The Flight from Positivism

Before the intellectual and political upheavals of the Sixties, many practitioners of the social disciplines had begun to convince themselves that they were well on the way to establishing a genuinely “scientific” method for the study of social life. But since that time, as the English sociologist Anthony Giddens has remarked, it has come to be widely agreed that “those who still wait for a Newton” of the social sciences “are not only waiting for a train that won’t arrive, they’re in the wrong station altogether.” The main aim of Professor Richard J. Bernstein’s survey The Restructuring of Social and Political Theory is to chart the course of this progressive disenchantment, and to ask whether it ought to be characterized as an intellectual advance or merely a failure of nerve.

Bernstein begins with what he calls “mainstream social science,” outlining the elements of the “scientific” orthodoxy which has recently been so widely repudiated. Such distinguished mainstream theorists as Robert K. Merton and Ernest Nagel took the subject matter of the social disciplines to be a realm of objective social facts. These were generally held to be logically separate from the values of the researcher, and capable of being investigated and described in a wholly neutral way. This empiricist assumption was often combined with emotivism, the doctrine which holds that all statements of value are reducible to expressions of emotion, and are thus devoid of any cognitive content. The outcome of linking these two beliefs was that the normative study of social and political principles was left for dead, while the white-coated social scientist was encouraged to get on with the accumulation of more and more facts.

The other central feature of the mainstream outlook was an account of the logic of explanation, one which embodied two of the most cherished theses of positivism. All explanations were held to be deductive in form: a puzzling fact was said to be explained if and only if it could be shown to follow from a known natural law. And the business of explanation was taken to be isomorphic with prediction: once a law was found to “cover” a particular fact, it could equally well be used to predict its recurrence.

Drawing on the work of Merton, Nagel, and others, Bernstein elaborates this vision of “empirical theory” in an opening chapter of great lucidity and fair-mindedness. The only quarrel one might have with it is that the influence of these assumptions, even in the course of the Fifties and Sixties, is perhaps overestimated. It is true that Bernstein quotes some fine dissenting judgments from Sheldon Wolin and Isaiah Berlin, both refusing to be bullied into agreeing that moral and political choices no longer formed a proper subject of rational debate. But he could have added that a similar commitment to normative political theory was also kept alive, even in the heyday of positivism, by two major schools of Anglo-American political thought. In England Michael Oakeshott and his followers continued to vindicate the rationality of their political values by reference to the political tradition itself. And in America Leo Strauss and his numerous disciples never ceased to insist on the immediate moral relevance of the classic texts in the history of political philosophy.

A second way in which Bernstein overstresses the hegemony of the mainstream model is through his decision to exclude history from his list of the social disciplines. Historians have always been more concerned than other social scientists with explaining individual voluntary actions—with questions like “Why did Brutus kill Caesar?” But it never seemed plausible to suppose that, in order to answer such a question, the historian has to look for an empirical law governing the assassination of heads of state, from which the explanation of the individual case can then be deductively derived. So it is not surprising that many philosophers of history refused to endorse the mainstream dogma that all valid explanations must include the citation of empirical laws. Many of them continued to insist on the propriety of ad hoc explanations in which an appeal is made to individual reasons rather than to general causes. R.G. Collingwood offered a classic defense of this viewpoint in The Idea of History, and a similar commitment was later upheld by such theorists as William H. Walsh, Michael Scriven, and William H. Dray. Bernstein claims in his survey of mainstream arguments that although he has “focused primarily on the work of sociologists, the same basic story could be told throughout the social sciences.” But this appears to overlook the important role played by the study of history in undermining positivist confidence.

