An increasing number of people believe that we owe it to future generations of people not to undermine their opportunities for a truly human life. When asked how it can be that we have moral duties to those who will be living in the future, the plain answer is that they are people. Persons have human worth and human rights, and actions of ours that undermine their opportunities for fulfilling lives, wrong them, – whether they live today or tomorrow.
I believe this common sense view is correct. However, it is faced with a number of philosophical, or conceptual, problems, – futurity problems they are called. Futurity problems stem from primarily two facts about future people: (a) that they do not yet exist, and (b) that their existence depends upon what we do now. In particular the second fact generates a hard philosophical problem.
So the basic question is whether we have any moral duties at all to people of the future. And this is the topic of my lecture. The further question – what kind of sacrifices are morally required of us? – calls for the assistance of social scientists and other scholars. Before I address the basic problem, I must, however, say a little about how we affect the lives of posterity, and what duties this places upon us. I shall now assume the common sense view that we do have such duties.
Our duties to future people will of course depend upon how we can affect them by what we do now. This we can do in three interconnected areas of policy: by ruining the environment, by changing the size and composition of the population, and through policies relating to various aspects of culture.
We have today become increasingly aware that our destruction of the environment eventually has serious impacts upon the quality of life in the future. Moral duties to future people therefore contribute to the justification of conservationist environmental policies. Exactly what natural resources we owe it to the future to conserve, is a further issue.
But clearly we cannot attribute to all future generations a right to a fair share of non-renewable resources, like oil. First, just as we cannot determine the fair share of a cake unless we know how many there are to share it, we cannot determine the share of the natural resources that we may justly spend, unless we know how many people there will be in the future. Since we do not know that number, we cannot determine the content of such a right. And secondly, even if the notion of a fair share had made sense in this context, the number of future people will be vast, to say the least, unless a catastrophe, like a total nuclear war, wipes mankind off the surface of the planet. The share allotted each will be extremely small: next to nothing. These two considerations show that future people cannot in general have a right to non-renewable resources. Still, we owe it to those in the near future, whose economy and technology we may assume will resemble ours, to save something, and even to channel a part of our surplus into the research for replacements.
But even though people in the far away future do not have a right to the non-renewable resources our society depends upon, they must have a right to the self-renewable natural resources that are the very conditions of any human way of life. We are morally required to leave them seas, forests, rivers, the soil, and the atmosphere intact. So, although considerations of distributive justice to future people may not put severe restrictions upon our use of oil, a concern for their right to a clean environment does. To discharge such duties is, we all know, a difficult task, which may require drastic shifts in our present way of living.
The quality of life in the future is also affected by the choice, or the lack of choice, of population policies. Though interconnected, environmental policies and population policies are separate factors. We could destroy the environment without having an increase in population, and we could, perhaps not realistically but at least in principle, pursue population policies that render the world uncomfortably crowded within the limits set by ecology. Our lifestyle could even, albeit less likely, result in a drastic decrease in population, for example by massive deaths, which would make life hard for the survivors by undermining their economy. Either way, the moral duty not to impose upon future generations hardship caused by population problems is, as Michael Bayles has argued, the ultimate justification of population policies (Bayles 1980, p. 14).
Finally, our choices of miscellaneous cultural policies have impacts upon future people. Our ability to develop a fair and functioning world-economy does, of course, greatly affect the standard of living in the future. So does technology. But quality of life is not just a matter of material goods. The arts and the sciences are important spiritual assets for any culture; political and legal institutions, and moral values, are even more crucial. Life outside legal and moral confines is, in my view, not truly a human life. Duties to posterity may therefore contain the positive moral duty to bequeath to it such cultural goods.
Since wars often greatly destroy all three, the environment, the population, and culture, moral duties to future people enjoin us to promote an international order aimed at a lasting peace. So clearly, if we do have a moral responsibility towards future people, this has vital consequences for present policies in almost all spheres of life.
