I always love that kind of argument. The contrary of a thing isn’t the contrary; oh, dear me, no! It’s the thing itself, but as it truly is. Ask any die-hard what conservatism is; he’ll tell you that it’s true socialism. And the brewers’ trade papers: they’re full of articles about the beauty of true temperance. Ordinary temperance is just gross refusal to drink; but true temperance, true temperance is something much more refined. True temperance is a bottle of claret with each meal and three double whiskies after dinner.

— Aldous Huxley, Eyeless in Gaza (London: Chatto and Windus, 1936) pp. 122–23. Cited in Hardcastle et. al. 2006.

Science has always had (…) a metaphoric function — that is, it generates an important part of a culture’s symbolic vocabulary and provides some of the metaphysical bases and philosophical orientations of our ideology. As a consequence of methods of argument of science, its conceptions and its models, have permeated first the intellectual life of the time, then the tenets and usages of everyday life. All philosophies share with science the need to work with concepts such as space, time, quantity, matter, order, law, causality, verification, reality.

— G. Holdton, Einstein, History and Other Passions (2000), 43; 137. Cited in The Scientist as Philosopher.

The Threat of Postmodernism

The tradition of philosophy, natural science, and social science as disciplines that aim at objective knowledge of the way the world is and the place of humans in that world, goes back to the beginnings of all three of these enterprises with Plato, Archimedes, and Thucydides. Modern science, which arguably began in the seventeenth century, appeared to be making steady progress in expanding knowledge and its practical application, first in the natural sciences, then in biology and medicine. During the Enlightenment period of the eighteenth century these achievements encouraged the overthrow of nonscientific and antiscientific traditions in religion and in culture more broadly. They held out the hope of successes similar to those of natural science in what became the social sciences. “Modernism” more generally came to describe this trend across the sciences, arts, and humanities that rejects tradition, extols reason, and seeks human improvement. (Rosenberg 2016, 307-308)

Some thinkers, especially those influenced by Continental philosophy, judged modernism to have fared poorly in the twentieth century, undermined by events in the recent past as an approach to understanding the world and our place in it. Modernism failed to provide either the beneficial outcomes many expected or a way to understand what actually happened in culture, politics, and human life. The fundamental problem, according to the late-twentieth-century movement known as postmodernism, was modernism’s philosophy, in particular its epistemology, the very ideas of absolutes in knowledge, meaning, or truth it inherited from Kant and Descartes and ultimately from Plato. Once the fixity of these categories is surrendered, there is no possibility of a final resolution of any intellectual matter. And this, on postmodernist views, is a good thing. (Rosenberg 2016, 308)

Postmodernism is a difficult doctrine to expound. In Chapter 7, it was noted that social facts are constituted by meanings agreed to and coordinated between people. According to John Searle, all social facts, events, states, institutions, rules, and practices exist through acts of interpretation of behavior, including behavior that comes to constitute meaningful speech when speakers address hearers who interpret the noises and marks that constitute speech and writing. But, asks the postmodernist, if human thought confers meaning, what gives the thoughts of humans the meanings they confer on behavior? Searle’s answer is that thoughts just have meaning; this is obvious to conscious introspection. But the philosophers who developed postmodernism reject this claim. Instead, they hold that what gives thought its meaning is the public vocabulary of words in a language that is used to express these thoughts. But since the statements constructed out of this public vocabulary need to be interpreted also, there is in fact nothing to the meaning or interpretation of any text but some other text, world without end. If everything is subject to interpretation (and an infinite regress of interpretations at that), then there is nothing extra linguistic and nothing to fix the independent truth or falsity of statements about anything. Indeed, talk of truth and falsity is just more text, more interpretation, advanced without any hope of being true but only of being accepted by someone else, under some interpretation or other. (Rosenberg 2016, 308)

