Introduction

The topic of obligations towards future generations was a fairly unpopular one among political philosophers until it became a hunting ground for theorists of social justice interested in Rawls’s savings principle (cf. Tremmel 2007), almost immediately followed by theorists of environmental ethics, who fielded the interests of future generations against the exploitation of natural resources by present generations (cf. e.g., Barry and Sikora 1978). In such areas of environmental politics as climate change, the appeal to the interests of future generations (“the children are our future”) is as commanding as considerations of health are elsewhere. Recently, future generations have also been introduced in theories of International Relations (e.g., Rendall 2007).

Theories of (re)distributive social justice (and that includes environmental justice) aim to de- and prescribe parameters of a formula: who owes what, when, where, to whom. Possibly more importantly, social justice theories try to delineate the “why,” i.e., they try to account for the choices of variables in that formula. When obligations between generations are discussed nowadays, it is usually in the context of such theories, and usually as generalized obligations between sets, i.e., generations, rather than individuals (cf. e.g., Van Liedekerke 2004; Beckerman 2006). In addition, generations are often assumed to contain (in sum) a near-infinite number of individuals, and some of the resources to be distributed are assumed to be finite but beyond that remain undefined—to list but a few heuristic simplifications (for more, see Tremmel 2006). Given that there are good reasons to simplify complex questions, as we shall see in a moment, it seems that characterizing obligations towards future generations in such condense terms is innocent while (in both senses of “while”) making the implied question “what do we owe?” answerable.

Yet such simplifications are not as innocent as they may seem to be. Together, they cast the relations between present and future humans in the mould of social liberal political thought,Footnote 1 with its assumptions of mutually advantageous cooperation, shared or collective responsibility for at least part of all natural resources and of the ultimate social product, and collective ownership of those resources (thus allowing involuntary taxation and redistribution). By framing the question of obligations towards future humans in terms of distribution of collective assets over generations, there is a risk that too little attention is paid to the creation of the context that defines and determines any specific question of distributive justice. As I shall try to show in this article, the social liberal perspective may be highly popular and even dominant, yet it is neither the most natural starting point for an analysis of obligations across generations, nor the most obviously legitimate one.

The context makes all the difference. If, for example, the next generation were also known with certainty to be the last, we might want to choose very different principles describing our obligations to “future generations” from those we would choose in the almost certain knowledge that there will be further generations. Another example: from a Marxist perspective, justice is the problem rather than the solution—it is the redistribution of scarcity rather than the creation of sufficiency, and there is no reason to assume that justice between generations is any different.

Simplification is helpful in so far as it allows a ceteris paribus assessment of the relation between the two or more variables or concepts that have not been “simplified away.” Hence, the results of any debate on aspects of obligations to future generations may be contradicted or may have to be amended once a new variable is included, e.g., once genetic engineering is assumed to be possible (Farrelly 2005), once generalized resources are replaced by specific depletable or irreplaceable resources (cf. Ball 1985), once animals are included (particularly when they are included as producers of goods and not merely as livestock or as subjects; cf. also Vallentyne 2005), or, the perhaps most famous source of complications in any debate on obligations towards the future, once generations are recognized as individuals. At that moment the famous Non-Identity Problem rears its head (Parfit 1984; cf. e.g., Carter 2002; Partridge 2002; Johnson 2003; Bell 2004; Mulgan 2006; Page 2006): we cannot call policy A (resulting in say, the non-existence of non-Aryans) more or less just towards the individuals who will exist in the future than B (say, the disappearance of all Aryans), at least not if A and B result in the existence of completely non-identical sets of future individuals Na and Nb; no one in A exists in B (and vice versa) and therefore no one in A is better or worse off than in B (and vice versa).

Simplification is helpful in gaining an insight in the relation between specific variables; it stops being helpful when conclusions that are only proven valid under ceteris paribus conditions, i.e., provisional conclusions, are confused with universally valid conclusions. In the present text, two relatively popular assumptions in intergenerational justice theory will be questioned: the inevitability of future generations and their costliness. Showing that these assumptions really are unrealistic simplifications helps us to become aware of a third simplification, viz., the idea that obligations between present and future humans are necessarily a matter of redistributive justice from present to future generations. Framing intergenerational obligations as obligations of distributive social justice is a political choice, one that voids, denies, or ignores among others the role of individual responsibility in the creation of future individuals. A complete and universally valid theory of intergenerational obligations should begin by asking whether rather than assuming that specific transactions between generations are a matter of social justice, or in other words, a collective responsibility, instead of an individual affair.

