor. The sociologist seeks knowledge of what is, not what ought to be. Weber states:
There is no (rational or empirical) scientific procedure of any kind whatsoever which can provide us with a decision here. The social sciences, which are strictly empirical sciences, are the least fitted to presume to save the individual the difficulty of making a choice, and they should therefore not create the impression that they can do so (quoted in Baert, 2005, p. 45).
Thus, Weber seems entirely committed to a kind of epistemic (if not ontological) fact/value dualism. Claims about facts must be justified by an appeal to empirical findings; claims about values can be justified only by appeal to some non-rational, arbitrary point, such as tradition or desire. We are unable to refer to experience when seeking epistemic support for value judgments. Baert argues,
Weber was right to point out that…people cannot resort to social science…in order to judge between competing values. This is precisely where Durkheim went astray. No study about social inequality could help to ascertain whether equality is worth striving for….Weber was right to argue that no reference to what is the case (das Seiende) could help clear up what ought to be (das Seinsollende) (Baert, 2005, p. 55).
Baert then gives the familiar argument that the attempt to ground ought-claims in is-claims leads to infinite regress. Durkheim allegedly commits this error by stating that the good is health, and health consists in normality (sickness, then, consists in deviancy). While normality is described as stability and harmony, Durkheim never explains why stability and harmony are preferable to change and conflict. According to Baert, this shows a sly bias toward conservatism in Durkheim’s thought. But more importantly, it shows that Durkheim actually grounds value claims in other values claims, not in facts. Baert claims, with Weber, that there is no empirical way to verify that stability is better for society than change. Value claims ultimately rest on non-empirical, non-rational commitments, and are revisable only in light of other non-empirical, non-rational objections.
Baert and Putnam on Facts and Values
One of Baert’s central criticisms of Weber is that he does not go far enough to do away with the notion value-neutrality. He argues that a pragmatist philosophy of social science would totally do away with the notion of value neutrality. Cognitive interests would then be seen as determinative both of the research questions and also the research process. Because cognitive interests vary—for instance, in some cases we want to explain the process of photosynthesis, in others we want to know whether to start a neighborhood organization—we should not expect to find a single method which is appropriate for answering all questions. Most importantly, we should not assume that the various methods used to answer questions about the way the world is are useful in answering questions about the way the world ought to be. There are “epistemological discontinuities” which indicate the lack of a “neutral algorithm” for handling any given question (Baert, 2005). As he states, questions of method always come back to questions of aims. The epistemic standards expected for a given claim depend on the purpose which the statement is expected to serve.
However, it seems to me that it is one thing to say that social science cannot furnish a neutral algorithm for the adjudication of both factual and evaluative claims; it is another to say that this therefore makes the evaluation of the two kinds of claims different kinds of affairs. I am more or less in agreement with Hilary Putnam that Baert’s doctrine—that there are epistemic discontinuities between fact and value claims—commits us to the situation described by Lionell Robbins in which debates about ends become a question of “thy blood or mine.” Putnam argues along the line of Amartya Sen—a social scientist of the first rank—that, “when we are dealing with any important value disagreement, we assume that facts are irrelevant at our peril” (Putnam, 2002, p. 78). Those who seek to inform ethical inquiry with social scientific study are not seeking to find a master-method which mechanically dispenses answers about means and ends. On Putnam’s view, they are rather insisting that the kind of objectivity our ethical claims require is the ability to withstand the sort of criticism that arises in an actual situation (Putnam, 2002, p. 94).
Catherine Elgin (1997) explains in considerable detail how value judgments be both relative and objective. She states, “The demarcation of facts depends on considerations of value, while assessments of value are infused with considerations of fact,” and further, that both kinds of judgment are “relative and objective” (Elgin, 1997). They are relative in the pragmatic sense (or in her jargon, the “constructive nominalist” sense) that they are driven by human needs and purposes. They are objective in the sense that they can do a better or worse job at serving those needs and purposes. Thus, for instance, econometric models can be relative, value-laden, and still objective, as can normative statements about how to interpret the results of econometric analysis.
