NOTE-1304: Intergenerational Justice

Contribution

A map of the philosophical debate over what the living owe future and past people, and an argument through it. It sets out why relations across generations differ from relations among contemporaries. It answers the doubts that future people can have rights. Its main work is to give a notion of harm that survives the non-identity problem, and to specify it. It ends by saying what a rights-based account cannot explain.

Key insight

We can wrong a future person whose existence depends on our choice, but only by a standard that does not compare her with how she would otherwise have been. On a threshold notion of harm, a person is harmed if she is brought into existence, or kept, below a level she is owed; she is then “worse off than she ought to be” (§3.1). That needs no counterfactual self, so the non-identity problem does not touch it. It also has a cost the entry states plainly. A person-affecting account of this kind “may specify no reason for preferring a future with people all of whom have lives far above the level of a sufficiency threshold to a future with no people” (§4.6).

Assumptions

• The Interest Theory of rights, not the Will Theory (§1). On the Will Theory, future people cannot exercise rights against us and so have none.
• A weak person-affecting view (§1). What matters is what is done to a particular person who exists at some time. “People who never exist and people who potentially exist do not count”. A strong person-affecting view would require the same person to be worse off than she would otherwise have been; impersonal views count states of affairs regardless of whose they are.
• Parfit’s genetic identity view frames the contingency (§1): who exists depends on which gametes meet, so most long-term policies change who will live. The European Commission’s estimate that the Erasmus programme produced more than a million babies is the entry’s illustration.
• The threshold is normative, not found (§4, §5).

Key results

• How intergenerational relations differ (§1).

  • No direct reciprocity. The entry notes accounts of indirect and transgenerational reciprocity, Scheffler 2013 among them.
  • A permanent asymmetry of power.
  • Future people’s existence, number and identity depend on our choices.
  • We cannot know future people as individuals.
  • Remote future people cannot affect our well-being, but they may set back the interests of present or past people “insofar as the latter have, or had, interests with respect to posthumous future states of affairs”.
  • Motivation is raised and set aside: compliance with duties to remote future people “is likely to be less reliable” (citing Scheffler 2025, 439–442). So is Gardiner’s “tyranny of the contemporary”.

• Doubts about future rights, answered (§2.1). That future people do not exist now shows they have no rights now, not that we cannot violate their future rights. No one has a right to exist, so we need not ascribe one. Not knowing future people as individuals does not block welfare rights such as subsistence, which rest on their being human (note 1).

• The procreation asymmetry (§2.2). No obligation to procreate for the sake of possible children; an obligation not to beget children whose lives would be miserable. It goes back to Narveson 1967 and 1973. Meyer defends it on the weak person-affecting view with a threshold: the reason “that persons can be worse off than they ought to be” speaks against creating a sub-threshold person, but not for creating an above-threshold one. Impersonal defences are noted: negative consequentialism, or a weak, overridable duty to create happy people.

• Four notions of harm (§§3.1–3.2).

  • (I) Diachronic: worse off at t2 than before t1.
  • (II) Subjunctive-historical: worse off at t2 than if the agent had not acted with respect to her at all.
  • (III) Threshold: caused to exist, or to be, in a sub-threshold state, avoidably and not minimized.
  • (IV) Disjunctive: (II) or (III).

  (I) and (II) cannot reach people whose existence depends on the act. (III) can, but is under-inclusive: stealing a rich man’s car would not harm him. (IV) fixes that.

• Four responses to the non-identity problem (§3.1): deny that such people have rights; rights without harm (Kumar); limit the problem to acts that are necessary conditions of existence; or the threshold conception, which Meyer holds.

• Parfit’s No-Difference View (§3.2). Parfit’s two medical programs (pregnancy testing, which affects actual people, and preconception testing, which affects possible people). The single threshold view is compatible with the No-Difference View. The disjunctive notion, plausibly read, makes cancelling pregnancy testing worse, because two harms apply there and one in the other case.

• Specifying the threshold (§§4.1–4.3).

  • Egalitarian: implies levelling down, and if pegged to the present generation, that it is in one respect worse for successors to be better off.
  • Prioritarian: excessive demands, given how many future people there may be, and no plausible cap on population.
  • Sufficientarian: the Positive Thesis (weighty reasons to secure enough) and the Shift Thesis (a discontinuity in the weight of reasons once enough is secured). Weak sufficientarianism, giving priority but not absolute priority below the threshold, is more plausible across generations. Below the threshold, a priority view.

• The currency (§4.4): basic needs or central capabilities. Not preferences, which future people may not share, and not resources, which track well-being badly.

• Rawls’s just savings (§4.5). Rawls never discusses the non-identity problem and mostly assumes a constant population. Save until just institutions are secure (the accumulation stage), then maintain them (the steady state). A Theory of Justice had the contractors care for their descendants; Political Liberalism drops that assumption for mutual disinterest plus the condition that they must want all previous generations to have followed the principle. Sidgwick, Ramsey, Rawls and Parfit all reject pure time discounting.

• Where rights run out (§4.6).

