Identity and (Legal) Rights of Future Generations, The

The Identity and (Legal) Rights of Future Generations Ori J. Herstein* I. Introduction What is the moral significance of “future generations”? Can they be the subjects of (legal) rights? And if so, do they have rights? The first part of this Article reflects on who or what in future generations is of moral significance, exploring three options: future generations as groups, as future individuals, and as types of future people. Lacking both intrinsic and instrumental moral significance, perhaps the most prevalent conception of future generations—generations, age groups,1 and birth cohorts2 (all taken as groups)—is ruled out as morally insignificant, thereby ruling out the notion that the rights of future generations are rights of generations. The notion that future individuals should be the focus of the moral concern with future generations, and therefore the subjects of the legal rights of future generations, is questioned for the straightforward reason that there are no future individuals, thereby putting in doubt the notion that the rights of future generations are rights of individuals. Even if not of intrinsic moral value, the most probable or promising subjects for the moral concern with the future are types of future people.3 Based on the determination that types of future people is the most promising object on which to project our concern for future generations, this Article explores how types of people (as opposed to particular individuals) can have (legal) rights and be the object of (legal) duties. Within the philosophical literature, the leading approach explaining how future generations can have rights argues that they have rights as tokens of types of people. Indeed, general norms confer rights on those actual individuals who meet the criteria of the normsubject type. As such, a right is conferred on individuals not as partic* LL.M., J.S.D., Columbia School of Law; B.A., LL.B., Tel Aviv University; Associate, Fried, Frank, Harris, Shriver & Jacobson LLP. For their comments I am grateful to Ittai BarSiman-Tov, Noa Ben-Asher, David DeGrazia, Michael C. Dorf, Robert A. Ferguson, Kent Greenawalt, Philip Hamburger, Miguel Herstein, Joseph Raz, Steven Smith, and the editors of The George Washington Law Review. Special thanks to Uri D. Leibowitz. 1 Comprising people of the same age. 2 Comprising people born during the same period. 3 The term “future people” refers to those people who will live in the future. September 2009 Vol. 77 No. 5/6 1173 1174 The George Washington Law Review [Vol. 77:1173 ulars, but as tokens of the general norm-subject type. Accordingly, future people may currently have rights, just not as particular individuals, but as particular instances of a normatively protected type of person. Here, this type-based approach is applied to the concepts of legal norms and legal rights, demonstrating how according to this approach future people can have legal rights as tokens of legally protected types. After presenting this account of the rights of future people and couching it in a jurisprudential context, this Article points out a possible deficiency in the approach’s metaphysical underpinnings. Assuming that future people can have rights (be it as individuals or as tokens of types) there are still reasons for doubting whether in most cases future people have any rights, deriving from the doubt whether future people will be harmed by most actions and choices in their prenatal past. According to what has come to be known as the “nonidentity argument,” actions and choices that are necessary parts of the causal chain leading up to the existence of a person cannot harm that person—had the act or choice not occurred that person would have never existed, and one is better off existing than not. Under the two prevalent theories of rights, the Will Theory and the Interest Theory, the nonidentity argument seemingly entails that future people have no rights. After exploring how this is the case, the conception of harm underlying the nonidentity argument is analyzed. Two types of interests future people may have in prenatal identitydeterminative events (constitutive interests and threshold interests) are explored as possible sources of certain rights future people may have—the nonidentity argument notwithstanding. II. How Are Future Generations Morally Significant or Who in Future Generations Should Be the Subject of (Legal) Rights? In the context of normative discourse, the terms “generation,” and more specifically “future generations,” are used in varying and often interchanging ways. Linguistically the term “future generations” may refer to all future people, some future people, all future generations (focusing on the generation as such, not on its members), some future generations (e.g., temporally closer future generations), certain future age groups or birth cohorts, future groups (e.g., future generations of Romanians), future members of a certain future group, types of future people, and so on. But which of these uses denote a future entity that is worthy of moral concern and as such should be the holder of the legal rights ascribed to future generations? Three options come to mind: future groups, future individuals, and types of 2009] The Identity and (Legal) Rights of Future Generations 1175 future people. The more prevalent manifestations of the first and second options are seriously questioned. The final option is adopted as the most promising of the three. A. “Future Generations” as Groups We may take the concern for the future as directed towards future groups. It is not uncommon to think of the rights of future people and future generations as group-rights, referring to the rights and interests of future generations as such (i.e., referring to the generation itself). In this view—focusing our concern with the future on how each future generation, as a generation, is faring—the term “generation” denotes a group most often comprising all people living during a certain period in time. Other prevalent future groups in the intergenerational justice debate are age groups and birth cohorts. Yet, although there are clearly future groups that are of moral significance, generations, age groups, and birth cohorts are not among them. A group may be of both intrinsic and extrinsic moral significance. A group’s most significant instrumental role—the source of its extrinsic moral significance—usually derives from the group’s contribution to the well-being and welfare of its members. Two interesting explanations of the intrinsic value of groups and social goods are those of Charles Taylor and Joseph Raz. Taylor points out that many intrinsic goods, such as art, authenticity, sophistication, or a fulfilling experience, depend on a background of meaning, which is provided by a culture allowing for the very intelligibility of such goods.4 The value of a specific culture derives not only from its instrumental role in facilitating goods but is also intrinsic—because many of these goods only exist and are only intelligible within, or in the context of, that specific culture.5 Moreover, cultures are inherently social; cultures are part and parcel of the society, community, or group they are a culture of.6 Therefore, the intrinsic value of the culture is also that of the society.7 Raz argues that assuming that an autonomous life is of intrinsic moral value and, seeing that certain social goods, such as subsistence, a sufficient array of acceptable choices, and possible life patterns are constitutive of an autonomous life (i.e., an autonomous life includes having 4 CHARLES TAYLOR, PHILOSOPHICAL ARGUMENTS 136–40 (1995). 5 Id. 6 Id. 7 Id. 1176 The George Washington Law Review [Vol. 77:1173 these social goods, not that these social goods facilitate such a life), it follows that such social goods are of intrinsic moral value.8 Are the groupings of future people into units of generations, age groups, or birth cohorts of moral significance? Should they be ascribed rights? I think not. The mere fact that several individuals are alive at the same time, and therefore a part of the same generation, constitutes a very “thin” or limited group and one lacking content (if it is a group at all) not unlike, for example, the group of people who eat tomatoes. Such aggregations of individuals have very few to no characteristics of the sort that contribute to the well-being or welfare of people, and are therefore of little to no (extrinsic) value or significance for their members. A group may be instrumentally significant for the well-being of individuals through its value-creating and value-facilitating roles, such as furnishing individuals with a culture, history, language, identity, sense of belonging, security, opportunities, public goods, social goods, patterns of valuable life choices, the fruits of collective action, etc. A generation, as a group, lacks all such instrumentally valuable aspects; it offers its members nothing, because it has too few attributes or characteristics and little content (of the above-mentioned nature). “Generation” really captures no less than “everyone” (alive at a certain point in time). People do not group, act, create, identify, or live in generations. It is other and often much smaller, denser, richer, and often oppositional groups, such as nations, religions, families, societies, clans, corporations, unions, clubs, or ideological movements that have instrumental value, according them with (extrinsic) moral significance. Perhaps with the continuing materialization and spreading of globalization groups such as generations may take on more content and meaning in people’s lives. But, as things are now, and as they have been for quite some time, this is not the case. For the same reasons I am hard pressed to identify any intrinsic moral significance in generations (as groups). A generation is nothing more than a shell, merely another way of referring to “everyone” alive at certain period, a group that is not a social good of intrinsic value or moral significance in either Raz’s terms (it does not include aspects of an autonomous life) or Taylor’s terms (it does not function as a background of meaning to intrinsically valuable goods). As just mentioned, the term “generation” is also often used to refer to age groups or birth cohorts. When taken as such, the term “generation” does not refer to the group comprising all people alive at 8 JOSEPH RAZ, THE MORALITY OF FREEDOM 203–07 (1984). 2009] The Identity and (Legal) Rights of Future Generations 1177 a certain period but to some of its subgroups, which would most likely temporally overlap with other age groups and birth cohorts. Even taken to refer to an age group or birth cohort, the term “generation” still has little role in determining the well-being of individuals. An examination of the impact of how well an age group is faring has on the well-being of its individual constituents must account for all members of the age group. It cannot merely account for those living at a certain place or belonging to a certain community in which membership in that particular age group is of significance to its members. If it did, we would no longer be referring to a generation but to a more discrete group—to a “generation of ___.” The same is true for birth cohorts. For this reason, I am hard pressed to identify an example in which an age group or a birth cohort makes up a group that is worthy of moral concern. Generally, age groups and birth cohorts have a minimal instrumental role (if any) in contributing to the wellbeing of their individual members. They capture little more than the fact that certain individuals were born in the same period or are of the same age. For similar reasons as generations, age groups and birth cohorts are groups of little if any moral significance. Consider the example of the “Free Love Generation.” That the Free Love Generation, as a group, fared well or was successful in having a positive and significant impact on society does seem to have value for its members (as well as for others), giving it some moral significance. But, this example does not demonstrate the instrumental value of an age group or birth cohort. The Free Love Generation is neither an age group nor a birth cohort as are, for example, “people born between 1921 and 1937” or “all people in their forties.” Not everyone who was in his or her teens, twenties, or thirties in the 1960s and 1970s, even only in the U.S. or Central and Western Europe, was part of the Free Love Generation, not to mention his or her contemporaries in other parts of the world. The use of the term “generation” in the name “Free Love Generation” is an example of a more finegrained use of the term than merely referring to an age group or birth cohort. Although many, if not all, who belong to the Free Love Generation also belong roughly to the same age group and birth cohort (clearly, being a certain age is a necessary condition for membership in that group) it is not a sufficient condition. There are other attributes that one must have at least partially exhibited to be part of the Free Love Generation, such as: living in the West, liking certain music, taking part in counterculture practices, opposing the Vietnam War 1178 The George Washington Law Review [Vol. 77:1173 and the Soviet occupation of Prague, etc.