But what are the implications of this “insight” for philosophy and science? First, we must surrender the “modernist” idea that there is a relationship between statements and facts in the world that confers truth on some statements and falsity on others. If there is nothing but texts and interpretations, there is nothing to compare texts to, nothing for true texts to correspond to, nothing to reveal the falsity of false ones. So, the notion that science—natural or social—is adequate or inadequate owing to the degree of its success in “mirroring” nature has no basis. Along with notions like adequacy and truth, notions like knowledge and method also lose their grip on a reality that could vindicate them. Instead, the way to understand the succession of scientific theories, models, and explanations is to trace the power, mastery, and hegemony over intellectual domains. (Rosenberg 2016, 308-309)

In some respects the postmodern approach is not so different from interpretationalism and constructivism as methods in social science. It treats actions, behaviors, practices, and combinations of them as institutions having meanings. The difference is that the meanings are not given by facts independent of meaning, but by other meaning-laden facts, and these in turn by others and so on. As each layer of interpretation is applied by some other person, class, race, gender, or other social group, meanings change, but none are final and none are right (and not because all are wrong, but because there really is no right and wrong, or at least none that is independent of anyone’s interpretation). (Rosenberg 2016, 309)

Postmodernism’s critique of social and natural science may be clearest in its attack on “essentialism.” Every discipline taxonomizes its domain, ordering phenomena into kinds, categories, and classes under which it will explain them. Classification is generally viewed as requiring us to identify the necessary and sufficient conditions for falling under a label, for being an instance of a kind, a member of a class. These will together constitute its essential property. The period table of the elements in chemistry is a neat example of apparently successful “essentialism,” the search for essential properties. Postmodernism argues that there are no essential properties, at least none in the social sciences. Essentialism is an error because it mistakenly supposes that there is a binary opposition in nature between things that satisfy the essential properties ordained by a theory, and other things that fail to do so and are therefore defective, disordered, diseased, broken, distorted, and so on. Categorizations and classifications are all arbitrary, matters of (endless) interpretation, and imposed not as a reflection of natural divisions but because their exponents have won some struggle for power in a discipline. So, such distinctions as male/female, right/wrong, democratic/undemocratic, and capitalist/socialist, as well as all other dichotomies (“binary oppositions”) of social sciences, are arbitrary reflections of local and temporary hegemonic interests. There is no fact of the matter about social phenomena for distinctions to get right. Science may claim, with Aristotle, to seek to “cut nature at the joints,” but at least in the sciences of man, there are no joints, no natural divisions, no uniquely correct ways to describe phenomena. (Rosenberg 2016, 309)

This doctrine has a great impact on a naturalistic approach to social science. In the absence of any neat system of classification into types, there can be no regularities, laws, models, or theories about how types are related to one another, and so no macro social science or science of individual behavior. But interpretative disciplines fare no better as purveyors of knowledge, since no single interpretation or narrative about the meanings of human events will turn out to be correct, to be a “totalizing metanarrative” in the postmodernist’s terms. Instead, there is a chaos, disunity, multiplicity, and irregularity in social lives, both because of the meanings, interpretations, and narratives driving, ordering, and directing the lives of social agents in competition with one another, and because any narrative that seeks to accommodate or encompass them all is itself just grist for another interpretation. It will come as no surprise that history is endlessly subject to “revision” and that no revision is more nearly correct than any other. (Rosenberg 2016, 310)

(….) Postmodernists take history seriously. They are especially attracted to the sort of history of concepts, ideas, and meanings developed in the work of Michel Foucault… Following German philosopher Nietzsche, Foucault developed “genealogies” tracing the historical emergence, growing influence, hegemony, and ultimately unraveling and extinction of important ways of thinking about people: reason, prison, pleasure, madness. Tracing genealogies reveals that contingency, arbitrariness, partiality, constructedness of things advertised by science or religion or government as objective, fixed, holy, or legal. Genealogies undermine essentialist metanarratives. But of course these genealogies undermine themselves as well, since genealogy is just another interpretation, itself subject to more interpretation. (Rosenberg 2016, 310)