After this brief introduction, I shall address the philosophical aspects of the problem, which concern futurity problems. I first discuss some futurity problems which perhaps are not very hard to solve. I then discuss consequentialism and deontology and their two major forms, utilitarianism and Kantian ethics. After mentioning some notorious problems that confront utilitarianism in our context, I put the two competing kinds of theories to a test. The experimentum crucis is the most intractable futurity problem: Parfit’s Paradox, I call it.
Our first futurity problem arises from the plain fact that future persons do not yet exist. Does it make sense to say that we have duties to people who do not exist, and that we infringe their rights? Assuming the notion of moral rights, they obviously will have moral rights when they exist, and these rights are correlated with moral duties on part of their contemporaries. The problem is whether their moral rights also are correlated with moral duties that we have now, – even before the holders of the rights have come into existence.
We can approach this problem through an analogy. Suppose, for the sake of argument, that only persons belong to the moral community, and that human fetuses become persons somewhere in the beginning of the second trimester. From fertilization until that phase, the fetus therefore has no human worth and no rights, so it cannot possibly be wronged by any damage inflicted upon it. Does it follow that some damage inflicted upon it in the first trimester could not wrong the person later to emerge from it? Certainly not! The damage may be carried along until the fetus has developed into a person. So this fetal person is wronged by the damage which was inflicted even before he became a person, and someone might complain on his behalf that his right to bodily integrity has been infringed. Of course, this presupposes that the embryo lives up to be a person.
First-trimester fetuses are indeed future persons of a kind. And since it makes sense to say that the rights of fetal persons have been infringed by what we do before they become persons, it is equally obvious that the actions we undertake now can infringe the rights of the future people who have not even been started. Again, this presupposes that they will exist. In Derek Parfit’s example, if “I leave some broken glass in the undergrowth of a wood,” which a hundred years later wounds a child, my act harms this child (Parfit 1984, p. 356).
Against this position, it might be objected that, unlike fetuses, people in the distant future cannot possibly stand in a moral communion with us. The child cannot, for example, sue me for compensation, and she cannot reciprocate in any way. But why must we assume that membership in the moral community extends solely to those in a reciprocal moral communion with us? Children usually are said to have rights against their parents, but hardly because they can be compensated when their rights have not been fulfilled, or because they later acquire a duty to care for their elderly parents. So if morality allows for the unconditional, unilateral respect for children’s rights, I cannot see that the rights of future people can be dismissed on account of a putative principle of reciprocal moral communion.
The counter-intuitiveness of denying future people rights can be brought out by an example: Suppose that country A launches a missile killing the innocent denizens of country B. Their right to life has been infringed. Now suppose again that country A launches the missile, only this time it follows an orbit in space before its kills the innocent denizens of country B two centuries later. If in the former case, this must surely also be an infringement of these future victims’ right to life. The fact that the missile hits its target two centuries after it was launched is morally irrelevant. By the same token, the present destruction of the environment and lack of population control will infringe the rights of those affected by it in the future, – unless we remedy the situation.
So it appears that this futurity problem, at least, does not call for a new kind of ethical theory. Our traditional concepts of moral rights and duties seem perfectly adequate. According to Hans Jonas, however, all forms of traditional ethics has delimited moral concerns to the community of contemporary human beings. Nature and future mankind are not included within its horizon. He sees Kant’s ethics as a representative example. Jonas believes that it cannot proscribe “that the happiness of present and proximate generations would be bought with the unhappiness or even non-existence of later ones” (Jonas, 1984, pp. 4, 5, 11). These are two different claims: a) That Kant’s ethics cannot proscribe buying our happiness at the expense of future people, and b) that it cannot justify that it is wrong to discontinue the human race altogether.
Against a) one might retort that Kant acknowledges both perfect and imperfect duties to other persons, and from what we have discussed thus far, nothing seems to prevent their application to future people. But Jonas acknowledges this (Jonas 1984, pp. 40-1). His criticism is based on the second charge above, that Kant’s ethics cannot impose upon us the duty to continue the human species. Jonas believes that the anthropocentric bias of all traditional ethics prevents them from justifying the moral duty to secure that there will be a future, since all duties toward the well-being of future people are conditional upon their existence (Jonas 1984, pp. 38-44).