The choice for the assumptions of inevitability and cost serves mainly to prove a point. The assumption that future individuals or generations manifest themselves involuntarily but certainly, and the assumption that they are recipients of goods only, for whom present generations only make sacrifices, rarely occur together in academic literature (cf. Young 2001). As a matter of fact, the latter assumption, detectable in the famous and oft-quoted Tragedy of the Commons (Hardin 1968), seems to be necessary to make precise statements on population policy possible, but it is an implicit denial of the first assumption. What matters, however, is that neither of these assumptions is innocent, let alone irrelevant. A different, and from a social liberal point of view more plausible, view of the inevitability and of the value of future individuals (to be developed in subsequent sections) suggest a fundamentally different philosophical and thereby political priority for questions of intergenerational distributive justice. Questions about the shape of intergenerational distributive justice (in cases like resource management, climate change or debt policies) may often be perceived as socially most urgent, but that does not mean that their being framed as legitimate collective responsibilities should be accepted as given.

Inevitable Individuals

One of the many variables often kept constant in work on obligations between generations is that of the inevitability of future individuals. Who each single individual may turn out to be, and whether each single individual will exist, may be determined by necessity or contingent factors—but that future generations and individuals will manifest themselves involuntarily yet certainly is usually taken for granted. Two fairly fresh examples stem from Jennifer Heyward and Axel Gosseries. Heyward (2008) demonstrated beautifully how the Rawlsian representation of future generations behind the veil of ignorance avoids the non-identity problem, while satisfying the demand of proponents of deliberative democracy that all affected by decisions should participate in decision-making, which is something that straightforward representation of future generations by present generations cannot. Heyward’s hidden assumption is that future generations’ interests must be represented even if they would never exist—hence they seem to have interests even when they will never exist. Axel Gosseries (2008), in his frank discussion of the still unrepaired holes in arguments for rights for future generations, acknowledges that rights can and perhaps should be granted to presently existing people as representatives of future generations, rather than (for now fictitious) future people—but again, the implication is that they seem to have present interests even if they will never exist.

Assuming the inevitable emergence of future generations may seem a natural assumption but it is not. The (in)evitability of the existence of a future individual does not just have consequences for the size and claims of future generations (and for the job prospects for philosophers of intergenerational justice), it also has causes that in their turn, at a prior stage, predetermine the range and character of our obligations—and this in two ways. First, the inevitability assumption obscures the fact that our obligations may be the result of deliberate individual behavior, not of risk or accident (a popular ground for risk-sharing cooperative institutions based on solidarity), and secondly, it thereby also changes the balance between distributive justice and other types of moral obligation.

Responsibility and choice play a major role in principles of justice: more often than not, they determine whether a claim to part of the social product or compensation for a disadvantage is justified or not. To assume that the existence of future individuals is inevitable results in the attribution of responsibility for (duties and obligations to, rights and claims of) countless future individuals on the collective shoulders of the present generation—where, as I hope to show, it ought not necessarily lie.

The assumed inevitability of future individuals also changes the balance between two types of justice: distributive justice and justice in exchange. Distributive justice is often equated with social justice; it is based on the assumption of shared responsibilities and shared resources, attributed hierarchically by (representatives of the) collective over members.Footnote 2 Justice in exchange is equated with market justice, fair pricing, and fair trade; it is based on the assumptions of individual responsibility, individual ownership, and equality between exchanging parties. Assuming future generations’ inevitability allows one to conceptualize their existence as no one’s work in particular, thereby as a collective responsibility, and thereby as a Inevitability makes distributive justice overshadow further types of moral obligation as well. On some views of justice, justice requires reciprocity; but since (or in so far as) future individuals do not exist, a reciprocal relationship with them is impossible, therefore justice between non-overlapping generations is impossible (cf. Page 2006). As Hillel Steiner frankly puts it, “future persons (…) have no rights against present persons nor, therefore, any rights that present persons save or conserve anything for them” (Steiner 1994, p. 261). On other views, like Brian Barry’s, reciprocity is not a necessary condition for justice. If justice is not the right word, ethical obligations of a non-reciprocal nature may nevertheless exist—their form may even (but need not) mimic that of principles of justice. That does not make the choice between justice and other concepts inconsequential: founding obligations to future generations in—say—natural duties that take precedence over justice (cf. Rawls 1999) may well make them stronger and more universal than any by definition controstice.

