Once we begin to speak of political life as a system of activity, certain consequences follow for the way in which we can undertake to analyze the working of a system. The very idea of a system suggests that we can separate political life from the rest of social activity, at least for analytical purposes, and examine it as though for the moment it were a selfcontained entity surrounded by, but clearly distinguishable from, the environment or setting in which it operates. In much the same way, astronomers consider the solar system a complex of events isolated for certain purposes from the rest of the universe. Furthermore, if we hold the system of political actions as a unit before our mind’s eye, as it were, we can see that what keeps the system going are inputs of various kinds. These inputs are converted by the processes of the system into outputs and these, in turn, have consequences both for the system and for the environment in which the system exists. The formula here is very simple but, as I hope to show, also very illuminating: inputs — political system or processes — outputs. These relationships are shown diagrammatically in Figure i. This diagram represents a very primitive „model“ — to dignify it with a fashionable name — for approaching the study of political life. Political systems have certain properties because they are systems. To present an overall view of the whole approach, let me identify the major attributes, say a little about each, and then treat one of these properties at somewhat greater length, even though still inadequately.
(1) Properties of identification. To distinguish a political system from other social systems, we must be able to identify it by describing its fundamental units and establishing the boundaries that demarcate it from units outside the system.
(a) Units of a political system. The units are the elements of which we say a system is composed. In the case of a political system, they are political actions. Normally it is useful to look at these as they structure themselves in political roles and political groups.
(b) Boundaries. Some of the most significant questions with regard to the operation of political systems can be answered only if we bear in mind the obvious fact that a system does not exist in a vacuum. It is always immersed in a specific setting or environment. The way in which a system works will be in part a function of its response to the total social, biological, and physical environment.
The special problem with which we are confronted is how to distinguish systematically between a political system and its setting. Does it even make sense to say that a political system has a boundary dividing it from its setting? If so, how are we to identify the line of demarcation? Without pausing to argue the matter, I would suggest that it is useful to conceive of a political system as having a boundary in the same sense as a physical system. The boundary of a political system is defined by all those actions more or less directly related to the making of binding decisions for a society; every social action that does not partake of this characteristic will be excluded from the system and thereby will automatically be viewed as an external variable in the environment.
(2) Inputs and outputs. Presumably, if we select political systems for special study, we do so because we believe that they have characteristically important consequences for society, namely, authoritative decisions. These consequences I shall call the outputs. If we judged that political systems did not have important outputs for society, we would probably not be interested in them. Unless a system is approaching a state of entropy — and we can assume that this is not true of most political systems — it must have continuing inputs to keep it going. Without inputs the system can do no work; without outputs we cannot identify the work done by the system. The specific research tasks in this connection would be to identify the inputs and the forces that shape and change them, to trace the processes through which they are transformed into outputs, to describe the general conditions under which such processes can be maintained, and to establish the relationship between outputs and succeeding inputs of the system.
From this point of view, much light can be shed on the working of a political system if we take into account the fact that much of what happens within a system has its birth in the efforts of the members of the system to cope with the changing environment. We can appreciate this point if we consider a familiar biological system such as the human organism. It is subject to constant stress from its surroundings to which it must adapt in one way or another if it is not to be completely destroyed. In part, of course, the way in which the body works represents responses to needs that are generated by the very organization of its anatomy and functions; but in large part, in order to understand both the structure and the working of the body, we must also be very sensitive to the inputs from the environment.
In the same way, the behavior of every political system is to some degree imposed upon it by the kind of system it is, that is, by its own structure and internal needs. But its behavior also reflects the strains occasioned by the specific setting within which the system operates. It may be argued that most of the significant changes within a political system have their origin in shifts among the external variables. Since I shall be devoting the bulk of this article to examining some of the problems related to the exchange between political systems and their environments, I shall move on to a rapid description of other properties of political systems.
(3) Differentiation within a system. As we shall see in a moment, from the environment come both energy to activate a system and information with regard to which the system uses this energy. In this way a system is able to do work. It has some sort of output that is different from the input that enters from the environment. We can take it as a useful hypothesis that if a political system is to perform some work for anything but a limited interval of time, a minimal amount of differentiation in its structure must occur. In fact, empirically it is impossible to find a significant political system in which the same units all perform the same activities at the same time. The members of a system engage in at least some minimal division of labor that provides a structure within which action takes place.
(4) Integration of a system. This fact of differentiation opens up a major area of inquiry with regard to political systems. Structural differentiation sets in motion forces that are potentially disintegrative in their results for the system. If two or more units are performing different kinds of activity at the same time, how are these activities to be brought into the minimal degree of articulation necessary if the members of the system are not to end up in utter disorganization with regard to the production of the outputs of interest to us? We can hypothesize that if a structured system is to maintain itself, it must provide mechanisms whereby its members are integrated or induced to cooperate in some minimal degree so that they can make authoritative decisions.