But such conditional moral duties to future people are not sufficient in our age, Jonas believes. The destructive power of technological civilization has convinced him that if we have no moral duty to safeguard man’s future existence, moral duties to future people will lack a foundation. He therefore argues that the duty to secure a future for mankind is the primary and unconditional duty of an ethics capable of addressing the threats we face today.
The source of Jonas’ worry is our ability to annihilate mankind through some major catastrophe totally destroying the biosphere. Here one could object that the violent annihilation of mankind, say by nuclear war, violates the rights of billions of people, and so is categorically prohibited by an ethics of respect for persons. However, Jonas believes that anthropocentric ethics could not prohibit non-violent terminations of mankind (Jonas 1984, p. 41).
A scenario Jonas would not welcome, and which he would claim cannot be proscribed unless there is a duty to continue the human race, is sketched by Gregory Kavka. It stems from the fundamental fact that we influence the number of future people by what we do. This kind of contingency of future populations upon our choices has the uneasy consequence that we may increase the level of our consumption and still provide for all the needs of future people, simply by producing fewer of them! Through a universal agreement not to have any children, we may even absolve ourselves from all moral duties to future persons, so that we can consume and pollute to our “heart’s content,” Kavka says, “without worrying about the effects on future generations of people” (Kavka 1978, pp. 193-4, 201).
The scenario would seem to sustain Jonas’ position that our first moral responsibility is to see to it that there will be a future mankind. No harm, let us assume, is done to any person, but the environment is depleted. But as I explained in my previous lecture, Humanism does invest nature with intrinsic value. Thereby it proscribes all policies that destroy the environment. Jonas’ “imperative of existence” (Jonas 1984, p. 43) is partly motivated by his belief that anthropocentric ethics of all kinds see our dealings with nature as ethically neutral. The arguments I gave to the opposite in the lecture seem therefore partly to undermine Jonas’ call for a new ethics.
Moreover, apart from nuclear war and other catastrophical actions already condemned by Humanism, there is no genuine reason to fear that mankind will bring upon itself its own extinction. Multiple interests, from the personal to the social sphere of life, are connected up with procreation. Bayles observes that just as “the present generation has an interest in there being another one”, it “will be in the interest of the next generation for there to be another, and so on.” (Bayles, p. 18) Notice, however, that such recursive considerations do not render unnecessary duties to future people in whom we have no such interests. The interests each generation takes in descendants, and even in their welfare, are in potential conflict with the rights of people in the distant future.
Jonas acknowledges that “we need have no fear of the permanence of the procreative drive”, but still he is not content to rely merely on duties to the future people who will exist (Jonas 1984, pp. 40-42). So it is important to stress that it is within the resources of Humanism to see the continuation of mankind as of intrinsic value. Kavka argues that from the fact that most of us have felt awe of man’s intellectual, artistic, scientific, or technological achievements, we “will not be indifferent to the continuation of these admirable human enterprises” (Kavka 1978, p. 197). This argument is clearly within the reach of Kantian ethics. Humanism therefore seems to deliver what Jonas searches for. Adding this final argument to all the above, I think the need for a new ethics has disappeared.
While Jonas charged all traditional ethics with an inability to ground an unconditional duty to secure the continued existence of mankind, Annette Baier charges that it cannot even ground conditional moral duties to the future people who will exist. She believes that
It took so long for the question of our duty to future persons to come to our attention because it took so long for ethics to free itself from theology, and to make morality concerned primarily and directly with the human good (Baier 1984, p. 219).
Once again, Kant’s ethics is taken to task. Baier claims that as a non-consequentialist theory, it cannot incorporate the so-called Person-Affecting Principle. The principle says that for any action to be wrong, it must affect some person or persons (usually other than the agent) for the worse (Baier 1984, pp. 217-218).
But Baier’s cursory discussion of Kant’s ethics fails to do justice to it. She neither considers Kant’s Formula of Humanity nor his doctrine of right (Kant 1797), both of which contain a deontological version of the Person-Affecting Principle. But her charge raises the right sort of question: what kind of moral theory, consequentialism or deontology, is best suited to account for moral duties to future people?