The inevitability axiom is questionable—obviously already by nature, clearly also because of cultural attitudes concerning procreation and education about fertility, and finally increasingly due to the development of technologies granting more and more control to presently existing individuals over their procreative activities. Future individuals are increasingly voluntary possibilities rather than inevitable necessities, and that makes the existence of the collective entity “future generation,” by definition, equally avoidable. Procreation itself has become a “human originated change” (to use a phrase coined by IPCC).

Assume for a moment that newly created individuals lay a prima facie valid claim on “our” resources. This would make children a burden and a problem. Liability, responsibility for causing a problem, makes a difference for the question who should solve it and carry the burdens of the solution (how much of a difference, when, and where, are different questions). Likewise, if or in so far as children are a benefit, responsibility for their creation makes a difference for who can claim those benefits. It is here, where the input for theories of justice is created, that the difference between distributive social justice and justice in exchange originates. We can perhaps, on the one hand, attribute the newly created individual’s talents to morally arbitrary accidents of birth, and the development of her talents to her environment, thus supporting the case for collective property and distributive justice. On the other hand, we also have to make room for the individual’s responsibility for the choices he or she makes, thereby for legitimate private property and for justice in exchange. For one, if responsibility were irrelevant, ethics itself would be voided, making shame and honor, punishment and reward, sheer fantasies.Footnote 3 Moreover, if individuals cannot be held responsible for their choices, the collective that allegedly enabled the individual to produce a benefit or burden, being made up of similar non-responsible individuals, could not be responsible for its contribution either—hence both the individual’s and the collective’s claim to benefits or burdens would be equally unfounded, and distributive justice would be as unjustifiable as justice in the exchange of private holdings.

Next to responsibility for the causal process of creating future generations, moral responsibility “after the fact” is equally relevant in assigning the “benefits” and “burdens” of future individuals. When, say, a traffic accident occurs, the person who caused the accident may be seen as primarily responsible for “righting the wrong.” Yet this does not clear bystanders of a prima facie moral responsibility to help, next to or even instead of the perpetrator. Assuming for the sake of simplicity that the victim bears no responsibility whatsoever for his fate, the Rawlsian notions of arbitrariness and solidarity seems to apply here, that is, since no one deserves his or her talents, disabilities, or any identifying traits, nor the (dis)advantages incurred by them, we ought to share such burdens and benefits fairly. In our example, no one deserves to be the victim of a traffic accident or suffer its consequences, hence (and other things being equal) we are all obliged to help the victim. To make a long story short: causal responsibility for the creation of a benefit or burden does not void, nor is it the same as, moral responsibility; the latter requires the availability of choice and of the means to act on that choice, and may be optional (granting a “privilege”) rather than a duty.Footnote 4 Since future generations are not traffic accidents, nor necessarily anything like traffic accidents, moral responsibility for their fate is neither evidently the collective’s, nor evidently a duty.

When we recognize the relevance of individual responsibility before and after the fact, we can also begin to distinguish between different sets of “commitments” (duties and freedoms) associated with different types of (causation of) future individuals. Merely to illustrate this point, I shall distinguish five relatively simple examples—please note that many more are imaginable (cf. Ekeli 2004). They serve only to illustrate, not to justify anything (Table 1).Footnote 5

Table 1 Some types of obligations to future individuals
Full size table

First of all, different actors can be causally responsible for the creation of a new individual, and if or in so far as they have the means and opportunity, they are also the first to be morally responsible for the consequences of creation. Under ideal circumstances, parents consensually choose to procreate voluntarily, out of love or out of prudence (Case 1). Let us assume the two to be mutually exclusive, with prudence being the considered judgment that creating a new individual is, overall, better than not doing so, while procreation out of love stands for any “unconsidered” decision to procreate: love for each other, love of children, a desire to allow one’s physical and mental potential to fully flourish, and so on.

