Andrew Brook

In Canada and the United States, we consume a huge amount of energy and other goods relative to other parts of the world and are totally dependent on large industries. Among the problems created by this way of life, the vast quantity of often dangerous wastes we produce is among the more difficult. A particularly interesting case study for the ethics of wastes and their management is the nuclear power industry. It presents some major waste management problems, problems that will require enormous amounts of money and labour to solve.

What to do about nuclear wastes is a policy question; policy questions always have moral questions at their heart. Those posed by nuclear energy tax the full resources of modern moral philosophy. In a classic study of two decades ago, Arthur Porter put it this way:

... an assessment of the value of nuclear power ... ultimately requires an examination of the acceptability to society of the risk and benefits of the technology, relative to other options. This process is, by definition, extremely difficult since value judgments of a particularly complex kind, transcending nuclear power per se, are clearly involved. Indeed, whose values are to be judged worthy and how this assessment is to be accomplished with justice are pertinent questions.(1)

All this is just as true today as it was when Porter wrote it. Of the wide range of cost/risk/benefit issues to which Porter alludes, we will concentrate on those resulting from the wastes created by nuclear power. The major wastes in the nuclear industry are the wastes from the mining and milling process used to create uranium fuel and the wastes left behind when this fuel is used to generate electricity in a reactor. We will focus on what are called high level wastes, the wastes created by burning fuel in a reactor. Our aim is to determine the values that should govern policy questions about these wastes and how to apply these values in a variety of contexts. The most central ethical question concerns our obligations to future generations.

Ethical decisions always underlie policy decisions. The former are often made in an analytic vacuum that we would not begin to accept for making design decisions or investment decisions. Similarly, cost/benefit assessments are often carried out with very narrow notions of what can properly be classed as costs or benefits. Here is how an ideal method for settling ethical issues in policy contexts might work. First we would collect the relevant facts: what are the problems, what are the possible solutions? We would next identify basic ethical principles such as fairness in the distribution of costs, risks and benefits over populations and times and liberty for the people concerned, and so on. Then we would lay out criteria for setting costs against costs, benefits against benefits, etc., both considered broadly enough to include full social costs and benefits, direct and indirect. Finally, we would apply the principles and criteria to the facts. We will carry this method out as fully as a short paper allows. We will sketch the relevant facts about the nuclear fuel cycle, identify relevant values, and lay out some of the relevant costs and benefits.

High level reactor wastes

The best known wastes in the nuclear industry are the high level wastes produced by nuclear reactors. In Canada, nuclear reactors are concentrated in Ontario, where they generate about 50% of the province's electricity. Quebec and New Brunswick also have reactors. Nuclear reactors generate electricity by setting up a controlled chain reaction in the radioactive component of uranium fuel, uranium 235. This fuel is manufactured into pellets held in tubes tied together into circular bundles and these bundles are inserted into long tubes inside the reactor. The chain reaction in this fuel generates an enormous amount of heat. This heat is used to superheat steam, which then powers turbines connected to electrical generators. In the course of the uranium being `burned', the industry term for the chain reaction, a number of highly radioactive materials come into existence inside the fuel bundles--radioactive strontium, cesium, americium, and so on--and the metal holding the pellets also becomes radioactive.

One important new mineral is plutonium, which is dangerous radiologically and also chemically--the radiation from a minute amount on the inside of the lungs can cause lung cancer and plutonium is more toxic chemically than almost any other material. Plutonium also poses a security risk; it can be refined into bomb grade material. Of the two atomic bombs dropped on Japan by the United States, one used refined plutonium, the other heavily enriched uranium 235.(2)

The radioactivity in spent fuel increases many orders of magnitude when burned.(3) Natural processes of radioactive decay restore the spent fuel to something like the radioactivity it once had but that takes 300 to 800 years (after about 500 years, the radioactivity of the fuel has decreased 200,000 times(4)). Different radioactive materials decay at rates that vary by orders of magnitude; plutonium, for example, has a halflife of about 24,000 years, which make wastes in which it occurs particularly long lived.(5)

Compared to the amounts of wastes produced by other industrial processes, the volume of high level wastes produced by a reactor is quite small, however. A pellet of fuel roughly the size of a large marble produces enough electricity to power an average house for a year. By comparison, it would take many tons of coal producing many tons of CO2 and a large amount of fly ash to produce the same amount of electricity. By comparison, all the high level wastes ever produced in Canada weigh less than 25,000 tons. In fact, all the high level waste produced by Canadian reactors in a history that is now about forty five years long is still stored onsite, first in large pools of water for six years or so, then in thick concrete containers as the fuel becomes less active. This form of storage is adequate to ensure that there is almost no release of radioactivity to the atmosphere so long as nothing goes wrong. The proviso, `so long as nothing goes wrong', is important, however; these methods need constant monitoring and maintenance. They are thus the very opposite of being a passive, permanent solution requiring no further human intervention.