It is often said that while consequentialism determines the moral rightness of action by its consequences, deontology either leaves the consequences entirely out of account, or determines moral rightness by considerations in addition to the consequences. But such characterizations do not set forth the distinctive features of the two theories. Perhaps a better way to characterize them is by what is more basic: consequence or duty.
In consequentialism, consequences are basic, in deontology duty. Consequentialism takes our moral duties to be determined, ultimately, by the non-morally good and bad consequences of action. I say ultimately, because consequentialism may also take into account consequences that are morally good and bad. But not at the basic level, since the consequentialist idea is that morality derives its content from the non-moral value of consequences. Hence, at the basic level, that action – within a set of available alternatives – is morally required which has the greatest net balance of non-morally good over non-morally bad consequences for some specific target group.
On the deontological view, on the other hand, moral duty, which is itself determined by one or more principles, determines what consequences are morally good and evil. Most deontological views see the non-morally good and bad consequences as morally relevant, but only in so far as they are morally good or evil. Rather than to describe all forms of deontology, I shall straight off present Kant’s version.
According to Kant, the binding force of all kinds of moral duty is expressed by the Categorical Imperative, although they relate to it in different ways. Of Kant’s different formulations of the principle, the so-called Formula of Universal Law is fundamental (Kant 1785). It requires us to act only on the maxims that qualify as universal laws (Kant 1788). Usually, the point of the formula is taken to be that we are constrained to act only on the maxims that all others can adopt. I agree that this is implied, but it does not, in my view, capture the essence of Kant’s principle.
On my reading, the Formula of Universal Law is a principle of human excellence. It enjoins us to model our conduct after the idea of a moral agent whose reason is in full control. This means that his superior, or conditioning, motives, never are sensuous, although sensuous motives certainly are allowed in the subordinate role. Still, the idea of a perfectly rational agent is too abstract. It cannot give us moral guidance unless we personify it. For this purpose we must design a moral ideal, which Kant says,
serves as the archetype for the complete determination of the copy; and we have no other standard for our actions than the conduct of this divine man within us, with which we compare and judge ourselves, and so reform ourselves, although we can never attain to the perfection thereby prescribed (Kant 1787, B 597).
Kantian autonomy, or
self-legislation, is to adopt the conduct of such a moral ideal as the moral
standard for one’s maxims. Since we are not perfectly rational, but finite,
biological beings with needs and desires, we are enjoined in the form of a
Categorical Imperative to act only on the maxims which qualify as rules of
conduct for this ideal agent. Hence, the content of our moral duties is
determined by the comparison with the moral ideal. Moreover, in order to live
up to the moral ideal, we must respect the humanity in all persons. This way we
get to the Formula of Humanity. Finally, we are asked to live up to the
full-blown social idea of interdependent persons united under universal laws:
All this does, of course, require a defence which cannot be given here. The essential point for our purposes, however, is that consequentialism determines our moral duties by comparing alternative actions with regard to the best consequences for some target group, whereas Kant’s ethics determines our moral duties by comparing maxims with the conduct of purely imaginary moral ideals. I believe this difference between them will prove pivotal for solving Parfit’s paradox.
But first I shall briefly characterize and discuss utilitarianism. Utilitarianism is the predominant form of consequentialism, and we get to utilitarianism by specifying the target group to include everybody affected. Consequentialism is neutral in this respect. A consequentialist might hold that only the consequences for him, for his family, for his country-men, or for his race, for example, matter. The utilitarian, on the other hand, is a universal consequentialist: the like consequences for all individuals affected by an action are to count for the same. What class of beings is included within the scope of moral concern depends upon the value-component: pleasure, preference-satisfaction, etc. I shall not go into this, but will refer to any such value as happiness.
Utilitarianism notoriously yields strange conclusions in the area of population policies. Two forms of utilitarianism are relevant: the total view and the average view. On the total view, we ought to maximize the total amount of happiness in the universe; on the average view, to maximize the average level of happiness, which is the total, divided with the number of affected parties. The total and the average view yield the same moral prescriptions as long as the size of populations is not affected by our choices. But in the cases in which they affect their size, the two versions of the theory yield different conclusions.