—attributes not shared by all those who were youths during the 1960s and 1970s. Furthermore, the reason we focus only on what is a handful of all age groups and birth cohorts when we speak of generations is that most such groups have no significant role in determining individuals’ well-being or are of no cultural, historical, or social significance. We may sort humanity according to decade, year, month, or even minute of birth. It is certain circumstances that make some age groups or birth cohorts significant and valuable and leave us indifferent to most other such groupings of individuals. For example, the reasons we focus on generations such as the Depression Generation, the WWII Generation, or the Baby Boomers Generation are that those groups have significant cultural, historical, political, and economic significance, differentiating them—as significant and valuable for their members and others—from other valueless groups, in which age or time of birth is also a condition for membership. Moreover, even these valuable birth cohorts and age groups are not universal. Such groups are culturally limited to including only some of those in the world who were born during the same period or are of the same age. It is their social context, limiting the membership in such groups, that makes them into significant groups. We are inching our way towards a familiar conclusion: certain groups and communities contribute to the well-being and welfare of their individual members and are therefore of extrinsic moral significance. Moreover, some groups may even be of intrinsic moral significance. As was just explained, however, a “generation,” taken to refer to the grouping of all individuals living during the same period or even only to a particular age group or birth cohort, is not such a valuable group. It is possible to divide a generation into various subgroups, searching for those that are of concern and therefore perhaps worthy of legal protection (in the form of legal rights). This exercise, however, would not redeem the moral significance of generations as groups. Rather, it would lead us towards morally significant groups which are not generations, such as nations, families, etc., entities that have a “lifespan” that does not necessarily correspond to that of a generation (as a group) and which delineate their members according to much richer properties than mere age, time of birth, or historical period of existence. In addition, although some perceive the rights of future generations in terms of group rights, at a minimum generations, age groups, and birth cohorts (taken as groups) lack some central characteristics 2009] The Identity and (Legal) Rights of Future Generations 1179 of groups. First, a group may have its own interests, rights, duties, etc.,9 which are not simply the interests, rights, and duties of its individual members. For example, there is a sense in which a military unit, as a group, can do extremely well (winning the battle) while that unit, as an aggregation of individuals, fares poorly (losing most of its members). In the first sense the unit, as a unit, does well as a military unit by achieving its ends. In the second sense the various individuals that make up the group fare badly, to varying degrees, and the aggregation of how well the individual interests of the various individual members of the unit are faring adds up to a negative sum. When we think of the unit as a group, we associate it with its own interests. In contrast, when viewing the unit as an aggregation of individuals, the interests of the unit are simply the function of the alignment of certain interests of its members. Therefore, a condition of groups is that the correlation between how well the group is faring and how well the interests of its members are faring is not tautological. If “how well the group is faring” is simply another way of referring to how well its individual parts are faring, then the group itself seems to have no interests of its own that are different from those of its individual members. If this is the case with future generations—taken as generations, age groups, or birth cohorts—there is no use in focusing on units of generations (as groups), because what directly matters are the individuals the group comprises. Moreover, the impact of how a group is faring has on the wellbeing of its individual members can be a much more complicated relation than that between the interests of an individual and an aggregation of individuals. For example, when the A-train breaks down, the inconvenience caused to the collection of people referred to as the “passengers on the train” is directly reducible to the inconvenience caused to the specific passengers. There is no sense in which “the passengers on the train,” as a group, is somehow benefited or harmed distinctly from what happens to each one of those individuals actually on the train. “The passengers on the train” is not a group. If one person on the train benefited from the train breaking down while all others were inconvenienced, we would not say that, although he individually benefited, he also suffered some setback as a member of the “passengers on the train.” The more accurate description is that some of the passengers were inconvenienced and others were not. In contrast, if one’s community (a sort of group) is destroyed in a way that, 9 See Joel Feinberg, The Rights of Animals and Unborn Generations, in PHILOSOPHY & ENVIRONMENTAL CRISIS 43, 56–57 (William T. Blackstone ed., 1974). 1180 The George Washington Law Review [Vol. 77:1173 indirectly, actually ends up also benefiting one, it still may be the case that the destruction of one’s community is also in some sense bad for that person. Generations, age groups, and birth cohorts do not seem to have interests that are distinct from the interests of their members. In what sense can a generation (taken as the group comprising everyone alive) be made better or worse off in and of itself? In the context of intergenerational justice the better understanding of the terms “generation,” “age group,” and “birth cohort” is as referring not to a group, but to all the individual people alive during a certain period or born during a certain time frame—referring not to the grouping of these individuals, i.e., not to the generation as such, but simply to a large number of individuals through some common temporal denominator they all happen to share. It is the interests of future people that seem important in these cases, and terms such as “generations,” “age groups,” and “birth cohorts” function as linguistic shortcuts for referring to such future individuals. To clarify, when referring to the future these terms do not refer to individual people; rather, they refer to a type of future person. For example, the “next two generations” refers to two types of future people, each defined by the time during which the individual tokens of which will live. To conclude, in the case of future generations the group-based use of the term “generation” is either confusing or wrong. If “future generation” refers merely to all people who will live in a particular time frame then by focusing on the generation as such, i.e., as a group, the term steers us away from the entities that are of moral significance—the individual future people. And, if the term “generation” refers to a group, it focuses the intergenerational justice debate on a type of group that is no more than a shell, void of any role in determining the well-being of its individual members as well as of any intrinsic moral significance. In addition, although there may be morally significant future groups, they do not include generations, age groups, or birth cohorts. Therefore, although it still may be the case that some of the rights of future generations are group rights, such rights are not the rights of generations (as a group). Future generations (as such) do not matter in intergenerational justice. B. “Future Generations” as Future Individuals At least for those with liberal and generally humanistic sentiments, “individuals” is perhaps the most natural answer to the question with whom or with what in the future should we concern ourselves when thinking of future generations. This is especially so, 2009] The Identity and (Legal) Rights of Future Generations 1181 considering that when thinking of future generations, at least from a moral point of view, we are in one way or another always concerned with the well-being and welfare of those people who will live in the future. There is at least one problem with this answer—there are no future individuals. An individual is a particular person with a certain personal identity making him or her into the particular person he or she is. Future people are not individuals as they have no personal identity; “they” do not yet exist and who “they” will turn out to be is in most cases indeterminate. At best there are potential future individuals, only a very small number of whom will actually end up coming into existence.10 A person’s identity depends on that person’s properties, be they psychological, physiological, or a combination of both.11 And, psychological/physical properties are, at least to an extent, causally determined—a person is born and develops his or her psychological makeup, personality, memories, character, and social identity as a causal consequence of factors such as the biological makeup of one’s parents, one’s own psychological propensities, how and in which society one grew up, and what one’s other life circumstances were, all of which are also, at least partially, effects of other causes. Seeing that many of the events that will take part in causally determining the individuality and identity of future people have not yet occurred, it follows that the individuality of future people is indeterminate. Therefore, there are no future individuals.12 That at no present point in time are there any particular future individuals to concern ourselves with (because “they” are indeterminate and nonexistent) entails that our concern for future generations See MELINDA A. ROBERTS, CHILD VERSUS CHILDMAKER: FUTURE PERSONS AND PREDUTIES IN ETHICS AND THE LAW 89–90 (1998) [hereinafter ROBERTS, CHILD]; see also Melinda A. Roberts, Present Duties and Future Persons: When Are Existence-Inducing Acts Wrong?, 14 LAW & PHIL. 297, 316–20 (1995) [hereinafter Roberts, Present Duties]. 11 For seminal contributions to the debate of personal identity and for different approaches to this concept, see generally ROBERT NOZICK, PHILOSOPHICAL EXPLANATIONS, 27–70 (1981) (discussing concepts of self); DEREK PARFIT, REASONS AND PERSONS, 245–80 (1987) (discussing different philosophical views of personal identity in the physical and psychological perspective); PERSONAL IDENTITY passim (Raymond Martin & John Barresi eds., 2003) (discussing personal identity); PERSONAL IDENTITY passim (John Perry ed., 1975) (discussing theories of personal identity and critiques). 12 One underlying assumption here is that the world is not fully deterministic. If it were, the fact that certain events that are determinative of the identity of future individuals have not yet occurred would not entail that there are no future individuals, because all such events as well as their causes would be already determined. 10 SENT 1182 The George Washington Law Review [Vol. 77:1173 cannot be understood as a concern with future individuals. If it were, it would necessitate that our concern for future generations is concern for nothing at all. The fact that actual individuals will inhabit the future does not undermine the proposition that there are no future individuals. When the individuals of the future will come into existence they will be actual people, never future individuals.13 And, in most cases, until “they” are born (or conceived) who “they” are is indeterminate. “They,” i.e., future individuals, at any present point in time do not exist (as individual people).14 C. “Future Generations” as Types of Future People The term “future generations” is best understood to refer to future types of people. A type of person is not an individual but a set of properties that more than one individual may exhibit. There are several benefits to looking to types of future people as the objects of intergenerational justice. Often when ascribing future generations with rights, future people are approached as comprised of generic individuals, i.e., of a single type of person who has interests common to all humans. This view corresponds to the conception of “future generations” as the aggregate of all future people. Indeed, some aspects of 13 Annette Baier analogizes future people to the dead, claiming that if it is possible to refer to the latter it is also possible to refer to the former. See Annette Baier, The Rights of Past and Future Persons, in RESPONSIBILITIES TO FUTURE GENERATIONS 171, 173 (Ernest Partridge ed., 1980). However, the dead and future people are not analogous—while the dead were once alive and are therefore identifiable as particular individuals, future people are indeterminate. 14 It is possible to argue that under a four-dimensional account of the world, future individuals do exist (for this observation I am grateful to David DeGrazia). In a four-dimensional framework, people (and other objects) exist in an eternal or timeless sense, deployed in fourdimensional space-time. In this worldview time is not a factor in attribution, similarly to how in a three-dimensional world distance in space between a subject and an object of value does not matter for attributing that value to that person. For example, breaking a confidence wrongs one regardless if it takes place nearby or in a remote land. In a four-dimensional account of time, future individuals do exist, because the directionality of causality does not determine when things come into existence—all things exist “all the time,” and therefore there could be future individuals to whom we, “at present,” can attribute rights and wrongs. For an account of fourdimensionalism in a similar context, see Harry S. Silverstein, The Evil of Death Revisited, in LIFE AND DEATH: METAPHYSICS AND ETHICS 116 (Peter A. French & Howard K. Wettstein eds., 2000) (arguing, based on a four-dimensionalist framework, that death and posthumous events may have value for one even though once one dies one no longer exists and therefore there no longer is anyone to whom to attribute [person-relative] value). One limitation of adopting a four-dimensional framework in this context is that, by way of metaphor, it uses a canon to shoot down a fly. Shifting to a four-dimensional framework entails numerous implications on many of our unrelated deeply held beliefs, making it, in all probability, a less than plausible option. This is especially true if used merely as a means of answering the “timing problem” of prenatal value; more so where an approach less disruptive to our conceptual scheme is available. 2009] The Identity and (Legal) Rights of Future Generations 1183 intergenerational justice may be similarly relevant to all members of future generations, such as, perhaps, a shared interest in a sustainable ecosystem. The philosophical debate on intergenerational justice may be cast in universal terms, focusing on the rights of all future individuals as humans, giving little consideration to the partiality we seem to show to different types of future people and future groups. The legal discourse on future generations can also be cast in terms of international human rights, once again focusing on what is owed to all future people and on the corresponding universal rights that all legal systems should provide. For this universal approach to future generations, terms such as “generations,” “age groups,” and “birth cohorts” (here denoting all individuals alive at a certain period, not a group) are useful due to their universality. Although a valid outlook in certain cases, this homogeneous conception of future generations does not capture the full range of what is morally significant in future generations. Not all future people have, or should have, the same rights or are owed, or should be owed, the same duties, at least not in all cases. Accordingly, the use of the term “future generations” is much more versatile, potentially referring to a variety of different future groups and types of people. When evoking “future generations” what is referred to is often a more discrete set than that of all future people alive during a certain period (such as the duration of a generation), referring to, for example, future people of a particular type or group, as expressed in the parlance “future generation of ___.” These uses may range from referring to all individuals who will inhabit the globe in any future time to future members of a particular family, lineage, line, religion, culture, nation, race, gender, citizenry, party, ideology, profession, caste, geographical community, ethnicity, and so on or simply refer directly to such groups. For example, we may concern ourselves with the future generations of the Cherokee Nation, future generations of Americans, future generations of Tudors, or our own (unconceived) future grandchildren. All these constitute different types of future people. The focus on types of future people