Mine/mill wastes

To make the fuel burned in reactors, uranium ore is mined and milled to extract the uranium and then fabricated into fuel. This process produces large quantities of low level wastes. The amounts of these wastes are huge--well over 150 million tonnes of these wastes now exist in Canada alone--and they may lead to more exposure to radiation than any other past of the fuel cycle.(6) Because the volumes are so huge, and also because much of the waste is fluid, these `low level' wastes may well be a more intractable problem than high level wastes. Currently, no good scheme for managing them has ever been devised. Thus they eminently deserve to be studied from the ethical point of view. In this paper, however, we will focus on high level wastes. One reason is that a promising scheme has been devised for dealing with these wastes and it is in need of careful ethical assessment.(7)

Long term management of high level wastes

High level wastes are contained in highly corrosion-resistance structures. Thus, they will not disperse for a long period of time, though still well before the radioactive materials in them decay to insignificant levels. Release via a massive uncontained explosion and fire on the model of the Chernobyl disaster is unlikely in Canada or the United States. What exploded in Chernobyl was active fuel in which a chain reaction was taking place within a reactors; all reactors outside the former Soviet countries have massive reinforced concrete containment shells. When the fuel is spent and removed from the reactor, there is no longer a chain reaction. The dangers become heat and extremely high levels of radioactivity. And the major risk is over the long term. Over time, the risk that the active monitoring of water tanks and concrete containers needed to ensure safety will decrease or even cease altogether obviously increases. If the monitoring and necessary repairs, etc., ceased, living beings could spend dangerous amounts of time around the concrete containers and any spent fuel not in containers in the medium term and in the long term these containers themselves would deteriorate, allowing the spent fuel within to dissipate.

A method for permanent disposal of spent fuel has been undergoing research and development for a couple of decades now. The method involves sinking mine shafts into stable, waterproof structures of plutonic rock in the Canadian Shield.(8) The spent fuel, which is itself highly resistant to corrosion, would be put into lead containers. These containers would placed in cavities lined with clay at the bottom of the shaft (clay lined to prevent any possible moisture penetration.) The shafts would then be backfilled with clay or concrete, effectively isolating the wastes from the environment and making any contact between them and living beings highly unlikely. The aim is to achieve a level of isolation such that there would not be more than one chance in a million per year that a maximally exposed creature would develop a fatal cancer or serious genetic defect, and that this isolation be assured for at least 10,000 years without active intervention. We receive much more radiation than this every year from the radium found in all rock, concrete and soil, from X-rays, from air travel, and so on.

Everything to do with nuclear power is contentious, but nuclear wastes have one feature that distinguishes them from other nuclear issues: they are already with us. Thus there is no longer any question about whether to bring them into existence. This means that many of the ethical issues that are central to decisions about nuclear power, issues arising from such questions as whether to build more reactors or phase out the ones we already have, do not arise. The principal ethical issue that remains is this: are we obliged to assume the costs of disposing of these wastes, and as permanently as possible, or is it permissible for us to pass at least some of the costs of dealing with them on to future generations? This issues breaks into two questions:

1. What are our obligations? and,

2. To what beings: just human beings, or all creatures, ecosystems, and the biosphere as a whole?

Values for the facts

Our task now is to identify values appropriate to the facts about wastes from the nuclear fuel cycle. Nuclear wastes have two useful features as a test case for ethical issues of waste management:

1. Nobody currently alive and nobody for a number of generations is going to benefit much from finding a more permanent solution to them; the current solution will protect us and a number of generations to come quite adequately. (Spent fuel does have one possible near benefit, but exploiting it is actually made more difficult by long term disposal, a point to which we will return.) Thus managing them does not give rise to difficult questions about distribution of benefits: who should get, or be allowed to use, how much of what, when and how?, and similar questions of distributive justice. Wastes raise primarily cost questions: given that they have to be managed for many thousands of years, who is obliged to assume the costs of doing so? This simplifies the ethical situation.

2. Because nobody alive now or for some time is going to benefit much from finding a more permanent solution than the one already in place, finding a more permanent solution will not be motivated by self-interest. Thus, the arguments in its favour have to be moral ones alone.

What are our obligations with respect to nuclear wastes? A number of issues need to be distinguished:

1. Principles. What are the general principles that should guide our ethical thinking about the disposal of nuclear wastes?

2. Scope. Are our obligations restricted to humans or do they extend to other creatures? To the environment? To the biosphere as a whole? To future generations as well as to the current one? Only to people, animals and ecologies close to us or over the whole surface of the planet?

1. and 2. cover the two issues identified at the end of the last subsection. Other partly moral, partly conceptual issues are also highly relevant:

3. Discounting. Are creatures of other kinds or far distant future generations of our own kind worth less morally than people existing now?

4. Cost/risk/benefit. Given that better solutions to a waste management problem tend to cost more money, what, all aspects of risk, cost and benefit considered, is the optimal expenditure on this problem versus other social problems and demands?

5. Moral assessment and risk assessment. How should risk assessments shape our moral assessment?

6. Uncertainty. Given that we can never be certain about any outcome in a complex industrial system and given that the further we project into the future, the more uncertain we become, how can we reach ethical conclusions in the face of such uncertainty?

7. Reducing risk vs. retaining benefits. What is the appropriate balance between reducing the risks contained in high level wastes and leaving open the possibility of exploiting the very considerable economic potential that the fuel rods still contain?

8. Procedural issues. What procedures would allow us to arrive at fair and democratic decisions, and who should have what roles in them?

1. Principles. What principles should govern our ethical thinking about deep disposal?(9) One is a principle of distributional equity: costs, risks and benefits must be distributed equitably, at a time and across time. A second is that liberty is a particularly central good (benefit), at least for members of the human species, and any reduction of liberty requires particularly powerful justification. We need to narrow both notions down.

The part of the principle of distributional equity of relevance here is this:

A. Fairness. Those who benefit should bear the costs.

It follows from this principle that, since the people now alive have reaped most of the benefits of the activities that have created the high level wastes we are considering, we have an obligation to bear the costs of disposing of them, and disposing of them permanently. I do not think that there is any way around this.

That may appear to settle the matter and so far as identifying our obligations on this issue, it does. But obligations can be overridden if there are conflicting obligations that are even stronger. We will take up this issue in 4. below. Furthermore, even if A. settles our intergenerational obligations, there is also a tricky interregional issue. Most of the beneficiaries of nuclear power live in or near large centres; people in the outlying regions have reaped few benefits. Yet any long term solution to the wastes will inevitably be constructed in some outlying region; that's where the appropriate rock structures are found and there is room to build a large mine complex.

