From Talcott Parsons, The System of Modern Societies. Englewood Cliffs, NJ: Prentice-Hall, 1971, pp. 4-8 (edited by SNK).

Action Systems and Social Systems

Talcott Parsons (1961-1971)

We consider social systems to be constituents of the more general system of action, the other primary constituents being cultural systems, personality systems, and behavioral organisms; all four are abstractly defined relative to the concrete behavior of social interaction. We treat the three subsystems of action other than the social system as constituents of its environment. This usage is somewhat unfamiliar, especially for the case of the personalities of individuals. It is justified fully elsewhere, but to understand what follows it is essential to keep in mind that neither social nor personality systems are here conceived as concrete entities.

The distinctions among the four subsystems of action are functional. We draw them in terms of the four primary functions which we impute to all systems of action, namely pattern-maintenance, integration, goal-attainment, and adaptation.

An action system's primary integrative problem is the coordination of its constituent units, in the first instance human individuals, though for certain purposes collectivities may be treated as actors. Hence, we attribute primacy of integrative function to the social system.

We attribute primacy of pattern-maintenance--and of creative pattern change--to the cultural system. Whereas social systems are organized with primary reference to the articulation of social relationships, cultural systems are organized around the characteristics of complexes of symbolic meanings--the codes in terms of which they are structured, the particular clusters of symbols they employ, and the conditions of their utilization, maintenance, and change as parts of action systems.

We attribute primacy of goal-attainment to the personality of the individual. The personality system is the primary agency of action processes, hence of the implementation of cultural principles and requirements. On the level of reward in the motivational sense the optimization of gratification or satisfaction to personalities is the primary goal of action.

The behavioral organism is conceived as the adaptive subsystem, the locus of the primary human facilities which underlie the other systems. It embodies a set of conditions to which action must adapt and comprises the primary mechanism of interrelation with the physical environment, especially through the input and processing of information in the central nervous system and through motor activity in coping with exigencies of the physical environment. These relationships are presented systematically in Table 1.

TABLE 1 [: Subsystems of the general system of action]


Primary Function




Pattern Maintenance*


Goal Attainment*

Behavioral Organism*


* These are the social subsystem's environment.

There are two systems of reality which are environmental to action in general and not constituents of action in our analytical sense. The first is the physical environment including not only phenomena as understandable in terms of physics and chemistry, but also the world of living organisms so far as they are not integrated into action systems. The second, which we conceive to be independent of the physical environment as well as of action systems as such, we will call ''ultimate reality" in a sense derived from traditions of philosophy. It concerns what Weber called "problem of meaning" for human action and is mediated into action primarily by the cultural system's structuring of meaningful orientations that include, but are not exhausted by, cognitive "answers."

In analyzing the interrelations among the four subsystems of action -- and between these systems and the environments of action -- it is essential to keep in mind the phenomenon of interpenetration. Perhaps the best-known case of interpenetration is the internalization of social objects and cultural norms into the personality of the individual. Learned content of experience, organized and stored in the memory apparatus of the organism, is another example, as is the institutionalization of normative components of cultural systems as constitutive structures of social systems. We hold that the boundary between any pair of action systems involves a "zone" of structured components or patterns which must be treated theoretically as common to both systems, not simply allocated to one system or the other. For example, it is untenable to say that norms of conduct derived from social experience, which both Freud (in the concept of the Superego) and Durkheim (in. the concept of collective representations) treated as parts of the personality of the individual, must be either that or part of the social system.

It is by virtue of the zones of interpenetration that processes of interchange among systems can take place. This is especially true at the levels of symbolic meaning and generalized motivation. In order to "communicate symbolically, individuals must have culturally organized common codes, such as those of language, which are also integrated into systems of their social interaction. In order to make information stored in the central nervous system utilizable for the personality, the behavioral organism must have mobilization and retrieval mechanisms which, through interpenetration, subserve motives organized at the personality level.

Thus, we conceive social systems to be "open," engaged in continual interchange of inputs and outputs with their environments. Moreover, we conceive them to be internally differentiated into various orders of subcomponents which are also continually involved in processes of interchange.

Social systems are those constituted by states and processes of social interaction among acting units. If the properties of interaction were derivable from properties of the acting units, social systems would be epiphenomenal, as much "individualistic" social theory has contended. Our position is sharply in disagreement: it derives particularly from Durkheim's statement that society--and other social systems-- is a "reality sui generis.''

The structure of social systems may be analyzed in terms of four types of independently variable components: values, norms, collectivities, and roles. Values take primacy in the pattern-maintenance functioning of social systems, for they are conceptions of desirable types of social systems that regulate the making of commitments by social units. Norms, which function primarily to integrate social systems, are specific to particular social functions and types of social situations. They include not only value components specified to appropriate levels in the structure of a social system, but also specific modes of orientation for acting under the functional and situational conditions of particular collectivities and roles. Collectivities are the type of structural component that have goal-attainment primacy. […W]e speak of a collectivity only where two specific criteria are fulfilled. First, there must be definite statuses of membership so that a useful distinction between members and nonmembers can generally be drawn, a criterion fulfilled by cases that vary from nuclear families to political communities. Second, there must be some differentiation among members in relation to their statuses and functions within the collectivity, so that some categories of members are expected to do certain things which are not expected of other members. A role, the type of structural component that has primacy in the adaptive function, we conceive as defining a class of individuals who, through reciprocal expectations, are involved in a particular collectivity. Hence, roles comprise the primary zones of interpenetration between the social system and the personality of the individual. A role is never idiosyncratic to a particular individual, however. A father is specific to his children in his fatherhood, but he is a father in terms of the role-structure of his society. At the same time, he also participates in various other contexts of interaction, filling, for example, an occupational role.

The reality sui generis of social systems may involve the independent variability of each of these types of structural components relative to the others. A generalized value-pattern does not legitimize the same norms, collectivities, or roles under all conditions, for example. Similarly, many norms regulate the action of indefinite numbers of collectivities and roles, but only specific sectors of their action. Hence a collectivity generally functions under the control of a large number of particular norms. It always involves a plurality of roles, although almost any major category of role is performed in a plurality of particular collectivities. Nevertheless, social systems are comprised of combinations of these structural components. To be institutionalized in a stable fashion, collectivities and roles must be "governed" by specific values and norms, whereas values and norms are themselves institutionalized only insofar as they are "implemented" by particular collectivities and roles.