Risk and time: from existential anxiety to post-enlightenment fantasy

Abstract

This editorial examines the conceptual and practical links between risk and time. It recognises that risk analysis may be regarded as both a product of enlightenment thinking, and a candidate for the means to deliver the ultimate enlightenment fantasy: to provide successive generations with a capacity to progressively control the world more effectively. It argues that, despite its significant achievements, risk analysis has fostered romantic fantasies that themselves generate risky futures. The discussion is illustrated with reference to emergency planning, the likelihood of nuclear accidents, long-term nuclear waste disposal and the exotic financial instruments that were implicated in the recent international financial crisis.

Keywords:

• risk
• time
• enlightenment
• fantasy
• normalisation of deviance
• futures trading
• interaction
• ethnomethodology

This editorial responds to an invitation¹ to contribute some thoughts on risk and time for the fourth of a series of special issues of this journal, published to mark the twentieth anniversary of the Royal Society’s 1992 report, Risk: Analysis, Perception and Management. It is almost a truism to begin by noting that risk analysis is, by necessity, future-oriented. It seeks to control for future possibilities, by adjusting exposure to undesirable contingencies and maximising the chances of desirable ones in the most cost-effective way possible. I will argue that, despite its significant achievements, risk analysis has fostered romantic fantasies about controlling the future. These fantasies can themselves generate risky futures.

Seen genealogically, risk is a style of thinking that was born, in its modern statistical form, during the nineteenth century (although it is possible to trace its lineage back to the late seventeenth century), and one which has made possible new ways of understanding and acting upon contingency: ‘risk thinking has brought the future into the present and made it calculable’ (Rose 1999, p. 246; see also Hacking 1990). Thus stated, risk thinking, as a set of techniques, brings a sense of authority and control, of science one might say; a safe basis for action and enterprise. In this way, risk analysis might be regarded as not only a product of enlightenment thinking, but also the means to deliver the ultimate enlightenment fantasy: the idea that successive generations will learn more about the world, and so become more able to manipulate it for human gain and contentment.

This enlightenment ambition has been brought into question in recent years by a diverse range of thinkers, but, I suggest, one can still find evidence of its influence in numerous areas of practice. First, let us consider some grounds for doubt. The idea of ‘colonising the future’ looms large in Tony Giddens’ (1991, 1994) theorisation of the nature of modernity. Rather than controlling the future, he means the restless dynamic by which contemporary societies reflexively make the world anew. In this way, modernity is a ‘risk culture’, and to live in that world has the feel of ‘riding the Juggernaut’ (Giddens 1990, pp. 151–154). This is not to say that the world has become more uniformly dangerous than it was for previous generations. Rather, in contrast to traditional societies, contemporary ones have become open to possibilities that bring not only new wonders but also new challenges and fears. Processes of innovation not only bring new gains, but also manufacture new risks that serve to undercut old certainties.

There is an urgent need to question whether the post-enlightenment lesson about human limits to control the future has been learned. It is a painful truth, given the intolerability in human sensibility of any sort of loss or order in the world. Giddens recognises this by placing existential anxiety at the heart of his eclectic theorisation. As Heidegger (1962) pointed out, human consciousness is shaped by the constant presence of the inevitability of death. The human need for the containment of death and the contingent, and the maintenance of order, is intimately related to risk, as risk threatens to puncture the carapace of basic trust that humans need in order to function in the world. Here, we can see the basis for a profound understanding of the sociology of risk. Such loss of order has been variously captured in terms of ‘breaching’ (Garfinkel 1967), ‘reality disjunctures’ (Pollner 1987) and ‘noetic crises’ (Maynard 2003). Examples of such risk-related social disruption include Strong’s (1990) account of the social impact of HIV/AIDS in the pre-antiretroviral era, and what he described as the ‘plagues of fear, panic, suspicion and stigma’ which accompanied the disease (discussed in Horlick-Jones 2011).

At its best, risk analysis makes possible the intelligent navigation of an imperfect and potentially frightening world, and the effective use of resources to deal with all manner of contingencies. However, if inappropriately used, risk analysis can itself be a source of risk. It can lead to fantasies characteristic of the worst romantic excesses of enlightenment thinking. Most worrying is the tendency for certain risk practices to ‘sweep under the carpet’ future risks, and to leave them hidden from sight (on the ‘risks of risk management’ see Power 2004, Horlick-Jones 2005)

Consider the enthusiasm, some 20 years ago, within the emergency planning community to embrace what was then seen as a fashionable move towards risk-based analysis. Emergency preparedness needs to deal with the range of possible contingencies, not only the most probable ones. One needs to plan for an envelope of contingencies, and where this is in conflict with the very reasonable idea of making the optimum use of resources, there is a need to adapt planning to allow an escalation of response in order to deal with low-probability/high-consequence occurrences. The dangers of being unprepared for rare events are clear; especially so after an extended incident-free period.

Similarly, although the likelihood (as indicated by modelling) of some severe contingencies may be extremely small – for example the melt-down of a nuclear power station – such events do happen, as illustrated by the recent Fukushima event. For some years, the basis of ‘engineering out’ the possibility of such events has been regulated in terms of the ‘tolerability’ of a certain, non-zero, level of risk (Health & Safety Executive 1992). This residual risk is expressed as a probability of failure per annum. It follows that if considered over a sufficiently long time period, some kind of incident will occur. In this way, the calculation of a tolerable level of residual risk amounts to ‘spreading’ the risk out over a time line until it produces a small yearly rate of exposure. Of course, when a failure does occur, no matter how unlikely, a whole incident will take place, not a tiny proportion of an incident. Designers seek to ensure that worse-case failures are limited in terms of their potential damage, but, historically, unforeseen factors can lead to more severe incidents, in which Fukushima follows in the tracks of Three Mile Island and Chernobyl.

