Many Anticipations

Anticipation comes in different guises. Some anticipations are explicit, meaning that the system knows that it has them. On the other hand, it is patently obvious that most anticipations work silently: they constrain the system’s behavior without the system being aware of them. This implies that the system knows only some fragments of its own identity. The main problem with such an extensive family of anticipations is that the different types of anticipation may work together and synthetically produce the system’s general anticipatory patterns, or they may conflict and eventually cancel each other out.

The widest distinction is between explicit and implicit types of anticipation. Explicit types of anticipation include foresights and expectations, while implicit types of anticipation are properties of the system, intrinsic to its functioning.


As far as explicit anticipations are concerned, the distinction between anticipation as a simple looking into the future and anticipation as the capacity to take account of the consequences of that looking, i.e. its  impact on current behavior, is worth considering. This distinction may appear to be trivial, yet many conflicts spring from a kind of blindness to the consequences of the actions performed.

The most efficient way to learn how to foresee each other’s reasons and actions is to devise forms of institutionalization of the agents’ expectations. Institutionalization lowers uncertainty, and less uncertainty raises confidence. The problem with institutionalization, however, is that it generates forms of blindness towards whatever does not match its internal codes. Institutionalized behavior may not be able to detect what futurists call ‘weak signals’, namely early and usually minor behavioral differences that may eventually grow and become new behavioral patterns.

Moreover, no description is able entirely to capture an anticipatory system. Side effects are a structural feature of anticipatory systems. By default, when the system carries out a particular activity, it uses only some of its internal resources. Side effects are due to the tension between the fact that the system’s dynamics characterize it as a whole (the equations of the system’s motion link all the variables defining the system) whilst the system’s functional activities require only some of its variables. The variables not involved in any particular functional activity are therefore free to interact with other systems in a non-functional way, and even in a dysfunctional way. A major consequence is that activities will in general have effects on a system other than those which are planned.