If we stigmatize this view as anarchistic, skeptical, subjective, arbitrary, antiscientific, the postmodernist will simply note that we are employing essentialist categories to construct a totalizing metanarrative in a vain attempt to gain some political supremacy. They may go on to diagnose our motives, but they will at least have the consistency to accept their diagnoses are simply more interpretations, without special standing. If you are unable to take postmodernism seriously, the postmodernist replies that you are on the right track. The ultimate conclusion of postmodernism is not to take anything seriously. Intellectual, academic, scientific, political debate are all built on false assumptions about truth and its attainability, knowledge versus ignorance, the correctness of some interpretation as opposed to another, the moral rightness of some alternative action, as opposed to another. Once we see that all these binary opposites are essentialist errors, we can recognize that debates in which they figure aren’t really about anything outside of themselves. (Rosenberg 2016, 310-311)

This book and almost all philosophy of social science, indeed, philosophy as a discipline, can no more take postmodernism seriously than it can refute postmodernism in its own terms. In what has gone before and in what follows, we simply assume it is false—a concept postmodernism won’t grant us. (Rosenberg 2016, 311)

Social Science and the enduring Questions of Philosophy

The problems of philosophy of social science are problems for both philosophy and social science. They are problems of philosophy because their ultimate resolution turns on the response to philosophical challenges that have been with us since Plato. They are problems of social science because social scientists inevitably takes sides on them, whether they realize it or not. Moreover, social scientists have defended competing and irreconcilable approaches to their own disciplines by appeal to philosophical theories. As noted in Chapter 2, the claim that philosophical reflection is irrelevant to advancing knowledge in social science is itself a philosophical claim. Social scientists indifferent to philosophy can embrace this view. But unless they argue for it, their view must appear to others to be sheer prejudice. However, an argument for the irrelevance of philosophy is itself philosophy, whether we call it that or not. (Rosenberg 2016, 311)

It should not really be surprising that the social sciences and philosophy are profoundly and indissolubly linked. Like the natural sciences, each one of the social sciences is a discipline that was once part and parcel of philosophy. Indeed, whereas the natural sciences separated themselves from philosophy in the 2,200 years from Euclid to Darwin, the social sciences became independent only during the nineteenth and twentieth centuries. In separating from philosophy, the natural sciences left questions they could not deal with for the philosophers: What are numbers and points? What are space and time? Is there substance? It has been easy for natural scientists to leave these questions to philosophy. They have been busy, especially in the centuries since Galileo, providing more and more detailed knowledge about large numbers of substances at widely separated points of space and time. As Thomas Kuhn noted, natural scientists have turned to philosophy and taken seriously questions about the foundations of their disciplines only during periods of crisis in the development of physics or chemistry. More often than not, the crises have been surmounted by a new piece of technology or a new nonphilosophical breakthrough. These scientific achievements have themselves had philosophical implications. (Rosenberg 2016, 311-312)

Since Newton, advances in physical theory have had a more profound impact on our view of philosophical problems than advances in philosophy have had on the natural sciences. Natural science has forced philosophy to come to terms with materialism, mechanism, first determinism and then indeterminism, relativity, evolution by natural selection, and so forth. Each revolution in the natural sciences has generated new problems for philosophy. (Rosenberg 2016, 312)

But that is not the case in the relationship between philosophy and social science. There have of course been new and original developments in each of the social sciences. But some of these innovations have not met with the uniform acceptance of social scientists that would force philosophy to take them seriously. And the rest of these innovations have not forced philosophy to address new problems in the way natural science has. The direction of influence between philosophy and social science still seems to be from philosophy instead of toward it. We can trace the leading ideas of almost all the social sciences back to the work of philosophers in the seventeenth and eighteenth centuries. This is not just a point about intellectual history. It shows that contemporary social science is much more bound up with the philosophical tradition than is contemporary natural science. (Rosenberg 2016, 312)

More than ever today, social scientists seem to be interested in philosophy, especially the philosophy of science. If Kuhn is right, that is a symptom of intellectual crisis. In the heyday of behaviorism after World War II, methodological reflection was out of favor among psychologists, economists, and other social scientists inspired by their optimism. The philosophy of science was treated as the last refuge of a social scientist incapable of making a “real” contribution to the discipline. It is a matter of some irony that confidence about the prospects for scientific progress was based on almost nothing but a philosophical theory—logical positivism, the latest version of empiricism. That doctrine goes back to the Enlightenment and probably to Plato’s contemporaries. (Rosenberg 2016, 312-313)