These are really quibbles, however, for Bernstein is not primarily concerned with this background, but rather with the way in which the authority of the mainstream model has suddenly collapsed in the course of the past decade. He finds the deepest roots of this intellectual revolution in the traditions of German phenomenology. Husserl is singled out as the seminal figure, but Bernstein mainly concentrates on the work of his pupil Alfred Schutz, whose Phenomenology of the Social World was first published in English in 1967. Schutz lays a special emphasis on the extent to which the social world is a domain of “typifications,”—i.e., meanings established in subjective relations between people—and is thus “constructed” (rather than merely investigated) by those who live in it. (For example, in our world everyone agrees that a red flag “means” danger.) This leads Schutz to insist that the mainstream emphasis on social facts involves taking the social world too much “for granted,” and fails to consider how its meanings come to be constituted and maintained. Schutz also argues—very much as Max Weber had already done more than a generation earlier—that an adequate explanation of a social action can often be furnished simply by recovering these intersubjective meanings and linking them to the presumed motives and intentions of individual agents. (If a red flag means danger, the person waving it presumably means to warn us.) This approach in turn poses an obvious challenge to the view that all social explanations are “nomological,” a matter of seeking causes and laws. So Schutz’s phenomenology—questioning the preoccupation with laws as well as facts—issues in a repudiation of both the key assumptions of mainstream social science.

The next challenge is said to have come from “Anglo-Saxon analytical philosophy.” This label is used to cover a rather heterogeneous body of writers, but Bernstein is chiefly interested in two developments. One is the analysis of social action undertaken by a number of Wittgenstein’s avowed followers, especially Peter Winch in The Idea of a Social Science. Here Bernstein presents us with something of a twice-told tale, partly because Winch’s account of actions as meaningful items within a shared “form of life” is curiously similar to Schutz’s, and partly because Winch’s views have already been so extensively discussed and criticized.

More interesting is the other issue Bernstein takes up, the impact of post-empiricist philosophies of science on the social disciplines. Here he concentrates on examining the prodigious influence exercised by Thomas Kuhn’s account of The Structure of Scientific Revolutions. Kuhn tells us that it is a mistake, even in the physical sciences, to suppose that we can ever hope to describe the facts we investigate in a neutral observation-language. All the observations and theories which go to make up the content of “normal science” are dependent upon a particular “paradigm,” a set of assumptions which serves to establish what are to count as facts, what are to count as problems worthy of investigation, and what are to count as their solutions. As Bernstein observes, these arguments have been taken to constitute a devastating critique of mainstream social science. If there are no brute facts at all, but only facts as filtered through the current paradigm, then the very idea of developing a neutral science of society appears to be discredited.

Finally, Bernstein studies the attack on mainstream social science which has arisen out of the recent revival and development of Marxist ideas, especially in the hands of the so-called Frankfurt School. Here he focuses in particular on the work of Jürgen Habermas, beginning with his rejection of the assumption that nomological science yields the only valid form of knowledge. According to Habermas’s account in Knowledge and Human Interests, there are three distinct forms of knowledge, each of which serves a corresponding “cognitive interest.” There is empirical science, which is directed toward a technical interest in control over nature. There is the type of understanding to be gained from the interpretative study of social action, which is said to serve the practical interest of improving our awareness of intersubjective meanings and symbolic structures. For example, a historian who attempts to recover a “world of traditional meanings” will tend to provide us with a self-understanding which connects us with that earlier world. And finally there is the “critical” knowledge that Habermas himself is mainly concerned to promote. This is said to transcend both the other forms of understanding and to serve the emancipatory interest of freeing us from false consciousness and the domination of forces beyond our control. With Habermas, the mainstream insistence on the validity of neutral analysis is rejected with the greatest deliberation and vehemence: the whole point of “critical theory” is said to be that of acting as a guide to revolutionary practice.

Anyone who wants to understand the profound changes that have overtaken the social disciplines in the course of the past decade ought to read Bernstein’s book. He examines an extraordinarily wide range of theories with the most scrupulous care, in each case adding a series of extremely perceptive criticisms. Many of the writers he discusses are chiefly famous in English-speaking circles for their obscurity, but he always manages to present their arguments with clarity and elegance. And in dealing with all these highly complex and disparate materials, he never for a moment loses control, but keeps the argument moving forward with an unfailing briskness of pace.