To see our way to the most ethical solution here, I think the word `inevitably' is important: if any long term disposal facility must be built in the hinterland, then there are only two choices: construction in the hinterland or forego a long term solution and continue the short term procedures now in place. If so, then disposal in the hinterland is the fairest long term solution possible, even if it is not entirely fair. If we adopted it, we would still be obliged, of course, to ensure that the costs to present and future beings in the area selected are kept as low as feasible, that benefits be maximized, and that any differential in costs is accompanied by fair compensation. The secure, well paid jobs that go with a waste disposal facility would be one compensatory benefit.

The aspect of the ethics of liberty of relevance to nuclear wastes is this:

B. Liberty. Our actions must infringe on the lives of other beings to the smallest extent reasonably possible.

This principle applies particularly obviously to human beings but it may well apply to many other beings, too, as we will see below. Here are some of the ways in which the management, or perhaps more accurately mismanagement, of high level wastes could restrict liberty: causing pain; damaging bodies; harming abilities; imposing significant protection costs; spoiling ecosystems; reducing opportunities; and so on. Some of the examples I just gave concerned freedom from (freedom from unnecessary pain, unnecessary costs, etc.) and some concerned freedom to (not harming the field of opportunities available). With this distinction, we can spell out the demand of liberty this way:

B'. We must choose the solution to the disposal of high level wastes that will reduce future beings' freedom from costs and harms and limit their freedom to pursue their life as they would live it to the minimum extent reasonably possible.

Note that even this longer version of the principle of liberty is weaker than some would argue it should be. It says nothing about enhancing freedom from or freedom to.

Of course, protecting the liberty of future beings requires that we restrict our own liberty in certain ways. In particular, the liberty to spend the resources needed to dispose of nuclear wastes permanently on things of immediate benefit to us must go. Thus, the demands of fairness and the liberty of future beings are in conflict with enhancing our own liberty. To resolve this conflict, we have to look deeper.

Here is a principle that can start to resolve the conflict:

C. Equal worth. Prior to considerations of morally relevant distinguishing qualities, each person has the same value as any other.

The argument for it is a general principle of rationality: unless we treat similar things as similar, it would be impossible to make general, comparative judgments about them at all. If so, it would be irrational to assign two relevantly similar people different moral value. This argument grounds the ethical principle of equal prima facie worth on a general consideration about what is required for rationality as a whole.

To be sure, discriminatory assignments of value can be justified--so long as there is a morally relevant difference. And we could be quite generous about what such differences might be.(10) Other kinds of partiality might also be allowable. For example, only a few people (including myself) are within my control; partiality toward these people will increase the chances of my doing what I can to procure and distribute things of value in this world. Similarly, special bonds of affection may justify special concern, if the loss of those bonds would seriously undermine my life having any point or purpose in my eyes, seriously undercut my self-confidence, and so on. All these would reduce my chances of doing something of value. (Williams has explored arguments for partiality to some people and also for partiality to humans over other beings, an issue directly related to Scope below.(11))

Some differences are not morally relevant, however. Self-interest, maximizing (or a desire to maximize) one's own benefits and minimizing one's costs simply because the benefits and costs would accrue to me, is not, for example. Here there is no relevant difference between me and anyone else; as the nineteenth century philosopher Henry Sidgwick put it, mere numerical difference makes no moral difference. Similarly, the collective self-interest of one's own generation is not by itself a morally relevant difference between one's own generation and future generations. In general, no one person (even if he is me) and no one generation (even if it is mine) have features that justify giving that person or generation preferential treatment in the distribution of costs and benefits.

Since we are the ones enjoying the benefits of nuclear energy, any passing on of the costs of the activities involved would constitute just such a discriminatory assignment of costs and benefits. From this we can conclude that we have a moral obligation to find a permanent, passive solution to the problem of nuclear wastes. If we were passing on additional benefits as large as the benefits we are enjoying and if we were also assuming half the costs of these benefits, the conclusion would not hold. But it seems unlikely that we are doing so. Of course, the principles of fairness is just an instantiation of the equal worth principle. But it is possible to argue for the latter in ways not open to the former, not directly.

It seems to me that the conclusion of the discussion above is clear: we have an ethical obligation to find a permanent, passive solution to the problem of radioactive waste management. A true ethical sceptic or someone with a strong interest in ignoring the problem of nuclear wastes might still try to wiggle out, but to refuse to accept such arguments as ethically binding would be pretty much to get out of the business of justifying courses of action, finding good and sufficient reasons for what we do, altogether. Moreover, we do care about these principles--they are deeply embedded in our view of how interpersonal relations ought to be governed, and therefore in our notions of self-respect and sense of decency.

Return now to the suggestion that liberty is a distinctively human good and therefore that the analysis above applies only to human beings. We take up the general issue of obligations to the nonhuman in the next section but on this specific issue, I think we can say the following: 1. Animals do care about their own liberty; think of how animals suffer when caged. 2. Disease, radioactive poisoning, etc., limit the liberty of animals just as much as of humans.