Baert is correct to point out that value judgments can only be criticized from the standpoint of some other value judgment, but this is a problem only if one had hoped to arrive at a categorical rather than merely hypothetical judgment. There is no reason to proclaim a priori that the field of overlapping beliefs shared between those in a community can not serve the same adjudicatory role as a neutral algorithm.
Holding facts and values to different epistemic standards leads to several unfortunate consequences in social science. First, it suggests that some issues are completely immune from empirical counterevidence. Such a belief rests upon two mistakes: first, an overly narrow conception of science, and second, a confusion of science-as-such with certain tools used in some areas of scientific inquiry. If we maintain that the differences between means and ends are matters of emphases, then we can liberate the social scientist from being a mere technician for the achievement of externally imposed ends. It would be paradoxical and perhaps tragic if agency and responsibility could be preserved only by renouncing the use of intelligently guided inquiry. If, as Baert argues, there simply is not a clear point in which we stop dealing with facts and start dealing with values, there can be no clear way to decide which of his two epistemic frameworks is most applicable.
Dewey’s argument against treating facts and values as fundamentally different appears in greatest detail in the chapter “The Construction of Good” in Quest for Certainty (Dewey, 1988). Practically speaking, statements about “what is” tend to play a different role in experience than statements about “what ought to be.” “What is” statements indicate some characteristic about the object as it is immediately given. To borrow Dewey’s example, “He ate an apple” is a description of an event taken in relative isolation from other events. It offers little direction for future conduct. But the statement, “The apple is edible” involves a judgment about the consequences that follow from the event. On Dewey’s view, this is all there is to the difference between facts and values. One involves taking things as they are; the other involves taking things in their relations to antecedents and consequences. It is the same with saying, “I value honesty” and “Honesty is valuable.” Or, as Larry Hickman (Hickman, 2007, pp. 158-161) puts it, facts and values are “products of discriminative inquiry,” and rather than being distinct classes, are “moments” in the same deliberative activity. The former expresses the event in its barest form; the latter expresses the event’s significance.
On Dewey’s view, science is the art of understanding and controlling the relations between antecedents and consequences. The difference between the scientific study of natural as opposed to social questions is due to the degree of complexity of the relations under investigation, not due to any antecedent division between “natural” and “social” objects. There is real peril involved in making inferences between what is and what ought to be. But the peril does not arise due to a basic difference between science and non-science, but between easy and difficult science. It is the peril we face any time we move from a relatively simple account of an isolated situation to a prediction about the multitudinous consequences arising from manipulating some part of it. Every significant step in the experimental method involves this leap from what we have observed to what we might observe if we make some alternation to present circumstances.
In is worth noting that Dewey and Peirce sharply disagree on this point. Peirce argues that his logic applies strictly to scientific reasoning, which in turn applies to matters of fact, not value. The grounds for this limitation are somewhat shaky, given Peirce’s own comment in “Fixation of Belief” (1992) that the failure to employ the method of science is most often due to obstinacy of lack of imagination. Yet in “Philosophy and the Conduct of Life,” Peirce is emphatic that science “has nothing directly to say concerning practical matters, and nothing even applicable at all to vital crises” (1998, p. 33). A. J. Ayer argues that Peirce’s position is “curious,” given that (1) it seems we can have genuinely serious doubts about ethical matters and (2) that these doubts are “bound up with our beliefs about matters of fact” (1968, p. 22). Perhaps Ayer finds Dewey’s position more coherent.
The mistaken dichotomy between facts and values gives rise to the dichotomy between means and ends. It is supposed that objective inquiry extends only to means, while ends are taken as final and beyond intelligent debate. But if Dewey is correct, then ends are simply means seen in light in of their conditions and consequences. A serious debate over whether one ought to do something is not won by appeal to a standard external to the situation, but by appeal to the connection between the actions in question and the preparations necessary to make it possible as well as the consequences which may follow. Only under such empirical scrutiny does the value of a value become apparent.