Pessimism about a thoroughly behavioral approach to human action drew many social scientists back to a preoccupation with philosophy after 1975. They found in the philosophy of science a number of theories ready to explain both why behaviorism failed in social science and why empiricism is inadequate as a philosophy of science. But that is what another tradition in philosophy and social science had been preaching steadily at least since Hegel in the early nineteenth century. (Rosenberg 2016, 313)

The social scientist’s preoccupation with philosophy of science seems to be another reason to identify the distinctive problem of the philosophy of social science as that surrounding the issue of progress and the allegedly invidious comparisons to natural science. But the practical concerns of the individual disciplines also make salient fundamental issues in epistemology, metaphysics, ethics, and logic. (Rosenberg 2016, 313)

The Unavoidability of Epistemology

The dispute about whether the goal of social science should be predictive improvement or increasing intelligibility is fundamentally a disagreement about the nature, extent, and justification of claims to knowledge. Of course, we’d rather not have to choose between seeking improvement in prediction and making human action more intelligible. Yet insofar as what we seek in social science is knowledge, the choice is forces on us. The demands of predictive improvement rest on a conception of knowledge as justified by its consistency with experience, and not just past experience. It is too easy easy to tailor a theory to be consistent with data that are already in. A theory that can tell us about the actual world must be composed of contingent claims, which the actual world could show to be false. A body of statements that actual events could not disconfirm would be consistent with whatever happens and thus explain nothing. (Rosenberg 2016, 313)

If increasing our understanding of the meaning of human actions improves our predictive powers, then of course there is no conflict. The kind of knowledge that the search for meanings provides will be the same as that which predictively confirmed claims provide. But as we have seen, there are serious obstacles in the way of achieving such predictive improvements in theories that take the search for meanings seriously. We have to decide whether the obstacles are surmountable. If we decide they are not, we face a forced choice between intelligibility and prediction. If we choose intelligibility, we are committed to a fundamentally different epistemology that does not require the same sort of justification for knowledge that prediction provides. Instead, the mark of knowledge that epistemology demands is some sort of certainty or necessity of connections that the mind can grasp. (Rosenberg 2016, 313-314)

Why not simply hold that the house of knowledge has many mansions, that there are many different sorts of knowledge? Social scientists may freely choose among them, for all are equally legitimate ways of expanding our understanding. Some social scientists are interested in knowledge that can be applied to informing social and individual policy and can be used to predict the consequences of planning or its absence. For them, predictions is crucial, and improvements in knowledge are measured by improvements in prediction. Other social scientists have interests to which improvements in prediction are irrelevant. For them, knowledge accumulates by increasing our detailed understanding of a culture or subculture from the inside. Predictive approaches and ones aimed at interpretation are equally valid “ways” of knowing that need not compete with each other. (Rosenberg 2016, 314)

This view sounds like an open-minded attitude of tolerance. But it is just a way of refusing to take seriously the problems social science faces. If there really are many different forms of knowledge, all equally valid, the question must arise: What do they have in common that makes them all knowledge? After all, the term knowledge has to stand for something; it can’t just be an arbitrary label for a heterogeneous collection of intellectual activities that have nothing in common. To suggest that religious knowledge, for instance, rests on revelation, that moral knowledge is justified by intuition, for instance, that scientific knowledge is empirical, that our of human action is based on introspective certainty, and that they are equally legitimate shows not so much tolerance as indifference to the claims of each of these approaches. It is the attitude that anything goes, that knowledge is whatever anyone cares to assert. If a social scientist chooses to seek one of these different kinds of knowledge, there must be a reason given for this choice. Surely it cannot be merely a matter of taste whether improvable generalizations or emphatic insight into intelligibility is the aim of a social scientist’s research program. It cannot be merely a matter of taste what the social scientist will count as good evidence for a theory or explanation advanced in the pursuit of inquiry. And when a social scientist chooses one goal but allows that all other epistemic goals are equally correct, she deprives her own choice of a rational foundation. (Rosenberg 2016, 314)