Badly motivated or not, Case 1 implies that the parents carry the primary responsibility, both causally and morally, for the consequences of their actions. It is only when means or opportunity to be responsible disappear, that other parties (“bystanders,” whom, for reasons of simplicity, I shall here call “society”) may become morally responsible in second instance—may, first of all because society too needs to have the means and opportunity required for actual responsibility. Secondly, love as a motive differs from prudence in that it is not so easy to argue that (for want of a better word) a love child could have happened to all of us as it is to argue that a prudence child could have been preferred by any of us. Perhaps one cannot argue with love, perhaps one can—and in the latter case, we could all, reasonably, have reconsidered our choice before acting on it. While it would have been unreasonable for us to reject a prudential argument for procreation under the same circumstances as the parents did when they decided, it would not be unreasonable for us to reject their (after all, “unconsidered”) argument from love. No child deserves its fate, hence the arbitrariness principle implies that society owes both types of children in Case 1 exactly the same—but for different reasons: where prudence ruled, we would have chosen as the parents did, so that the arbitrariness principle implies a duty to assist, a duty to carry the parents’ responsibility should they be unable to, while where “irrationality” ruled, we could easily have chosen differently, making society’s assistance in the parents’ stead more a privilege than a duty.

In Case 2, we assume that parents can choose but do not: a child is created totally involuntarily, i.e., by accident or through sheer stupidity. It happens. While parents in such cases could have been better prepared (and while they do therefore carry some moral responsibility), they accidentally were not—and since we may assume that this could happen to anyone of us, society’s second instance moral responsibility again takes the shape of a duty.

When parents are mentally totally incapable of choice (Case 3), we can no longer call them even causally responsible for their acts—rather, it is nature doing its work. The individuals’ being incapable of carrying responsibility, a fate that could have happened to us all, society is now no longer morally responsible in second but, by default, first place.

Case 4 is defined by the absence of the means to carry responsibility, voiding any room there might otherwise have been for choice. One example would be a materially impoverished society where procreation, e.g., as an investment in old age, is a matter of survival—unlike prudence in Case 1, procreation in such societies does not simply make anyone better off relative to a more or less acceptable baseline, nor does it even necessarily guarantee that such a baseline will be established; rather, it is simply a necessary condition for any kind of life. Another example is a society with an oppressive culture, prescribing child-bearing and/or fertility at the price of exclusion. In Case 4, it is again unreasonable to attribute even causal responsibility to parents—they have, by assumption, no choice—nor can we this time blame nature; both moral and causal responsibility lie with society.

Finally, in Case 5, children are born out of non-consensual relations where no natural or social causes prevent one party’s voluntary choice: rape. Here, both the causal and primary moral responsibility for the consequences of creating a new individual are the father’s. (In attributing causal and primary moral responsibility to the father for the fate of a child born out of rape, I do not regret imposing Western liberal values on the vast majority of humankind, who attribute responsibility for rape differently.) The biological mother’s moral responsibility at best implies that what she does for her child is based on a privilege; she has no duty to carry the burdens of an act that—mildly put—could reasonably and as a considered judgment have been rejected. (I disregard the painful question how and even if the father’s responsibility should be implemented when, e.g., the mother objects.) Assuming that the mother does, however, have the means and opportunity to decide whether or not to give birth to the child, she remains morally responsible, in second instance, for the consequences. Society comes third, its moral responsibility being a duty again—since being a woman and being raped could have happened to all of us.

A word of caution is required. The six cases are based on an extremely oversimplified picture of reality, even if it is more subtle than simply assuming that future humans will necessarily exist. I assume that the line between voluntary and involuntary behavior is sharp where, in reality, it rarely is. The same goes for the distinction between having and not having the means to be responsible. The world of these five cases is divided between a father, a mother, and an abstract entity called society—where reality complicates matters considerably by, for one, possibilities unthought-of in ages before (IVF, sperm and egg donors, surrogate mothers and so on) and, secondly, distinctions between several social spheres each with their own responsibilities—like family and extended family, community, civil society, and the state. Also, I have in a sense made the questionable assumption that might implies right: I have assumed that the ability to control procreative capacities implies at least primarily exclusive moral responsibility for the use of those capacities. To assign primary responsibility to society (or any other agent), despite the individuals’ actual control over their capacities, seems to presume the repugnant idea of collective or corporate management (or worse, ownership) of the individual’s reproductive organs.

