1
Theory and Practice, Skepticism and Liberalism
The first half of my title assumes that theory and practice are sufficiently distinct to permit consideration of relationships between them. This assumption is sharply controversial. Melding the two concepts, thinkers in the praxis tradition seem to exclude the possibility of a relationship between them. Other writers might replace my conjunction with a hyphen. There are no practices that are innocent of theory, and all theories presuppose some practice of which they are the, or among the, theories. Theories and practices form pairs not unities, but no instance of either of them can be identified or usefully discussed apart from the instance of the other with which it is paired. Numerous others who have reflected on this topic endorse one or another version of the assumption implicit in my title, and some among them further insist that, properly understood, theory and practice are not only categorically distinct but disjunctive in the sense that theory can play no role in practice.
We have to consider these disagreements. But we do not have to accept the view, which might be suggested by the persistence of the disagreements, that reflection about theory and practice can yield nothing better than a list, perhaps with additions, of stipulations. It is not, I think, tendentious to say that the parties to the controversies I have mentioned are advancing theories about theory and practice. Their theorizing occurs in concepts and makes use of ideas that have not been and almost certainly could not be fully appropriated by any of the past, present, or future contributors to it. There is reason for impatience with the topic of theory and practice, for doubt about the richness of its yield, but there is no reason to think that theorizing about it has been or must be merely and hence fruitlessly sectarian.
Similar concerns arise about, and an analogous stance is appropriate concerning, the other elements in my topicâthat is, liberalism and the roles of theory and of skepticism in liberal practice. There is a tendency to think that it is with âliberalâ and âliberalismâ as Hobbes said it was with âgoodâ and âevilââthat is, that these words âare ever used with relation to the person that useth them: there being nothing simply and absolutely so; nor any common rule ⌠to be taken from the nature of the objects themselves.â1 This tendency of thought commits an exaggeration. Leaving aside the historianâs question of just when liberalism emerged, the term is now widely and often vigorously used of ideas and people who hold them, of institutions and policies, of governments and nations. Features prominent in liberalism do give us reason to doubt that we can significantly reduce the diversity of recognizably liberal idioms; certainly there is reason to doubt that we can fully resolve the disputes among self-designated liberals and between them and opponents of liberalism. But this circumstance, commonplace in moral and political philosophy, is hardly reason for dismissing the disputes as spurious or for rejecting the possibility of contributive thinking about them.
There is an easy transition from the foregoing remarks to some preliminary comments concerning skepticism. In its âAcademicâ and âPyrrhonianâ variants,2 respectively, classical general skepticism denied or doubted the possibility of knowledge when âknowledgeâ was taken to require true, warranted, or at least reasonable beliefs and when the beliefs pertinent to knowledge claims were (or are now usually) taken to consist of integrated sets of general propositions. Let us say, at least provisionally, that such an integrated set of truth-evaluable general propositions is a necessary feature of a theory. On this construal of theory, if we embrace general skepticism we will thereby have answered a main question concerning theory and practice. We will either (1) have to deny that theory can or should play any affirmative role in practice or (2) we will have to say that theory can and should guide practice despite the fact that theories cannot be true (dogmatic skepticism) or that thus far no theories are known to be true (undogmatic skepticism). On the widely received position that true or at least well-warranted answers to questions that practice confronts are the distinctive contribution of theory to practice, the stronger of these two conclusions will follow. Either way, the classical forms of general or doctrinal skepticism seem to require us to reject the idea that theory is necessary to practice, and dogmatic general skepticism requires us to reject the possibility that theory can contribute to the quality of practice.
According to Miles Burnyeat, the Pyrrhonian skeptics of classical antiquity drew very definite practical conclusions from their epistemological doubts and denials. Unable to resolve the issues of moral and political life, they sought âtranquillityâ by withdrawing as far as possible from it.3 If we regard skepticism as a theory about the possibility of true or contributive theory, their theory of theory and practice had material implications for both theory and practice. As Burnyeat puts it, there was no âinsulationâ between their theory of theory and practice and their practical judgments.4
Burnyeatâs account suggests several contrasts with more recent thinking on these topics. As he remarks, a number of modern and contemporary thinkers have reached dogmatically skeptical conclusions about moral and political theory but have put a thick layer of insulation between these conclusions and their thinking about moral and political practice. The âfirst-orderâ judgments of moral and political practice are not diminished by the unavailability of true âsecond-orderâ theories of the kind traditionally sought by moral and political philosophers. What is increasingly called âcom-monsense moralityâ gets along nicely without the assistance of such theories. By denying that practice has any need for such theory, these theories of theory and practice insulate practice from the effects of their own skepticism about theory.
An alternative to (variant of?) the position just sketched is provided by âantifoundationalism.â Instead of saying with the classical skeptics that we cannot establish the truth of theories, or that we do not know whether we can do so, antifoundationalists contend that these epistemological questions are misbegotten and that we should address ourselves to other issues.
In Burnyeatâs terms, antifoundationalism might be characterized as putting a double layer of insulation between skepticism and practice. If we adopt this position, we insulate practice not only from (truth-seeking) theory but from that species of theory about theory and practice that is epistemology. Antifoundationalism is a theory about theory and practice which makes not only theory of or about practice but theory about theory and practice (or, more exactly, all theories about theory and practice other than its own) irrelevant to theory and practice.