We can maximize total happiness either by increasing the level of happiness in a fixed population or by increasing the number of happy members. The latter implies that we can increase the total happiness by adding members to the population, even though the overcrowding etc. results in less happiness for each member, to the extent that the reduction cancels out the total increase. Parfit has argued that the total view commits us to increase the population until we get “some enormous population whose members have lives that are not much above the level where life ceases to be worth living” (Parfit, 1984, p. 388). This is clearly not acceptable.
The average view does not fare much better. A population of, say, 100 people with an average happiness score of 10 units of happiness, has the total happiness of 1000 units. Suppose that the population increases to 600 people, but that the average happiness in consequence decreases to only 2 units. Now the total happiness is 1200 units, but on the average view, the score of 10 is preferable to 2. So it then prescribes a smaller population than the total view. However, as Peter Wenz explains, that is not an asset:
In the world as we know it, maximizing average utility would require drastically reducing the size of the human population [...S]uch a diminution of the human population could be accomplished only by totalitarian methods of population control over the next several generations (Wenz 1988, p. 202).
These totalitarian methods may even, as Parfit remarks, include killing vast numbers of people (Parfit 1984, p. 386). Hence, the average view also leads to morally unacceptable conclusions. Utilitarians have attempted to remedy the situation by drawing a subtle distinction between future people and possible people. But even if that distinction could be sustained, the theory is not immune to other repugnant conclusions. Instead of going into this debate, I shall argue against its root idea: consequentialism itself.
Think about the many coincidences that led to exactly you being born: that your parents met and that they decided to have a baby at the time they did. Had they waited, another pair of gametes would have met. But then you would not have existed. Rather, they would have conceived another child in your place, – as different from you as a sibling. And if the economic situation had been different, your parents might not have met at all. Parfit asks us to consider whether, “even if railways and motor cars had never been invented, I would still have been born?” (Parfit 1984, p. 360). I think the answer is plain: the existence of each particular person is contingent upon a vast number of past actions and events: had they not occurred, none of us would have been born.
This also applies to future generations. What particular persons who will be born depend upon what policies are adopted now. Say that we have a choice between a policy which is environmentally responsible and checks the population growth, and a laissez-fair policy which is reckless in these respects. The choice between them affects what particular individuals will be born. Thus, each policy would lead to different individual persons being born in the two respective possible populations. In fact, after about two centuries, there is not going to be one single person contained in the one population who would also have existed in the other. But then the particular persons who will be living two centuries from now as the result of our choice of the reckless laissez-fair policy would not have existed if instead we had chosen the responsible policy.
The outcome of our choice of the laissez-fair policy is that the quality of life is not very good for our future people; they live a sort of MadMax-life. Their lives are not so miserable that they regret they were born, but not very fulfilling. Can they blame us for our choice? Have we wronged them?
Thomas Schwartz has argued that they cannot complain. He believes that it makes no sense to say that someone has been wronged unless some particular person is made worse off than he would have been otherwise as a result of what is done to him. The point is that this version of the Person-Affective Principle clearly does not apply in our case. Nobody has been made worse off by our choice than they would have been had we chosen differently, for then none of these miserable particular persons would have existed. Schwartz concludes that it does not make sense to say that we have wronged them: so we have no moral duty to conserve the environment for the sake of future generations of people. Our only duty in this respect is to those of our contemporaries “who would like our distant descendants to enjoy a clean, commodious, well-stocked world” (Schwartz 1978, p. 13).
I have here presented Parfit’s paradox. But Parfit himself does not endorse Schwartz’ counter-intuitive conclusion. Rather, Parfit urges us to give up the Person-Affecting Principle (Parfit 1984, p. 363). I disagree with both of them; I believe that it makes sense to say that a particular person has been wronged – even though he would not have existed had we chosen otherwise. The problem is how to account for the intuition. First we need a diagnosis of the problem, – and then a cure. The aim of my argument is to show that consequentialism is responsible for the problem, while Kantianism solves it.