Taking more and more of these exceptions and similar subtleties of life into account forces us to admit that attributing responsibility for the creation of a new individual is a matter of practical rationality—judgments will have to differ from case to case and will have to take countless factors into consideration.

More subtlety will thus make it less easy to decide precisely for which consequences of procreation individuals are responsible and for which others are, whether that responsibility is a matter of justice or other duties and, where justice is relevant, whether it is a matter of distributive justice or justice in exchange. Yet this supports rather than undermines the case for the relevancy of responsibility, and thereby for the thesis that moral obligations between generations are not necessarily obligations of justice, let alone collective obligations to institute a form of distributive justice.

The Value of Future Individuals

The existence of future individuals is, at least in part, a matter of parental choice, therefore a choice for a perceived or expected advantage. Particularly advocates of population policies, however, tend to conceptualize future generations as overall burdens on society. Contrary to the intuition of many parents, such scholars assume that future individuals either pose a threat (i.e., they will have the power to demand and get retribution in our old age) or merely form a burden on the resources of the presently living, since they apparently detract from our resources but give nothing back (for more subtle overviews, see e.g., Mulgan 2006, or Tremmel 2006). While some authors add a degree of subtlety by pointing to limited reciprocal relations between coming and going generations, creating chains of reciprocity (Howarth 1992), as a rule, the parents and children are nevertheless expected to somehow benefit from procreation, while the rest of society pays. Garrett Hardin’s Tragedy of the Commons (1968) is the classic illustrationFootnote 6: here, selfish parents get all the free-rider advantages of polluting the environment with children.

Not only does this distribution of benefits and burdens seem prima facie unfair—it is also based on an oversimplification of both the nature and actual distribution of benefits and burdens. If a theory of justice ought to be complete (Rawls 1999), then we need it to include the attribution of moral responsibility not only for the burdens imposed by future generations, but also for their benefits to each and every party involved. Parents, children, family, society, and other entities are all potentially both recipients and distributors of advantages and disadvantages to others (cf. Young 2001). What follows is again a rough and ready overview of only some of the effects of procreation for only a limited number of potentially affected parties, merely for the purpose of illustration; completeness is not the object.

Parents as recipients, first of all, can obviously benefit in numerous ways from the creation of future individuals, but it would be too easy to call this the whole truth. Children may bring joy and pride, but they may also turn out to be disappointments or worse; they may be an insurance against the frailty of old age, or not. They may be an answer to social, peer, and family pressure, but in this respect as well, the costs may outweigh the benefits. In so far as children are created as producers of future benefits, they are a risky investment, opted for because the alternative appears less attractive. In another sense, as benefits in themselves rather than as producers of benefits, children are perhaps less risky investments: barring PND, parents are almost certain to enjoy all the benefits of parenting—the miracles of birth, growth, and development, the opportunity to care and foster responsibility, and so on. Unlike the instrumental advantages of children, these inherent rewards seem to come with guaranteed benefits (though still at a cost).

One step (or more) removed from parents and their benefits, members of the future individual’s extended family derive basically the same benefits—both instrumental and inherent—from their creation, although to a lesser degree (cf. Binmore 2005). Only siblings are possibly worse off: a new brother or sister often means fewer resources for them, which in turn may result in reduced prospects in life. Then again, siblings are social capital; the benefits may outweigh the burdens.

Society as a whole draws less clear benefits from future individuals, unless the continuation of a society’s existence beyond the life span of those presently living is seen as a good per se (i.e., regardless of its meaning to individuals—in which liberals are far more interested). While on average able to contribute less, and initially merely putting an extra strain on resources in impoverished societies, their existence seems more beneficial for others in relatively rich welfare states: through taxes and jobs in care, new individuals can contribute to the wellbeing of the latter. Those same rich societies, however, will first have to invest in feeding, clothing, housing and educating the new individuals. All in all, the benefits of new generations of citizens can only come about as a result of a prior commitment to make costs, and that prior commitment will be either based on the inevitability of “costly” future individuals, or on an expected inevitable need for their contribution.