accounts for the variety of future people. Future people will most likely be a diverse collection of individuals of different and even conflicting interests, interests that may make different demands on current generations. When the head of the Society for the Preservation of Spanish Heritage speaks about safeguarding the treasures of Spanish culture for the sake of future generations, she is not referring to the same set of future people the King of Jordan is referring to when he espouses the benefits peace in the Middle East 1184 The George Washington Law Review [Vol. 77:1173 will offer future generations. The same may be true for different legal systems in approaching the interests of “future generations” and in enacting norms that regulate conduct affecting the interests of future generations. In addition to the multiplicity of the individuals making up each specific generation, the heterogeneity of future people is further enhanced by intergenerational differences. Separated by years and millennia, future people will likely show great and diverse differences, just as they have differed in the past. Beyond the variety of future people, when theorizing on the relation between current legal systems and future generations, one must also take into account the diversity and even parochial nature of legal systems, often causing legal systems to differ in their interests, regard, and treatment of different members of future generations. Different legal systems do not, and to a degree should not, similarly concern themselves with the same future people. In other words, legal systems, even those functioning during the same era, may treat like cases differently, conferring on future people different legal rights. For example, when oriented towards future generations, the German legal system may, depending on the context, concern itself with future members of the German citizenry, the German nation, all future Europeans, etc. These (future) subjects the German legal system is oriented towards, or should be oriented towards, are not necessarily shared by other legal systems, which may focus on radically different (future) subjects when forming norms affecting the interests of future generations. It is therefore fruitful to break down the generalized use of the term “future generations” into smaller, more concrete groups and types of future people so as to correspond better to the way legal systems should and often do discriminate among different future people when taking the interest of future generations into account.15 15 Such partiality exercised by legal systems is not necessarily morally wrong. First, some moral obligations are agent relative. On agent relativity, see Bernard Williams, Consequences and Integrity, in CONSEQUENTIALISM AND ITS CRITICS 20 (Samuel Scheffler ed., 1988). For example, a father may owe certain duties towards his children, duties he may not owe to others. The same is often true for the relation between a family and its members, a country and its citizenry, etc. On occasion, this type of reasoning seems also applicable to the moral regard owed by different current moral agents to various and different members of future generations. For example, it is plausible that in certain respects the U.S. has a stronger duty towards future Americans than it does towards other types of future people. Second, partiality shown by legal systems may also be justified on the grounds of feasibility and division of labor. Normative reasoning must take reality into account when making practical prescriptions, ascribing different duties to different agents and political entities. Some legal systems are more suitable and better positioned than others to carry out certain moral tasks and to take care of the interests of certain members of future generations. Such circumstances may, 2009] The Identity and (Legal) Rights of Future Generations 1185 It is true that some legal systems have provisions that are universal in their regard and applicability to all people (at least in relation to all those alive at a particular moment in time). Such norms aspire to protect what is owed to all future people, applying what is in a sense a human rights approach. We may use the term “generation” (not as a group) to refer to this very broad type of person. But norms of a universal scope or extension are an anomaly within most legal systems. It is the essence of laws to make distinctions, and the vast majority of legal norms make distinctions between types of people. A further benefit of focusing on types of people rather than on individuals or generations is that in many cases the rights of future people are collective rights, not individual rights. Rights are a function of interests: Q has a right to y if Q’s interest in y is sufficient reason for holding others to be under a duty to y.16 Now, although people often have interests in social goods, maintaining and generating such goods is often burdensome on many. Therefore, the “cost” of many social goods is often too great to be justified by the interest of a single individual. Rights to social goods usually derive from the cumulative force of the interests of many individuals. Therefore, many rights for social goods are in fact collective rights,17 deriving from the alignment of the interests of many. A type of person may be instantiated by many actual individuals, together making up a collective or aggregation of many individuals whose interests give rise, in concert, to a collective right. Therefore, the rights of future generations for many social goods are most likely the rights of collectives or aggregations of future people. And, a collective or collection of individuals is a number of individuals who are all tokens of a type of person, which comprises the set of shared attributes by which we refer to those distinct individuals collectively (e.g., “redheads” or “the people on the bus”). Many if not most of the interests of future generations are in social goods, such as the promotion and perseveration of knowledge and culture, public health, technology development, energy resources, sustainable economy, etc. These are all goods that require vast resources and collective action. For example, assume that in the future people will live, on average, much longer than we do. These people will redue to considerations of feasibility and division of labor, justify partiality in treatment given by legal system x to person Q over person P. It should be noted that neither of these two arguments justify all legal partiality. Clearly some legal systems discriminate unjustly. 16 See RAZ, supra note 8, at 166. 17 See id. at 207–09. 1186 The George Washington Law Review [Vol. 77:1173 quire Social Security for a longer period than we do. This may mandate that we, in the present, funnel even more resources into our failing Social Security system in order to secure the well-being of future people. The interests of no single future person can justify ascribing us with a duty to sacrifice our own welfare in order to further strengthen the Social Security system for his or her individual benefit. The combined interests of all those Americans of the next few generations who will live to see the age of, let’s say, 150, however, may have sufficient force to justify such a duty. Therefore, the right of future generations for a robust Social Security system must be a collective right, where at present the collective or set of individuals with the property of longevity is captured by the notion of a type of person (expected to have many tokens in the future).18 A further benefit of thinking about the future in terms of types and not individuals is epistemic. Even assuming that future individual people do exist, in many respects their needs, life circumstances, wishes, and very identity are beyond our capacity to foresee. Because of this, tailoring rights for future individual people will often fail as a normative tool for regulating conduct. First, not knowing whom they are directed towards, some such norms will be very difficult to construct. Second, seeing that moral and most forms of legal liability require responsibility, which in turn assumes foreseeability, and considering that foreseeing what will harm or wrong future individuals is practically impossible in many cases, rights ascribed to future individuals will often be a dead letter—most violators of such rights legally could not and morally should not be held responsible for such violations. To conclude, focusing on types of future people avoids some of the epistemic and metaphysical difficulties that arise in the case of future individuals. It also allows for capturing several aspects of future generations (all of which are relevant to the notion of the rights 18 To clarify, the notion of rights of a collective (when the term “collective” is taken to mean an aggregate of individuals, not a group) should not be confused with the notion of group rights. A group right is the right of the group as such, reflecting the interests of the group. A right of a collection of people of a certain type, who are an aggregation of individuals, is a “collective” right in that it derives from the collective force of the personal interests of all the individuals that make up the collection. There is no sense in which a collective right somehow derives from the interest of the collective in and of itself, as if it were a group. There is a difference between an alignment in the interests of several individuals, which weighed together justify a collective right, and the notion of a group’s right. The rights of a group derive at least on one level from the group’s own interests, not (at least not directly) from the interests of the particular individuals it comprises. 2009] The Identity and (Legal) Rights of Future Generations 1187 of future generations) such as their diversity (as well as the diversity of rights-conferring normative systems) and the collective nature of many of the rights of future generations. III. Two Puzzles Concerning the Rights of Future People The moral significance of future people may warrant affording them the protection of legal rights. Still, there are at least two puzzles with the notion of the rights of future people. First, accepting that types of future people are the key subjects (along with certain groups) in the concern for future generations raises the question: how can types of people (as opposed to particular individuals) have (legal) rights and be the object of (legal) duties? Second, putting the nonexistence of future people aside and assuming that future people can be attributed with rights (be it as tokens of types or as individuals), it is not clear whether actual future people, once they come into existence, in fact have any rights in relation to most of the prenatal events and choices that are often thought to be the focus of the rights of future people. A. The Structure of Legal Norms and the Problem with Types Having (Legal) Rights A right is always a right of someone or something. Therefore, one wonders how can future people, who do not exist and are not yet determinate, have rights—in whom does the right vest and to whom is the corresponding duty owed? If I shoot an arrow into a crowded wood, I normally have a duty towards all those in the wood not to do so, and they have a corresponding right against me. That the arrow will only land in five minutes and only then will it be determined whom exactly among those particular individuals in the wood the arrow will hit is immaterial to the fact that when I shoot the arrow I abridge my duty not to cause injury, which I owe to the person who actually ends up getting hit. The indeterminacy at the time I released the arrow of who among those at risk will be harmed does not suggest that the person who will end up injured does not have a right that I not shoot an arrow at him. Although who among those in the wood will be injured is indeterminate at the time I shoot the arrow, whoever it turns out to be is already in existence at that time—there is an actual person to whom I am liable, even though it is still neither known nor determined who exactly that is. In contrast, if the arrow takes a hundred years to land, then the identity of the specific future victim was not only indetermi- 1188 The George Washington Law Review [Vol. 77:1173 nate when I shot the arrow, but no such person existed at the time. Unlike the previous example, where it was determined which actual individuals were in danger even though it was indeterminate who among them will be hurt, here it is the very existence of those who are put in danger that is indeterminate. To whom then did I have a duty not to shoot the arrow and who had a right that I not do so? The answer seems inescapable—no one. No one existed at the time and because there are no free-floating rights, i.e., rights categorically are the rights of someone or something, it seems that no one had a right that I not shoot the arrow.19 To clarify, nonexistence should not be confused with not being born. There is a sense in which one may exist before one’s birth. For example, it seems sensible to claim that a fetus, at a stage in which it is still not yet a person, is far enough developed to establish an individual future person—the person the fetus will evolve into. Such a future person already has DNA, often a name, personal history, sex, parents, etc. Unlike the case of the evolved fetus, however, future people are mostly indeterminate and as such do not exist as individual people. Seemingly one way of overcoming this puzzle is arguing that we owe the duty not to any individual future person but to a type of person, which in the example above would be loosely defined as “people in the wood who will be injured by the arrow.” And, while the individual person who will be injured may not yet exist when we abridge “her” rights, the type of person does. Yet, one cannot but wonder how can types have rights? How can there be a duty to a type of person? After all, types (of people) are not actual individual people and a right is always the right of some actual particular entity, such as a person, corporation, animal, collective, or group. Rights, by their nature, benefit the rightholder;20 as such, the rightholder must be capable of having interests.21 Having legal norms conferring rights on types (of people) and conferring corresponding duties owed to types (of people) seems to conflict with this position. Types cannot be said to have interests—after all, a type is 19 For versions of this puzzle, see Ruth Macklin, Can Future Generations Correctly Be Said to Have Rights?, in RESPONSIBILITIES TO FUTURE GENERATIONS, supra note 13, at 151, 151–55; Galen K. Pletcher, The Rights of Future Generations, in RESPONSIBILITIES TO FUTURE GENERATIONS, supra note 13, at 167, 167–70; Mary Anne Warren, Do Potential Persons Have Rights?, in RESPONSIBILITIES TO FUTURE GENERATIONS, supra note 13, at 261, 261–73. 20 See, e.g., Feinberg, supra note 9, at 51; F.M. Kamm, Rights, in THE OXFORD HANDBOOK OF JURISPRUDENCE AND PHILOSOPHY OF LAW 476, 477–83 (Jules Coleman & Scott Shapiro eds., 2002). 21 RAZ, supra note 8, at 180; see also Feinberg, supra note 9, at 51. 