Following Dewey, Elizabeth Anderson argues that fact/value dualism has harmed both philosophical and social scientific inquiry. She states, “On the part of philosophy, it has led to serious neglect of the importance of evidence and experience to the justification of ethical principles. On the part of the human sciences, it has led to a failure to reflect on what it takes to justify the ways they classify phenomena” (Anderson, 1998, p. 15). On Dewey’s view, investigation into means is not merely an analysis of whether an end is attainable; rather, it is an attempt to fully comprehend the end as end. It is tempting to see the difference between the two activities as the difference between thinking about what is to be done versus how. However, on Dewey’s view analysis of means is a crucial component of reasoning about what is to be done. That is, analysis of means is indispensible to deliberation about the significance and desirability of an end. He argues that we have not really comprehended the end until we see it as a series of “what nexts” (Dewey, 1982, p. 29). We do not understand it as end, much less whether it is possible to achieve. We do not fully know what the ends are until we know how they can be brought about, and what consequences they may beget.
Dewey echoes nearly word for word Durkheim’s claim that “every means is an end” (Durkheim, 1982, p. 86). Like Durkheim, Dewey holds that judgments of ends have the same cognitive status as judgments of means. However, unlike Durkheim, Dewey does not believe that the standards for judging means can be furnished externally. Durkheim posits a universal criterion for health and sickness, namely, generality and deviance, which bear no internal relation to concrete societies. Dewey argues that the criteria for evaluating ends comes from within the situation itself. Dewey provides a unique contribution to the Weber-Durkheim debate in sociology. He defends the critical potential of social science both from Weber’s non-cognitivism and Durkheim’s submission to external standards.
Ethical questions about what we ought to do are forever entangled with questions about what kinds of persons we are and can be. By extension, questions about social policy are bound up with practical questions regarding what exists and what is possible. This argument builds on Dewey’s discussion of the valued—what we actually want—and the valuable—what we want after careful consideration. Dewey does not deny that there is a difference between questions of facts and questions of values, or questions of what is and what ought to be. By extension, there is a useful distinction to be made between human nature and human conduct. The former concerns questions about existing conditions, things over which we have little control. The latter concerns questions about things which are presently malleable and undetermined. Human nature captures our more durable qualities, those which do not change quickly or easily. They exert a powerful influence on our understanding of the world and its possibilities. However, these natures are to a great extent the result of past conduct. They are established through accretion. Their durability is the product of habit and custom having worn deep channels in our private and public ways of being. Our conduct flows like a river through its channels; though the river bank shapes the flow of the river, we do well to remember that it is also the product of that river, rather than existing antecedently.
So much the same may be said for the relationship between facts and values. The facts are what they are, and we must be cognizant of them as we shape our values. However, so much that is said to be a matter of fact—for instance, that humans act so as to maximize their satisfactions, or that poverty leads to educational failure, or that children with single mothers are more likely to suffer illnesses—is the product of past valuations. On Dewey’s view, it is not so much the “old Adam” that causes humans to strive for their personal good above all others, but a mode or organization that isolates individuals and forces them to wage financial or physical battle against one another. Competition is no more indigenous to human nature than cooperation—a point one should recommend both to the capitalist and the socialist. There is no outward economic system which best expresses our inner natures. However, there might be economic systems which better resolve the concrete problems arising from interdependence.
The separation of the processes by which means and ends are evaluated creates a rift between social science and the public, or between expert and lay criticism. The rapid multiplication and increasing technical sophistication of research designs, methods, and reporting styles make much social research inaccessible to non-specialists. A constant danger looms in the development of science. If social science remains an occult practice open only to the initiated, democratic deliberation over social policy quickly gives way to technocratic manipulation by a priestly, privileged class (Manicas, 1998, p. 11). As Dewey remarks in “The Sources of a Science of Education” (1984), people do not cease to desire power the moment they become scientists. If we assert both that what ought to be depends largely on what is and that social science gives us vital insight into what is, then we risk excluding the vast majority of citizens from discussions of public ends.
