Part I
Overview of Contemporary Sociological Theory
This essay systematically introduces the reader to eleven (plus one) theoretic viewpoints from which human social phenomena are currently being investigated. The introduction is systematic in the sense that it depends on a single, multidimensional, framework for differentiating as well as integrating all eleven viewpoints.
In developing this framework, my aim has been to discover useful ways in which different types of theory complement each other, rather than ways in which they contradict each other. There are two reasons for choosing this emphasis. First are the impressions that no single current viewpoint adequately represents the full range of sociological interest; that none is wholly “right” about all aspects of social life; that each is “right” about some of these aspects;1 and that when it comes to doing a particular research, sociological analysts often draw on several different viewpoints for hypotheses and interpretations. These apprehensions suggest that it may not be necessary, after all, to be unswervingly loyal to one particular theory; that one may be eclectic and be neither renegade nor indecisive; that a systematic assessment of complementarities among theories may be useful in understanding the theories themselves and in applying them to empirical research. In other words, one may reasonably prefer “the Jain logic of ancient India, with its doctrine of syadvada: that every proposition is true only up to a point, in a manner of speaking, in certain respects, … [as] in the Jain legend of the seven blind men and the elephant.”2
The second reason for emphasizing complementarities among different viewpoints rests on the rule of parsimony: when a scientific discipline has developed several theories bearing on the same problem, their number should be reduced by rejecting some and/or integrating some or all of them into a single theory expressing their common and complementary features.3 Although emphasis on contradictions (especially contradictory predictions) between theories is required when deciding which theories to reject and which to retain, when one seeks, instead, to integrate several theories that seem equally acceptable, it is more useful to focus on their complementarities. It has been said that sociology has “too much theory.” If this is so, it may be due not only to the fact that “competing conceptual systems and theories have not been adjudicated,”4 but also to the fact that the identities and complementarities among them have not been sought out. In principle at least, the reduction of theory can proceed by condensation as well as by elimination. Obviously, the result of any theoretic synthesis must incorporate at least its predecessors’ identities and complementarities, although it need not incorporate or seek to resolve their incompatibilities. “In an important sense, new scientific theories do not ‘refute’ the old ones but somehow remake them; even scientific revolutions preserve some continuity with the old order of things.”5
Having thus chosen to concentrate on a synthetic integration of theories, my discussion is not intended to be another “summary” or “review”; still less is it intended to be a “critical essay.” I shall not deal with any particular theory in full detail, and I shall not attempt systematically to evalutate or to proselytize for any theory. My aim here is severely limited: to expose and integrate the distinctive features of several current theoretic viewpoints. And it should be noted in passing that the distinctive features of a particular viewpoint need not be either its most emphasized features or its most persuasive features.
But granting this, there remains the problem of selecting among the undoubtedly large number of possible dimensions according to which the viewpoints in question might be classified. In the present case, I have tried to select dimensions that refer directly to the sociological subjectmatter as such, and have avoided dimensions that refer, instead, to non-sociological thought systems. That is, I have rejected philosophy of science descriptions like “positivistic,” “phenomenological,” “nominalistic,” and “systemic,” and I have also rejected discipline descriptions like “ecological,” “biological,” and “psychological.” Needless to say, I have also avoided the relatively uninformative practice of simply listing theoretic viewpoints according to the chronological order of their publication or according to the names or nationalities of their principal authors. Although several sorts of terms (including some just mentioned) will be used here to describe specific theoretic viewpoints, the terms that will be used to describe the classification of such viewpoints are of a more primitive order, more intimately tied to possible observations and less intimately tied to existing conceptual systems. I have asked, in short, What kind of direct observations docs each theory imply?—rather than asking, What kind of non-sociological conceptual framework does each theory resemble?
This stress on the observations (and the empirical generalizations based on them) that theories indicate, of course, allows the scheme to suggest conclusions not only about theories, but also about the design and interpretation of empirical research—a possibility that will be made more explicit in the “Conclusions” of this discussion.
