If there is any one concept that separates Luhmann's work from most sociologists, It Is his dependence on the Idea of contingency. Contingency, however, is the bastard child of science, blamed on a lack of understanding, on rules operating though unknown. True contingency, as Luhmann pictured it, of scientific laws which produce state-spaces of possibilities rather than a single unequivocal and predictable outcome, is a new conception that many scientists hesitate to accept. Applying concept of complexity as the totality of possible interactions available to a system is one way of avoiding the problem because it suggests a state-space of causes with a one-on-one relationship to a state-space of effects thus saving the deterministic view that underlies the scientific paradigm. Even Luhmann's fuzzy interpretation of complexity tends to lend support to that view. However, the bulk of his work stresses what we might call true contingency, or causes which lead to a multiplicity of possible effects. If we interpret his work using a more formal meaning of complexity as a special kind of hierarchical structure with predictable properties which arise out of that structure, we find that we arrive at descriptions of social activities that mirror the kinds of activities we actually experience closer than descriptions tied to more deterministic assumptions.

In the first place a truly contingent event takes place whenever there are among the state space of possible outcomes two or more which have equal probabilities of occurring. There is no need for external force, to cause one outcome rather than another, for if only one outcome is possible then the choice can Just as well be made by pure chance. Another case is when the chance occurrence of an unrelated event modifies the probabilities such that a specific outcome is chosen. In both of these cases we can say that an uncontrolled choice has been made. Neither case represents an example of a determined outcome since even knowing all of the relevant facts does not lead to a faithful prediction of the outcome. The chance occurrence of an unrelated event Is not a relevant fact.

The difference is a matter of control. Given a contingent event in an environment with a great deal of variety the outcome is self-controlled when the choice from among probabilities is determined by forces within the system Involved. it is an externally controlled event if the choice is determined by events external to the system. Thus in order for a system to be self controlled It must contain within Its own repertoire at least as much variety as there is in that part of the environment which it must normally face. This, of course, is a restatement of Ashby's law of requisite variety

bulletThe Reduction of Variety
bulletReturn to Table of Contents