The current strategy for mitigating this danger has been to give authority in all technical matters to experts, and authority in all moral matters to laypersons. Or philosophers. Or preachers. Each party wins something. The expert’s isolation from the broader discussion of aims insulates her from criticism about her assumptions and purposes. The layperson is spared from the difficult task of clarifying the full significance and entailments of her projects. She is allowed to “think big,” which, as the comparative sociologist Philip Foster once commented, is too often an excuse for not thinking at all. As Dewey remarks, moral discussion becomes popularly associated with lame complaints about what is and exhortations about what ought to be (Dewey, 1989, p. 232). In such an atmosphere, absolutism is allowed to thrive, as any ideal can be propagated so long as the collateral consequences of pursuing it remain unexamined.
However, the mere
aggregation of information may not produce epistemic benefits. Individuals may hold narrow or partisan
agendas, or may make unwarranted generalizations from their own
If experts assume undue authority over ends, democracy unravels into technocracy. If experts are excluded from discussions of ends, public discussion becomes dominated by narrow, obstinate, entrenched interests. The solution lies, I think, not in a more clever system of checks and balances between two competing groups, but from the reconstruction of society so as to eradicate the forces which sustain the division. As Philip Kitcher notes, one of the major problems facing democracy is that, “citizens are both extremely poorly informed and divided from any serious engagement with the needs, interests, and aspirations of one another” (2006, p. 1220). At the heart of the problem is the failure to educate people so as to be able to use and critique the methods for examining public problems, or to even form a public which can understand and respond to its own problems. To a large extent Dewey’s, Democracy and Education (1985) is an attempt to sketch a way of remedying this situation. This requires educators to emphasize the logic and methods of social inquiry, teaching not just history but also historiography not just social science but also qualitative and quantitative analysis. Only through the democratization of the means of social criticism can the tension between expert and lay authority be resolved.
Perhaps more than any other writer, John Dewey worked to understand the full potential of intelligent inquiry to transform human relationships. He worked to develop William James’ observation that moral philosophy cannot be made up in advance. Yet now, as in Dewey’s time, the scientific method has been used primarily in the service of predetermined values. It has only occasionally entered into our evaluations of those values themselves.
This paper suggests that a better understanding of the relationship between facts and values can lead to a more direct line between social science and social progress. I have not meant in any way to denigrate the place of conceptual labor, the kind that has long gone on under the title of “philosophy.” However, I have insisted that systematic observation and experiment are unavoidable. People will act in some way or another. Experts and laypersons will deal with each other in one way or another. Matters will be left private or managed publically. And these interactions will produce consequences, some pleasing and others not. The more we can know about these consequences and how to control them, the more we can speak intelligently about them.
Anderson, E. (1998). Pragmatism, Science, and Moral Inquiry. In R. W. Fox & R. B. Westbrook (Eds.), In Face of the Facts: Moral Inquiry in American Scholarship (pp. 10-39). Cambridge: Woodrow Wilson Center Press and Cambridge University Press.
Anderson, E. (2006). The Epistemology of Democracy. Episteme, 3(1-2), 8-22.
Ayer, A. J. (1968). The origins of pragmatism: studies in the philosophy of Charles Sanders Peirce and William James. San Francisco,: Freeman.
Baert, P. (2005). Philosophy of the Social Sciences: Towards Pragmatism. Cambridge: Polity Press.
Baez, B., & Boyles, D. (2009). The Politics of Inquiry: Education Research and the "Culture of Science". Albany: SUNY Press.
Beatty, J. (2006). Masking Disagreement Among Experts. Episteme, 3(1-2), 52-67.
Biesta, G., & Burbules, N. (2003). Pragmatism and Educational Research. Lanham, MD: Rowman and Littlefield.