The long-term storage of radioactive or other dangerous waste is another contentious area where certain cost–benefit calculation technologies, that purport to provide a value-free, scientific, basis for decision-making, serve to obscure future risks. The use of discounting procedures result in the scale of future risks and the value of the lives of future persons being much reduced in comparison with current values (e.g. Shrader-Frechette 1993).

The elementary point that needs to be made here is that the only way of being sure about avoiding future risks is to avoid doing the activity in the first place.² I made this observation in my evidence to the recent UK Parliamentary inquiry into the risks associated with energy systems, and was pleased that it was recognised in the conclusions of the resulting report (House of Commons 2012, p. 36):

Changes in energy policy in Germany after Fukushima illustrate an important point, which was also highlighted by Professor Tom Horlick-Jones who stated that ‘no matter how safe an industry can be made, … ultimately its acceptability [depends on] whether the society in question can live with possible failures’.

Another important lesson of Fukushima lies in the potential for institutional creep to occur, in which, in the absence of failures, the level of ‘acceptable’ risk rises implicitly over time. This took place in the case of the Challenger space shuttle disaster, and was very effectively analysed by Diane Vaughan (1996) in terms of her notion of the ‘normalisation of deviance’; as professionals in routine contact with risk issues (about which they are regarded as expert) over long, incident-free, periods of time can arrive at a shared appreciation of the acceptability of certain practices that, with the benefit of hindsight, might be regarded as inherently risky. My colleague Mike Levi and I recently (2013) used this sort of analysis in investigating the Fukushima accident, and the extent to which the incident could be regarded as some kind of a corporate crime. We concluded that the most plausible way of understanding the failure is in terms of the development of ‘taken for granted aspects of organisational life that created a way of seeing that was simultaneously a way of not seeing’ (Vaughan 1996, p. 394). In such circumstances, technical risk-based safety procedures can amount to little more than ‘fantasy documents’ (Clarke 1999).

The tendency towards risk analysis becoming a rhetoric that serves to obscure a source of residual risk that is portrayed as negligible was responsible for perhaps the most significant failure in recent years. I have in mind the 2008 global financial crisis; the consequences of which have yet to become fully apparent. The risk objects at the heart of it were financial products that entailed trading in possible futures. It is difficult to find a better example of how ostensibly sophisticated risk-management techniques produced a fantasy of control within a community of professionals, who were in the thrall of a collective exercise of normalisation of deviance. Gillian Tett’s (2009, pp. 113–114) gripping account of the developing events conveys the nature of the crisis beautifully:

The schemes became more creative still when banks started creating these products not out of bundles of mortgage loans, but out of derivatives made of mortgage loans. The idea was borrowed from the world of credit default swaps, and as with corporate-loan-based CDOs,³ these derivatives versions of CDOs enabled investors to place bets of whether mortgage bonds would default or not….They would lead to a frenzy of speculation, all based on the fundamental premise that the default risk bundles of mortgages had been virtually erased by the process of bundling and then slicing them into tranches….the risk had been so effectively dispersed that the chance the banks would ever take a hit from it seemed so remote as to unfathomable.

I bring this editorial to a close with a plea for modesty, addressed to risk managers everywhere. For scholars, there is similarly a need for limited ambitions and, I would argue, for investigations of risky situations that are grounded in the fine detail of interaction, and the pre-occupations to which actors are demonstrably oriented. As a corrective to the aspirations of grand theory in risk scholarship, I bring attention to a paper on the social world of futures traders (or, as she notes, self-styled ‘masters of the universe’) that was written, not long before her untimely death, by the late Dede Boden (2000). Although Boden was at the forefront of appreciating how globalisation is subtly changing the nature of social interaction, she was resolute in locating her analysis of trading processes – events that are extended in space and time – in the here and now of everyday interaction:

Local/global qualities interact constantly. On satellite and cellular phone, Goffman’s ‘interaction order’ and Garfinkel’s ‘irremediably local practices’ go global….(brokers) are thus acutely aware of, responsive to and occasionally depend upon more subtle interpersonal cues such as volume of voice, posture, gesture, pace, direction of gaze and general choreographed demeanour of other brokers in their vicinity and beyond….How the future and present interpenetrate is what futures trading is all about. (Boden 2000, pp. 184–195).

Acknowledgements

I am most grateful to Patrick Brown, Bob Heyman, Lorenzo Marvulli and Ana Prades for their comments (which were both supportive and incisive) on an earlier draft of this editorial.

Notes

^1. There is a certain grim irony in me receiving this invitation. At the time of writing, I am receiving heavyweight treatment for cancer for the second time in my life. Six years ago, I was diagnosed with, and treated for, breast cancer (Horlick-Jones 2011). Some ten months ago, metastases were detected. There is no cure for this illness. It is a question of whether the condition can be stabilised, and for how long. At present, risk is far from simply an academic or theoretical concern for me, and time is something that seems in short supply. I hope readers will forgive this short biographical note. I should say that these circumstances served to concentrate my mind on more analytical questions concerning risk and time. I hope the resulting text is helpful.

^2. Of course, this underlying logic is used by critics when invoking the precautionary principle as a basis for demanding severe constraints on the use of certain technologies.

^3. I have not annotated this extract with notes so as to avoid losing the power of the prose. In essence, the banking community had developed products (e.g. collateralised debt obligations, or CDOs) that had abstract relationships (often based upon complicated mathematical formulae) to underlying assets. Many of the these products functioned by trading upon future prices of commodities, options on purchasing or selling products at certain prices at some point in the future, or entailed institutions swapping future obligations. Tett’s (2009) detailed account suggests that these products were developed in an introspective environment that was weakly regulated, so making possible the pre-conditions for fantasies of invulnerability to take root and proliferate.