2. Scope. Questions of scope come in a number of dimensions. To how many kinds of creatures do our obligations extend, just to humans or also to other creatures, the various ecologies, and the biosphere as a whole? Over what space--to people, animals and ecologies close to us or across the whole planet? And over what span of time--just to us or to future generations, too? We are most apt to reach ethical judgments that can be defended if we take the widest possible scope: All beings, not just humans; everywhere on the planet, not just in our own communities; and at all times, not just now and in the immediate future. If something has the capacity to be harmed, it has interests, and if it has interests, the principle of equal worth can be applied, mutatis mutandis, to argue that it would be unfair to load a discriminatory mix of costs and benefits on it. Quantifying `discriminatory' over members of various species, the species themselves, their ecosystems, etc., will be a difficult task but it will be an unavoidable one if we wish our waste management decisions to be ethical over the scope of the whole biosphere.(12)

3. Discounting. Even if we do consider everything everywhere and at all times, does it all have to be given equal consideration? This is the difficult issue of discounting. In economics, discounting is clearly justified. An economic benefit in the distant future is worth less to me now than one immediately available to me. Is there anything comparable in ethics? Is it ever ethical to value far distant beings, beings far in the future, beings very different from us (earth worms, say, or bacteria) less than we value ourselves and the human beings immediately around us?

This issue tends to break down into two subissues. One concerns human beings far distant from us in space or time, the other concerns all other beings. In connection with the first, I cannot see any ethical justification for valuing one human being less than another, no matter where or when that human being may live. By contrast, we might be able to make a case for valuing some forms of life less than others: even if we want to keep plentiful examples of each kind of life, even down to bacteria, etc., it is far less clear that each individual in many lifeforms deserves the same concern as we extend to human individuals.

All this raises a number of difficulties for the principles enunciated earlier. For beings where preservation of genetic material may be more important than protection of the wellbeing or even the existence of individuals, liberty is at most a far smaller concern. Likewise, with such beings we switch from the worth of individuals to the worth of types, something anathema to the ethics of human beings. And so on. When one combines these complications with the uncertainties that still infect the notion of discounting for the future in general, I think we rapidly find ourselves completely at sea. For these reasons, I will restrict my discussion of discounting to human beings.

For distant human beings, especially human beings distant in time, we might try to use the economist's notion of discounting to justify extending less concern to them than we extend to human beings living now. It would argue that future generations, like other things in the future, are worth less than the equivalent things right now. This is the principle that a bird in the hand is worth two in the bush. In financial contexts it can be argued for on a number of bases: uncertainties or probabilities, rates of return on present wealth, and so on. The economist's notion probably has no relevance to moral questions. One discounts the economic value of the future on the basis of its value for us (now); one considers the moral worth of future generations on the basis of their value, period--their value for anyone, including themselves.

There are, however, other discounting principles. Some people might argue that far distant people, just by being so far distant, are worth less per head than people currently alive. While this does not seem a morally relevant difference, being mere location in time, two other discounting principles are perhaps more plausible.

1. Nature of future persons. The farther we move into the future, the greater the probability that persons alive then will be different from us in one or another of a number of relevant ways: they may be immune to radiological damage and chemical poisoning; though still at risk, such risks may no longer affect their interests; their forms of social life may be so different from ours that their moral worth is reduced; they may be such moral monsters that they would not merit our moral concern. These all seem to be very remote possibilities.

2. Existence of future persons. The farther we move into the future, the greater the probability that no persons will exist at all. This claim is sound but ignores the fact (as does the first) that significant damage from nuclear wastes could occur within a few hundred years if nothing more is done, a period short enough to reduce the probability considerably.

To see the real moral force of these discounting principles, it is essential to distinguish between epistemic possibility (`It may, I just don't know') and real probability (`It may; reliable calculation reveals a chance of . . ., which is more than insignificant'). If we just don't know, we can ensure that we have met our obligations (though we don't, of course, know that we have them either) only by acting as though there will, in the future, be people, and people relevantly like us. And even if we do know that there is some real probability that there won't be, this probability must, I think, be fairly high before our obligation to act as though there will be is significantly reduced--though this point is controversial. So long as there is any significant probability that there will be relevantly similar people, there is a probability that we have obligations. Thus any discounting principle of which I am aware seems to have only a negligible effect on our obligations to future generations.

In addition to the question of temporal scope and discounting, there is an important ethical issue that arises with spatial scope and consideration of beings elsewhere on the planet. Canada is not the only country with nuclear waste disposal problems. Indeed, compared to the problems faced by the United States, Russia or any other bomb producing country, our problems are pretty small. Both the United States and Russia have literally hundreds of times as much high level waste in storage as we do. If we can find a passive, permanent solution to our own high level wastes, do we have an obligation to use it to help other countries deal with theirs?

The issues here are practical and political as well as ethical. On the practical side, there is the issue of transportation and the dangers inherent in moving highly toxic wastes such long distances. On the political side, there is the fact that both the United States and Russia have been massively less responsible in their management of high level wastes than Canadians have been, Russia in particular. Nevertheless, a question does arise about whether we have obligations to peoples in other parts of the world as they face their nuclear waste problems.

4. Cost/risk/benefit. Earlier we looked at the implications of principles of fairness and protection of liberty for the management of nuclear wastes. Another principle is also important, indeed can overrule the demands of fairness and liberty in some contexts. It is the broadly utilitarian principle that we should strive to get the maximum benefit, that is to say, to do the greatest good, for each expenditure of resources and incurring of risk. The argument for this principle is very simple: anything else wastes resources. The mechanism for applying it is cost/risk/benefit analysis.

Cost/risk/benefit analysis is the activity of assessing the benefits of a proposal, the costs of achieving those benefits, and the risks involved. More precisely, it consists of analyzing financial risks--financial costs times probability of incurring them; nonfinancial risks--harms times probably of incurring them and costs of rectifying them--; and both the probability and the size of the benefits that would accrue.