Since Bernstein is such an expert and congenial guide, my chief complaint is that the pace is sometimes altogether too brisk. In the account of critical theory, it is a pity that there is no discussion of either Adorno or Marcuse, especially since their influence in English-speaking countries has surely been far greater than that of Habermas. And in the survey of German phenomenology it would have helped to be given a much fuller account of the tangled history of the movement. Very little is said about the place of Max Weber in the story. Yet his concept of an interpretative sociology has been well known in England and America at least since the Thirties, largely because of the translations and commentaries of Talcott Parsons and Edward Shils. Moreover, it is arguable that Weber’s emphasis on the importance of verstehen, on the recovery of “subjective meanings,” has acted as one of the most potent solvents of the belief in a purely nomological social science. For it has served to direct attention away from the mere investigation of regularities, and to underline the importance of studying the concepts used by “acting individuals” in the understanding of their own social experience. Nor are we told in any detail about the influence of Husserl. Yet his concept of a transcendental phenomenology—allied with Heidegger’s antiscientism—appears to have had a strong impact on the development of interpretative social theories, especially in America. A word could perhaps have been said in this context about the writings of Hannah Arendt, as well as the more recent school of “ethnomethodologists” inspired by the work of Harold Garfinkel.¹

But it is in the discussion of the analytical tradition that one chiefly feels the need for a more detailed treatment. One of the most interesting features of recent moral philosophy has been the repudiation of any form of emotivism, and the revival of the Kantian attempt to connect the moral with the rational point of view. Bernstein’s failure to discuss this development is especially surprising, since it has already helped to bring about a dramatic resurgence in normative political theory. R.P. Wolff has made a strong appeal to Kant’s ethical theory in his essay In Defense of Anarchism; Robert Nozick has offered a reworking of a Lockean theory of natural rights in his Anarchy, State and Utopia; and John Rawls has turned to Rousseau as well as Locke and Kant in developing his Theory of Justice, carrying on their contractarian approach in a systematic treatise of tremendous range and power. After being pronounced dead, normative political theory in the grand manner is very much alive again. It is curious that Bernstein doesn’t seem at all inclined to join in the celebrations.

These are not very serious criticisms, however, for it is obvious that any attempt to chart the complex cross-currents of recent social and political philosophy will be bound to be highly selective in its coverage. The more important question is whether Bernstein is right in his assumption that the three schools of thought on which he has chosen to concentrate have in fact made the greatest contribution to the “restructuring” process he describes. Here I do feel an element of disagreement with his account. Although his survey is exceptionally acute and illuminating as far as it goes, he ought I think to have left space for the inclusion of at least two further themes.

One of these is the role of hermeneutics. Bernstein continues to endorse the Weberian belief that the understanding of social life must be a matter of putting together the explication of social meanings with the investigation of causes and regularities. So he tends to overlook the contribution of those who have sought to revive the more radical thesis of hermeneutics as enunciated by Dilthey and his followers: the thesis that the study of social life is a study of meanings and interpretations as opposed to a study of causes.

In commenting on the theories of Peter Winch and A.R. Louch,² Bernstein appears to dismiss this commitment as something of an extravagance. But it has clearly gained an increasing measure of support in the course of the general flight from positivism. Hans-Georg Gadamer has recently revived the claims of hermeneutics in their most imperialist form in his major treatise on Truth and Method, first published in 1960 and now available in English. According to Gadamer the goal of the human sciences is not merely to achieve “understanding,” but also “to understand the universe of understanding” through engaging in an essentially historical critique of our prevailing conceptual schemes. He insists that this process of self-interpretation cannot be combined with, since it stands in complete contrast to, the nomological methods of the natural sciences. What is said to be at issue is “an experience of truth that transcends the sphere” of scientific methodology, and “resists any attempt to change it into a method of science.”

A similar commitment to a purely hermeneutic approach has recently been defended by some of the most prominent social theorists in England and America. Clifford Geertz has insisted with great eloquence in The Interpretation of Cultures that the discipline of anthropology must not be seen as “an experimental science in search of law,” but rather as “an interpretative one in search of meaning.” And Charles Taylor has argued in The Explanation of Behavior that any account of human action must always focus on the concept of intentionality, and that all such explanations are irreducible in principle to a causal form. Moreover, Taylor has explicitly appealed to Gadamer’s authority in his more recent work, arguing that “the sciences of man” are fundamentally hermeneutic as opposed to nomothetic in character.³

