Arnold Rose’s “A Deductive Ideal-Type Method”
Helen M. Hacker Originally published 1951
To the Editor:
In the July, 1950, issue of the American Journal of Sociology Arnold M. Rose proposes in “A Deductive Ideal-Type Method” that sociologists endeavor to apply to the phenomena of their discipline the deductive method, which has proved so rewarding in theoretical economics. This deductive method, in his definition, consists in formulating significant tautologies involving “a few basic and manipulable variables” within the limits of assumptions based on observation. From such tautologies or truisms, logical deductions may be made which have predictive value. Dr. Rose then cites the equation of exchange as an example in economics of the empirical fruitfulness of a “practical truism.” Standing in the way of the utilization of the deductive method in sociology, he further states, are the twin difficulties that sociologists are agreed as to neither their subject matter nor the basic unit of their science. Nevertheless, he proceeds to supply five sociological examples of truistic propositions which have important empirical implications.
The purpose of these comments is to criticize neither Dr. Rose’s proposal nor his illustrations but to suggest that the employment of deductive ideal types is by no means absent from the history of sociological thinking, though perhaps not always linked to relevant, verifiable assumptions, and that a recent striking example is furnished by the structural-functional approach as developed by Parsons, Merton, and others. Moreover, this theory also obviates the two difficulties mentioned above, in that it holds sociology to be the science of institutional structure and its basic unit as the “actor-situation.”1 Basic to sociology, says Talcott Parsons, must be a theory of social systems. He defines a social system as “a system of social action involving a plurality of interacting individuals,” and he clarifies action as meaning “motivated human behavior.”2 Consider his prescription of “the functional prerequisites of the social system:”
“Functional requirements of the maintenance of any such pattern system or pattern line of change can be generalized to a certain degree. In the first place, of course, a social system must somehow provide for the minimum biological and psychological needs of a sufficient proportion of its component members. On a more strictly social level, there seem to be two primary fundamental foci of its functional prerequisites. One lies in the problem of order, in the problem of the coordination of the activities of the various members in such a way that they are prevented from mutually blocking each other’s action or destroying one another by actual physical destruction of the organism, and, on the other hand, they are sufficiently geared in with each other so that they do mutually contribute to the functioning of the system as a whole. The second focus is on adequacy of motivation. The system can only function if a sufficient proportion of its members perform the essential social roles with an adequate degree of effectiveness. If they are not adequately motivated to this minimum level of contribution to the system, the system, itself, of course, cannot operate. A variety of further elaborations of the problem of functional prerequisites can be worked out from these starting points.”3
Obviously, the statements in this paragraph constitute a series of truisms derived from the definition of a social system. They may be recast in the form of the following proposition: If a social system consists of social action involving a plurality of interacting individuals, then the conditions permitting such social action must be met, i.e., the survival of a sufficient number of individuals to engage in social action, kinds of action which do not impede further action, and motivations adequate to continuing nonblocking action. One may therefore predict that any internal or external condition which destroys more than the needed number of persons to carry on a social system, which leads to an excessive amount of mutually opposing action, or which negates previously effective sources of motivation will lead to the downfall of that system. Thus the attention of the investigator is directed to specific crucial empirical facts in evaluating the factors making for the success or failure of social systems.
The question may be raised, however, of whether this truism meets Dr. Rose’s criteria of relevance and usefulness. The variables included are perhaps too general to dictate predictions specific enough for empirical verification. Even if one had the most precise knowledge of the structural and functional categories of a social system, one could not predict that a given change in a behavior pattern or shift in motivation would necessarily be dysfunctional to the system. Professor Merton has called attention to the importance of “functional alternatives, equivalents, or substitutes."4 This concept permits the realization that certain functions, even if indispensable to a social system, may be performed by a variety of cultural forms. His distinction between manifest and latent functions also complicates the problem of assessing the effects of change in “particular conditions and process” upon a social system. An item may be manifestly functional and latently dysfunctional, or manifestly dysfunctional and latently functional. Extended observation may be required to determine which is the case. If the observer is also a participant in the social system under study, his own identification of himself with it will render it difficult for him to decide, in the short run, whether a given cultural item is functional or dysfunctional to the effectiveness of the system. The notion of “function” thus assumes a subjective character. What is functional for one group may be dysfunctional for another. Even if the observer attempts to view a social system from an ethically neutral vantage point, his decision concerning the functionality or dysfunctionality of a set of conditions for any individual or subgroup implies a definite value as to what constitutes the “good life"—unless he employs their goals as a touchstone.