Wittgensteinâs position, which some regard as a version of this view, will concern us below. For now, however, the more striking point is the lack of success of antifoundationalists in their campaign to banish epistemological concerns from moral and political philosophy. Writers of this persuasion may have moved epistemology to the wings of the general philosophical stage, but the concerns traditional to it remain prominent in the domains of morals and politicsâperhaps particularly so in the work of self- or widely other-designated liberal thinkers. Leaving aside the early history of liberalism, writers such as Isaiah Berlin, Bruce Ackerman, and a bevy of neo-Kantian thinkers led by John Rawls have rejected utilitarianism and what Rawls call perfectionism and insisted on the ineliminable plurality of conceptions of the good and the good life. The voluntarism, subjectivism, or agent-relativity of these views has provoked vigorous rejoinders from other neo-Kantians (for example, Alan Gewirth) and from neo-Aristotelians (Alasdair MacIntyre), neo-Hegelians (Charles Taylor), and communitarians (Roberto Unger), who claim that moral and political theory can and should provide interpersonally and even interculturally valid specifications of the good. And because some of the liberals who have abandoned the search for a true or a best theory of the good have nevertheless tried to circumscribe pursuit of the conceptions of good that people in fact form with general theories of the right, of justice, and of rights, their efforts have prompted skeptical reactions from a diverse array of writers who deny the possibility, the usability, or the desirability of such theories. Very little of this argumentation extends its reach to general skepticism or the âdogmatismsâ that have opposed it, and to this extent these thinkers may show the influence of antifoundationalism. But within the domains of concern here, including within and about liberalism, we continue to find a range of views extending from insistent skepticism about the possibilities of true or otherwise contributive theorizing to soaring aspirations for this mode of reflection. If antifoundationalists claim that the epistemological flavor of this work reveals it to be archaic or even reactionary, they must be pressed to defend the certitude with which they advance their own theories of theory and their own theories of theory and practice.
I
A familiar construal of liberalism, at least of its anglophone variants, gathers the four terms of this essay under the philosophical rubric of empiricism. Most versions of this view posit a hierarchy descending from the general to the particular with epistemology (and sometimes metaphysics) or âfirst philosophyâ governing moral and political theory, which in turn governs moral and political practice. Empiricism correctly determines how we know and what we can and cannot know. Accepting and thinking within these determinations, liberalism as moral and political theory determines how we can know about morals and politics, what we can and cannot, do and do not now know about them. Liberals as moral and political agents act within the determinations of liberalism and attempt to implement their findings and conclusions in moral and political practice.
A slightly modified but forceful articulation of these connections was presented by Bertrand Russell in 1947. After identifying Locke as the founder (âso far as the modern world is concernedâ) of both empiricism and liberalism, Russell asked:
What has theoretical philosophy to say that is relevant to the validity or otherwise of the Liberal outlook? The essence of the Liberal outlook lies not in what opinions are held, but in how they are held; instead of being held dogmatically, they are held tentatively, and with a consciousness that new evidence may at any moment lead to their abandonment. This is the way in which opinions are held in science, as opposed to the way in which they are held in theology. ⌠Science is empirical, tentative and undogmatic; all immutable dogma is unscientific. The scientific outlook, accordingly, is the intellectual counterpart of what is, in the practical sphere, the outlook of Liberalism.5
As with Locke, Hume, Mill, and numerous others, Russell then identifies empirical science with empiricist philosophy: âThe empiricistâs theory of knowledgeâto which, with some reservations, I adhereâis halfway between dogma and scepticism. ⌠Scientific theories are accepted as useful hypotheses to suggest further research, and as having some element of truth in virtue of which they are able to colligate existing observations; but no sensible person regards them as immutably perfect.â6 His conclusion is that âin our day as in the time of Locke, empiricist Liberalism ⌠is the only philosophy that can be adopted by a man who, on the one hand, demands some scientific evidence for his beliefs, and, on the other hand, desires human happiness more than the prevalence of this or that party or creed.â7
It may strike some students of liberalism that, as with related accounts from the same period (accounts by âcold war liberalsâ as they are now, usually with derision, commonly called), Russell modifies understandings that were predominant in the thinking of earlier and some later liberals. As with Russell (and ignoring the transition from âpsychologicalâ to âlogicalâ empiricism), many nineteenth- and early twentieth-century liberals regarded empiricism as the philosophical position that best accounted for and regulated natural science; in at least partial contrast with Russell, however, for them the startling successes of science provided the basis for a robust optimism about moral and political matters. Beginning with the sensationalist-associationist psychologists and early utilitarians, at least in part in John Stuart Mill, and continuing among some of the founders of the Liberal party in Great Britain (for example, Graham Wallas), the ambition of many liberal thinkers has been to bring empiricist-based scientific thinking and investigation to bear on social life, expecting thereby to achieve improvements in that realm at least approximating those Newtonian science had made possible in the understanding and control of nature. If present-day liberals are less likely to celebrate science, aspirations of this sort remain alive among them.
This difference between Russell and his predecessors and successors represents a phenomenonânamely, proponents of a single general philosophical position drawing divergent practical inferences from itâwhich is important to the concerns of this essay. The point for immediate attention, however, is the agreement between Russell and most of his empiricist liberal predecessors concerning the proper relationship between theory and practice.
Empiricism gives an account of our access to reality, and a further account of science as the means of achieving knowledge and truth. Russell puts more stress on the limitations on scientific knowledge than do most of his predecessors (albeit he denies that he does so out of the âsceptical intentionâ of Hume, who of course anticipated him in this emphasis), but he has only minor reservations concerning the main elements of the empiricist epistemology and metaphysic and no doubt that science, as construed by empiricism, gives us the best knowledge we can have.
Theory, then, is construed as ...