Finally, assessing the benefits and burdens of existence for any new individual him- or herself takes us, obviously, to the core of Parfit’s non-identity problem and, beyond that, to questions like whether existence or life is good in itself. I want to avoid those questions here. We may not be able to harm future individuals, i.e., we cannot make them “relatively” worse off because there is nothing “relative” to compare them to, but according to Carter (2002) and others, we can make them suffer (for further circumvention strategies see Howarth 1992; Huang 1997; Beekman 2004). We can cause the existence of an individual A so ill and handicapped that her life is just barely worth living, or even individuals B whose life is utter torture but who cannot end it themselves and cannot communicate their fate to others who could end it. A’s and B’s suffering counts for its own sake. By the same token, we may not be able to benefit future individuals relative to how they could fare otherwise (they cannot be but who they are), but we are certainly capable of benefiting them in absolute terms. In sum then, future individuals can be recipients of benefits as much as of burdens. There is a caveat: present humans create the next generation and are responsible for its fate, and they partly shape the conditions under which subsequent generations can exist, but it is the next generation and not we who have the means and opportunity to create a third; the next generation carries the causal and primary moral responsibility for the third. We can neither harm a remote future generation nor make it suffer—but we can burden next generations with a choice between creating no subsequent generation or one that will be (not “worse” but) badly off.Footnote 7

In conclusion: to assume that future individuals only benefit from us while present individuals pay, or that individual procreators benefit while society as a whole pays, is an unjustifiable oversimplification.

Implications for Intergenerational Morality

The existence of future individuals is not inevitable; causal and moral responsibility for their existence and fate can usually be attributed. In addition, parents and children are not necessarily only recipients of benefits, nor does society necessarily inherit burdens only (as assumed in Hardin’s Tragedy). With these prejudicial assumptions removed, both the priority and the remit of principles for intergenerational redistribution changes. For instance, since causal responsibility matters, the Rawlsian notions of arbitrariness and solidarity no longer justify attributing prima facie moral responsibility for the fate of every future individual to the collective. Distributive justice takes second place, as a rectification device, to justice in exchange. The politically urgent questions that and how many natural resources and artificial capital “we” owe to future generations are certainly not voided but they do, logically, follow rather than precede the question “are existing individuals sufficiently empowered to carry the responsibility for their progeny?” The following seven duties all logically precede and override demands of intergenerational redistribution—which does not necessarily imply that they have political priority.Footnote 8

  1. (1)

    If personal responsibility is indeed a good thing, then having a real choice in whether or not to procreate is a good thing too. From this follows, as a necessary condition for any justifiable obligation to future generations, a collective and individual duty to emancipate individuals, that is, to provide not only adequate sexual education and access to contraceptives, but also systemic preconditions for individual responsibility like a relatively enlightened sexual morality, and a socio-economic structure allowing both a life of personal sufficiency and room for new individuals. Obviously poverty and the prospect of continuing poverty induce involuntary procreation, but above a certain level we may expect that a decrease in the proportion of young members of a population will also provide an economic incentive for procreation. The emancipation duty can therefore also be rephrased as a duty to prevent the need for intergenerational distributive justice, i.e., a duty to prevent society from itself becoming responsible.

  2. (2)

    Assuming now that a social structure is in place that allows individuals to voluntarily choose whether or not to procreate, we can and must hold parents primarily responsible for their decisions and the consequences thereof (cf. Velleman 2008; Steiner and Vallentyne 2009). Hence, a parental duty to procreate if and only if their future child’s wellbeing can be reasonably assured takes precedence over all of society’s real and imagined obligations as well as—after all, justice is cold and jealous—over love and stupidity.

  3. (3)

    It does not follow automatically that parents are also obligated to care for their child; for this we need the additional premises that no child deserves its lot in life (cf. Velleman 2008) and that those causally responsible for this fate are also morally primarily responsible. This results in a parental duty not to impose avoidable suffering on their future child. Whether there is also a concurrent duty to benefit the child whenever possible, i.e., to maximize its wellbeing (with consideration for the effects on others), probably depends on the measure of wellbeing: e.g., a resourcist view of the good life need not support such a duty if it understands the good life in terms of sufficiency.

  4. (4)

    Responsible parents carry responsibility not just for the fate of their child but also for its impact on the wellbeing of third parties — hence the “parents pay” principle: a parental duty to compensate any third parties unfairly disadvantaged by the creation of their child.Footnote 9 There is no concurrent right to be compensated by third parties (individuals or society as a whole) for any unsolicited advantages bestowed on the latter (cf. Steiner and Vallentyne 2009), although third parties may be free to do so. After all, to demand an appropriate compensation for sending someone an unasked-for new car or a twenty-volume encyclopaedia is considered absurd.