Blanshard, B. (1978). The nature of thought. London, New York,: Allen & Unwin, Humanities Press.
Bohman, J. (1991). New Philosophy of Social Science: Problems of Indeterminacy. Cambridge, Mass: MIT Press.
Bohman, J. (1996). Public deliberation : pluralism, complexity, and democracy. Cambridge, Mass.: MIT Press.
Bohman, J. (2006). Deliberative Democracy and the Epistemic Benefits of Diversity. Episteme, 3(1), 175-191.
Ciaffa, J. A. (1998). Max Weber and the problems of value-free social science : a critical examination of the Werturteilsstreit. Lewisburg, NJ: Bucknell University Press.
Dewey, J. (1982). Human Nature and Conduct. In J. A. Boydston (Ed.), John Dewey: The Middle Works, 1899-1924 (Vol. 14). Carbondale: Southern Illinois University Press.
Dewey, J. (1984). The Sources of a Science of Education. In J. A. Boydston (Ed.), John Dewey: The Later Works, 1925-1953 (Vol. 5). Carbondale, IL: Southern Illinois University Press.
Dewey, J. (1985). Democracy and Education. In J. A. Boydston (Ed.), John Dewey: The Middle Works, 1899-1924 (Vol. 9). Carbondale: Southern Illinois University Press.
Dewey, J. (1988). The Quest for Certainty. In J. A. Boydston (Ed.), John Dewey: The Later Works, 1925-1953 (Vol. 4). Carbondale, IL: Southern Illinois University Press.
Dewey, J. (1989). Liberating the Social Scientist. In J. A. Boydston (Ed.), John Dewey: The Later Works, 1925-1953 (Vol. 15, pp. 224-238). Carbondale and Edwardsville: Southern Illinois University Press.
Durkheim, E. (1982). The Rules of Sociological Method and Selected texts on Sociology and its Method (W. D. Halls, Trans.). New York: Free Press.
Durkheim, E. (1983). Pragmatism and Sociology. Cambridge: Cambridge University Press.
Elgin, C. Z. (1997). Between the absolute and the arbitrary. Ithaca, N.Y.: Cornell University Press.
Goldman, A. (2001). Experts: Which Ones Should You Trust. Philosophy and Phenomenological Research, 63(1), 85-110.
Hickman, L. A. (2007). Pragmatism as post-postmodernism : lessons from John Dewey (1st ed.). New York: Fordham University Press.
Howe, K. (2003). Closing Methodological Divides: Toward Democratic Educational Research. Boston: Kluwer Academic Publishers.
James, W. (1978). The Moral Philosopher and the Moral Life. In J. J. McDermott (Ed.), Writings of William James. Chicago: University of Chicago Press.
Joas, H. (1993). Pragmatism and Social Theory. Chicago: University of Chicago Press.
Kitcher, P. (2006). Public knowledge and the difficulties of democracy. Social Research, 73(4), 1205-1224.
Manicas, P. (1998). John Dewey and American Social Science. In L. Hickman (Ed.), Reading Dewey. Bloomington: Indiana University Press.
Peirce, C. S. (1992). The Fixation of Belief. In N. Houser & C. Kloesel (Eds.), The Essential Peirce: Volume I (1867-1893). Indianapolis: Indiana University Press.
Peirce, C. S. (1998). Philosophy and the Conduct of Life. In N. Houser & A. Tienne (Eds.), The Essential Peirce: Volume 2 (1893-1913) (pp. 27-41). Bloomington and Indianapolis: Indiana University Press.
Phillips, D. C. (2000). The Expanded Social Scientist's Bestiary. Lanham, MD: Rowman and Littlefield.
Putnam, H. (2002). The Collapse of the Fact/Value Dichotomy. Cambridge, Mass: Harvard University Press.
Stone, G. (1994). Dewey on Causation in Social Science. Educational Theory, 44(4), 417-428.
 My description of these views follows Elizabeth Anderson, “Pragmatism, Science, and Moral Inquiry” (1998).