Sometimes benefits are so great or costs relative to a benefit so high that neither an analysis not additional moral judgment is needed to decide what should be done. For example, no competent cost/risk/benefit analysis of hospitals is going to question whether we should keep them--though there can certainly be disagreement about how many we need, what kind, and where, as we are seeing! We view health as a benefit that can be traded off against other things only down to a certain level (though that level may be falling right now). At the other extreme, no competent public authority would allocate $1m to fix a few potholes. But benefits and/or costs are often not so clear.

In particular, better management of wastes is a benefit that we have historically viewed as readily tradeable for other benefits, lower debt loads for example. For this reason, we tend to make decisions about waste management on a comparative basis: for a given expenditure, what is the greatest benefit we can attain, and at what overall cost of resources and harms? I think that this is the right approach and that we should do it on the widest possible basis. In considering radiological wastes, we should take into account, for example, the fact that nonradiological toxins have no halflife. In considering an expenditure of resources to dispose of radioactive wastes, we should consider what else we could accomplish for the same money. And not just for us; also for people elsewhere and well into the future, and for beings of other kinds. (We are part of the 5% of the population of the world that uses 40% of the world's resources; we can afford to be generous.)

Here is one central issue: for a given cost, at a given probability or risk level, what is the greatest obtainable benefit? To meet our obligations to future generations with respect to the hazards in radioactive wastes would cost a lot. Would the benefit conferred on living things as a whole from expending resources on these wastes be as great as the benefits we could confer by expending the resources in some other way?

Some answers to this question might well cancel obligations arising from application of the principles of fairness and liberty. Suppose that the answer to the above question is no, that achieving a permanent solution to nuclear wastes would not be a benefit as great as we could achieve by expending the same resources in other ways. Suppose further that the greater benefit can be secured with as much fairness and protection of liberty across species, space, and time as would be true of a permanent disposition of nuclear wastes. In this situation, our obligations with respect to nuclear wastes would either be cancelled or greatly reduced. Resources can be expended only once, and we can obtain a greater benefit with as much fairness, etc., by expending them another way. Therefore, we are obliged to expend them the other way. If we have spent them some other way, we cannot expend them on nuclear wastes. But ought imply can. Therefore, our obligation to expend the resources on nuclear wastes is either cancelled or at least greatly reduced.

Whether the answer is no is a complicated, partly philosophical, partly factual question: philosophical to the extent that we would need criteria for comparing costs and ranking benefits; factual for reasons that are perfectly obvious. A couple of examples of possible alternative expenditures of the relevant resources may serve to highlight not only the complexity but also the practical reality of both the philosophical and the factual issues. There is increasing evidence that one unhappy byproduct of modern medicine is what one might call pollution of the human gene pool. The exponential increase and spread of diabetes is one example. Certainly passing on a sound gene pool would be a great benefit to future generations. An even simpler example is CO2 and other `greenhouse gas' emissions; all fossil fuel power generation releases vast quantities of CO2 and also SO2, as well as fly ash, particulate carbon and even radiation. (Some authorities suggest that the radon and other radioactive materials released when some coal is burned in a normal operation exceed the amount of radioactivity likely to be released from a nuclear power plant during the most seriously likely breakdown or accident.) Or what about the benefits of winding `big energy' options such as nuclear, fossil and hydro generation down and substituting `small' options such as conservation and solar and wind generation?

Would the benefit of improvements in the gene pool or in levels of greenhouse gas and other emissions or in conversion to small energy options be as great as the benefit of freedom from the dangers of radioactive poisoning from reactor wastes or uranium tailings? And could they be secured for a similar expenditure of resources? I doubt that anyone knows the answer to these questions. But the questions must be asked. As Bodde has put it, "our duty is to create solutions to the disposal of radioactive wastes in the context of other threats to human existence, rather than in isolation"--and not just to human existence!(13) Rational public policy requires answers to thousands of similar questions. Ask yourself: What would the balance be if the comparison were with renewable energy? And what would it be if the comparison were with heavy metal contamination? (Heavy metals have no halflife and retain their toxicity forever.)

These hypothetical cost/benefit questions are still artificially simplistic. The real cost/benefit questions (and hence the real moral questions) also include:

i. When we allocate resources to create a new benefit, how should the new benefits in turn be used? For example, should we use it to expand wealth or to enhance health, safety or make other gains in liberty?

ii. Then distributional questions arise: additional wealth or liberty for what beings, over what space and what time? We are clearly permitted to consume some benefits, but it is equally clear that if we are to honour the principles of the equal worth of every person and the worth of all that lives, we cannot consume as much as we like unless we can leave others with the same largesse. Within these parameters, however, lies an enormous range of options, both with respect to us and others alive now, us and future generations, and us and other kinds of being.

iii. Technical questions, such as whether the principle of equal worth can be realized in equal cost/benefit ratios, or whether equal cost/benefit distributions (so that everyone gets the same amount of each, and not just the same ratio) are required. This question too needs to be asked for nonhuman as well as human beings.(14)

And so on, almost without limit. Public policy that is rational from the cost/benefit point of view (and therefore public policy that meets even the necessary conditions of moral soundness) is extraordinarily complicated--something that will come as no surprise to those familiar with the deep and sometimes unavoidable vacuum in which many policy decisions are taken.(15)

It is perhaps worth noting, before we leave this topic, that cost/benefit issues may also be relevant to our previous discussion of probability and obligations. There I argued that the probability of there not being relevantly similar people, or people at all, must be determinate and fairly high before our obligations are significantly reduced. This argument may have to be modified in the following way. If the costs of meeting these obligations are high enough that great benefits could otherwise be achieved, our obligations may fall faster than would otherwise be the case. The same might be argued, mutatis mutandis, for other creatures, ecological systems, etc.