2009] The Identity and (Legal) Rights of Future Generations 1189 nothing more than a category, an abstraction, a set of properties. The incapacity of types to have rights puts into question the validity of the notion that types of future people can have rights and be owed corresponding duties. One suggestion is that norms can confer (legal) rights not on types per se but on future people as tokens of types of people (not as a particular individual person). In various versions, this type-based approach to the rights and interests of future people has had several advocates in the recent philosophical literature on future people.22 Here I introduce the puzzle and the type-based approach in jurisprudential terms, applying them to the concepts of legal norms and legal rights. After presenting the type-based account in the most favorable light, I close this Section with what seems to me the main difficulty with the types-based approach. 1. The Structure of Legal Norms23 Norms comprise a norm-subject and a norm-act24—respectively determining whom the norm addresses and the prescription the norm conveys, such as conferring a duty or right. Norms may be general or particular. A particular norm is directed at a specific subject; a general norm is directed at a type, applying to whoever exhibits the characteristics that make up the type. A particular norm may in principle address more subjects than a generalized norm with a narrowly defined norm-subject. Therefore, although a generalized norm would normally apply to more individuals than a particular norm, the difference between the two types of norms is that of form and not of the number of norm-subjects. In addition, there are rights-conferring norms and duty-conferring norms, both of which may be general or 22 See, e.g., Caspar Hare, Voices from Another World: Must We Respect the Interests of People Who Do Not, and Will Never, Exist?, 117 ETHICS 498, 498–523 (2007); Rahul Kumar, Who Can Be Harmed?, 31 PHIL. & PUB. AFF. 99, 110–15 (2003); Jeffrey Reiman, Being Fair to Future People: The Non-Identity Problem in the Original Position, 35 PHIL. & PUB. AFF. 69, 83–86 (2007). 23 My concern here is with prescriptive norms, i.e., norms that prescribe actions and function as reasons for action. Not all norms are prescriptive. There are, for example, constitutive norms, which contribute to the constitution of practices, such as the rules of a game. See GEORG HENRIK VON WRIGHT, NORM AND ACTION: A LOGICAL ENQUIRY 6–7 (1963). Although following constitutive norms is a condition for performing the practice those norms constitute, for example playing a game, such norms do not prescribe that one ought to follow them, only that one need follow them in order to perform the practice. I will treat legal norms as prescriptive. In the interest of brevity, I will refer to prescriptive norms simply as “norms.” 24 Id. at 1–16; Andrei Marmor, The Rule of Law and Its Limits, 23 LAW & PHIL. 1, 9 (2004). 1190 The George Washington Law Review [Vol. 77:1173 particular. Duty-conferring norms take one of the two following forms: • Individuals A, B, and C [particular norm-subject] ought to y [norm act]. • All Qs with feature O [norm-subject type] ought to y [norm 25 act]. Rights-conferring norms may also be particular or general and are structured a little differently from duty-conferring norms: • Individuals A, B, and C [particular norm-subject] have the right to y [norm act]. • All Qs with feature O [norm-subject type] have the right to y [norm act]. Rights-conferring norms are directed at a rightholder as a normsubject. Duty-conferring norms are directed at those who are subject to the duty as norm-subjects. Rights-conferring norms are related to, give rise to, or function as a source of duty-conferring norms because—as is widely accepted—by definition rights have corresponding duties.26 While legal norms may have either a particular or a general norm-subject,27 the law does (and mostly should) favor general norms. In fact many, if not most, legal norms have a general form, not directed towards particular individuals. Most legal norms that refer to particular individuals function as applications of more general norms to particular instances or are limited and guided by general norms that set the standards or rules for the creation of the particular norm. Legal judgments, administrative decrees, executive orders, or military commands are often norms that expressly apply to particular individuals (for example, “Q must pay P damages in the sum of $1.”). A parMarmor, supra note 24, at 9. It is generally accepted that rights are related to duties. See, e.g., JOSEPH RAZ, ETHICS IN THE PUBLIC DOMAIN 254–60 (1994) [hereinafter RAZ, ETHICS]. In Hohfeldian terms such rights are labeled “claim-rights,” always correlating to a duty in another (or others). See Wesley Newcomb Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning, 26 YALE L.J. 710, 717 (1917). There is a debate as to the primacy of rights over duties or of duties over rights. Some claim that duties ground rights, see, e.g., id. at 740, while others claim that rights give rise to duties, see, e.g., JOSEPH RAZ, THE CONCEPT OF A LEGAL SYSTEM 225–227 (2d ed. 1980) [hereinafter RAZ, LEGAL SYSTEM]; Neil MacCormick, Rights in Legislation, in LAW, MORALITY, AND SOCIETY: ESSAYS IN HONOUR OF H.L.A. HART 189, 189–209 (P.M.S. Hacker & J. Raz eds., 1977). For a discussion see Kamm, supra note 20, at 480–82. Here I assume the latter position, although nothing here turns on this choice. 27 A norm may also exhibit degrees in the generality or specificity of its norm-act. Marmor, supra note 24, at 9. I do not explore this aspect of norms here. 25 26 2009] The Identity and (Legal) Rights of Future Generations 1191 ticular norm may also function as a general norm through normcreating doctrines, such as stare decisis. Legal systems better succeed in adhering to the ideal of the rule of law when mostly comprised and controlled by general norms rather than by particularistic ones. The ideal of the rule of law is best understood to prescribe that people should be ruled by law and that the law should be such that it could guide people.28 As such, the ideal of the rule of law is best realized when a legal system is most able to regulate conduct and guide behavior—its primary function.29 Particular legal norms make it harder for people to guide their behavior in light of the law because such norms are directed at particular instances, and offer no determinative guidance and often little predictive guidance as to how other cases, even similar ones, would be regulated. Therefore, general legal norms make for a functionally better legal system just as a sharper knife makes for a better knife. 2. The Rights of Future People: General Norms and Types of People That a norm does not address any particular individual but a type of person normally raises no difficulty. The duty or right conferred by a norm-act vests in any particular instance or token of the general norm-subject type—in other words, in any individual by whom the characteristics making up the norm-subject are instantiated. For example, if a regulation confers a duty on those making over a million dollars a year to pay 50% of their income as tax, then that duty only applies to those individuals who actually make over a million a year. Until some actual individual makes a million a year, the norm applies to no one in particular, only to an empty set or type of person. The same is also true for rights-conferring norms. When such norms have a general norm-subject, the right only vests once it is attached to some particular subject that meets the criteria of the normsubject. For example, one only has disability rights if one is disabled according to the legal definition of disability. In a world where no particular person is disabled, the rights of the disabled do not vest in any particular person. As things are, there are actual disabled people in each generation, and it is in them that disability rights vest. If no one were disabled, these rights would have vested in no one. JOSEPH RAZ, THE AUTHORITY OF LAW 213 (1979). Id. at 214; Marmor, supra note 24, at 5; see also LON L. FULLER, THE MORALITY OF LAW (1965), reprinted in ESSAYS IN JURISPRUDENCE AND PHILOSOPHY 343, 349–51 (H.L.A. Hart ed., 1983). 28 29 1192 The George Washington Law Review [Vol. 77:1173 Therefore, while legal norms should mostly take a general form, i.e., be directed at types of people, it seems that the duties and rights conferred by such norms only vest in particular individuals that actually meet the criteria of the general norm-subject type. Thus, even if a norm has a general form, its norm-act does not vest in the type making up its norm-subject but only in the particular actual instances of that general norm-subject, which are the tokens of the norm-subject type.30 Hence, for such norms to have actual (as opposed to potential) normative force, they must apply to some actual norm-subject.31 This last feature of general norms raises a possible problem for norms purporting to confer rights on future people. Unlike the case of disabled people, there are no particular individual future people— as was argued above, most future individuals are nonexistent. For this reason there are only types of future people. It is also for this reason that norms that confer rights on future people may only take the form of a norm with a general norm-subject, referring not to particular individuals but to types. If the normative force of a general norm, be it a duty- or right-conferring norm, is only realized when the characteristics of its norm-subject instantiate in an actual individual, however, it is not clear how norms may confer rights on future people who are not particular individuals. Seeing that future people are nonexistent, there are no specific, actual, particular, individual “them” for rights to vest in, once again raising the question: how can future people have rights? Therefore, while the preferred structure of legal norms as general norms fits well with the notion of future generations suggested above—as types of people and not particular individuals—it raises a conceptual hurdle for legal norms purporting to confer rights on future people. The problem is exacerbated by the fact that our sense that we should respect the interests of future people is at least one main motivation for our concern for future people and for enacting legal norms that confer duties on contemporaries to promote the wellbeing and welfare of future generations. We often think future people have moral rights that deserve legal recognition (in the form of legal rights for future generations and corresponding legal duties on contemporaries). If future people cannot be the holders of rights (legal or moral), because rights cannot vest in “them,” however, we must not only discard the idea of future people having legal rights, but we also lose what we take as perhaps the main (moral) reason for ascribing 30 31 See, e.g., Macklin, supra note 19, at 152. Id. 2009] The Identity and (Legal) Rights of Future Generations 1193 legal duties (even ones without corresponding rights)—to protect and even better the lot of future generations. The leading approach in the philosophical literature to resolving this puzzle does not reject the notion that rights only vest in particulars but advocates acknowledging that, when prescribed by general norms, rights vest in particulars as tokens of the type that such rights vest in. Rahul Kumar, addressing this very puzzle as it relates to moral rights, argues that moral rights are ascribed by general principles—conferring moral rights to types of rights-holders, not to particular individuals.32 In fact, it is often thought that moral principles are necessarily general or universalizable33 (which is Kumar’s assumption as well). In other words, that person Q1 has a moral right to y entails, or rather derives from, the fact that for every Qx it is true that Qx has a right to y. The question we are concerned with here is whether general norms can confer legal rights on future people. According to the structure of general norms, as laid out above, it seems that they can. Whatever the validity of Kumar’s assumption as to the logic of morality and the nature of moral norms as general or particular (a contested matter), legal norms may be tailored in either of the two forms.34 After all, the generality or particularity of legal norms is subject to the norm’s source. For example, where a legislature is considered (according to a rule of recognition)35 the source of legal norms, the structure of such norms is subject to the “will” of that legislature. As we saw, the logic of general norms entails that general rights-conferring norms do not confer rights on types in and of themselves—after all, a type is merely a category or a set of (at times normatively significant) characteristics. In addition, general rights-conferring norms do not confer rights on individuals as particulars. Rather, such norms confer rights on particular individuals as tokens of a general type. Another way of putting this point is that general norms do not apply to individuals because of who they are (in terms of having a specific personal identity) but as a function of what they are (tokens of a type). We respect the rights of certain individuals because they are of the type of person the right-conferring norm protects, and we do so regardless of 32 Kumar, supra note 22, at 111; see also Baier, supra note 13, at 173; Ernest Partridge, On the Rights of Future Generations, in UPSTREAM/DOWNSTREAM: ISSUES IN ENVIRONMENTAL ETHICS 40, 58 (Donald Scherer ed., 1990); Pletcher, supra note 19, at 168. 33 Ci Jiwei, The Logic of the Generality of Moral Principles, 38 PHIL. Q. 305, 305 (1987). 34 See VON WRIGHT, supra note 23, at 6–8. 35 H.L.A. HART, THE CONCEPT OF LAW 100–10 (2d ed. 1994). 1194 The George Washington Law Review [Vol. 77:1173 the specific personal identity and individuality of those individuals. What matters for the vesting of the right is that the rightholder, whoever she is, exhibits the attributes that make up the norm-subject. Applying the logic of general norms, the proponents of the typebased approach to the rights of future people hold that the fact that future person Q is nonexistent or, in other words, does not have an individual identity, does not entail that we cannot wrong Q or that Q cannot have rights—violating the rights of a type wrongs a particular individual once these rights vest in that actual individual, whoever “he” or “she” may turn out to be. When this occurs, the particular individual is wronged as the type of person he or she is, regardless of his or her individuality and particularity. In fact, according to this approach the particularity of the rightholder plays no role in the application of general norms.36 Jeffrey Reiman’s account of the rights of future people is particularly helpful in trying to explain this notion that people are wronged as tokens of types of people, regardless of their individual identity or particularity.37 Reiman distinguishes between a person as a “particular” and a person’s properties, explaining that even two people who are completely identical in both their “personal properties”38 and their “worldly properties”39 are still distinct, because they are still two different particulars.40 In technical terms, the two particular individuals are numerically different.41 Future people are not particular individuals—in Reiman’s terms “they” only have properties.42 For example, “they” may live in a polluted world or a well-preserved world; “they” may have congenital deformities or be born healthy. We can only refer to future people in terms of “their” properties (not “their” particularity) and, as such, we can respect the rights of future people by manipulating “their” properties. For example, preserving the environment would improve the worldly properties of future people and depleting the earth’s resources would make “them” worse off— regardless of exactly who “they” will turn out to be as individuals. Therefore, even though future people do not exist as particulars, we 36 This type of reasoning is second nature for the jurist. It is logic not unlike the logic applied in discerning whether a precedent applies to a subsequent case. 37 Reiman, supra note 22, at 83–86. 