That does not mean that once we have made a choice, we should not accept or tolerate other choices and other methods as possible alternatives. For our best views of what constitutes knowledge are fallible. Having made our epistemic choice, we could be wrong. But the fallibility of our choice does not entail either that it is the wrong choice or that there is no more evidence for it than for its competitors. (Rosenberg 2016, 314-315)

If we choose to seek predictive improvement or intelligibility of our theories as the mark of knowledge, we must allow others to identify other goals, because for all we know, we might be wrong about what constitutes knowledge. But if we don’t have reasons to support our choice, and perhaps also to oppose theirs, then our choice is not rationally justified. (Rosenberg 2016, 315)

That is what makes epistemology unavoidable for those who hold that the aim of social science is to provide knowledge. Indifference to the issues of epistemology is sometimes a cover for contempt. Some natural scientists, secure in their conviction about what the right methods for attaining scientific knowledge are, express great tolerance about the appropriate methods in social science. They often decline to endorse their own methods as appropriate for the study of human action and social institutions. On their view, “anything goes” in social science. But without good reason to show that human behavior and its consequences are so different from natural phenomena that scientific methods are inappropriate for its study, this attitude is a contemptuous one. It simply disguises the view that the “soft” sciences don’t provide knowledge at all, just the free play of competing speculations, which succeed each other on grounds of fashionableness instead of justification. If social science is to provide knowledge, it cannot be indifferent to what constitutes knowledge. Nor can it accept a permanent agnosticism about claims of incomplete theories of knowledge. (Rosenberg 2016, 315)

Science and Metaphysics

I have argued that the epistemic choice of predictive improvement as a mark of increasing knowledge must make us dissatisfied with intentional approaches to the explanation of human behavior. Similarly, an unswerving commitment to such strategies of explanation will seriously weaken the claims of prediction as an epistemic goal of social science. (Rosenberg 2016, 315)

Either of these alternatives raises fundamental questions about human beings and our place in nature, questions that have always been the special province of metaphysics. For the social scientist, taking sides on these metaphysical questions seems just as unavoidable as it is for matters of epistemology. The interpretative philosophy of social science that exempts the study of humankind from the methods appropriate in the study of the rest of nature must provide an explanation of this exception. And the naturalistic philosophy that absorbs social science into this paradigm must explain away an equally recalcitrant fact about people. (Rosenberg 2016, 315) [See Müller-Kademann 2019 on uncertainty]

Interpretative philosophy of social science teaches that the goals of natural science are inappropriate in the study of human behavior. Another set of aims not recognized in the natural sciences must be substituted. By analyzing the way social science actually proceeds and showing that it cannot proceed in any other way, we may be able to illustrate why the goals of natural science are wrong for the study of humans. But the question is left open of why that is so. Why must the study of humans be different from every other science? It must be because of some fact about us, in particular about our minds, thoughts, consciousness, and the facts of intentionality on which interpretation trades. (Rosenberg 2019, 316) [See Müller-Kademann 2019 on uncertainty]

If, as Descartes held, the mind is a substance quite different from the rest of nature, operating in accordance with different principles, then we have the beginnings of an explanation of why the human sciences cannot proceed in the way the study of matter does. Metaphysical differences dictate scientific differences. Descartes argued that mind is distinct from the body on the grounds that it has properties no chunk of matter could possibly have. His most famous argument was that our minds have the property of of our not being able to doubt their existence, whereas no part of our bodies, including our brains, has this feature. I can well imagine what it would be like to wake up discovering I was missing a limb or even that my skull was empty. But I cannot imagine discovering that I have no mind, for who would make this discovery if I had none? Thus my mind has a property my body lacks: indubitable existence. Accordingly, the mind cannot be part of the body. (Rosenberg 2016, 315)

But this dualism runs into the gravest difficulty with the evident fact that our mental states have both physical causes and physical effects. It is hard to see how something nonphysical can have such relations. [See Stapp 2017] For causation is preeminently a physical relation that involves pushes and pulls. [not on the quantum level, things become far less ‘physical’] It requires the transfer of kinetic energy, which is a function of mass and velocity—that is, matter in motion. But the interpretationalist can turn this mystery to advantage. The impossibility of causal relations between mind and matter explains why a predictive science of human behavior modeled on natural science is impossible: no causation, no laws; no laws, no prediction. [See Stapp 2007, 2017 and Müller-Kademann 2019]