5. Moral assessment and risk assessment. Costs and benefits and other aspects of resource utilization always have risks attached: there is a risk that expending resources will not achieve the targeted benefit, that something in the process will have unintended side effects, and so on. How should assessment of risks guide our ethical assessments? In my view, no matter how well technological risk assessment based in probability calculations is done, it is never settles the ethical questions. Of course, this does not imply that risk assessments have no role to play. Even if we should always do a separate ethical assessment, we must also ground it in the best knowledge of the facts available. This is what the regime of risk assessment can provide. Nevertheless, after all the facts are in, we still have to apply our ethical principles and make an independent determination of what we ought to do.

Here is an easy example to demonstrate that level of risk and level of ethical seriousness are not the same thing. Consider the spate of children run over by school buses every year. With these accidents, increased moral acceptability is not linear with decreasing risk. Reducing the frequency of such accidents would not be enough to make the ones that remain morally acceptable. No such accidents are acceptable. That implies that the only acceptable level of risk in this case is the lowest level we can attain. In short, risk and moral acceptability are not in a linear relationship.

Here is another way in which risk assessment does not settle the ethical questions. In Canada, we generate electricity primarily in three ways: nuclear fission, burning fossil fuels (coal, oil or natural gas), and hydro power. Compare nuclear to fossil fuel generation. These technologies clearly have very different harm/probability profiles. The harms that nuclear power can inflict are catastrophically severe, as Chernobyl demonstrated, but the probability of incurring them is generally thought to be quite low, at least in Canada. By contrast, with fossil fuel, we are virtually certain to incur the harms they can produce, harms such as the CO2 and SO2 emissions and release of radium and other radioactive and/or toxic elements already mentioned and also depletion of a nonrenewable resource. But these harms are generally considered less significant than those that a nuclear disaster can inflict. Whether we are right to consider the harms inflicted by burning fossil fuels relatively less significant might be debated, but the point I want to make here is this. Only by a separate ethical assessment in the light of our various visions of the good life for ourselves and other kinds of being can one assess the high harm/low probability risks of nuclear power against low harm/high probability risks such as we find with fossil fuels.

Other philosophically difficult questions arise about risks, too. For example, how does diminishing level of risk work, or rather, how should it work, as a discounter of costs? It is quite clear that it does do so: in connection with the nuclear industry we have been able to accept some potential costs that would be quite horrendous because we have been able to convince ourselves that the risk of having to pay them is quite low. Since we cannot avoid running some low risks of high costs in whatever we do, it seems clear that some discount factor must be morally acceptable. But how much a cost can be discounted as the probability of risk moves to very small figures is totally unclear. It is an important question in connection with the nuclear industry generally and in connection with our relation to future generations in particular. The principle of equal worth is unlikely to be able to help us with it.

Second, is there such a thing as an absolutely unacceptable cost, however low the risk and whatever the benefit? Given the Doomsday aura that surrounds certain aspects of the nuclear industry, this is a question that some people have asked. Even the principles, let alone the criteria and facts, for an adequate answer probably do not exist.

A third starts from the idea that the depletion of nonrenewable resources is itself a cost being imposed on future generations. Classical economic cost/benefit analysis does not seem well suited to thinking about this kind of cost, there being no possible benefit that could compensate for it. Criteria for assigning cost values to questions of depletion are needed, and might be difficult to find.

6. Uncertainty. We can never be certain about any outcome in a complex industrial system and the further we move into the future, the more uncertain we become. How can we reach ethical conclusions in the face of such uncertainty?

One of the most ethically vexing aspects of the nuclear industry or any other big, technologically complex system flows from the limits of knowledge. Though I have spoken about many factual issues above as though the relevant facts are known with a high degree of certainty, in fact that is not always the case. It would have taken many detailed qualifications to identify even the most important limits to certainty but that does not mean that they do not exist. This compounds the problem of risk assessment: not only must we try to assess the level of risk in each relevant context, risk being a matter of the size of a danger times the probability of it occurring, we must also recognize that there are uncertainties built into these calculations so large that our conclusions could be seriously in error.

There is probably no good ethical way to take account of this limitation. The best we can do from the moral point of view is to err on the side of caution. Where we are dealing with the health and safety of large numbers of humans and other living things over large ecologies and geological periods of time, we can be sure that we have met our obligations in the face of uncertainty only if we build the worst outcomes that are at all likely into our analyses at all relevant points.

7. Reducing risk vs. retaining benefits. As well as posing very serious risks of radiological and chemical poisoning, spent nuclear fuel also houses a huge potential benefit. As was noted earlier, one of the results of fissioning uranium is an even heavier and more fissionable metal, plutonium. Plutonium does not exist in nature but substantial amounts of it are contained in spent reactor fuel. Since it is just as good a reactor fuel as uranium, it would generate enormous amounts of additional energy if it were to be `burned' in its own controlled chain reaction. This gives rise to a tricky ethical question.

On the one hand, we want to protect future generations from the risks posed by spent fuel--unprotected contact and diversion for weapons. On the other hand, we do not want to cut them off from a potentially huge benefit. What is the optimal balance between reducing these hazards over a geological time frame and leaving this potential benefit available to future generations? In the light of the diversion risk, should we make it impossible for future generations to use this resource at all? It does have potential to be used to make bombs--it could diverted and purified into weapons grade material.

8. Procedural issues. My comments on procedural ethics will be very sketchy; the topic requires a paper of its own. In connection with the management of nuclear wastes, a number of important ethical issues arise that have to do with procedures. Who should make the decisions? What procedures will allow all stakeholders to have a fair say? What procedures will generate the best decisions? (These are not necessarily the same.) How can we prevent the majority from tyrannizing minorities; should communities have veto power, for example, over the location of a facility in their area? (This is a basic problem of all democratic decision making.) Can compensation justify overruling a community's or an individual's objections? And how should the wishes of people with views very different from ourselves be taken into account?