38 “Personal properties” are properties of a person. 39 One’s “worldly properties” are properties made up of the surrounding circumstances in which one exists and which pertain to one. 40 Reiman, supra note 22, at 83. 41 Id. 42 Id. 2009] The Identity and (Legal) Rights of Future Generations 1195 may affect “them” by impacting “their” personal properties and “their” worldly properties as tokens of types.43 Naturally, so long as the attributes of the type are not instantiated by an actual particular individual, making the type worse off wrongs no one. It can only wrong the actual future person who comes into existence as a token of that type; however, and this is key, the particular individual who actually ends up wronged is not wronged as the particular person he or she is. In fact, according to the type-based approach, particularity does not matter at all since rights are not conferred on particulars. To conclude, the type-based approach entails that future people can presently have (legal) rights even though future people do not presently exist. Moreover, the line of thought according to which future generations must be approached as types of people via general norms fits well with the ideal of the rule of law, which prefers general norms, as well as with the structure of rights-conferring legal norms. However, although the type-based approach to explaining how future people can have rights (even though “they” do not exist) is the leading resolution to this puzzle and fits nicely into the jurisprudential concept of “legal rights,” I remain skeptical as to its validity. The type-based approach relies on a suspect distinction between an individual as a particular and his or her properties (the latter supposedly being determinative of the type that person is a token of). This distinction assumes that there is a sense in which individual Q is Q beyond the properties of Q that make up Q’s identity. Consider Reiman’s position that “[e]ven if two twins were truly identical, having exactly all the same properties, they would still be two different particulars. They are, as philosophers sometimes say, numerically different.”44 The assumption here is that what makes the two identical individuals distinct from each other is not their properties—after all, they are identical45—but something we may call their “particularity.” The notion that there is more to a person’s identity than his or her properties (worldly, personal, or relational) stands in contrast to the main trend of the modern philosophy of personal identity. Going all the way back at least to John Locke,46 this tradition has viewed personal identity as determined by properties and not by some etheFor a similar view, see Hare, supra note 22. Reiman, supra note 22, at 83. 45 Discounting for the twins’ relational properties. 46 See JOHN LOCKE, AN ESSAY CONCERNING HUMAN UNDERSTANDING 241–57 (Prometheus Books 1995) (1694). 43 44 1196 The George Washington Law Review [Vol. 77:1173 real substance or soul.47 Further, the type-based solution relies on the notion that we can attribute a right to a token of a type without also attributing the same right to an actual particular individual, a distinction that seems to fall apart if we reject the notion, as many have, that there is something constitutive of individuals beyond their properties. Therefore, while intuitively appealing, especially for those whose intuitions are formed by the law (which functions on categories and types), the types-based approach to the rights of future people stands on shaky metaphysical grounds. B. The Nonidentity Problem and the Rights of Future Generations The second puzzle concerning the notion of the rights of future people derives from what has come to be known as the “Nonidentity Argument.” Briefly, the nonidentity argument concludes that future people cannot be made worse off by (most) acts and choices that occurred prior to their birth, so long as those particular people would have never been born (or conceived) had those events not occurred.48 This casts doubt over whether people have any rights in relation to prenatal events that are determinative of their existence. Therefore, putting aside the problem of the nonexistence of future people, which is addressed above, and assuming that future people can be attributed with interests and rights, there is further reason to doubt whether future people in fact have any rights. First, this Section explains the function of rights as it is captured by the two leading theories of rights—the Interest Theory and the Will Theory. The Section next introduces the nonidentity problem and explains how, according to both rights theories, it suggests that future people have no rights. Then, it analyzes the conception of harm assumed by the nonidentity argument. Finally, two alternative conceptions of harm people may suffer (the nonidentity argument notwithstanding) are introduced; conceptions of harms that may, after all, allow for grounding certain rights in future people. 1. The Function of Rights—The Will and the Interest Theories There are two competing theories on the function of rights, explaining what rights do for the rightholder. According to the Will (or Choice) Theory of rights, rights protect the rightholder’s exercise of choice.49 Divided according to Hofeldian terms, a holder of a claim47 48 49 See generally sources cited supra note 11. See PARFIT, supra note 11, at 355–56. For accounts of the Will Theory, see H.L.A. Hart, Legal Rights, in ESSAYS ON BEN- 2009] The Identity and (Legal) Rights of Future Generations 1197 right has a power over the duties of others—she may waive another’s duty or the remedy owed for breaching the duty, hold another to his duty, or make a claim against another based on his failure to meet his duties.50 A holder of a privilege- (or liberty-) right is under no duty not to y and therefore is free to choose y.51 A holder of a power-right is at liberty to choose to create or change the existing normative arrangement and consequentially mold the rights and duties of others.52 Finally, an immunity-right entails that others do not have the power to manipulate (some of) one’s rights and duties.53 All these types or aspects of rights protect the choice of the rightholder. Under the Interest Theory of rights, person Q has a right to y if Q’s interest in y is sufficient reason for holding others to be under a duty to y.54 Thus, a right furthers the interests of its holder and only vests in a rightholder when there are sufficient reasons for ascribing others with duties to further or respect those interests.55 Although earlier in the discussion the Interest Theory was assumed to be the background theory to various propositions regarding rights, here I do not argue for or against either of the two theories.56 THAM: STUDIES IN JURISPRUDENCE AND POLITICAL THEORY 162, 183–84 (1982) [hereinafter Hart, Legal Rights]; MacCormick, supra note 26, at 192; Alon Harel, Theories of Rights, in THE BLACKWELL GUIDE TO THE PHILOSOPHY OF LAW AND LEGAL THEORY 191, 194 (Martin P. Golding & William A. Edmundson eds., 2005); NEIL MACCORMICK, H.L.A. HART 115 (2d ed. 2008); H.L.A. Hart, Are There Any Natural Rights?, 64 PHIL. REV. 175, 175–91 (1955), reprinted in THEORIES OF RIGHTS 77–90 (Jeremy Waldron ed., 1984). 50 Harel, supra note 49, at 192. 51 For an account of Hofeld’s theory, see JOEL FEINBERG, SOCIAL PHILOSOPHY 55–67 (1973); see also Harel, supra note 49, at 192–93; MacCormick, supra note 26, at 193. 52 Hofeld defines the state of having a power as a right to create or change the claims, privileges, powers, and immunities of others (as well as one’s own). Harel, supra note 49, at 193. 53 Id. 54 RAZ, supra note 8, at 166. For discussions of the Interest Theory, see MacCormick, supra note 26, at 192; Harel, supra note 49, at 195. 55 The Interest and the Will Theories of rights purport to capture the structure and role of rights, not the grounds justifying rights. The grounds justifying a right under the Interest Theory must give sufficient reason for holding others to the duty to further the rightholder’s relevant interests. RAZ, ETHICS, supra note 26, at 243. The grounds justifying a right under the Will Theory must give sufficient reasons for giving one control over the duties and rights of others. Kamm, supra note 20, at 481. In morality, grounds are the purview of normative ethics, which in the context of grounds for rights is often divided into deontological, consequential, contractual, and natural rights-based reasons. See Harel, supra note 49, at 198. In the law, grounds are found in authority. Therefore, although the function, structure, and logic of legal rights is similar to that of rights in general—including moral rights—legal rights and moral rights do not share the same grounds. Moral rights have moral grounds while legal rights spring from legal sources. See RAZ, ETHICS, supra note 26, at 238–60. My concern here is with the legal rights of future people. 56 For a discussion on the merits of both rights theories, see Harel, supra note 49, at 193–97; Kamm, supra note 20, at 481–95; MacCormick, supra note 26, at 189–209. 1198 The George Washington Law Review [Vol. 77:1173 Rather, I ask what implications the nonidentity argument has on the rights of future people according to each one of the two theories. 2. The Nonidentity Problem The nonidentity argument denies that an act, event, or choice that is a necessary part of the causal chain leading up to a person’s birth (or conception) can make that same person worse off.57 It is a fact that actions, choices, and events can determine the identity of the individuals who will be born in the future. Many actions and choices have this effect, entailing that had people acted or chosen differently history would have taken a different course, subsequently causing the birth of different individuals, born to different parents, with different DNA, personal histories, memories, etc.58 Such identity-determinative choices and actions or events have been labeled “different-people choices”59 (or actions). According to the nonidentity argument “different-people choices” (or actions) cannot harm the individuals whose existence is determined by those very events, so long as those individuals’ lives are worth living. Had the choice not been made the individuals at stake would have never been born and, if their lives are worth living, they are better off existing having the life they have than never having been born at all. The opposite is true for people who do not have a life worth living, i.e., they are made worse off by being brought into existence.60 PARFIT, supra note 11, at 359. The multiplicity of identity-determinative actions, events, and choices derives from what Gregory Kavka calls the “precariousness” of existence. Gregory S. Kavka, The Paradox of Future Individuals, 11 PHIL. & PUB. AFF. 93, 93 (1982). It derives from the fact that practically all events have a growing ripple effect that determines the identity of more and more future people as times goes by. Id. at 93–94; PARFIT, supra note 11, at 351–52; ROBERTS, CHILD, supra note 10, at 89–90; Christopher W. Morris, Existential Limits to the Rectification of Past Wrongs, 21 AM. PHIL. Q. 175, 176–77 (1984); Jonathan Glover, Future People, Disability, and Screening, in BIOETHICS 429, 429–31 (John Harris ed., 2001); Matthew Hanser, Harming Future People, 19 PHIL. & PUB. AFF 47, 47 (1990); Alan Carter, Can We Harm Future People?, 10 ENVTL. VALUES 429, 430 (2001). 59 PARFIT, supra note 11, at 355–56. 60 It can also be argued, in contrast, that according to the nonidentity argument one categorically cannot ever be made either better or worse off by being brought into existence. This position turns on accepting that nonexistence has no value for one, and therefore existence does not make one better or worse off. In line with this approach, some argue that being alive or existing is a precondition for person-relative value, see JOSEPH RAZ, VALUE, RESPECT, AND ATTACHMENT 77–123 (2001), or that person-relative value cannot be attributed to “nonexistent people,” see Ori Jonathan Herstein, The Non-Identity Problem: Harm, Historic Injustice and the Value of Nonexistence (2008) (unpublished dissertation, Columbia University) (on file with author). According to this line of reasoning, the notion that people can be made better or worse 57 58 2009] The Identity and (Legal) Rights of Future Generations 1199 In denying that future people can be made worse off by identitydeterminative actions, choices, and events that took place in their prenatal past, the nonidentity argument entails that many of the events, acts, and choices we ordinarily categorize as wrong, because they seem to harm future people unjustly, are either permissible or forbidden for reasons that are not person-affecting.61 Hence, if we have duties not to perform such acts (such as depleting energy resources, causing pollution, and perpetuating historic injustices), these duties do not derive from the danger of wronging (or otherwise adversely affecting) those future individuals whose identity is determined by such acts.62 While highly abstract, the nonidentity argument has generated extensive debate in the fields of practical philosophy and legal theory, including discussions on the rights of future people, which is the focus of the remainder of this Article.63 off by being brought into existence is a muddle, because there is no state or condition of personal value, of any degree or polarity, for “nonexistent people.” It follows that an act that brought a person into existence did not make that person better or worse off than he or she would have been otherwise, because had the act not taken place he or she would never have existed, and as a consequence “he or she” would have had no state of well-being of any kind. According to this argument, as long as one accepts that not being born (or conceived) means never existing, people are never harmed nor benefited by identity-determinative acts. In this approach, prenatal identity-determinative acts are merely necessary stages in the causal chain leading up to the existence of those individuals. From the point of view of a future individual, such acts are always neither harmful nor beneficial, regardless of whether the lives of such individuals are worth living. Accepting this second reading of the nonidentity argument does not diminish the problem it poses for normative theory, because it entails that identity-determinative acts never harm future individuals, even when the lives of those individuals are not worth living. 