The conversion of this document from MS Word to html has seriously degraded the image quality. The original document is available as RTF.
Individual risk management – climbing a tree, riding a bicycle, driving a car... - is a balancing act in which the potential rewards of a course of action are weighed against potential adverse outcomes. Unfortunately, both rewards and "accidents" come in a great variety of incommensurable forms. We do not undertake a formal probabilistic risk assessment before crossing the street.
Figure 1.
At the level of institutional risk management the impossibility of quantifying interpersonal comparisons of utility (do I like Macdonald's hamburgers more than you like Mozart operas) presents a further - thus far insoluble - problem.
Yet another difficulty associated with institutional risk management is that it usually focuses exclusively on risk reduction – the bottom loop of Figure1; the rewards of risk taking in institutional settings are usually systematically undervalued.
Different risks are managed differently. It is important to be clear about the nature of the risk one is dealing with. Figure 2 partitions risk into three categories. The overlap between the categories acknowledges the existence of grey areas at the boundaries. Psychiatrists, for example, deal routinely with people who construe reality idiosyncratically, people who perceive threats that "normal" people cannot see. But the concept of normality can embrace a range of perceptual filters. Even where risks are clearly perceptible, either directly or with the help of science, it can be difficult for "normal" people to reach a consensus about what is normal and what is not. Agreement about virtual risks is much more elusive.
Figure 2.
Participants in debates about virtual risks frequently question their opponents’ grip on reality and denounce those who disagree with them as "mad". With such risks filters are all; the perception of virtual risk comes uncomfortably close to hallucination - defined as "perception in the absence of external stimuli".
Figure 3: a typology of perceptual filters.
With risks such as global warming, genetic engineering, mobile telephones etc the egalitarian response is to insist that if you cannot prove it is safe, you must assume that it is dangerous and ban it. The individualist response is to insist that if you cannot prove it is dangerous you are entitled to assume that it is safe; publish what you know and let the shopper decide. The hierarchist – big business, big government – wrings his hands and tries vainly to accommodate both. Optimality is not on offer.
When risks to health and the environment are scientifically unresolved, people are liberated to argue from their pre-established beliefs, convictions, superstitions and prejudices. When science cannot settle an issue, or even attach probabilities to it, the prospect of institutional risk managers being able to chart a course that is optimal for a collectivity is remote.
Question for discussion:- are the potential social and economic consequences of the Millenium Bug virtual risks?
References – the above is drawn from the following publications by John Adams
Risk, UCL Press, London 1995.
What do mad cows, Brent Spar, the NHS, and contaminated land have in common? in What Risk?: Science, Politics and Public Health, Roger Bate (ed) Butterworth Heinemann, 1997.
Virtual risk and the management of uncertainty, in Science, Policy and Risk, The Royal Society, London 1997.
A Richter Scale for risk? Interdisciplinary Science Reviews, vol. 23, no. 2, 1998.
Computer Science Home Page