![]()
Part I
Political theory and political science
![]()
1 Deduction, explanation, and social scientific inquiry (1969)
I. Introduction
The purpose here is to explore certain aspects of the philosophy of science which have serious implications both for the practice of social and political science and for understanding that practice. The current relationship between social science and the philosophy of science (or the philosophy of the social sciences) is a curious one. Despite the emergence of a considerable body of literature in philosophy which is pertinent to the methodological problems of social science, there has been a lack of ostensive ties between the two areas. A justified concern with the independence of social scientific research has contributed to a tendency toward isolation which is unfortunate in view of the proliferation of philosophical problems which necessarily attends the rapid expansion of any empirical discipline. Although in the literature of contemporary social science there are frequent references to certain works in the philosophy of science and to philosophical issues relating to methodology, these are most often in the context of bald pronouncements and shibboleths relating to the nature of science, its goals, and the character of its reasoning. But what is most disturbing about the fact that social scientists have little direct and thorough acquaintance with the philosophy of science is not merely that there has been a failure to carefully examine the many logical and epistemological assumptions which are implicit in social scientific inquiry, since this task might normally and properly be considered to be within the province of the philosopher of science. It is more significant that this unfamiliarity has, paradoxically, obscured the extent to which certain doctrines originating in the philosophy of science have been uncritically insinuated into the enterprise of social science as informing categoricals.
Much of the theoretical literature in political and social science, as well as attempts to explicate the epistemic features of social scientific investigation, rest on a belief that the activity of the natural scientist is an appropriate model for understanding and prescribing the role of the social scientist. Although it may be granted that social and natural science must be distinguished in terms of such characteristics as technique and subject matter, this is often understood as explaining certain inherent or temporary limitations of social science. Generally it has been assumed that with regard to the logic of explanation, social science must be fundamentally symmetrical with natural science if it is to count as science. Although there are a number of common features that would support the idea of symmetry, such as a mutual empirical orientation and concern for systematic explanation and description, and although philosophy, science, and everyday life may be analytically as well as existentially differentiated as modes of thought and activity, this view is misconceived to the extent that it tends to postulate a hierarchy of epistemological structures with impermeable boundaries and fails to take account of some of the more generic aspects of explanation. As a consequence of this misconception, social scientists have been reluctant to move beyond the realm of natural science when seeking a model of inquiry even when the requirements of explanation in the two spheres seem hopelessly disparate. The result has been principally either an imperious attempt at assimilation or an assignment of social science to some inferior position within the general class of activities understood as science. Both positions are seriously defective, and a more adequate formulation is possible only by breaking down some of the alleged constraints which are assumed to be endemic to scientific inquiry.
A conception of social scientific explanation can be elaborated which will not only support the case for logical asymmetry between the natural and social sciences and at the same time preserve the characteristics essential to any mode of empirical inquiry but will better illuminate what most social scientists in fact do in producing explanations as well as what sort of explanations are required by the character of the phenomena with which they are concerned.1 But the problem of supporting the contention that natural science is inadequate as a paradigm governing the conception of methodology in the social sciences is complicated by the fact that what is often construed by social scientists as natural science is actually an ideal typification of the logic of science which is the invention of certain philosophers of science. This philosophical reconstruction has been accepted in one form or another by many social scientists, especially those concerned with the problems of theory construction, as an adequate representation of the character of all scientific explanation and as a prescriptive norm for inquiry. What the social scientist accepts as his model is, then, itself a model and one that, although popularly received for many years, has recently been severely criticized within the philosophy of science itself not only as a reconstruction of the logic of scientific explanation in general but even natural science in particular.
A thorough critical analysis of this model, the so-called deductive model, is a necessary prologue to any further evaluation of the thesis of logical symmetry between explanation in the natural and social sciences as well as an explication of the substantive character of social scientific explanation, and such an analysis is the principal, although limited, concern of this chapter. Consideration of the deductive model also provides a vehicle for examining certain more general problems about explanation and the relationship between social science and the philosophy of science. Since social scientists, and especially political scientists, have been influenced, either directly or indirectly, by the deductive model and other aspects of the philosophy of logical empiricism from which it emanates, there is not only the obvious question whether social scientists have correctly understood this construct and this school of thought but the question of the validity of logical empiricism, even if correctly understood, as an approach to the philosophy of science (and social science). Finally, and possibly most important, there is the problem of the extent to which the philosophy of science may be understood as an enterprise charged with establishing a theory of science which can serve as an authoritative guide for substantive empirical inquiry and a norm for the evaluation of particular explanations.