61 PARFIT, supra note 11, at 378. 62 See id. 63 The nonidentity argument has been applied to a large variety of scenarios involving, for example, reproduction, intergenerational justice, human cloning, and fiscal policy. All these cases involve acts or choices that create or determine the identity of people and are, therefore, deemed not harmful to those people who are created as a consequence of such actions and choices (even though those acts appear to harm or to be able to harm the individuals they create). The extensive literature on the nonidentity problem demonstrates how this problem raises important questions and difficulties not only in ethical theory but also in specific practical contexts of policymaking and justice. As far as legal theory is concerned, the nonidentity problem has often been discussed in the context of “wrongful life” torts. In such cases, a baby is born with a congenital illness, disability, or deformity. The defendant is usually a medical caregiver who negligently failed to discover the risk of the congenital disability or illness. Had the risk been revealed in time, the parents would have aborted the pregnancy. For an excellent survey of the approaches to wrongful life in various jurisdictions, see Ronen Perry, It’s a Wonderful Life, 93 CORNELL L. REV. 329, 329–44 (2008). Suits of this type have been brought before numerous courts in multiple jurisdictions, and have almost always been rejected for reasons of nonidentity. Id. Common-law and civil-law jurisdictions that have accepted, to varying degrees, wrongful life torts are in the clear minority; 1200 The George Washington Law Review [Vol. 77:1173 3. The Nonidentity Problem and the Rights of Future People Rights have the quality of being the rights of someone (a rightholder)—they are not free-floating. Accordingly, the duties that correspond to rights are directed duties, i.e., duties towards someone or something (including, at least in part, the rightholder). In addition, they include California, Connecticut, Maine, New Jersey, Washington, Israel, and The Netherlands. Id. at 337–38, 340, 344. Aside from the wrongful life context, the nonidentity problem has also been discussed in other situations pertaining to reproduction. See, e.g., Eric Rakowski, Who Should Pay for Bad Genes?, 90 CAL. L. REV. 345, 345 (2002). The implications of the nonidentity problem have also been extensively evaluated in the field of bioethics, especially in the context of a variety of assisted reproductive technologies. See, e.g., Philip G. Peters, Jr., Harming Future Persons: Obligations to the Children of Reproductive Technology, 8 S. CAL. INTERDISC. L.J. 375, 375 (1999); Richard A. Posner, The Ethics and Economics of Enforcing Contracts of Surrogate Motherhood, 5 J. CONTEMP. HEALTH L. & POL’Y 21, 23–24 (1989); M.A. Roberts, Cloning and Harming: Children, Future Persons, and the “Best Interest” Test, 13 NOTRE DAME J.L. ETHICS & PUB. POL’Y 37, 57–61 (1999); Roberts, Present Duties, supra note 10, at 298–303; M.A. Roberts, Supernumerary Pregnancy, Collective Harm, and Two Forms of the Nonidentity Problem, 34 J.L. MED. & ETHICS 776, 776 (2006); John A. Robertson, Liberty, Identity, and Human Cloning, 76 TEX. L. REV. 1371, 1411–18 (1998); John A. Robertson, Procreative Liberty and Harm to Offspring in Assisted Reproduction, 30 AM. J.L. & MED. 7, 13–19 (2004); Justin Trent, Assisted Reproductive Technologies, 7 GEO. J. GENDER & L. 1143, 1161–62 (2006). The problematic implications of the nonidentity argument also appear to extend to issues of intergenerational justice and social policy. For example, Neil Buchanan has recently explored the implications of the nonidentity argument for questions of fiscal policy. Neil H. Buchanan, What Do We Owe Future Generations?, 77 GEO. WASH. L. REV. 1237, 1270–73 (arguing that, considering that in all likelihood the quality of life of future generations will increase, current fiscal policies may in fact be benefiting future generations supererogatorily). The nonidentity problem has also been applied to population ethics, see PARFIT, supra note 11, at 371–77; Derek Parfit, On Doing the Best for Our Children, in ETHICS AND POPULATION (Michael D. Bayles ed., 1976); Thomas Schwartz, Obligations to Posterity, in OBLIGATIONS TO FUTURE GENERATIONS 3, 3–13 (R.I. Sikora & Brian Barry eds., 1978), as well as to environmental questions, see Clayton Hubin, Justice and Future Generations, 6 PHIL. & PUB. AFF. 70, 70–83 (1976). The nonidentity argument can also be easily applied to many other issues of social policy, such as energy conservation, maintaining future generations’ Social Security, etc. The nonidentity problem has also been applied in the historic justice context. See, e.g., Ori J. Herstein, Historic Injustice, Group Membership and Harm to Individuals: Defending Claims for Historic Justice from the Non-Identity Problem, 25 HARV. BLACKLETTER L.J. (forthcoming 2009), available at http://ssrn.com/abstract=1108130; Leo Katz, What to Compensate? Some Surprisingly Unappreciated Reasons Why the Problem Is So Hard, 40 SAN DIEGO L. REV. 1345, 1362–64 (2003); Rahul Kumar & David Silver, The Legacy of Injustice: Wronging the Future, Responsibility for the Past, in INJUSTICE IN TIME: RESPONDING TO HISTORICAL INJUSTICE 145–58 (Lukas H. Meyer ed., 2004); Lukas H. Meyer, Historical Injustice and the Right of Return, 5 THEORETICAL INQUIRIES IN L. 305, 306–11 (2004) [hereinafter Meyer, Historical Injustice]; Lukas H. Meyer, Past and Future: The Case for a Threshold Notion of Harm, in RIGHTS, CULTURE, AND THE LAW: THEMES FROM THE LEGAL AND POLITICAL PHILOSOPHY OF JOSEPH RAZ 143, 149–57 (Lukas H. Meyer, Stanley L. Paulson & Thomas W. Pogge eds., 2003) [hereinafter Meyer, Past and Future]; Ernest Partridge, The Future—For Better or Worse, 11 ENVTL. VALUES 75, 81–82 (2002); Amy J. Sepinwall, Responsibility for Historical Injustices: Reconceiving the Case for Reparations, 22 J.L. & POL. 183, 216–18 (2006). 2009] The Identity and (Legal) Rights of Future Generations 1201 rights are for the benefit of someone (and at least partially are for the benefit of the rightholder). The two competing theories of rights attempt to capture the role of rights in practical reasoning and explain in what sense rights are for the benefit of the rightholder, protecting either her interests or her choices. a. The Will Theory of Rights, the Nonidentity Problem, and the Rights of Future People The Will (or Choice) Theory of rights contends that rights protect the exercise of choice.64 In this theory, it seems that if future people had a right to bar certain identity-determinative prenatal actions they would nevertheless always waive such rights and forgo the corresponding duties of their predecessors. The reason is that future people (whose life is worth living) would prefer to be born and suffer the consequences of the past (such as living in a highly polluted world) rather than never be born at all.65 Even though future people cannot actively waive their rights, considering that they do not yet exist (which is obviously a precondition for having a will and for making choices), presumably any rational person in such circumstances would choose to waive such rights.66 Therefore, future people have no rights. This rejection of the rights of future people is not persuasive under the Will Theory of rights. It is not clear whether a waiver given under the threat of nonexistence is valid, especially under a theory that purports to protect choice. A choice made under threat is suspect as coercive or extortionist.67 In a sense, it is no choice at all. If the function of rights is indeed to protect choice and the free exercise of will, a coerced waiver of a right is most likely void of normative force. Nevertheless, there is another, deeper obstacle in the way of the notion that future people have rights—it is not clear how one can choose not to exist. The act or choice over which such a right could be exercised is always already made when one comes into being. If Q’s 64 See HANS KELSEN, INTRODUCTION TO THE PROBLEMS OF LEGAL THEORY 44–45 (Bonnie Litschewski Paulson & Stanley L. Paulson trans., Oxford: Clarendon Press 1992) (1934); Harel, supra note 49, at 194; Hart, Legal Rights, supra note 49, at 183–85; MacCormick, supra note 26, at 192. 65 PARFIT, supra note 11, at 364–66. 66 Notice that this position is not susceptible to the cultural critique that what constitutes a life worth living is a cultural term, because the requirement here is that any rational person would rather have a life worth living than not—not what constitutes a life worth living, which is to a degree culturally determined. 67 See Jefferson McMahan, Problems of Population Theory, 92 ETHICS 96, 124–27 (1981) (criticizing the concept of an individual waiving his prior right because he will be glad he was conceived). 1202 The George Washington Law Review [Vol. 77:1173 having a choice is predicated on Q’s existence then Q cannot ever have a choice as to his or her own existence.68 This entails that future people have no rights in cases of nonidentity, because future people cannot have a choice as to whether a prenatal event occurs or not. Because rights protect choice and in said circumstances there can be no choice, it follows that there can also be no rights. It is possible to speculate whether people, as rational beings, would choose to exist or not under various harmful circumstances. Beyond the problem of coercion, however, I find basing or ruling out a right of a future person on a hypothetical choice less than entirely convincing, especially according to the Will Theory of rights. Hypothetical choices may have normative force in justificatory schemes that are primarily motivated by fairness and formal justice as underlying values.69 Such choices may also be compelling when speculating about what an actual particular person would have wanted. Considering that the Will Theory of rights assumes that rights are for the protection of choices and the exercise of will, however, it is not at all obvious that a hypothetical choice made by a “hypothetical person,” which is a pure abstraction, can waive rights. What will is protected under such circumstances? Proponents of a Will or Choice Theory of rights tend to justify actual rights based on basic values such as individuality, selfdetermination, autonomy, and personhood,70 none of which future people possess. Future people have no personhood, individuality, self, or autonomy, and lack the very capacity to make choices. Valuing choice entails valuing not only that a choice is made, or even that the right choice is made, but also that the choice is a function of the free exercise of will, i.e., it is an outcome of choosing. In the case of future people we have no choice but to assume a “hypothetical choice” to waive one’s rights, which does not seem to have much to do with the above-mentioned underlying values. Therefore, assuming that rights protect the choices and the free exercise of the rightholder’s will, there is good reason to doubt whether future people ever have rights. 68 One should not confuse the choice that some people make to stop living with the muddled notion of having a choice whether or not to exist. 69 See JOHN RAWLS, A THEORY 70 Harel, supra note 49, at 194. OF JUSTICE 10–14 (rev. ed. 2003). 2009] The Identity and (Legal) Rights of Future Generations 1203 b. The Interest Theory of Rights, the Nonidentity Problem, and the Rights of Future People As explained above, the Interest Theory of rights entails that person Q has a right to y if Q’s interest in y is a sufficient reason for holding others to be under a duty.71 As such, rights protect and promote the rightholder’s interests, aspiring to make the rightholder better off and to protect her from being made worse off. When considering the effect that different-people choices have on the interests of a future person, the focus must be on the value that person’s existence has for that person. This is because different-people choices are identity-determinative; therefore, when assessing whether such prenatal choices, acts, and events make one better or worse off, the relevant point of reference is the value one’s very existence has for one and not how such acts, choices, and events impact some discrete aspect of one’s life. According to the nonidentity argument future people cannot ever be made worse off by acts that are necessary, causal conditions for their having been brought into existence (assuming that such people have a life worth living). Under these circumstances, such prenatal events never harm one. In fact, it is possible to make a cogent argument that one is made better off by events without which one would have never existed—because one is better off existing than not existing. It follows that in nonidentity cases future people have no interest in the conferring of a duty on others not to perform the acts without which those particular future people would never have existed. This may include prenatal actions and choices that are widely thought harmful to future generations and are often considered to be the objects of the rights of future generations. For example, wasteful energy practices, irresponsible fiscal policies, overpopulation, historic injustices, and acts that give rise to certain wrongful-life claims are all determinative of the identity of many future people. If people do not (or rarely) have an interest in not ever having been born (and may even have a positive interest in being born), it follows that because rights protect interests of rightholders, and because having a right is a function of having interests strong enough to justify holding others to a duty, future people do not have rights against the performance of prenatal actions and choices that are determinative of their existence, and currently existing people have no 71 RAZ, supra note 8, at 166. 