Some will find that such an argument proves too much, for it seems to them beyond doubt that our desires and beliefs have environmental causes and behavioral effects. They may adapt Descartes’s argument to a less controversial but still sufficiently strong argument against naturalism. We may grant that mental states have causes and effects, but the sort of causation involved is not physical and does not consist in generalizations we may improve in the direction of laws. Indeed, the causal relations between mind and matter are singular and irregular. But they reflect logical or conceptual relations between the intentional content of the mind, the statements describing what we believe and want, and descriptions of action. It is these conceptual connections that force a study of meanings on us as the only way to come to grips with the mind and action. (Rosenberg 2016, 316-317)

The explanatory power of such a doctrine rests in large measure on its initial metaphysical assumption that mind is distinct from the body and not a part of the physical world. Unless interpretationalists are content to leave unexplained the distinctiveness of social scientific method, they must face the challenge of substantiating the metaphysical view. (Rosenberg 2016, 317)

The naturalist has the same problem in reverse. Naturalism holds that the mind is a natural object, thus explaining the appropriateness of methods drawn from the natural sciences to its study. As we saw in Chapter 4, that is no easy matter. We have yet no plausible explanation for the most basic naturalism rests on: how physical matter can have intentional content, how one arrangement of matter—the brain—can represent other arrangements of physical matter. Yet if the mind is the brain, that is what our beliefs and desires will be: my belief that Paris is the capital of France must be an arrangement of neurotransmitters at the synapses of a particular part of my cerebrum. Without invoking someone or something to interpret this physical arrangement, it seems impossible to explain how it could represent some state of affairs obtaining in France, thousands of miles from my brain, involving large areas of space and complex legal facts about them. This mystery is just as great as the dualist’s mystery of how nonphysical events in the mind can have physical causes and effects. (Rosenberg 2016, 317)

Merely announcing that the mind is the brain will not make it so. And even if the mind is the brain, we need to understand exactly how it can be, if we are to employ this bit of metaphysics in the explanation of why some methods will be more appropriate than others in the study of the mind and its effects. (Rosenberg 2016, 317)

It would be understandable if impatience with these matters leads some to say that how the brain represents is a matter of science, not metaphysics, and is therefore better left to scientists than philosophers. But his response fails to recognize that science is in fact continuous with metaphysics. Our fundamental conception of the nature of reality and our substantive study of it are on a continuum, and each heavily influences the other. Consider the impact of Newtonian mechanics on metaphysics—determinism, materialism, corpuscularism. Consider the way in which commitment to such metaphysical views led to the expansion of the domain of Newtonian science in the absence of factual evidence of determinism, materialism, corpuscularism. The explanation of the nature of reality that Newtonian metaphysics provided underwrote its scientific strategy long before the evidence for its predictive powers became overwhelming. And finally, reflect on the fact that the overthrow of Newtonian physics had equally strong ramifications for metaphysics and indeed for epistemology. The situation is the same in social science. (Rosenberg 2016, 317-318)

The role of metaphysics may be, in fact, more critical here. For if the social sciences do not have much at present to show in the way of predictive success, then we need an explanation of why they don’t—and perhaps cannot—or we need an explanation of why they will ultimately provide such knowledge. Either sort of explanation so greatly transcends narrow factual matters that it must be metaphysical. (Rosenberg 2016, 318)

Moreover, solving the problem of how the brain actually represents requires first a solution to the puzzle of how it could possibly represent. For without a solution to the conceptual problems of intentionality, we have no hint of where to begin in searching for a solution to the factual problem of connecting psychology and neuroscience. What is more, naturalism needs to solve the metaphysical problem of representation if it is to take our intentional explanations seriously here and now, not in some happy future time when neuroscience has established itself. For in the absence of such solutions, naturalism loses out to interpretative social science as the approach most suited to the study of intentional creatures like us. (Rosenberg 2016, 318)