 While the authors mentioned above are my objects of attention in this essay, it is worth mentioning that there have been several recent attempts by others to work on issues in the philosophy of social science from a pragmatic perspective. For some of the diverse approaches which go unmentioned in this paper, see for instance Benjamin Baez and Deron Boyles (2009), Gert Biesta and Nicholas Burbules (2003), Kenneth Howe (2003), Hans Joas (1993), Denis Phillips (2000), and George Stone (1994) , to name just a few.
 One of Putnam’s central examples about the entanglement of facts and values is the role of epistemic values in science. He argues that epistemic values (such as verifiability, beauty, simplicity) can be usefully distinguished from ethical values. This seems to me to be an instance of Putnam avoiding the central problem, which is to figure out how we ought to settle our ethical beliefs. It seems rather obvious that epistemic and ethical values are at least as “entangled” as epistemic values and facts.
 This is a considerable departure from some other efforts to bring moral and aesthetic experience into the domain of empirical investigation. For instance, Brand Blanshard (1978) is primarily concerned with whether aesthetic experiences (such as finding a rose to be beautiful) can be examined as psychological phenomena. But while it is clear that he believes that psychology can investigate aesthetic and moral experiences, it is less clear what role psychology can play in evaluating the value of those experiences, that is, “what we ought to appreciate” (1978, p. 121).
 James Bohman presents a somewhat different argument for abandoning the distinction between facts and values as well as similar distinction between explanation and criticism or description and evaluation (1991, pp. 186-231). Following Weber, he argues that social science must seek to understand human values. However, following Habermas, he objects to Weber’s insistence that social science cannot rationally criticize those values. Bohman’s main objection is that Weber’s fact/value dualism relies on a problematic non-cognitive account of values (1991, p. 230).
Bohman has also argued, again following Habermas, that “no normative conception of politics or law can be developed independently of a descriptively adequate model of the complexity of contemporary society,” including the “asymmetries of competence, expertise, and the availability of information” (1996, p. 13). That is, any normative statement about social conditions must reference actual rather than merely ideal conditions. It subsequent writings he has outlined an “epistemic difference principle,” which holds that any asymmetries in deliberative resources ought to benefit the least well off the most. One might argue that the merits of this principle are to be determined by its use in actual rather than ideal societies (Bohman, 2006).
For a partial defense of the Weberian position regarding the impossibility of scientifically informed social criticism (other than “technical criticism”), see chapter five of Jay Ciaffa, Max Weber and the Problems of Value-Free Social Science (1998). Ciaffa tries to separate Weber’s fact/value dualism (which he endorses) from his ethical noncognitivism (which he rejects).
 Though I focus on social scientific research, these arguments are applicable to most scientific inquiry. For an interesting discussion of the tension between scientific experts and the lay public, see John Beatty, “Masking Disagreement Among Experts” (2006). Beatty does an especially good job of dissecting the features of communication between scientists and the lay public, such as the simplification of complex procedures and the attending withholding of information.
Alvin Goldman’s “Experts: Which Ones Should You Trust” (2001) discusses some of the various ways in which novices may be able to assess the veracity of expert testimony. Its chief accomplishment is the presentation of a clear definition of expertise (possession of an “extensive fund of knowledge” and “a set of skills or methods for apt and successful deployment of this knowledge to new questions in the domain” (2001, p. 92). The rest of the paper presents a rather tedious elaboration of the obvious—i.e., novices can assess the opinions among experts and meta-experts as well as evaluate track records and credentials. Yet it appears by the conclusion of the article that Dewey is again waiting at the end of the road. Goldman eventually arrives at a series of practical questions (though he proposes no answers): “What kinds of education, for example, could substantially improve the ability of novices to appraise expertise, and what kinds of communicational intermediaries might help make the novice-expert relationship more one of justified credence than blind trust” (2001, p. 109). In short,, the lay/expert question is best posed as an educational and social problem of enabling a citizenry to be able to conduct social inquiry.