1204 The George Washington Law Review [Vol. 77:1173 corresponding duty not to perform such actions. It is improper to consider as harmful actions and choices commonly perceived to wrong future generations, because many of these actions, such as energy squandering, irresponsible fiscal policies, or overpopulation, are identity-determinative. A more sophisticated version of the Interest Theory springs from the seemingly puzzling fact that there are many rights that ascribe duties that cannot be justified by the interests of the rightholder alone.72 As pointed out above, an essential feature of rights is that they are for the benefit of the rightholder—they protect or promote his or her interests.73 Nevertheless, this does not entail that a right must be justified solely by the weight or significance the protected interests of the rightholder have for him or her. Often the significance of one’s right is much greater than the importance the particular interests (protected by the rights) have for one. Raz gives the following example: Imagine two people with an equal interest in having [a] shirt. It is clear that the one who owns it should have it. But as their interest gives them equal claim to it, this can only be because the rightholder’s right to the shirt is a reason for giving it to him which is greater than his interest. His right does not merely reflect his interest, it adds to it an additional, independent reason.74 This is puzzling because it indicates that often the stringency and even the very existence of a right are not fully explained by the normative weight of the rightholder’s interests that it protects. This is explained by the fact that rights may be justified not only by the interests of the rightholder but also by the interests of others, when the interests of others are served by the protection and furthering of the interests of the rightholder. It is the combined force of all these interests that justifies the right and gives it its weight and stringency.75 Even this more expansive version of the Interest Theory of rights, however, does not solve the puzzle of the nonidentity argument. The justification of rights remains predicated on the strength or significance of the interests of the rightholder. The fact that occasionally the degree of importance of protecting and furthering the interests of the rightholder partially (and often largely) derives from the weight of the 72 See RAZ, supra note 8, at 44–57 (considering the validity of the imposition of legal duties by authorities to protect certain interests of rightholders). 73 Id. at 44–45. 74 RAZ, ETHICS, supra note 26, at 31. 75 See id. at 35–40. 2009] The Identity and (Legal) Rights of Future Generations 1205 interests of others in the protection of the rightholder’s right-protected interest does not change the fact that in order for Q to have a right to y, Q must have some interest in y. The problem of nonidentity rules out future people having any interest, no matter how minute, in (many) prenatal identity-determinative actions and choices not taking place. To conclude, according to both the Will and Interest Theories of rights, future people (mostly) have no rights in relation to the occurrence of events that predate their birth and are determinative of their identity. This does not necessarily rule out having duties towards the future. A legal system may establish various duties mandating that contemporaries bring about a future populated by prosperous individuals. The reasons for enacting such duties, however, cannot derive from the rights or claims of future people or from the desire to make future particular individuals better off. Moreover, in the absence of a harm (or a benefit) to future individual people it is quite difficult to justify, morally and legally, limiting the resources and liberty of contemporaries. This is especially true for liberal democracies strongly dedicated to the harm principle—basically the notion that liberty should only be curtailed in order to prevent harm76 or to benefit individuals.77 4. Accounts of Possible Harm to Future People The path towards some amelioration of the destructive implications the nonidentity argument has for the notion of the rights of future generations passes through a better understanding of the limitations of the conception of harm assumed by the nonidentity argument. The nonidentity argument holds that harm is comparative and aggregative. This is a valid conception of harm, but it is not exhaustive of all types of harm. For example, the “Threshold Conception of Harm” challenges the assumption that all harm is comparative; the “Constitutive Conception of Harm” challenges the notion that all harm is aggregative. Both these conceptions of harm may serve as a basis for justifying certain rights in future people (under the Interest Theory of rights). These two conceptions of harm are applicable to 76 JOHN STUART MILL, ON LIBERTY 69 (Penguin Classics, 1985) (1859) (“That the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.”); H.L.A. HART, LAW, LIBERTY, AND MORALITY 27–33 (1963). 77 See RAZ, supra note 8, at 400–29 (arguing that Mill’s harm principle needs to be expanded to account for government’s duty to promote the autonomy of people). 1206 The George Washington Law Review [Vol. 77:1173 both the notion of the rights of future people as individual rights or as rights of tokens of types. a. The Conception of Harm Assumed in the Nonidentity Argument (i) The Comparison-Based Conception of Harm The nonidentity argument assumes a conception of harm that assesses whether an act benefited or set back the interests of a person based on whether the act made that person better or worse off.78 This is determined by comparing the well-being of that individual in two states of affairs: one in which the act takes place and one in which it does not.79 In assessing the harmfulness of y to Q we ask whether Q would have been better or worse off had y not taken place.80 If Q is worse off for y taking place, it follows that y harmed Q. The opposite is true for when y benefits Q. (ii) The Aggregative Conception of Harm The nonidentity argument also takes an aggregative approach to harm, calculating the harmful or beneficial impact an event has on one according to the overall effect or impact that event has on one.81 If harm is a setback of interests,82 then, according to the aggregative conception of harm, y harms one if the impact y has on one is, all things considered, negative. Therefore, even if an act happens to further some of one’s interests it is still deemed harmful (rather than beneficial) to one if the event’s negative impact outweighs the event’s beneficial consequences. 78 See Meyer, Past and Future, supra note 63, at 148–49. See George Sher, Compensation and Transworld Personal Identity, 62 MONIST 378, 382–83 (1979). 79 80 Meyer, Past and Future, supra note 63, at 147–48. For more examples of counterfactual approaches to assessing harm comparatively, see, for example, JOEL FEINBERG, THE MORAL LIMITS OF THE CRIMINAL LAW 51–55, 101–02 (1984) [hereinafter FEINBERG, MORAL LIMITS]; Joel Feinberg, Wrongful Life and the Counterfactual Element, in FREEDOM AND FULFILLMENT: PHILOSOPHICAL ESSAYS 3, 6–7 (1992) [hereinafter FEINBERG, Wrongful Life]; RAZ, supra note 8, at 414. 81 Feinberg, Wrongful Life, supra note 80, at 4–5. See id. at 31–36 (discussing the potential enlargement of the definition of harm to include a setback of interests); see also FEINBERG, supra note 51, at 26. 82 2009] The Identity and (Legal) Rights of Future Generations 1207 b. A Threshold Conception of Harm—A Non-Comparative Approach to Harm Some writers hold that certain states of affairs are harmful to people regardless of whether those individuals were better or worse off by being brought into such a state.83 In this view, causing someone to exist in such a state is per se harmful to that person. This type of approach requires setting a threshold delineating a minimum level of personal well-being or welfare below which a person is deemed harmed. In the philosophical literature on the nonidentity problem some (for example Lukas Meyer) have advocated this approach.84 In the legal sphere, a good example of this approach is found in an influential opinion of Justice Aharon Barak of the Israeli Supreme Court.85 Meyer suggests what he calls a “subjunctive-threshold interpretation” of harm, which entails that “an action (or inaction) at time T1 harms someone only if the agent thereby causes (allows) this person’s life to fall below some specified threshold.”86 In this view, if a past event causes one to have a life that falls below this threshold, then it entails that the future person was harmed by this event, even if that individual’s life is worth living and she never would have existed had the harmful event never taken place. Elizabeth Harman also seems to have this type of conception of harm in mind in stipulating that “[a]n action harms a person if the action causes pain, early death, bodily damage, or deformity to her, even if she would not have existed if the action had not been performed.”87 Haavi Morreim claims that certain harms reside in a substandard state of affairs itself, not deriving from how that state of affairs is evaluated in comparison with how things could have been.88 83 See, e.g., Jeff McMahan, Wrongful Life: Paradoxes in the Morality of Causing People to Exist, in RATIONAL COMMITMENT AND SOCIAL JUSTICE 208 (Jules L. Coleman & Christopher W. Morris eds., 1998); Meyer, Historical Injustice, supra note 63. 84 Meyer, Historical Injustice, supra note 63, at 308. 85 CA 512/81 Zeitsov v. Katz [1986] IsrSC 40(2) 85. See generally Perry, supra note 63, at 369–76 (discussing the opinions in Zeitsov). 86 Meyer, Historical Injustice, supra note 63, at 308; see also Lukas Meyer, Intergenerational Justice, in THE STANFORD ENCYCLOPEDIA OF PHILOSOPHY (Edward N. Zalta ed., 2008), http://plato.stanford.edu/archives/spr2008/entries/justice-intergenerational [hereinafter Meyer, Intergenerational Justice]. For other discussions of the threshold account, see McMahan, supra note 83; Seana Valentine Shiffrin, Wrongful Life, Procreative Responsibility, and the Significance of Harm, 5 LEGAL THEORY 117, 123–24 (1999). 87 Elizabeth Harman, Can We Harm and Benefit in Creating?, 18 PHIL. PERSP. 89, 93 (2004). 88 E. Haavi Morreim, The Concept of Harm Reconceived: A Different Look at Wrongful Life, 7 LAW & PHIL. 3, 23–24 (1988) (suggesting setting the threshold for harm according to a 1208 The George Washington Law Review [Vol. 77:1173 Joel Feinberg suggests an ad hoc revision of the comparative conception of harm to account for the possibility of certain harms involved in nonidentity cases.89 What these various thresholds require varies from one account to the other. One may take a universal approach, delineating a strict minimum standard that applies to all humans.90 Another approach, which I find more sensible, is context sensitive, taking into consideration how in different communities and eras different goods and capacities are required for a minimally good life. Considering the changes societies go through over time, a threshold dependent upon social context limits our ability to foresee what the threshold would require for far-removed future generations.91 This is true even if the grounds for determining the changing standards are universal and stable, focusing on values such as “a decent life,” “a life worth living,” an “autonomous life,” and so on. The threshold approach has its limitations. First, the threshold may be too low. Meyer’s version only recognizes harm to future people whose lives are not worth living,92 which is a relatively low threshold. Another example of such an approach is that of Michael Freeman, who argues (focusing on procreative ethics) that certain children have a right not to be born because their lives would be awful; therefore, there is a corresponding duty on parents not to conceive.93 Freeman’s threshold seems to apply only to cases where people are better off never existing at all, people whose lives will not be worth living.94 Low thresholds of harm leave many deserving cases uncovered. For example, in many cases of historic injustice those demanding justice for harms deriving from historic wrongs often have lives that are still worth living.95 In addition, Meyer’s “subjunctive-threshold interpretation”96 of harm is both over- and underinclusive as a general account of harm. comparison to how things should have been as opposed to how they could have been as a matter of causality). 89 FEINBERG, MORAL LIMITS, supra note 80, at 101. 90 See, e.g., Morreim, supra note 88, at 24. 91 B. Barry, Justice Between Generations, in LAW, MORALITY, AND SOCIETY: ESSAYS IN HONOUR OF H.L.A. HART, supra note 26, at 268, 274–75. 92 See Meyer, Intergenerational Justice, supra note 86. 93 MICHAEL FREEMAN, Do Children Have the Right Not to Be Born?, in THE MORAL STATUS OF CHILDREN 165, 172–84 (1997). 94 Id. at 168–72. 95 Ori J. Herstein, Historic Justice and the Non-Identity Problem: The Limitations of the Subsequent-Wrong Solution and Towards a New Solution, 27 LAW & PHIL. 505, 505 (2008). 96 Meyer, Historical Injustice, supra note 63, at 308. 2009] The Identity and (Legal) Rights of Future Generations 1209 First, it deems harmful certain acts that improve one’s well-being but still fail to lift one above the threshold, even in cases in which it is unclear why improving counts as harming, such as when reaching the threshold is either impossible or requires conduct that seems supererogatory. Second, Meyer’s conception deems choices not to improve another’s well-being as not harmful, so long as these choices promise the affected person a level of well-being higher than the threshold. But there are cases in which not improving, even beyond a threshold, seems harmful—for example, wealthy parents who give their children only the bare minimum. According to Justice Barak, the relevant comparison for assessing the harm involved in a wrongful-life tort, which raises the problem of nonidentity, is not between life and nonexistence, but between the value of the plaintiff’s life with the impairment and without it.97 Morreim offers a more restricted threshold than Barak, pointing to “minimally decent human welfare,”98 which seems to require more than just having a life worth living, especially considering that Morreim counts blindness, deafness and retardation as falling below the threshold.99 Both Barak and Morreim offer a higher threshold than the one implicit in the approaches of Meyer and Freeman, as Barak and Morreim do not merely look to a “life worth living” (Meyer) but to a “normal life” (Barak) and to a life that is not severely impaired (Morreim). The main problem with a more expansive threshold is that it finds harm in situations where we think there is none—for example, in the case of those born with relatively minor negative deviations from the norm (assuming we set the norm as the threshold) or people born into less than ideal circumstances, which nevertheless allow for valuable lives. An alternative is to tailor the threshold not according to any specified minimum but rather to give it a relational form. For example, the threshold may require that no person will have less than ten percent of what others have. Such a threshold would most likely have an egalitarian component built into it. For example, the threshold could require that the higher the average level of well-being, the higher and more demanding the threshold.100 An example of this approach is Ronald Green’s, given in the context of procreation ethics. Green ar97 98 99 100 CA 512/81 Zeitsov v. Katz [1986] IsrSC 40(2) 85, 116. Morreim, supra note 88, at 24. Id. at 28. Meyer, Intergenerational Justice, supra note 86, § 4.1. 