120 Impasse and Resolution
have just read a functional explanation which makes it explicit that the model being used is a self-regulating one and that what is to be explained is how religion functions as a variable for providing equilibrium or adjustment as an effect; for example, that the belief in the existence of "Nats" among Burmese Buddhists, or in witchcraft among the Azande, functions to maintain social solidarity. The criticism of such an explanation often states that the belief has not really been explained. After all, the believer in "Nats" or witches is certainly not saying that he has acted in a way which will maintain social solidarity by believing in them or by performing a ritual which will protect him from their evil influences. The reason he believes in them, or performs certain rituals which are related to them is because he believes both that they exist and that they affect his life. If I asked a Shaivite why he believes in Shiva, I would certainly be surprised to hear that he believes in Shiva because Shiva is a symbolic representation of "perceptual sets" formed in early childhood and that, furthermore, such a belief reinforces both emotional and intellectual needs and the social structure as an unintended consequence.
The mistake we make in such critical replies is that the criticism is beside the point. We do not usually travel all the way to India, Africa or Burma, spend our lives reading ancient texts to report what is self-evident, that Azande believe in witches, Burmese Buddhists believe that "Nats" exist, and some Hindu's who belong to the Shaivite tradition believe in the existence of Shiva.
If a functionalist theory is explicit about the aims of the explanation, it should come as no surprise that an Azande or a Hindu does not make the same kinds of functional statements. Most scholars who are interested in explaining the function of religion are interested in the unintended consequences of the beliefs that are held and the rituals that are performed. What the functionalist is interested in is not primarily a report about these beliefs and rituals (as important as they are for an explanation), but the unintended consequences which these beliefs and rituals have for providing feedback into the regulation of the system. And if the functionalist can discover such unintended consequences, then it is surely odd to criticize the explanation because the believer did not report them when asked about the particular beliefs held. Once again, such an explanation would certainly advance our knowledge of religion.
If, however, we wish to take this approach to explain religion as providing feedback in a system of relations which is self-regulating, then
Functional Explanations of Religion    121
it will become necessary to answer several important questions. Robert Brown has come up with three questions which require an answer. The first is, what property is being maintained in a steady state? Secondly, what are the internal variables and can they or their effects be measured? Finally, what are the external conditions which are assumed to be constant?21 No functionalist theory constructed on the model of a feedback system that I know of has come close to meeting the above requirements. In fact, I am not certain about how we could possibly meet them.
Let us imagine, for the moment, that it is conceivable that we could meet the requirements. We could begin by claiming that social solidarity or social equilibrium was the property which was being held in a steady state. We would then have to specify which of the variables or which single variable works to provide the steady state, and we would also have to provide a range of values to determine when the system is in equilibrium. To compound the difficulties, we would also have to specify which external threat of whatever kind could prevent the function of a unit from providing the necessary effects of maintaining the system.
We have discussed the difficulties of establishing whether a religious unit is a sufficient or necessary condition for maintaining a social system. In a self-regulating model, if the unit is a necessary condition for maintaining the system, we will then have to specify the exact relations which it maintains; since it would be most difficult to uphold the notion that a religious belief system, for example, functions to maintain all the relations in a social system. On the other hand, if we state that the unit is a sufficient condition, we then fail to explain exactly how the religious unit functions to satisfy the requirement. If we move to a class of units which jointly function to satisfy the needs of a society, we are reduced to trivial if not invalid conclusions. And no one as far as I know has been able to specify the external conditions which are constant or the degree of variation which will allow a social system to continue to maintain itself. Functionalism is at best a heuristic device, a strategy, for interpreting religion; heuristic devices are neither true nor false, they are in the domain of "your guess is as good as mine."
I believe that this conclusion is fully confirmed by the current debate among Marxists and other social scientists regarding the validity of functionalism in the corpus of publications, ever since Marx, which attempt to "explain" political/economic history.
Jon Elster, for example, offers us the following strong definition of a functional explanation: "On my definition then, an institution or a