  5. (5)

    Society has no direct duties towards parents (other than to guarantee truly free choice on procreation), but it has a duty to take over the parents’ responsibility where the latter fail. It is morally responsible, as guardian in loco parentis, for the fate of the child — carrying secondary responsibility after the parents for “voluntary” future individuals and primary responsibility for “involuntary” children. In addition, it is responsible for the fate of third parties. In all these cases, the arbitrariness principle kicks in again — no child and no third party deserves parents who fail their obligations.

  6. (6/7)

    In causal terms, parents and society carry their respective responsibilities only for newly created individuals, metaphorically speaking “the next generation.” It is this next generation that is, in turn, causally responsible for the fate of its successors—and so on. This chain of causal responsibility, combined with the arbitrariness principle (we could have ended up in any generation) has two implications.

First, (6) each generation (individual parents as well as society in their respective capacities) is morally responsible for providing the next with the capacity to fulfill its duties as potential future parents or guardians; hence each generation has a duty to provide the next generation with as much procreative freedom as possible, up to the level of freedom that it enjoyed itself, and it is free but not obliged to do more. Thus, above and beyond the level of strict sustainability, every generation is at liberty, not obliged, to provide for (research into) more efficient use of resources and the development of new resources. By procreative freedom, for the record, I mean the presence of the means and opportunity to be responsible for the consequences of the creation of an indeterminate number of future individuals. It does not imply a duty to reproduce (cf. Steiner and Vallentyne 2009), not even up (or down) to the level required to maintain a viable and sustainable reproduction rate within society; the slow and natural extinction of humanity remains an option. It does imply a duty to make such a sustainable level of reproduction an option—but the choice whether or not to use that opportunity lies with the individuals. Wombs and testes are not collective property.

Secondly, (7) each generation has a duty to inspire in the next generation a similar sense of duty as prescribed by principle 6.

So where does all this leave the idea of intergenerational distributive justice, of society as a steward of the property of an intergenerational collective? It is still very much alive, but not as we know it. Let me highlight four pertinent consequences of logically prioritizing individual responsibility and justice in exchange over intergenerational redistribution.

First, on the “parents pay” view of duties towards future individuals, parents, and sometimes in their stead society have a responsibility to guarantee any new child a fair and sufficient share of society’s legitimate property, i.e., the benefits and burdens of social cooperation. (They also owe it, by the way and by the same token, the opportunity to contribute his or her fair and sufficient share to those benefits and burdens.) This fair share may be assigned by any principle of distributive social justice, and measured by any metric; we’ll leave that choice undetermined. What matters is that those principles and metric take account of not just the direct day-to-day needs of individuals, but also of long term interests as expressed in, for instance, the Rawlsian and Nozickian plan of life. Such a plan of life will include (see duties 6 and 7 above) at least the opportunity to consider procreation. Thus, the existing and new individuals to whom the collective attributes shares in its property (natural resources and social product) are to be conceived of as more complete individuals than most theories of justice do today. Humans do not just hope or aspire (or not—if they so choose) to a career in this field, perfection in that activity, leisure through specific hobbies or sports, enlightenment through religion or art, and so on—they also aspire (or not) to being good, caring parents. In yet other words: the individual subject of social justice is a representative of members of the next generation, and through those members, a representative of all generations to come.

Secondly, understanding the option of procreation as part of a human being’s legitimate goals in life, as part of his or her interests, implies that a principle of intergenerational distributive justice still needs to be part of a complete theory of justice, albeit in a roundabout way. It may also seem to imply a bias in favor of this being the popular No Worse Off (NWO) principle—the present generation should leave future generations no worse off than it is itself (cf. e.g., Steiner and Vallentyne 2009). Yet, NWO is too simple: it prohibits, for instance, any investment however, small at the cost of one generation to the benefit, however, great, of later ones (cf. Wissenburg 1998), where even a Kantian deontologist would wish to argue for such investments provided the aim were not the maximization of welfare as such but the promotion of future individuals’ perspectives of autonomy.

