IN DEFENCE OF POSIWID - the purpose of the system is what it does.

There have been many discussions about this observation made by Stafford Beer. It arises because the assumed purpose of a system and what the system actually does may not be identical. ?Systems often display unexpected behaviour that does not entirely fit the purpose intended by programmers and designers. Hence Stafford Beer’s observation.

?I’d like to offer some thoughts that examine Beer’s famous dictate from an epistemological perspective. The examination shows that ascribing a purpose to the observed behaviour of a system has definite advantages and reveals the wisdom of using POSIWID as an epistemological approach to investigating systemic behaviour.

In a world where everything is intrinsically interconnected, meaningful knowledge cannot be gained unless the number of connections to be examined is constrained in some way. In fact, it could be said that all epistemological approaches require some constraints or filters to permit access to meaningful knowledge.

This, of course, also applies to systems thinking when considered as an epistemic tool. The data examined in systems thinking is restricted by the definition of a system boundary, by procedural rules, by relying on circular causation and simply by seeing the world in terms of systems. ?Without these constraints ‘systemic insights’ would not materialize. Paradoxically, constraints are an enabling device, making knowledge acquisition possible. Of course, some constraints applied excessively could also limit the number of insights that are possible. Constraints should therefore be applied selectively and in agreement with the intended objective of any investigation. However, meaningful knowledge always requires some constraints. ?

Observing what a system actually does and taking this behaviour to be the system’s purpose is a useful form of constraining the information to be employed for a subsequent systemic investigation of how the system produces the behaviour. This ascribed purpose will limit the components of the system that can fit rationally into the model being built to explain (and reproduce) the observed behaviour. In fact, this process will? define the system’s boundary. System boundaries can, of course, ?be established in various ways. ?One example would be building a systems map. However, such maps can produce a number of different boundaries for the same system. The use of POSIWID fixes the system boundary simply by asking which variables contribute (or do not contribute) to the achievement of the identified objective. ?It could be argued that ascribing a purpose to observed systemic behaviour is a more realistic way to determine the boundary of a system since it is related to actual observation.

Beer’s POSIWID, by introducing useful and realistic constraints is an enabler of systemic insights and knowledge and in this context an excellent systemic tool that helps anchor the findings of systemic investigations to an essential part of observed reality.

H.J. Hodann (16-07-2024)