Annex A

1. Background Information

In 1983 the Royal Society report on risk (Royal Society 1983) made a distinction between:

• Objective risk: measured by the experts
• Perceived risk: the layperson’s evaluation of the risk

These perspectives are discussed below and then an alternative view is presented which challenges this distinction.

2. Objective Risk: Quantifying Risk

Risk Assessment is a major policy analysis technique which has come to dominate environmental policy decision-making. It is carried out by natural scientists, who believe that risk is a phenomenon that exists in some quantity and can be verified with objective, replicable measurement:

“As a probability in the sense of statistical theory, risk obeys all the formal laws of combining probability.” (Royal Society 1983)

Scientists aim to refine their methods and collect more data on probabilities and magnitudes to make increasingly accurate estimations of risk (Adams 1995).

Risk = probability ´ magnitude

Approaches to quantifying risk:

a) Actuarial approach (usually applied to natural hazards): 'expected values' - relative frequency of an event averaged over time, extrapolated from statistical data from previous years (e.g. floods)

b) Probabilistic risk assessment (usually applied to technological hazards): attempts to predict the probability of safety failures of complex technological systems, often in the absence of sufficient data. Using fault tree or event tree analysis the failure probabilities for each component of the system are assessed and combined. The same 'product' is generated as in actuarial approaches, i.e. an estimate of how many undesirable events one can expect over time

c) Toxicology/epidemiology (usually applied to long term chemical hazards): causal relationships have to be explored and modelled explicitly (i.e. the relationship between a physical risk agent such as dioxins, and the physical harm observed in humans) - based on toxicological (animal experiments) and epidemiological (comparison of population exposed to risk agent with population not exposed to risk agent) evidence.

Expressions of risk

a. Individual risk:

• is one of the most widely used expressions
• is usually expressed as a mortality rate (the fraction of the population at risk that suffer death per unit time, i.e. annual number of deaths divided by the total population at risk)

b. Death per unit measure of activity:

• relates to a specific activity as risks are related to the level/ type of activity
• measured in deaths per unit time of activity (e.g. passenger hours for transport deaths)
• a further measure for occupational risks has been developed - Fatal Accident Rate - which is the number of deaths per man-hour (sic) of exposure

c. Loss of life expectancy:

• useful for evaluating the delayed effect due to exposure
• can compare age-specific death rates of those 'exposed' with the norm

d. Frequency against consequence:

• used in the case where the magnitude of an event may vary and result in different consequences
• often shown in the form of a graph

e. Costs and benefits:

· the economic approach transforms physical harm/undesirable effects into 'utilities' – and attempts to determine the optimal trade-off between the benefits of risk taking and the costs

Risk assessment has a number of aims, depending on the context:

• to reduce risk
• to supply the public with more objective (technical) information with the intention of offsetting emotional and uninformed thinkers (Fischer 1995)
• to find the optimal trade off between the benefits of risk taking and the cost (to determine for example whether a particular chemical should be banned, or whether a proposal for a hazardous industry should be allowed to go ahead).

3. Perceived Risk: Psychological Perspectives

The mainstream position is that perceived risk, the layperson's anticipation of future events, is subjective and therefore inaccurate. As lay people have tended to remain unswayed by formal risk assessments, this has served to reinforce the view of risk experts that the public are irrational.

Factors that are considered to affect risk perception:

• several studies have concluded that people are prepared to accept a much higher level of risk from voluntary activities such as skiing, than those they will tolerate form involuntary risks that provide the same level of benefit (by roughly a thousand times)
• high impact, low probability, events are 'overestimated' by the public (e.g. the layperson tends to worry far more about air travel than driving the car to the airport, despite the fact that the latter is far more risky)
• familiarity tends to lead people to underestimate risk - whereas new risks are likely to be overestimated
• the scale of institutions and technologies tends to affect the degree to which the risk appears to be voluntary or involuntary. The less control and trust people feel the more they are likely to fear the risk.
• distance from the threat has some influence on risk perception
• if the danger comes from another country the risk will again, be perceived to be less voluntary and therefore greater
• poverty may yield a lower level of perceived risk as the relative benefits may be higher
• similar activities/ associations affect the level of perceived risk (e.g. nuclear power and nuclear weapons)

Numerous efforts have been made to measure the perception of risk. In the 1970s, Slovic developed psychometric tests to produce a cognitive map of risk perception. Such tests are based on the premise that risk perception is quantifiable and predictable.

4. Problems with Formal Risk Assessment: Scientific Uncertainty

Since the early 17th century, scientific investigation has been dominated by the philosophy of positivism. This posits that there is an external reality driven by laws; that the scientist remains detached; and that through a process of ‘reductionism’, the complex components of the world can be broken down, analysed, understood, and used as a basis for predictions. Thus the conventional view of science is that it is objective, unbiased, and value free. Scientific theories are built on the (neutral) observation of empirical 'facts'.