Not only does the value identification of the observer make problematic an objective determination of the functionality or dysfunctionality of given cultural items, but it also may render the manifest or latent character of a function equally dependent upon his perspective. Thus Professor Merton5 designates as a latent function of the political machine, not fulfilled by other alternatives and often unrecognized by political reformers, the personal service which the political boss gives to members of minorities who are fearful of formal agencies. But to whom, besides the political reformer who is now being informed of it, is this activity of the political boss latent? While it may not be his prime purpose, the boss knows what he is doing and frequently defends his existence in just these terms. To the individual he helped, also, such activity may appear as manifest, while the vote-catching aspect of the boss’s behavior is latent. Both participants, not accustomed to taking a total view of the social system, may regard the tactics of the political boss as manifestly functional to themselves, without seeing, as a sociologist might, whether these actions contribute to the maintenance of the social system as a whole and/or the welfare of which, if any, of its participants. And now we are led back to the original problem of deciding which activities are functional and which dysfunctional, without receiving any great aid from the distinction between manifest and latent functions.
These considerations cast doubt upon the scientific utility of functionalism. Instead of formulating a priori assumptions concerning the “needs” of social systems, we can simply observe the activities carried on by participants in a social system. We can study the social processes which motivate certain kinds of behavior and the consequences of this behavior, without invoking the notion of function. If, for example, the training which the child receives in the family orients him to goals which conflict with those of groups which he enters as an adult, then this consequence can be noted without castigating the family as “dysfunctional."
In support of this suggested rejection of the functional approach as unrewarding, let us examine three possible meanings of the term “functionalism.” The first is that of reinforcing, implementing, or contributing to a common end. This is the meaning accepted in the above discussion, and its difficulties have been indicated. It leads to propositions which are circular, obvious, or nonverifiable. A second meaning is that two items mesh or interlock. All that this implies is that they are compatible, that they can coexist, and that such imputations are subject to empirical investigation and are not in the present state of knowledge to be determined in terms of psychological or cultural theory. For example, the fact that idealist philosophies may be compatible with either “radical” or “reactionary” political views is a matter of experience and not deducible from the “nature” of such philosophies.
The third meaning of “functionalism” is the semimathematical one of dependence or invariant relationship. Two things vary together; one is a function of the other. In verifying such mutual dependencies, the language of functionalism is excess baggage. One can seek to establish social laws without reference to “needs,” “compensatory mechanisms,” etc.
Thus one must reluctantly conclude that, despite the seeming promise of the structural-functional theory as a tautology capable of yielding practical predictions, it does not appear useful in its present form. It must still be a painstaking matter of empirical investigation to determine the minimum conditions for survival of any social system, to determine which roles are essential and how they must be performed. Then there are the additional problems of isolating the “contribution” of any item to the ongoing system (except by “thinking it away”), of verifying latent functions and dysfunctions within the confines of one society. It is the belief of this writer that the notion of function is an unnecessary “intervening variable” in establishing invariant relationships among social phenomena, though it may prove its value in directing the attention of sociologists to certain relationships which they might otherwise have overlooked.
1 Talcott Parsons. 1949. Essays in Sociological Theory Pure and Applied. Glencoe, IL: Free Press, pp. 32 and 39.
2 Ibid, pp. 6, 33.
3 Ibid, p. 6.
4 Robert K. Merton. 1949. Social Theory and Social Structure: Toward the Codification of Theory and Research. Glencoe, IL: Free Press, pp. 35-36.
5 Ibid, p. 79.