118 Impasse and Resolution
My argument leads to the following conclusion. A second inspection of Spiro's "revised" explanation of religious beliefs reveals that it will be difficult to create a scale for measuring beliefs which are "more" or "less" satisfying. Nevertheless, such a scale might be created and tested. The deeper problem is that the qualification "more or less" removes the necessary condition of the religious belief system as satisfying a requirement of need in a society. For to say that in the context of a specific social system certain beliefs are "more" satisfying clearly does not exclude the possibility that certain alternative beliefs mgy become more effective than the beliefs presently held. In brief, we have not explained why the religious belief system is a necessary condition for satisfying the need, and thus we have not shown that there are no functional equivalents for a particular religious belief system in a society.
I believe these problems illustrate the inherent logical difficulties which functionalist theories must resolve before they can be accepted as adequate explanations of religion. If the critique I have presented is not convincing enough, there is a final methodological move which may clinch my argument. This involves the notion that a social system is a self-regulating system.
Both Ernest Nagel and Robert Brown have examined the problems of this model in their analysis of functionalist explanations. According to Nagel, "functional statements are regarded as appropriate in connection with systems possessing self-maintaining mechanisms for certain of their traits, but seem pointless and even misleading when used with reference to systems lacking such self-regulating devices."19
Robert Brown describes a self-maintaining system, a system that has negative feedback in the following way:
A self-persisting system is commonly taken to be a system which maintains at least one of its properties in an equilibrium position despite variations in the other properties, either inside or outside the system, to which the presence of the first property is causally related. This ability to maintain a property in a steady state while its causal factors vary within certain limits depends on the system containing certain devices. These must be self-regulators in the sense that they must register any significant variations in the state of the property which is being maintained and must compensate for these variations in such a way as to preserve
Functional Explanations of Religion   119
the property within a range of permissible values. The simplest example and the one most favored by recent authors, is the thermostat which increases or reduces the heat throttle according to whether its thermometer registers above or below a set value.20
The model for many contemporary theories of religion is something like the above description of a system having negative feedback. This model often permits the functionalist to argue that the criticism of functionalism is basically irrelevant because the argument is directed against attempts to explain the origin or religion. By use of a negative feedback model, many functionalists claim that this is not what functionalism is all about, even though some of the fathers of functionalism made this mistake. The aim of functional theories of religion is not to show why a religious unit comes into existence, but how it functions and persists, what role it plays in a self-regulating system. Religious units in a social system are to be understood as variables in a self-maintaining system. This is an important distinction, but the success of the model for explaining religion is negligible.
Self-regulating systems are often assumed in the explanation of religion. A brief example must suffice to illustrate how the model is used. What we wish to explain is how a particular religious ritual works in the self-regulation of a social system. We assume for example, that both social equilibrium and personal stability are being maintained, since according to our observations both are functioning adequately. The religious ritual is explained if we can show how it functions to satisfy the requirements of equilibrium in the society and the stability in the individual. The religious ritual is explained, then, if we can show the relations of the ritual as reinforcing not only the "perceptual sets" formed in early childhood but also how the satisfaction of intellectual and emotional needs, in turn, reinforces the social structure of the society in which the religious ritual is a unit.
The use of this functionalist model has often been misunderstood. The explanation begins with religion present as a variable and then attempts to explain how it works in providing the self-regulation of the system. Failure to see the significance of this kind of explanation has often led to criticisms which are wide of the mark. The common mistake is to accuse functionalists who use self-adjusting models of not explaining why religion occurs, i.e. they fail to explain what causes religion.   This often takes the following form.   Let us assume that we