The other major topic which Bernstein overlooks is the influence of structuralism. He remains content to assume that the explanation of social change must always include at least some reference to the motives and intentions of individual agents. But the most striking achievement of the French structuralists has been to issue a strong challenge to precisely this belief. Michel Foucault has sought to demonstrate in The Archaeology of Knowledge that although the “discourses” or belief-systems of any given society will always be “in perpetual transformation,” such changes will always be “anonymous and without a subject,” since they will not be the outcome of anyone’s conscious intentions or strategies. And Louis Althusser, while disdaining the title of structuralist, has recently sought to found a complete Marxist theory of society on the same structuralist assumption that “history is a process without a subject.” Focusing on the cliché that men make their own history, Althusser asks how it comes about that the men who make their history are themselves made. He answers that they are made by the existing “practices” of their societies, and that these rather than the actions of allegedly voluntary agents form the only appropriate subject of inquiry for a social scientist. Such frontal assaults on the newly fashionable Weberian categories of subjective meaning and social action have already begun to have a marked impact even on a number of English-speaking social scientists. Such theorists as Barry Hindess and Paul Hirst have largely adopted Althusser’s program, while Perry Anderson has begun to issue a remarkable series of historical studies in which the underlying methodology appears to be a distinctively Althusserian combination of “the autonomy of the political” with the thesis of economic determinism.⁴

It would belittle Bernstein’s achievement to leave the impression that he is merely concerned to tell the story of the decline and fall of mainstream social science. Like any good storyteller he has a moral to convey, and his book is subtly organized to make this clear. The analytical philosophers are said to have arrived most confidently at the crucial category of social meaning, but without managing to develop their insights in a systematic way. The phenomenologists are said to have achieved a more systematic approach, but without making the necessary transition to a critical perspective. Among current social philosophers, the only one who has thoroughly assimilated both these approaches, as well as moving beyond them to the statement of a theory which is at once “empirical, interpretative and critical,” is said to be Jürgen Habermas.

The moral of the story is thus that Habermas is its hero. It is true that Bernstein offers a number of extremely acute criticisms of his recent work. But the ideal of “critical theory” is still said to be the one most worthy of our allegiance. Bernstein is particularly attracted by Habermas’s current efforts to evolve a theory of what he calls “communicative competence.” According to Habermas, even the most ideologically distorted forms of communication may be said to imply and anticipate a model of “unrestrained” and purely rational discourse. Once we grasp the nature of this “ideal speech situation” underlying our current reports of social life, we can hope to overcome the deformations our society imposes on our perceptions, and thus arrive at a fully rational and “emancipatory” critique of our prevailing beliefs and practices. A genuinely reflective understanding of our language is thus taken to be the key to unlocking our false consciousness, enabling us to see beyond the confines of our present “form of life” to an untrammeled vision of “truth, freedom, and justice.”

At this point, however, it is hard not to feel assailed by some rather serious doubts. Does Habermas’s project really make sense? If there is one lesson we have thoroughly learned from Quine, Kuhn, and the post-empiricists, it is that the evidence we cite in favor of our beliefs is not sufficient to determine them. What we are prepared to count as evidence depends on the beliefs we already hold. But this appears to leave us in a world of rival and perhaps incommensurable systems of thought, each of which will be susceptible of rational defense, while none of them will be capable of being appraised from a genuinely neutral point of view.

We can readily agree with Bernstein that the only way out of such a world will be to formulate “a more comprehensive theory of rationality” by which we can then hope to discriminate “those structures which are presumably fixed, permanent and a priori from those which have specific historical roots and causes.” But how can such a comprehensive theory be evolved? How can we possibly hope, by using (as we are bound to) our own local assumptions and canons of evidence, to construct a theory which is then employed to criticize those precise assumptions and canons of evidence? Doesn’t it begin to look as if the project of “critical theory” amounts to little more than a version of the Indian rope-trick?

I do not mean to imply that there is no hope of answering such relativist doubts. I am only suggesting that, if Bernstein wishes to hold up Habermas’s work as a model to other social theorists, he will have to provide a much fuller account of how such doubts can in fact be allayed. Admittedly this is to ask him for nothing less than a further book. But the present one is so valuable that such a continuation could only be immensely welcome.