Of course, one can always opt for the view of Skinner and other materialists who refuse to take intentional states seriously in the first place. Among philosophers, this view has had some currency. Though they hold no brief for the explanatory variables Skinner adopted, they agree that intentional states have no role in adequate scientific explanations and will, in the long run, suffer the fate of notions like “phlogiston,” or “demonic possession.” They will simply disappear from the best explanations of behavior. Such eliminative materialists have their own metaphysical problems, distinct from those of naturalists hoping to accommodate intentional phenomena to, instead of eliminate from, the natural sciences. Perhaps the most serious of these problems is the sheer implausibility of saying our actions are not caused by our desires and beliefs, and that we don’t have sensations or thoughts. This view is so implausible that its denial is often viewed as close to an a priori truth and the most basic premise of interpretative social science (see Chapters 8 and 15). In fact, eliminative materialists have tried hard to render consistent their view that such concepts will disappear from scientific explanations with our first-person convictions that we do have such intentional states. The details need not concern us here [see Dupré 2001, 5-6]. But the argument is as much a piece of fundamental philosophy as that required to justify naturalism or interpretationalism as a method in social science. (Rosenberg 2016, 318)

So, all sides of the dispute about social science and their goals and methods have a metaphysical mystery to deal with. Naturally, social scientists cannot be expected to cease their work and turn to the philosophy of mind. But they have taken sides on these questions by choosing methods that are underwritten by answers to these questions. They cannot pretend that the issues do not concern them and will not in the long run have an impact on the direction of research in the social disciplines. (Rosenberg 2016, 318-319)

Individualism and Instrumentalism

Those who hope to skirt metaphysical issues about the mind fix the agenda of the social sciences to macrosocial facts free from psychology and individual action. They must face equally fundamental questions addressed initially in the philosophy of science and eventually in metaphysics and epistemology. (Rosenberg 2016, 318-319)

Reductionists and methodological individualists face the problem that at least some large-scale social phenomena, their descriptions and their explanations, resist explanation and description in terms of the components that make them up. This fact is hard to reconcile with the reductionism characteristic of the physical sciences. Moreover, the obvious explanation, that such phenomena somehow reflect supra-individual agencies, is difficult to accept or even make sense of if society is composed of individuals and nothing else. Therefore, individualists must search for another way to explain the resistance of social facts to reduction. One strategy is to explain away reference to irreducible wholes as a mistake. That is, however, unconvincing to those not already wedded to individualism. Another tactic is to treat macrosocial theories, not as true or false claims about the world, but as useful instruments, tools for systematizing data, and not to be taken seriously. (Rosenberg 2016, 319)

This approach, however, raises questions that instrumentalism has always faced in philosophy: If these instruments are so good, what is the explanation for their usefulness? And more important, why can we not produce theories that are both good as instruments and true? Are there computational or cognitive limitations on us that prevent us from producing theories in social science that seem, like theories in natural sciences, to be more than just good instruments? Or are all theories natural and social merely tools for systematizing observations? Whichever move the individualist makes leads straight into the philosophy of science and thence into epistemology and metaphysics. (Rosenberg 2016, 319)

The holist is no better off. Holism may justify its extravagant ontology by the instrumental success of holistic theories. But it cannot rest with such justification. It too must explain how social facts, made up of the behavior of individuals, can nevertheless be distinct from individuals. Such explanations are plainly a part of metaphysics. And holism must explain how we can have knowledge of such facts when all that ever meets our eyes is the behavior of individuals. Unless holism takes such questions seriously, its position collapses into the individualist’s instrumentalism and faces the same questions it does. (Rosenberg 2016, 319-320)

Philosophy and the Moral Sciences

Probably little needs to be said to convince us that moral philosophy has a profound bearing on the social sciences and vice vesa. The social sciences were, in fact, at one time known as the moral sciences, and they remain the disciplines that help us decide matters of policy, private and public. The twentieth-century trend, evinced in economics and other disciplines, of divesting the social sciences of a moral voice has never met with general agreement, and through the vicissitudes of the century, the plea for value neutrality has sometimes been reduced to the opinion of a small minority. The majority view that social science cannot be morally neutral is faced directly with the matter of what moral and social prescriptions ought to be offered. (Rosenberg 2016, 320)

