The Field Guide › Paper
Simon's essay is the second pillar of the complexity-foundations set and, for anyone thinking about organisations, the more directly useful of the two: it supplies the concepts of hierarchy and near-decomposability that let one treat an organisation as a complex system rather than a machine. Simon begins where Weaver (whom he cites) leaves off, concerned with organised complexity, and defines a complex system loosely as one made of many parts that interact in a non-simple way, such that, given the parts and the rules of their interaction, it is not trivial to infer the behaviour of the whole (a stance he sums up by saying that in the face of complexity an in-principle reductionist may be at the same time a pragmatic holist). His central claim is that complexity in nature overwhelmingly takes the form of hierarchy: systems nested within systems, each level composed of subsystems down to some arbitrarily chosen elementary unit. Crucially, Simon widens hierarchy beyond the authority-relation sense the word usually carries (he notes that in real organisations the formal chart exists only on paper, and the flesh-and-blood organisation has many relations other than the lines of authority) to mean any system decomposable into successive subsystems, whether or not one part commands another. He then makes three moves. First, an evolutionary argument, carried by the parable of the two watchmakers Hora and Tempus: the one who builds his watches from stable sub-assemblies finishes vastly faster than the one who must start over from loose parts whenever he is interrupted, so that among possible complex forms the hierarchies built from stable intermediate forms are the ones that have time to evolve. Second, and most useful for organisations, the property of near-decomposability: in a hierarchy the interactions within a subsystem are stronger than those between subsystems, so that in the short run each part behaves almost independently of the others and in the long run each depends on the others only in an aggregate way (his worked example is heat flowing between the rooms and cubicles of an insulated building). This is what makes a complex system analysable at all, since one can study a part without tracing every interaction to every other part. Third, a discussion of description, distinguishing state descriptions (blueprints) from process descriptions (recipes) and observing that nearly-decomposable systems are richly redundant and so can be described economically, an 'empty world' in which most things are only weakly connected to most others. For a corpus about organisational complexity, near-decomposability is the load-bearing idea: it warrants treating teams, units and functions as semi-autonomous subsystems with their own internal dynamics rather than as cogs whose every motion is set from outside, and it underwrites later work on loose coupling, modularity, resilience and the nested adaptive cycles of ecological systems (Holling), while cautioning that a system which is genuinely non-decomposable would tend to escape our understanding altogether.