II. The deductive model
The persistent, pervasive, and yet largely unexamined assumptions that the logic of any enterprise claiming the title of science must be symmetrical with that of natural science and that there is universality in the logic of empirical inquiry, as well as the conception of the logic of explanation in natural science accepted by most social scientists, have for the most part been derived either explicitly or implicitly from certain formulations in the philosophy of science. In seeking to emulate natural science, “the model taken for imitation has been the artifact of the philosopher of science rather than the actual practice of scientists.”2 This formulation, the deductive model, has become the basis of the modern “textbook” view of science. Although in many instances it is difficult to judge whether social scientists have drawn their model of inquiry directly from the philosophy of science or whether they have merely unreflectively accepted some popular version of that model, the latter hypothesis seems more tenable in view of the numerous and basically unexplicated statements in the literature of social science to the effect that “the task of science is the reasoned interpretation of experience through the discovery of valid generalizations and the application of such generalizations to particular events.”3 The precise meaning of such assertions is seldom unpacked; they are inserted as ritualistic formulae. But whether the conception of natural science which is entertained is once or twice removed from that activity, it is clear that social scientists have rarely attempted either to defend in any rigorous manner the idea that there must be symmetry in the logic of inquiry or to systematically support the notion of the exemplary character of natural science. Yet, more specifically, there has been a failure both to justify the deductive model as a proper reconstruction of the logic of natural science and to demonstrate its applicability to empirical science as a whole, including social science.
The argument to be advanced here is that the deductive model is deficient in most relevant respects. It presents an incorrect, or at least otiose, reconstruction of the logic of natural science, and there has been no adequate demonstration that it is applicable either as a norm for social scientific inquiry or as an explication of social scientific explanation. But unfortunately this argument can be set against social scientists only indirectly, since they themselves have not developed a defense of the deductive model. The battle must be joined on the ground of the philosophy of science, and this would be entirely the proper ground for such a discussion if the question were merely one of developing a general structural description of social scientific explanation, that is, if it were not for the extraordinary fact that much of theory construction in the social sciences takes the philosophers’ deductive model as a standard, however poorly approximated and understood. This assertion may invite the charge of setting up a straw man since the defense of the deductive model in principle as well as its applicability to social scientific explanation has been conducted essentially by philosophers of science while there is little in the way of explicit advocacy by social scientists.4 But although a separate treatment would be required to thoroughly support the intellectual history of the influence of the deductive model and logical empiricism on social science, it is assumed that this influence is not a matter of serious dispute and that it can be readily recognized in a variety of works dealing with methodological problems of social and political science as well as popular treatises on theory construction and specific empirical studies.