In recent years, moral philosophy has been as much a consumer or importer of theories and findings from social science as it has been a producer of and exporter to the social sciences of moral theories about what is right, good, required, prohibited, or permitted. This tendency has reflected the same doubts about a distinction between facts and values that has animated the opponents of value-free social science. There now seems little difference between the language of arguments in political philosophy and welfare economics, for instance. But the philosopher seems less constrained by economic orthodoxy. Political philosophers are prepared to consider the possibility of interpersonal comparisons and perhaps even cardinal utility, notions that have no place among modern mathematical economists. But for those theories to gain acceptance, the arguments that economics has mounted against them must be disposed of. This is certainly a task to be faced by social scientists as well as philosophers who reject the constraints of Pareto optimality. (Rosenberg 2016, 320)

So here the situation is reversed. Social scientists need to concern themselves with moral philosophy both because they cannot avoid ethical issues and because they may have more to say about them than we might expect. In fact, they may be able to provide the kind of information philosophy needs in order to advance and improve its own moral theories. It is, accordingly, an intellectual duty to provide this kind of help. The duty comes with the claim that social science provides, inseparably, normative and factual knowledge. (Rosenberg 2016, 320)

In a way, the moral responsibilities of a normatively committed social science make the classical problems of epistemology and metaphysics even more compelling. As we have seen, choosing between competing methods of pursuing social science heavily tilts our choices about moral theories. Naturalism makes a consequential theory more inviting. Antinaturalism is more sympathetic to a theory of rights and duties than to one of general welfare. So choosing between these moral points of view makes the epistemological and metaphysical problems behind the competing methods even more pressing than their purely intellectual or academic fascination might make them. (Rosenberg 2016, 321)

But even those who hold that social science is at its best free from value judgments and subjective impurities must face moral problems distinctive of social science. These problems are the constraints that ethics places on our research methods, the steps we take to communicate them and their impact on others, as well as the very questions we decide to pursue as social scientists. The moral neutrality of our theories, methods, and epistemic goals, if they are indeed neutral, does not extend to us, the social scientists who pursue these goals. We make choices either self-consciously or by default. The choices seem better made as a result of serious reflection than sheer inadvertence. And such reflection takes the form of moral philosophy and applied ethics. (Rosenberg 2016, 321)

The first thing one learns about moral philosophy is that, like the other divisions of the subject, it too is wracked with controversy and disagreements both fundamental and derivative. Yet in contrast to the case with other areas of philosophy, we cannot remain agnostic for long about these disagreements because they have an immediate bearing on our conduct and its effects on others and ourselves. (Rosenberg 2016, 321)

Conclusion

This introduction is meant largely for social scientists. Its aims have been three: to introduce the traditional problems of the philosophy of social science; to connect these problems with the methodological, factual, and moral choices that social scientists themselves make; and to show how the problems bring together the day-to-day research agenda of the social scientist with the most central, deepest problems of philosophy. (Rosenberg 2016, 321)

The first aim reflects my belief that current controversies in the philosophy of social science are almost always new versions of traditional debates. Sometimes it is difficult to recognize this fact because the jargon has changed and the participants themselves often mistakenly think they have discovered a new issue. Today’s argument between interpretational social science and naturalistic social science reflects the same issues that were debated among Weber and Durkheim, Dilthey and Comte, Mill and Marx, Hegel and Hobbes. That does not mean that current disputes are condemned to perpetual gridlock. Rather, it means that traditional insights bear a continuing relevance. (Rosenberg 2016, 321-322)

The third aim, of making the social scientist see the seriousness and the relevance of questions that daunted Plato, Descartes, Hume, Kant, and their discipline, reflects the conviction that the search for knowledge is all of one piece. But this conviction is also the basis of another aim, which could animate an introduction to the philosophy of social science. This is the aim of encouraging philosophers to recognize the bearing of work in the social sciences to their traditional concerns. If social scientists take sides on philosophical issues in their work, then the findings, theories, and methods of these disciplines must test, and eventually inform, the thinking of philosophers. (Rosenberg 2016, 322)