In this view, the goal of science is to obtain a clear reflection of nature, free from any social and subjective influences which may distort the 'facts' or theoretical understandings derived from them. However, many sociologists of scientific knowledge challenge the view that the object and the analyst can be separated from one another; they argue that science itself is ‘constitutively social’ - i.e. that social factors influence scientific outcomes. Analysis of environmental problems comes under particular scrutiny. Some of the reasons for this are:

• scarcity of environmental data
• the reductionist principle of science is inappropriate for complex ecological systems (Hannigan 1995).
• environmental scientists are often working on the 'margins of observability' - e.g. subtle and barely tangible forms of damage or change (Irwin 1995).
• choice comes into what to observe and how to observe it (different 'facts' can be generated)
• by observing the facts, the facts themselves might be changed (cf Heisenberg’s uncertainty principle, which shows that the act of measuring the location of a particle alters the position of a particle in an unpredictable way)
• different interpretations are possible, based on interpolation and extrapolation principles (especially predictions)
• numerous feedback mechanisms exist which generate a huge degree of uncertainty
• assumptions are frequently made that ecological systems function in a linear and equilibrium fashion. Attention is increasingly turning to the non-linear and multiple stable states of ecosystems

Adams (1995) stresses that the greater the degree of scientific uncertainty, the more we are guided by assumption, inference and belief. This opens the door to heightened contestation of environmental problems (Irwin 1995), compared with issues where scientific ‘facts’ have less variability.

It has been suggested by the philosopher of science, Thomas Kuhn (1970: ), that "an intrinsic feature of scientists' work is their selective inattention to evidence which does not conform to their picture of reality". Kuhn (1970: ) also identified what he termed "the dogmatism of mature science" which he defines as a deep commitment to a particular way of viewing the world and practising science within it. He suggested that:

• Most science proceeds within the existing paradigm (the ‘rules of the game’ that inform scientists of the questions that they may legitimately ask and the techniques that can be used to search for answers), which he calls 'normal science'.
• Unexpected discoveries are resisted and innovation is suppressed until a time when the anomalous findings become particularly stubborn and striking.
• This causes an intellectual crisis and forces scientists to raise questions about established beliefs and procedures. A paradigm shift results (e.g. from Newtonian to Einsteinian physics).

Other authors have developed this theme with reference to actual case studies. Latour and Woolgar (1979) for example, show exhaustively that ‘facts’ become stabilised only through a process of social negotiation among scientists who have a stake in the outcome (Bird 1987). As Bird (1987:257) explains:

“According to their view, scientific consensus is reached not when the facts “speak” for themselves, but rather when the political professional and economic costs of refuting them are such that further negotiation becomes untenable”.

As environmentalists themselves often use scientific evidence as a basis of their argument, critiques of science may also destabilise their arguments and grounds for protest. Some authors (e.g. Bird 1987) argue that, rather than relying on science as a foundation for action, we should instead rely on inter-subjective, socially negotiated moral truths achieved in the interests of (environmental) justice. In this view, science can be used more honestly as a tool, rather than a discipline that is in touch with some absolute reality (e.g. Pretty 1994).

Science, then, may not be able to present us with an accurate measure of environmental risk for a number of reasons, and risk methodologies have themselves become a source of contention. Some of these problems and concerns are:

• Records of past risks are not an accurate guide to the future because people respond to risk, thereby changing it. For example, insurance companies consult past claim experiences in calculating premiums they charge to cover future risks. This in turn affects people’s risk taking behaviour (Adams 1995).
• The historical data record itself is unreliable. Deaths are most accurately recorded, but injuries are often under-recorded.
• It is often difficult to isolate causes (e.g. for cancer), so specific experiments may be used to test for a direct relationship between a dose and a response. Problems exist with such dose-response experiments. For example, observations on animals are of uncertain relevance to humans, partly because the high doses necessary to provide statistically significant results in animals are then extrapolated to the much lower doses to which humans are normally exposed. Underlying assumptions - such as whether the relationship between dose and response is linear, supra-linear, sub-linear, or threshold - further increase the uncertainty. Thus a variety of conclusions might be drawn from the same experimental data, depending on the assumptions used in extrapolating to lower doses.
• Long latency periods and the possibility of synergistic effects between different chemicals further increases the levels of uncertainty in calculating risk.
• In economic analysis discounting is necessary, but the choice of discount rate (intended to reflect how we value the present over the future) is fairly arbitrary. Further, numerous problems exist with trying to monetize the benefits, such as a clean environment.
• It has also been suggested that risk assessment itself is biased in favour of the dominant techno-industrial system and its values. A change in language from danger and hazard to risk defuses the threat. It prioritises 'expert' decision making over the public who are affected (Fischer 1995).