Some of the most prestigious theoretical literature in political science contains numerous phrases such as “scientific method,” “scientific enterprise,” “scientific credo,” “the rigorous logic and rules of science,” “scientific reasoning,” and “scientific rules of procedure,” but little concrete elaboration is presented. Despite professions of faith in the unity of science and an enumeration of attributes often associated with natural science such as “systematization,” “quantification,” and the search for “regularities,” there is a singular absence of argument which would support the assumption of the logical equivalence of social and natural science, and there are few explicit references to either the philosophy of science or the actual practice of natural scientists which would give tangible meaning to these vague assertions about the defining characteristics of scientific activity.5 When David Easton, in The Political System, challenged political science to become truly scientific, it was far from clear what he meant in substantive terms. Notwithstanding the plea for the development of an empirically grounded “causal theory,” which in his view was the only indication of “the attainment of reliable knowledge,” and despite his later pronouncements about the “gargantuan strides” which have been taken during the intervening period toward the goal of a social science that would compare favorably with the natural sciences, the meaning of “science” in the sense of a logic of inquiry has remained nebulous.6 Neither in 1953 nor in the 1960s has Easton either defended the accuracy of his conception of natural science or detailed the philosophical view on which this conception is based. Yet it was assumed that the character of this anticipated theory would be “deductive” and that although a theoretic framework which “might reach the stage of maturity associated with theory in physics” where “from a few basic premises, empirically derived, it has proved possible to formulate deductively a whole body of intermediate theory and from this in turn, to predict the occurrence of empirical events,” might be a long way off, such a “theoretical system” was to be the goal and standard of inquiry in political science.7
For political scientists, the meaning of “science” is seldom interpreted beyond such formal, and essentially empty, equations of explanation and “generalization” and statements to the effect that explanation can be understood as “simply the subsuming of a particular event or class of events … under a more general law or hypothesis,” or “deriving it from some more general proposition that describes a regular sequence which we believe to hold true in the world of politics.”8 It is from the philosophy of science, and from a rather well-defined group of scholars within that discipline, that political and social scientists have ultimately derived their views about explanation and the requirements of scientific logic. The works of such philosophers as Carl Hempel and Ernest Nagel9 have become philosophical primers for social scientists and are continually cited as presumably unimpeachable authority to support statements about the purpose and function of science, the symmetry of natural and social science, and the logic of scientific explanation. The indefeasibility of this philosophical perspective and the incorrigibility of its representation of science is taken for granted, and it is assumed that from this body of literature can be extracted the fundamental principles which must govern and give direction to empirical inquiry in political science.10
As Eugene Meehan has recently noted, the deductive model has become “an albatross around the neck of the social scientist”11 in the sense that it provides a set of demands which are not only impossible to realize but often irrelevant to the requirements of explanation in social science. But the answer to the problem is not to substitute a “systems paradigm” for the deductive model as Meehan has suggested; these constructs are not logically comparable. Before embarking on any extended criticism of the deductive model, it is necessary to precisely determine its character and to understand its present status in the philosophy of science. Meehan, perpetuating a common misconception in social science, tends to view the deductive model as if it were essentially a procedural recommendation for scientific inquiry; but the model, as presented by most of its proponents, was never conceived either as a methodological norm in the sense of how to go about the task of explanation or as a reconstruction of the process of explanation. Meehan also maintains that this model has been “rarely criticized” in the literature of the philosophy of science, but, on the contrary, every indication is that at present the weight of argument is against the deductive model, and that the revolt against the model as well as other aspects of logical empiricism “has now lasted long enough to produce something of a counterrevolution.”12 Even the defenders of the deductive model recognize that the “global nature of the attack gives the current controversy its special interest.”13 Yet the fact that the impact of the debate has scarcely been felt in the social sciences is only a symptom of the cultural lag in the relationship between these two disciplines.
As the term “deductive model” is used here, it refers to a quite specific formulation which in its essential characteristics is embraced and advocated by a number of philosophers but most prominently associated with the work of Hempel. The Hempelian thesis of deductive nomological explanation is most fundamentally an argument for the “methodological unity of all empirical science”14 since it attempts to reduce the criteria of scientific explanation to one universal logical pattern. It asserts that “all scientific explanation,” as opposed to the manifold everyday senses of “explain” and other specialized meanings of the term, “involves, explicitly or by implication, a subsumption of its subject matter under general regularities; that it seeks to provide a systematic understanding of empirical phenomena by showing that they fit into a nomic nexus.”15 More specifically, it is argued that an event or phenomenon that is to be explained, or strictly speaking, the statement describing the phenomenon or event (the explanandum), must be deducible from premises (the explanans) which are assumed to be true and contain, in addition to certain statements of fact or initial or antecedent conditions, at least one general law expressing empirical regularities. Empirical laws are in turn explained in the same manner by deductive subsumption under more inclusive laws or comprehensive theories which consist of a deductively joined system of axioms and theorems. The proponents of the model are unequivocal in their assertion that “science looks for laws, because without them neither explanation nor prediction is possible” and that the sense in which the term “deduction” is to be taken is the strict one of following the rule of modus ponens of formal logic, that is, that laws function as the “premises of a deductive argument” and “the conclusion may be deduced from the premises because the premises logically imply the conclusion. And the premises logically imply the conclusion because the corresponding conditional, if the premises are true then the conclusion...