The term “theory,” in its most minimal and unqualified sense, will be taken to mean any set of symbols that is claimed verifiably to represent and make intelligible specified classes of phenomena and one or more of their relationships.6 In one very important qualification, however, a theory is said to be “of” a particular class of phenomena when the theory is constructed in such a way as to offer an explanation of phenomena in that class (i.e., when the latter is treated as explanandum or dependent variable, regardless of what the explanans or independent variable may be.)7 Thus, all theories of the class of phenomena called “social” treat phenomena in that class either as causal results of antecedent phenomena, or as logical deductions from higher order abstractions, or as both.8 Through such theories of the social we are informed about why or how social phenomena occur—what brings social systems, societies, social actions and interactions, into existence, and what keeps them working. It is exclusively with this variety of social theory that I am concerned here.
There is, however, a second variety of such theory, which, despite the fact that it will not figure in the following discussion, should be mentioned for the sake of contrast. This second variety treats the social class of phenomena either as cause of postcedent phenomena, or as a higher order abstraction from which other deductions are made, or as both.9 Through such theories we are informed about the role of social phenomena in other phenomena—for example, in individual perception, individual personality formation, individual decision-making, and in physically changing the atmosphere and general face of the earth and other planets, etc. In brief, this second variety of theory treats the social as a class of phenomena-that-explain while the first variety treats the social as a class of phenomena-to-be-explained. For denotative clarity, the term “sociological” will be applied here to the first variety only. In this terminology there can be social, but not sociological, theories of individual personality or behavior.
It is obviously possible to have an extremely large number of sociological theories so defined—extending to all possible classes of all possible explanatory phenomena in all their possible relationships.10 Even if one pays attention only to variability in the classes of possible explanatory phenomena and leaves aside variability in their relationships, a huge list can readily be imagined, with places for theories that explain the social with psychological, physiological, geographical, and technological phenomena; with electromagnetic, chemical, and gravitational phenomena; and so on and on. Of course, somewhere in this gigantic array, one would also find social theories of social phenomena wherein one subclass of social phenomena would be offered as the principal explanation of another.11 Obviously, therefore, when all the theories under consideration are presumed to define a common class of phenomena-to-be-explained (in this instance the social), one major source of variation among them lies in the differences in the classes of phenomena that they propose as principal explaining phenomena.12 The discussion below will take this variation as one of its two central foci. My aim in this connection, then, is to construct reasonable categories for analyzing differences in the kinds of things that sociological theories take as phenomena-that-explain.
The second central focus of discusssion arises from the unfortunate and often confusing observation that sociological theories purporting to share the same class of phenomena-to-be-explained do not always do so. I refer, of course, not to the plain fact that some are theories of stratification and some are theories of socialization (for example), but to something more essential: there exist different ways of defining “the social”—fundamentally different opinions of what qualities are required before a given phenomenon can be called social.13 The second focus of this paper, then, is on variation in the sort of phenomena thought to be eligible objects of explanation for sociological theories.
To put both foci succinctly, the discussion that follows seeks to discover the answers provided by sociological theories to two questions: (I) How is the social defined? (2) How is the social explained (i.e., by what classes of phenomena)? Both questions are required, and joint answers to them will be sought here, because it seems wholly inadequate to differentiate theories in terms of the single, loose, and indefinite question of what they are “interested in,” or what their “approach” is, or what they “deal with,” although this is often done. The question is indefinite because, given that all scientific theories require both an explanandum and an explanans, it is apparent that two quite different theories may be equally “interested” in the same class of phenomena. That is, one theory may be interested in it as explanandum, while the other theory may be interested in it as explanans. The distinction between kinds of theoretic interest is therefore crucial. It is chiefly on the basis of an effort to maintain this distinction between how the social explanandum is defined and how it is explained that the present discussion is formally distinguishable from most earlier ones at classifying sociological theories.14
Definitions of the Social
All definitions of the social seem to have in common at least one statement clearly setting it apart from other phenomena: a social phenomenon is always defined in terms of interorganism behavior relations. That is, it seems generally agreed that a social phenomenon is constituted by the regular accompaniment of one organism’s behavior by at least one other organism’s behavior.15
But granting this as part of the boundary of common definition distinguishing sociological theory as a whole from other theory, there remain several dimensions of possible variation within that boundary. For example, the “organisms” in question may be defined as humans, termites, bees, bats, monkeys, wolves, wild dogs, gorillas, geese, fish, etc., or any of a great variety of interspecies combinations. The criterion of “regularity” may be defined in relatively simple or complex terms (analogous to the varieties of mus...