This last question is pressing with respect to the geological deep disposal proposal. The facility is likely to be built in northern Ontario. If it is, aboriginal people will be far more affected by the facility than by any other aspect of the nuclear fuel cycle. Yet, not only have they not received many benefits from the power thus generated, they tend to view the world and our place in it in a way that is quite antithetical to big power or any other system that imposes heavy environmental demands. How should these people enter the decision making process, in particular the process of choosing a site?

The current proposal

We have now discussed the eight considerations introduced earlier: principle; scope; discounting; cost/risk/benefit; moral assessment and risk assessment; uncertainty; reducing risk vs. retaining benefits; and procedures. Let us now ask how well the current proposal to bury spent fuel in plutons in the Canadian Shield does when assessed against these desiderata.

1. Deep geological disposal of high level wastes probably achieves fairness and protects future liberty better than any other proposal.

2. Deep disposal seems likely to provide protection of the widest scope compared to other concepts, in all three dimensions, that is, kind of being, time, and place.

3. Because of its long time frame, deep disposal discounts future generations of people and other living things less than any other concept.

4. Concerning costs and benefit, the deep disposal concept clearly forces those who benefit to assume the costs. On the issue of achieving the greatest benefit for a given cost, however, I wonder if the deep disposal concept may not be less than optimal. I suspect that greater social and ecological goods could be attained by building a less secure facility and expending some of the funds in other ways. I have in mind things such as other protections for ecosystems, population control, and of course the perennial demands of feeding the hungry, helping people find sustainable ways to provide for themselves, and achieving a fairer distribution of resources.

5. Because deep disposal reduces risk as far as is reasonably achievable, given current knowledge, it does as well at satisfying our moral demands for risk reduction as any concept is likely to do.

6. At least half a billion dollars worth of research have been done in connection with the deep disposal concept. Thus this proposal probably reduces the uncertainties in our knowledge farther than other concepts.

7. Recovering the plutonium contained in spend fuel buried in plutonic rock would be expensive but not impossible. So a balance is struck between minimization of harm and retrievability of possible benefits, though one favouring minimization of harm.

8. Since most of the important procedural issues will arise as the proposal is considered in the public arena and decisions are made about commencement, site, etc., it is too early to say how the proposal stacks up procedurally.

Application to other issues

The methodology we have used for determining our obligations in connection with nuclear wastes could be applied to many other issues, too. Within the nuclear industry, some nonenvironmental issues to which it could be applied include: uranium mining and an economy based on exploitation of natural resources; marketing nuclear technology; terrorism and the weapons risk; consumption vs. conservation in industrial economies; the creation of a technological `priesthood' having esoteric knowledge not possessed by the rest of us and having a lot of power in society by virtue of their control of large industrial complexes; and other industrial and social structure issues.

The moral analysis offered above can be applied to other social and environmental issues. I have used it, for example, in connection with land use policy in British Columbia. The issues such as soil depletion, biomass depletion, and diversion of land from agriculture to housing, transportation and industry have essentially the same moral structure as the issue of nuclear wastes. They are all situations in which gaining a benefit for ourselves now will impose large costs on future generations of humans and other beings with little by way of compensating benefits unless we do something. Yet the costs of doing anything effective would be quite high and yield little if any direct benefit to us. The moral issue is also the same: Do we have an obligation to assume costs of our current activities such as these, saving future generations from having to do so?

More analyses are needed before definitive moral conclusions can be reached about nuclear energy and the environment, both factual and criterial analyses. Much has been done--a number of fairly well founded standards now exist, for example, and a promising proposal for a high level waste disposal facility is under consideration. But more remains to be done and we still do not have an integrated ethical framework within which to think about the industry and what surrounds it as a whole.

The biggest problem may not be the lack of a clear ethical framework, however. In many areas of social policy, what we are obliged to do is actually quite clear. The problem is to muster the personal and political will and the resources to do it. I think that we know what we should do about nuclear wastes: find the most permanent, passive solution to them that comes at a reasonable cost, given the costs of other potential benefits, and that will not make future exploitation of the fuel impossible. How do we marshall the will and the resources to do it?(16)


* Editor's Note: The AECB studies may be obtained by writing to: Atomic Energy Control Board, 280 Slater St., P.O. Box 1046, Ottawa K1P 5S9. Atomic Energy of Canada Ltd. documents can be obtained from: AECL Research, Whiteshell Laboratories, Pinawa, Manitoba R0E 1L0. The material you wish to obtain should be identified by name and date or number as indicated above.

Delete most pre1980 suggested further readings and add:

Environmental Impact Statement on the Concept for the Disposal of Canada's Nuclear Fuel Wastes Atomic Energy of Canada Ltd 10721, COG-93-11, 1994. The background document is entitled The Disposal of Canada's Nuclear Fuel Waste: Public Involvement and Social Aspects, M. Greber et al., AECL 10712, COG-93-2, 1994.

Schrader-Frechette, K. Ethical dilemmas and radioactive waste: a survey of the issues. Environmental Ethics 13(4) (1991), pp. 327-43.

Biographical Note:

Andrew Brook is Professor of Philosophy and Director of the Institute of Interdisciplinary Studies at Carleton University, OTTAWA Canada K1S 5B6. At various times, he has been a member of the Atomic Energy Control Board Panel on the Long Range Management of Uranium Tailings and consultant to Ontario Hydro, Atomic Energy of Canada Ltd., the Ministry of Natural Resources of Ontario and various consulting firms on ethical issues in resource and waste management. He was an expert witness before the Canadian Environmental Assessment Agency panel on the AECL proposal for a geological deep disposal facility for high level nuclear wastes in northern Ontario.

