Simulation Home Introduction Uses Definitions Approaches Design Implementation Evaluation Examples Bibliography

Simulation is an alternative to simple speculation. Alternatively, we might say that simulation is a means for evaluating speculation. Rather than simply imagine, argue and debate how the different parts of social, cultural and individual life interact together to form that gestalt 'wholism' anthropologists so value, you can use a simulation to examine, explore and evaluate such interactions within a context. Simulations are often used to represent 'real' events, people and things, ranging from rituals and objects representing supernatural entities and forces, to calendars, schedules and essay outlines. In this unit we will look at some of the issues underlying simulation, how it has been used in anthropology, and at a range of simple computer-based simulations and how these have been used in anthropology. These range from relatively concrete simulations of physical objects or situations of cultural importance, to more abstract simulations of social, economic and political processes. Simulations are most often of processes where one or more entities or objects is involved (this is complex though. See Definitions). Simulations most useful in two situations. Firstly, where the actual process is either impossible or impractical to enact, and secondly, as a means for modelling situations where we are sufficiently ignorant to preclude a more systematic account (which in social and cultural anthropology is most things). Bateson observes in his "Things every schoolboy knows" (Bateson 1987) that logic does not adequately describe or account for causal processes. Simulations are generally based on a presumption of causality, although not necessarily monotonic causality (what happens is caused, but the actual outcome was not the only possible outcome, or indeed the only outcome). Most simulations are a culmination of a series of local causal models interacting with each other. Part of the problem with simulation is the justification - just how can we interpret the results? The usual response to this is to claim a weak interpretation for simulation. That is, concurrence that simulation is a way of exploring possibilities, and should not be too closely identified with the inspiration for the simulation. This leaves us in the position of claiming on the one hand that simulation is a good method for investigating certain kinds of situations, but that we cannot really generalise from the results of the simulation to strongly support understanding of these same situations. This is sometimes the case, but should not be considered an absolute. Sometimes we are justified in generalising, sometimes we are not.