1210 The George Washington Law Review [Vol. 77:1173 gues that if a child endures significantly greater suffering or disability than the “reasonably expected health status of others in the child’s birth cohort,” that child is harmed per se.101 The threshold may also be influenced by a nonreduction principle, according to which a current generation harms a future generation if the current generation causes the next generation to have a lower level of welfare or well-being than the current generation had or “inherited.”102 The problem with this sort of approach is that it implies an expectation of human progress or at least of some stability between generations, allowing the previous generation to adhere to a nonreduction principle. This expectation is suspect. Progress and even stability are controversial ideas: aspects of the world sometimes worsen rather than improve, making it impractical and even unfair to require intergenerational nonreduction as a threshold requirement for not harming. In any case, accepting that people have interests in a threshold level of well-being and welfare, and acknowledging that the frustration of these interests is harmful simpliciter (as opposed to comparatively), allows for such interests to ground rights even in cases of nonidentity. These rights would correspond to a duty to secure future people a minimum of welfare and well-being, considerations of nonidentity notwithstanding. c. Non-Aggregative Approaches to Harm The aggregative conception of harm, which is assumed by the nonidentity problem, has also been challenged. There are at least two general types of explanations or accounts for how y can harm Q while still benefiting Q overall or on balance. First, harms and benefits may not be symmetrical; they do not always offset each other. Second, even if harms and benefits are equally functions of furthering and setting back of interests, they may still not always be aggregatable. (i) The Asymmetry Between Harms and Benefits Seana Shiffrin claims that harm and benefit are not symmetrical.103 According to her, harm and benefit are not two poles on a sin101 Ronald M. Green, Parental Autonomy and the Obligation Not to Harm One’s Child Genetically, 25 J.L. MED. & ETHICS 5, 8–10 (1997). 102 See Brian Barry, Sustainability and Intergenerational Justice, in FAIRNESS AND FUTURITY: ESSAYS ON ENVIRONMENTAL SUSTAINABILITY 93, 106–07 (Andrew Dobson ed., 1999). 103 Shiffrin, supra note 86, at 123. Shiffrin argues against what she calls “counterfactual comparative views [of harm].” Id. at 135. In effect, however, she really attacks the notion of 2009] The Identity and (Legal) Rights of Future Generations 1211 gle scale measuring what is good or bad for an interestholder.104 To demonstrate this, Shiffrin offers what she calls the “rescue example”: A saves Q from drowning but has to break Q’s arm in the process. Although it is clear that A benefited Q and Q is better off than he would have been had A not saved him, Shiffrin maintains that Q is still, in a sense, harmed by the rescue in a way that is not offset by its benefits.105 Shiffrin’s main concern lies with explaining why and when the blame for the harmful byproduct of an otherwise beneficial action and the cost of rectifying that harm should be laid at the door of the party that caused it. In doing so Shiffrin shifts the focus of the nonidentity problem from identifying the harm involved in nonidentity cases, which she thinks is not a real problem, to ascribing blame and liability. Shiffrin’s conclusion is that causing lesser harm to one in the process of preventing a greater harm to that same person is not wrongful.106 Causing someone a lesser harm in the process of conferring a greater benefit to him or her, however, without his or her consent, is wrongful.107 It seems that Harman agrees with this line of reasoning, arguing harm as an “all things considered” concept more than attacking the comparative aspect of the definition of harm. See id. at 121, 127–30. 104 Id. at 121. 105 Id. at 125. For more on the concept of “beneficial harm,” see JOHN HARRIS, WONDERWOMAN AND SUPERMAN: THE ETHICS OF HUMAN BIOTECHNOLOGY 92–94 (1992). 106 Shiffrin, supra note 86, at 135. 107 Id. Shiffrin argues that there are some fundamental differences between benefiting and rescuing from harm and that the case of wrongful life, where a person is wrongfully made worse off by being brought into existence, is more analogous to the former than the latter. She does so by posing two cases: the rescue case and a case of supererogatory benefiting without the consent (positive or negative) of the beneficiary, where the benefiting action also causes the beneficiary a lesser setback. See id. at 121–30. In the latter type of cases we feel that, even if one benefits another, if the benefit is bestowed without the consent of that person then the setbacks that person suffers in the process of receiving the benefit are not expunged by the overall benefit, but persist as possibly actionable harm. According to Shiffrin, these types of cases are better analogized to wrongful-life cases than the rescue case, because in bringing one into existence we do not save one from a greater harm (never existing is not a harm) but rather bestow the benefit of existing at the price of a relatively minor harm (this analogy is only valid for people whose life is worth living). Id. at 139. Shiffrin points out that benefiting through harming, without the consent of the benefited/ harmed party, is morally problematic and we often think that the benefactor is liable for the harms caused to the beneficiary regardless of the overall benefit bestowed on her. Id. This is not so in the rescue case, in which one saves another from a greater harm, as opposed to benefiting one. In the rescue case we do not recognize any blame for the lesser setbacks or any corresponding claim against the rescuer for those setbacks. Shiffrin points out some key factors that generate the difference between the two cases. First, she claims that preventing harm is more significant morally than benefiting and the failure to prevent is more objectionable morally than the failure to improve. Id. at 134–35. Therefore, 1212 The George Washington Law Review [Vol. 77:1173 that even though benefits may outweigh harms and make one better off overall, they do not outweigh the harms in the sense of making them permissible—benefiting does not justify harming in the process of benefiting.108 Shiffrin’s account of how harming through benefiting is possible also explains how people may have an interest against the occurrence of acts that bring them into existence, even when those acts benefit them overall. Such interests can, when sufficiently weighty, justify holding others to a duty not to set back those interests, conferring a corresponding right on the interestholder. Under such circumstances, even though the violation of the right makes the future person better off overall (being a necessary condition for his or her existence), it is still adverse to that person’s interests in not being harmed through beneficence. Therefore, Shiffrin argues, one may be wrongly harmed by acts that are necessary conditions for one’s own existence, even if one has a life worth living and was therefore benefited overall by the prenatal acts that determined one’s existence.109 (ii) Non-Aggregatable Interests James Woodward proposes a rights-based answer to the nonidentity problem, according to which one may be wronged via a violation of one’s rights even if the wrongful act benefits one overall.110 For example, not keeping a promise may wrong someone even if keeping the promise would cause the promisee to be worse off. Woodward, in his exchange with Derek Parfit over the nonidentity argument, offers several compelling examples of people suffering wrongs despite not suffering harm.111 One such example is of a person who is denied a one is obliged to prevent more than one is obliged to help, making blaming one for the byproduct of preventing less morally attractive. This is even more so in the case of bringing one into existence, which is a benefit that does not even improve one’s well-being (for nonexistence has no value). Second, she points out that causing harm is worse than not benefiting, which is another reason why the benefactor is more blameworthy for the harmful byproduct of her benefiting actions than the rescuer who tries to prevent harm. Id. at 135. In addition, the absence of consent seems more objectionable in causing harm for a greater benefit than in harming while saving. Id. Therefore, because wrongful-life cases (in which one’s life is worth living) are more like the benefiting case than the rescue case, it seems permissible to lay the blame for the harm caused by existence-determinative acts on those responsible for such acts. Id. 108 Harman, supra note 87, at 100. 109 Shiffrin, supra note 86, at 137–39. 110 James Woodward, The Non-Identity Problem, 96 ETHICS 804, 804 (1986). 111 Id.; James Woodward, Reply to Parfit, 97 ETHICS 800, 800 (1987). For a similar approach, see Carson Strong, Harming by Conceiving: A Review of Misconceptions and a New Analysis, 30 J. MED. & PHIL. 491, 507–508 (2005). 2009] The Identity and (Legal) Rights of Future Generations 1213 plane ticket based on racist reasons. It seems that she is still wronged by the discrimination even if the plane eventually crashes and her life is spared thanks to the airline’s bigotry.112 Another example is of a doctor who amputates one of his patient’s limbs to save the patient’s life. The doctor does not wrong the patient not only because he makes her better off overall but also because the patient agreed or would have agreed to the procedure if asked. Had the patient not agreed and the doctor proceeded with the operation regardless, it is less clear that the fact the patient was made better off overall would negate the wrongness of the doctor’s actions. Woodward explains such lingering evils as a type of rights violation or a harmless wrong.113 Woodward’s account of the origin of such rights contains a nonaggregative approach to harm that is different from the asymmetrical approach suggested by Shiffrin. Woodward believes that the rights-based analysis he applies in the context of a beneficial yet wrongful act also applies to nonidentity cases—an identitydeterminative act may make the violation of the rights of some future person inevitable or highly likely, regardless of whether the future individual would have been worse off had the action never occurred (because he or she would never have existed). According to this approach, what is wrong in such cases is the creation of the future person (a rightholder) under conditions that include the violation of those rights. At least one difficulty with the harmless wrongs approach arises from the very problem we are grappling with here—assuming rights protect the interests of the rightholder, then people who have a life worth living do not have an interest in not being born. Hence, it seems that in cases of nonidentity, no right is abridged because there is no right in the first place. Where there is no harm there can be no rights violation. But there is more to Woodward’s account, because it is premised on an Interest Theory of rights that is coupled with a nonaggregative approach to harm. Woodward contends that some setbacks and advances of one’s interests do not lend themselves to aggregation. He explains: [t]he “space” of people’s interests has, as it were, many dimensions rather than just one and lacks a natural metric. People do not in all circumstances regard a setback to one interest as fully compensatable by a sufficiently large gain to some other interest. Relatedly, people’s choices with respect 112 113 Woodward, supra note 110, at 810–11. See id. at 811–13. 1214 The George Washington Law Review [Vol. 77:1173 to important interests often reflect a satisfying rather than maximizing strategy. . . . To show that an action violates a moral requirement and is wrong, it will often be enough to show that it adversely affects some specific interest protected by a right; that the action affects other interests in a way that is, on balance, beneficial will not automatically cancel or compensate for this violation.114 This suggests that at its core Woodward’s approach is grounded in a notion of harm that may give rise to rights and duties, the nonidentity problem notwithstanding. Yet, Woodward does not explain exactly what the various “dimensions” of interests he has in mind are and when and why they are not open to aggregation. In trying to tie this last knot in the argument, two notions of nonaggregativity come to mind. First, interests may stand for incommensurable values. This would entail that, on occasion, the value the setback of some of one’s interests has for one may be incommensurate with the value the advances of some of one’s other interests have for one. In other words, not all setbacks and advances to one’s interests lend themselves to mutual tradeoffs and aggregation. A second approach may focus on how certain interests may be constitutive of an individual’s identity, and therefore setting them back harms one in a way that is constitutive and not open to aggregation. I have attempted to develop such an approach at considerable length elsewhere.115 Developing the former approach goes beyond the parameters of this Article. Fully developed, both approaches will explain how certain interests of future people can give rise to certain rights for future generations, the nonidentity argument notwithstanding. Therefore, perhaps future people do have certain rights after all. IV. Conclusion Intergenerational justice is peculiar—on a practical everyday plane most people strongly believe posterity deserves our care and consideration. This is particularly apparent during “historic events,” such as the signing of peace treaties, declarations of war (e.g., “the war to end all wars”), enactments of constitutional provisions, and even during presidential elections. If there is one thing on which everyone seems to agree, it is that we owe much to future generations. In contrast, on the theoretical plane, explaining how the most basic prac114 115 Woodward, supra note 111, at 802–03. Herstein, supra note 63. 2009] The Identity and (Legal) Rights of Future Generations 1215 tical concepts, such as “interest,” “harm,” “duty,” “wrong,” and “right” apply to posterity seems almost intractable. The focus of this Article was on the concept of the rights of future people. First, the Article explored to whom or what in the future should rights be ascribed. The conclusion was that types of future people, as opposed to individual future people or future generations as groups, are the most promising candidates. Thereupon, the Article turned to exploring several difficulties with the notion of attributing rights to future people, where “future people” was taken to refer first to types of future people and later to future people as tokens or as individuals. The outcome of this discussion was inconclusive. First, although the type-based approach offers a promising explanation of how future people can have legal rights, the key distinction between a person as a token of a general type and a person as an individual is less than stable. Second, the nonidentity argument proves a formidable foe to the notion that future people have any rights, whether we take the rights of future people to be the rights of future individual people or the rights of future people as tokens of types. Although the Article does manage to mount several responses to this problem, they are only partially developed. In some ways this Article raises as many hurdles in the way of attributing rights to future people as it removes. Nevertheless, beyond suggesting and analyzing possible answers to the problems it raises, this Article has aspired to clarify some of the theoretical hurdles standing in the way of attributing (legal) rights to posterity, hopefully contributing to bridging the gap between theory and what many people hold to be true—that posterity has rights.