Consider an analogous example: a couple suffering from AIDS has decided to have a baby, fully aware of the chances that their child will be born HIV-positive. The child has a good life before he gets sick and dies of AIDS at the age of 10. One would think that the child could blame its parents for having breached their duty to him, – a moral duty corresponding to his moral right to a good start in life. But suppose it was replied on behalf of his parents that he has nothing to complain about, for if they had not conceived him, he would not have existed at all. What premise does this reply rest upon, and how can we explain that it is mistaken?
First we should exclude disturbing considerations, such as harm done to other members of society. We achieve this by focusing only on the couple and their child. On the consequentialist view, his parents could only be said to have wronged him if they chose an action with worse consequences for him than any other action available to them. But all alternatives would result in the child not coming into existence, so it cannot be said that the parents could have chosen another action which would have had better consequences for him. The conclusion is that the parents did not fail to do what was morally required of them, and so cannot be blamed for having wronged their child.
This shows that the source of the paradox is the consequentialist view that the morally required action is determined by that action within the available set which has the best consequences for a target group, here consisting of the child.
On the Kantian moral theory, on the other hand, our moral duty is determined through a comparison of the action done with an ideal: namely with what a perfectly rational free agent would have done in the situation. The parents’ action is compared with what perfectly rational parents would have done for their children.
Since such ideal parents would act from respect for the humanity in their children, and would take responsibility for having conceived them, they would do everything in their might to protect the capacities and promote the abilities which are necessary for their offspring creating fulfilling lives for themselves. This ideal, incorporating the wisdom of our moral tradition, represents conduct that real parents morally ought to display. The claim is not that real parents have to be motivated by respect in order to fulfil their duty to their children. Nature has implanted tender feelings in us with that function. The point is that the content of parental duties is determined by the ideal.
Accordingly, any action by which parents undermine their children’s opportunities for a full human life would be contrary to their moral duties. I argued at the beginning of the lecture that it does not matter if the damage is inflicted before the fetus becomes a person. The child in our case therefore is entitled to blame his parents for having breached their moral duty to him on the grounds that they failed to live up to the moral standards set by the envisioned conduct of the ideal parents.
This moral judgment is not based upon a comparison with how the same child would have fared on alternative actions available to his parents. Rather, the comparison is with an ideal of thought: the parents wronged the child because their action made him worse off than what the ideal parents would have made any child of theirs.
If the Kantian reasoning here solves Parfit’s paradox in the AIDS-example, it also solves our futurity problem. The people who will live as a result of the reckless laissez-fair policy can complain even though they would not have existed had we chosen the responsible policy. Our moral duty is determined by what ideally rational agents would have done. Acting from respect for the humanity in all persons, they would surely have avoided endangering the conditions of a truly human life for any person, – future or present. By falling short of this moral ideal, we fail in our moral duty to future people, regardless of who they will be. Since these are moral duties of justice, our breach of them now infringes rights they will acquire then.
I conclude that we ought to persist in our common sense view that we do have a moral responsibility to the people who will live in the future, and that unless we change our ways, we infringe their moral rights.
Baier, Annette (1984), “For the Sake of
Future Generations,” in Regan, T. (ed.), Earthbound: New Introductory Essays
in Environmental Ethics (
Bayles, Michael (1980), Morality and
Population Policy (
Jonas, Hans (1984), The Imperative of
Kant, Immanuel (1787), Critique of Pure Reason, trans. N. Kemp Smith, (London: Macmillan, 1990).
Kant, Immanuel (1785), Grounding for the Metaphysics of Morals, trans. J.W. Ellington, in Kant’s Ethical Philosophy (Indianapolis: Hackett Publishing Company, 1983).
Kant, Immanuel (1788), Critique of Practical Reason, trans. L.W. Beck (New York: Macmillan Publishing Company, 1956).
Kant, Immanuel (1797), The Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge University press, 1991).
Kavka, Gregory (1978), “The Futurity
Parfit, Derek (1984), Reasons and Persons
Schwartz, Thomas (1978), “Obligations to
Wenz, Peter S. (1988), Environmental
* Lecture in partial satisfaction of the requirements for the degree
of doctor philosophiae,