Whatever (other) rule may come to guide society’s obligations to future individuals through present ones, it will, thirdly, have to be accompanied by a savings principle. Society, after all, is a parent too, or (see duty 5) at least a guardian for children whose parents, voluntarily or not, fail to take responsibility for them. The difference with a Rawlsian savings principle, however, is that it would not be a tax punishing non-procreators but would instead consist in a rule for estimating the number of “orphans” society represents here and now, in addition to the “non-orphaned” already represented by their parents.

Fourth and last, acknowledging the role of responsibility in procreation means that theorists of intergenerational justice will face new questions. The Parfitian questions how many people and who there should be, and what kind of life we wish to create for them, may be prior to the more action-oriented question how much precisely we owe to future generations in terms of natural resources, capital, security, or health, but both are in their turn preceded by the primary responsibility of individual parents for their individual progeny. In more accessible terms: deciding what to collectively save for the future comes second to asking what future is desirable per se, and third to asking, individually, whether it is prudent to procreate or not. It is the last question that determines the size of future generations’ rightful (or wishful) claims and the gap between promises we can and cannot fulfill.

This amended mission for intergenerational redistribution is of course based on two debatable premises: that there actually is a legitimately obtained social stock to provide a safety net for uncared-for children, and that it is a good thing to be responsible. As to the first premise: carrying responsibility, in this case the collective’s responsibility in loco parentis for the wellbeing of uncared-for children, does not logically imply that one also has the means to act accordingly—that is precisely what created the uncared-for in the first place. In a world where responsibility counts, where the products of labor belong to the producing individuals proportionate to their contribution (cf. Vallentyne 2007; Steiner and Vallentyne 2009), the creation of a collective stock, a “social product” created by e.g., taxation, is notoriously hard to justify. If there is any perspective for a legitimate collective stock, it may well lie where left-libertarians like Vallentyne et al. (2005) see it: in corrective taxation of those advantages individuals gain when they use more than their fair share of nature’s collectively owned resources. One could, perhaps, argue that any natural resources I appropriate include a taxable part X that I took to be unclaimed, but that by rights belongs to the voiceless unborn or the uncared-for.Footnote 10 Still, apportioning fair shares to individuals who represent both their own and the interests of their future offspring brings with it the further problem of ascertaining how much resources not just individual Y deserves but also that individual’s offspring. It will be clear that both the origins of the collective stock and the metric for its allocation remain controversial topics.

The second major premise supporting the “parents pay” view of intergenerational obligations is that it is an intrinsically good thing for an individual to be responsible, a premise utilitarians (for instance) might question. I do not have the room to defend this axiom adequately here; let me just point out that the capacity to take personal responsibility is not only a necessary condition of a liberal view of the good life, but also a necessary condition for a minimal version of autonomy, i.e., personal autonomy, the simple ability to choose (Waldron 2005; cf. Wolff 1990; also Jonas 1979)—and thereby a necessary condition for attributing an entity the ethical status of subject.

In this article I have focused exclusively on the implications for the philosophy of intergenerational justice of a more realistic conceptualization of individual responsibility and the value of future individuals. Policy implications have been mostly ignored—I have not talked about whether and how irresponsible parents should be sanctioned, for instance, or whether prioritizing individual responsibility implicitly allows burdening future generations, “other people’s children,” with debts or with a seriously diminished range of choices in life styles. I want to end, nevertheless, by highlighting one important “practical” consequence. The “parents pay” view of duties towards future generations limits the freedom of choice of a society (liberal democratic or other) when it comes to population policy and “procreation management.” There is no denying that even in liberal democracies, “the collective” pre-empts many of the individual’s choices, including those in the area of procreation: states shape the educational system, the labor market, the economy, social security, and child benefit systems. Unlike Platonic totalitarianism with its breeding programs, liberal democracy accepts that there are limits to the duties and burdens it can legitimately impose on its citizens. Individual desires may sometimes have to give way to the interests of the collective, but in a liberal democracy individuals are never sacrificed for the good of the community. Liberal social and political institutions aim to serve the interests of individuals, a mission that is incompatible with a view of individuals as possessed by the collective, and a mission that is therefore also incompatible with the idea that procreation is first of all a collective and only in second place an individual concern—quite the reverse is true. To make a long story short: the “parents pay” view asks that collective choices affecting the individual lives of real-existing human beings take as their primary aim respect for all existing individuals rather than only those who would procreate; in other words, “parents pay” calls for the emancipation of individuals (including procreative autonomy) rather than the redistribution of dependence.