1. Royal Commission on Electric Power Planning, A Race Against Time (known as the Porter Commission Report after the Chair of the Commission, Arthur Porter) Toronto: Queen's Printer, 1978, p. 153.

2. Here is what is meant by saying that uranium is enriched. In nature, more than 99% of uranium is of the isotope 238, less than 1% is 235, the fissionable material. In bombs, enough of the 238 is removed to raise the portion of 235 to roughly 70% of the total. The fuel in most reactors is enriched, too, but the ratio of 235 is increased only to about 7%, not to 70% or more as in bombs. Candu reactors use unenriched fuel, in which the fissionable 235 makes up only about .7% of the total. Among other things, this makes Candu fuel more difficult to use as raw material for bombs.

3. An order of magnitude is ten times larger or smaller than the initial amount. So if X is two orders of magnitude larger than Y, then X is 100 times (10x10) larger than Y.

4. `300 to 800 years": see the report of the Advisory Panel on Tailings of the Atomic Energy Control Board, "An Appraisal of Current Practices for the Management of Uranium Mill Tailings" (AECB 1156). `200,000 times less': Environmental Impact Statement on the Concept for the Disposal of Canada's Nuclear Fuel Wastes Atomic Energy of Canada Ltd 10721, COG-93-11, 1994. Unless otherwise noted, facts cited concerning high level wastes are taken from this document.

5. A halflife is the time it takes the radioactivity in a material to diminish by one half. Thus plutonium is half as radioactive after 24,000 years, one quarter as radioactive after 48,000 years, one eighth as radioactive after 72,000 years, and so on.

6. American Physical Society, "Report to the American Physical Society by the Study Croup on Nuclear Fuel Cycles and Waste Management," Review of Modern Physics 50 (January 1978), pp. 1-186.

7. There is an article on the ethics of managing low level uranium mine/mill waste by the same author in earlier editions of Contemporary Moral Issues.

8. Unlike most rock, plutonic rock has few cracks or fissures and is highly impermeable. Over one thousand possibly suitable structures have been identified.

9. At this point, some ethicists would introduce the leading metaethical positions such as libertarianism, contractarianism, utilitarianism, and various egalitarian and deontological approaches. An interesting example is contained in Ch. 6 of the first reference document for Environmental Impact Statement on the Concept for the Disposal of Canada's Nuclear Fuel Wastes Atomic Energy of Canada Ltd 10721, COG-93-11, 1994. The background document is entitled The Disposal of Canada's Nuclear Fuel Waste: Public Involvement and Social Aspects, M. Greber et al., AECL 10712, COG-93-2, 1994. Some of this moral background can also be found in R. Gaizauskas, "A Philosophical Examination of Our Responsibility to Future Generations." Project for the AECB (AECB: Ottawa, 1977). I have adopted a different strategy.

10. We could even go so far as to weigh personal attributes differentially without violating the principle, though many would consider these morally irrelevant: for some people and in some contexts, having Beethoven's powers may make him more valuable than John Smith. Likewise for other kinds of genius, sensitivity, and maybe even personal beauty, grace or charm.

11. Williams, B. Ethics and the Limits of Philosophy Cambridge, MA: Harvard University Press 1985, see esp. Ch. 6.

12. Holmes Rolston mounts an argument along these lines connecting interests to moral worth in Environmental Ethics,

An argument diametrically opposed to the thrust of his and my analysis has enjoyed some currency in libertarian and contractarian circles. It urges that we have no obligations to future generations, to other creatures, or even to many contemporary persons, as follows:

1. We have obligations only where someone has rights; rights are conferred by explicit or implicit agreement of others.

2. Only actual, not possible, persons can enter into agreements and thereby gain rights. Very few even of them have entered any such agreements with me or people with whom I share rights.

3. In particular, no future person could enter agreements or have rights; to contract, you have to be alive.

4. Hence no future person can have rights now.

5. Hence we have no obligation now to future persons.

6. Likewise, mutatis mutandis, for other species, most other people alive now, and so on.

Even if we accept 1. to 3., however, 4. does not follow. First, a person's rights may extend further in time than the person does. Thus any actual person, no matter at what time he or she is actual, may well have rights now. Second, the argument as a whole assumes that we have no obligations to people and other beings that are not conferred by agreement. Against this view, a powerful argument can be mounted that people and other beings have intrinsic value, value that does not depend on whether others acknowledge it or bind themselves by agreement to respect it.

13. Bodde, D. L. Radioactive wastes: pragmatic strategies and ethical perspectives. In D. Maclean and P. G. Brown, eds. Energy and the Future Totawa NJ: Rowman and Littlefield 1983, p. 121.

14. To some extent, this is merely an academic question. Some level of cost is unavoidable to any being who has to service a body, so any strict adherence even to the ratio interpretation will achieve at least some measure of distributional equity.

15. In some contexts, at least three further cost/benefit questions would have to be addressed. Fortunately, they do not affect our obligations with respect to nuclear wastes. (i) Are total benefits of an activity greater than minimum total costs? With respect to nuclear wastes, the answer is almost certainly yes. (ii) Are sufficient total resources available at all? The answer is again yes, though the real question here is whether sufficient resources can reasonably be made available; and were the answer not yes, we would have to go into that very difficult question in a serious way. (iii) Will future generations be able to secure the same benefits for themselves much more cheaply than we can? Although the answer is probably no, the whole question can be rejected, if my earlier argument is sound, because it is we who are reaping the benefits of nuclear power and so there is no sound moral reason in general that future generations should be made to assume any of the costs.

16. I would like to thank Christine Koggel and especially Wesley Cragg for very helpful comments and suggestions.