Safety is defined in dictionaries as freedom from the risk of harm or danger. Since all human activities involve some risk, to a greater or lesser extent, there is no such thing as zero risk any more than there is complete, unbridled freedom, and hence there is no absolute safety. Mr. Justice Wright, in his Reasons for Judgment in dismissing claims by Energy Probe and others concerning the Nuclear Liability Act, expressed it thus:
"The level of adequate protection need not, and almost certainly will not, be the level of 'zero risk'. ... [S]afe is not the equivalent of risk-free .... [An activity] can hardly be considered 'unsafe' unless it threatens ... a significant risk of harm."
Parliament, in the new Nuclear Safety and Control Act, effectively defines safety as the prevention of an unreasonable risk to health, i.e., there is no requirement for zero risk or for minimizing risk. Thus both legal and legislative precedents deny the myth of zero risk. (The technical definition of risk is provided at the start of Canadian Approach.)
If zero risk is unrealistic, we have to decide what level of risk is acceptable. One soon realizes that the answer is not simple: nobody wants to accept a risk, however small, unless there is some associated benefit. What is an acceptable level of risk for an activity yielding major benefits should be unacceptable for one yielding trivial benefits: a simple example illustrates how acceptability depends on both safety and benefits. The risks from many activities are acceptable to society as a whole because of their substantial benefits, even if the risks are unacceptable to some individuals who are, or perceive themselves to be, at greatest risk, e.g., people living close to an airport. Thus there can be no single "acceptable risk", and the risk that would be acceptable for any particular activity should depend on the expected benefits. (The costs involved in achieving an activity can be regarded as negative benefits so that "benefits" here are net benefits.) Another consideration in deciding what risk is acceptable is how the risk/benefit ratio for the activity proposed compares with the ratio for alternatives that achieve the same objective by different means. Is there another way of getting the same benefits at lower risk?
This leads to the Alternatives and Consequences Principle (Chapter 13), applied here to decision-making on risks. Instead of deciding first what risk is acceptable, one should determine what alternatives can provide the desired objective, then estimate the consequences expected for each one in terms of benefits and risks, before deciding which gives the best balance of risks and benefits. The risk associated with the alternative selected becomes the "acceptable risk" for that objective.
In any discussion of risk we must recognize a major difference between the risk associated with various activities as judged by professional risk analysts and by members of the general public. Some analysts argue that their estimates are the "real" risk, and that they have an obligation as professionals to do their best to protect their clients, the public. On the other hand, it is sometimes argued that public policies should be based not on the analysts' risk but on popular perception of the risk. This would have stopped automobiles and electricity before they started since, in their early days, both were perceived to be much more dangerous than they actually were. Automobiles had to be preceded by flagmen and electric wall-switches bore a sign:
"The use of Electricity for lighting is in no way harmful to health, nor does it affect the soundness of sleep".
Neither extreme view of risk is defensible. Public perceptions cannot be ignored if acceptance is to be gained for nuclear energy, or any other technology. Equally, the best estimates by qualified analysts cannot be ignored in assuring responsible protection of the public. Both must be considered in arriving at safety policies.
To study what affects people's perceptions of risk, Paul Slovic and his associates asked four different groups in the U.S. to rank thirty different activities and technologies in order of their threat to life. Two of the three lay groups (members of the League of Women Voters, college students and members of an "Active Club") perceived nuclear power to be the riskiest, while the fourth (experts in risk assessment) ranked it twentieth, below aviation and railroads, for example. When asked to estimate the fatalities to be expected from these activities and technologies in an average year, the experts predictably exhibited a closer correspondence with actuarial values than the lay groups, but this did not explain the difference over nuclear power. When asked to estimate how much worse things would be in a particularly bad year, lay groups considered that the fatalities from almost all the activities and technologies would roughly double, but that those from nuclear power would increase approximately a hundredfold.
Thus people's perception of the risk from nuclear power involves a dread of possible disaster. Slovic's studies did not examine whether the lay groups knew of, but disbelieved, inquiries that show such a disaster to be extremely improbable in North American reactors. According to those who conducted the study, human nature is such that people confronted with evidence of a risk tend to accept or reject either the evidence or the cause of the risk, depending on their existing beliefs:
"New evidence appears reliable and informative if it is consistent with one's initial belief; contrary evidence is dismissed as unreliable, erroneous, or unrepresentative. Thus, depending on one's predispositions, intense effort to reduce a hazard may be interpreted to mean either that the risks are great or that the technologists are responsive to the public's concerns. Likewise, opponents of a technology may view minor mishaps as near catastrophes and dismiss the contrary opinions of experts as biased by vested interests."
The public (in the U.S.) overestimates the annual deaths from botulism by a factor of about one hundred. I have argued that if the public is aware of the actual, actuarial, death rate and wants to expend a disproportionate share of its resources on preventing botulism, that is its right. However, risk analysts should do their best to ensure that the public is aware of the actual risk. An article in the New Yorker provided a more illuminating example. In a survey 75 per cent of "Americans" believed that the U.S. spends too much on foreign aid. To the question on how big a share of the federal budget currently goes to foreign aid, the median answer was 18 per cent; and on how much would be appropriate, the median answer was 5 per cent. Actual expenditures are less than 1 per cent. This further illustration of the don't-confuse-me-with-facts-my-mind-is-made-up principle has to be recognized in considering that public attitudes depend on more than quantifiable information. Expressed differently, these are examples of the public's insistence on its right to be wrong according to experts.
To a very large extent nuclear safety is concerned with the release, actual or possible, of radioactive materials. Any industrial activity is prone to conventional accidents, such as falls, burns, etc., and the nuclear industry is not exempt from these. However, it is the possibility of exposure to radiation that arouses public concern. With some nuclear applications, such as the transport and use of radioisotopes for medical and industrial purposes, the concern is that individuals may be exposed to direct external radiation from an insufficiently shielded radioactive source. To most people, however, nuclear safety is about the release of radioactive material from nuclear power plants, either in routine discharges or in the event of a serious accident to the plant. The concern here is that people in the surrounding area may be exposed to dangerous levels of radiation through ingestion or inhalation of the radioactive material, or by external radiation from a cloud passing over them or from deposited radioactive material.
Most people living in the neighbourhood of CANDU reactors do not appear to be troubled by the risks associated with the normal operation of these reactors. However, even they are not immune from the fear of a severe accident. We may not believe in ghosts but we still avoid walking through graveyards at night.
| Go to Contents of Chapter or of Book |