ABSTRACT
This chapter addresses the initial challenges of designing a risk system for an industrial application. We argue that the widely accepted norms in this arena are scientifically indefensible. They indicate a system with a character which cannot easily be understood as organisationally pertinent. They seem to accept, without real question, definitions, measurement scales and combination rules which are so robust as to border on meaningless. Nonetheless these are applied to a discretised list of isolated events the relationships between which are insufficiently explicated to support their subsequent aggregation into priorities based on an indeterminate local judgement process. We discuss a completely revisionist approach, beginning with a logical four-part schema which deconstructs the accepted norms into its rightful constituent analyses. These being the semantic clarification of the risk object, the organisationally grounded abstraction of its referents, the statistically valid measurement of its properties and the imperative to formulate its outcomes in business material terms. Following this analysis, we speculate on some of the reasons why the impoverished norms persist. We propose four main reasons concluding that, whatever the cause, the approach to risk it produces is hugely postmodern and lacking in any qualities that would warrant the definition of a system.