5. The Cultural Construction of Risk

Increasingly, then, the view that a separation can be maintained between objective risk and subjective risk is being challenged. Indeed, this is recognised in the 1992 Royal Society report on risk (Royal Society 1992). It stems, in part, from the inability of science to provide certainties. No interpretation of risk is correct - all are biased as they confer different meanings on different situations, events, objects and relationships (Hannigan 1995; Dake 1994). This leads Plough and Krimsky (1987) to argue that the view (of risk experts) that the public are incapable of digesting complex technical findings - and are therefore left to fall back on irrational judgements - misses the point. In an uncertain world, the premises upon which rational arguments are constructed vary. It is not that some arguments are rational and others irrational, but that our belief systems rest on different and often competing rationalities. This view has been labelled cultural theory and it “illuminates the world of plural rationalities” (Adams 1995: ix). Where there are unresolved debates (for example between the nuclear industry and environmentalists) cultural theory seeks explanation, not in further scientific analysis, but in analysing the different premises and rationalities from which the groups are arguing. Various categorical systems have evolved within this theory. Plough and Krimsky (1987), for example, make a broad distinction between:

· technical rationality - the mindset that puts its faith in empirical evidence and scientific method

· cultural rationality - emphasises the opinions of traditional and peer groups over those of the experts. It focuses on personal experience rather than depersonalised calculations

A more sophisticated categorisation has been developed by, amongst others, the anthropologist Mary Douglas (for a summary of which, see Adams 1995: especially chapter 3). In this view, a functional interdependence exists between people’s view of the world and their adherence to a particular set of social relations. Five types of social relations are identified - namely hierarchical, individualist, egalitarian, fatalist and autonomous. The first four of these, it is suggested, have corresponding views of nature - namely nature as perverse or tolerant, benign, fragile and capricious respectively. Thus it is often possible to discern people’s ‘worldview’ and how this affects their risk perception.

6. Classifying Philosophical Perspectives

Proponents of the subjective/objective risk division, and those believing in the socially constructed nature of all risk, may be seen to have different underlying philosophies. Some key terms are:

Ontology: Theory of what exists/ the nature of reality

Epistemology: Theory of knowledge/ what we can know

Ontology and epistemology may be viewed in terms of realism and relativism.

The view that there exists an independent, external reality is ontological realism. The view that we can attain a precise reflection of reality through our study of it is called epistemological realism.

The view, common to all social constructionists, that nature can never be known exactly - because our knowledge and understanding of it is affected by our culture, politics, worldviews, scientific paradigm etc. - is called epistemological relativism. The more extreme view that there is no common reality - that we all live in distinctly different worlds, that there is no external reality and therefore no foundation to knowledge - is called ontological relativism.

It is evident that the technical and cultural rationality perspectives put forward by Plough and Krimsky (1987), can be categorised according to these different underlying philosophies. Technical rationality is grounded in a faith in science, to accurately reflect risk, and may therefore be said to be both ontologically and epistemologically realist. Cultural rationality, on the other hand, lacks faith in science and the ability of scientists to discern reality, and may therefore be seen to be grounded in epistemological relativism (and either ontological realism or relativism).

References

Adams, J. (1995) Risk, UCL Press

Bird, E. (1987) The social construction of nature: theoretical approaches to the history of environmental problems, Environmental Review, 11, 255-264

Dake, K. (1992) Myths of nature: culture and the social construction of risk, Journal of Social Issues, 48 (4), 21-37

Fischer, F. (1995) Hazardous waste policy, community movements and the politics of Nimby: participatory risk assessment in the USA and Canada. In Fischer, F and Black, M. (eds) Greening Environmental Policy: The Politics of a Sustainable Future, Paul Chapman Publishing

Hannigan, J.A. (1995) Environmental Sociology: A Social Constructionist Perspective, Routledge

Irwin, A. (1995) Citizen Science, Routledge

Kuhn, T. (1970) The Structure of Scientific Revolutions, University of Chicago Press

Plough, A. and Krimsky, S. (1987) The emergence of risk communication studies: social and political context, Science, Technology and Human Values, 12 (3-4), 4-10

Pretty, J. (1994) Alternative systems of inquiry for sustainable agriculture, IDS Bulletin, 25.2, 37-48.

Latour, B. and Woolgar, S. (1979) Laboratory Life: The Social Construction of Scientific Facts, Beverly Hills

Royal Society (1983) Risk Assessment: A Study Group Report (Royal Society)

Royal Society (1992) Risk: Analysis, Perception and Management (Royal Society)

Back to tlr