  • The person-affecting account entails the procreation asymmetry and gives no right to be created.
  • Above-threshold concerns are, at most, weaker justice claims: environmental conditions, cultural goods, ways of life. Hence the admission quoted under Key insight.
  • So “widely shared concerns about the continuation of human life on earth at a high level of well-being” need another ground. Meyer cites McKinnon (2025) on Scheffler’s “afterlife conjecture”: the prospect of extinction “undermines the conditions under which we value, because many of our projects, evaluative commitments, and affective attachments presuppose the continued existence of future people”. Such concerns reflect “our participation in an ongoing chain of generations” and our “evaluative, emotional, and practical dependence on them”.
  • Meyer’s own proposal (after Baier 1981 and Thompson 2009) is an obligation with no correlative right. The living owe respect to valuable goods their predecessors bequeathed, and to their contemporaries’ future-oriented projects. So they must not wilfully destroy those goods, or the social practices on which pursuing such projects depends. “While future people belong to the beneficiaries, the obligation is owed to both present and past people.”
  • Recent relational and environmental accounts are listed as reinforcing this, among them Casal’s “Conservationist Sufficiency”.

• Historical injustice (§5). The threshold and disjunctive notions carry over. Past injustice can set what counts as sub-threshold for people who would not have existed without it (Waligore). Sher’s subsequent wrongs and De-Shalit’s transgenerational communities give further routes.

Claims

id	claim	strength	support
C1	The lack of reciprocity, the asymmetry of power and our ignorance of future individuals do not prevent future people having rights against us	moderate: argued, given the Interest Theory; the Will Theory denies it	§§1, 2.1
C2	The non-identity problem defeats counterfactual and diachronic notions of harm for people whose existence depends on the act	strong: definitional, and widely accepted	§3.1
C3	A threshold notion of harm escapes the non-identity problem and is weakly person-affecting	moderate: argued; it inherits the problem of fixing the threshold	§§3.1, 4
C4	A disjunctive notion of harm is preferable to either single notion, at the cost of the No-Difference View in its theoretical form	moderate: argued from cases	§3.2
C5	The threshold is best specified as weakly sufficientarian, in basic needs or capabilities	moderate: argued by elimination of egalitarian and prioritarian rivals	§§4.1–4.4
C6	Rights-based accounts cannot explain concern that humanity continue at a high level of well-being	strong on its own premises: the entry’s account gives no reason to prefer a flourishing future to none	§4.6, note 15
C7	The living have an obligation, owed to present and past people, not to wilfully destroy inherited goods and the conditions of future-oriented projects	weak to moderate: sketched, with sources	§4.6

Concepts

• non-identity problem: if an act is a condition of a person’s existence, she cannot be worse off because of it than she would otherwise have been.
• weak / strong person-affecting view: only people who exist at some time count / an act harms only if the same person is worse off than she would otherwise have been.
• impersonal view: the value of states of affairs is not reducible to their effects on particular people.
• threshold notion of harm: being caused to exist, or to be, below a normatively set level of well-being.
• disjunctive notion of harm: harm on the threshold or on the subjunctive-historical notion.
• No-Difference View (Parfit): it makes no difference to what we should do whether the size and composition of future generations depend on our choice.
• Shift Thesis: once enough is secured, the weight of reasons to benefit someone further changes discontinuously.
• afterlife conjecture (Scheffler, as glossed here): much of what we value presupposes that people will live after us.
• non-correlative obligation: a duty with no matching right in anyone.

Connections

• Narveson (LIT-1769), unread. Named as the start of the debate on whether a person-affecting view can account for the procreation asymmetry (§2.2). Meyer’s weak person-affecting view keeps Narveson’s half of it: no reason to create the happy.
• Parfit (LIT-469; LIT-1768), both unread. The genetic identity view, same and different people choices, the two medical programs and the No-Difference View come from Reasons and Persons. The 2017 paper is listed among works holding that intergenerational ethics must be impersonal or mixed (§1).
• Scheffler (LIT-1754; LIT-1712), unread. Death and the Afterlife is cited for indirect reciprocity (§1), the possibility of extinction (§1) and concerns rights cannot capture (§4.6, pp. 60–63, 72–73, 80–81). The afterlife conjecture is given through McKinnon and Scheffler 2025. Why Worry About Future Generations? is in the bibliography but not cited in the text.
• Pitcher (LIT-1740), not cited. §1’s point that present or past people can be set back by posthumous states of affairs is the reading QUESTION-301 attributes to Pitcher.
• Gardiner (LIT-1681) is named and set aside, as a structural dynamic rather than an account of what is owed (§1).
• Ramsey (LIT-1733): optimal savings without distributional concerns; rejects pure time discounting (§4.5).
• Longtermism (LIT-1684) is absent. The entry cites none of Greaves, MacAskill, Bostrom or Ord. Its impersonal side appears only as views it reports and does not adopt.
• No empirical work on motivation. The psychology the question cites (Schubert et al., LIT-1748; Caviola et al., LIT-1749) is not cited.

Bearing on the record

• What the living owe the far future (QUESTION-299). The entry fills in the question’s second ground, the good of future people, and shows it is not as empty as the question’s summary says.
  • A weak person-affecting view can state harms to actual future people despite the non-identity problem, by a threshold.
  • But it owes nothing to merely possible people, and by its own account gives no reason to prefer a flourishing future to an empty one.
  • It meets the question’s third ground, what the living value, at exactly that point. Meyer reaches for Scheffler, and for an obligation owed to present and past people.
  • So, on the person-affecting side, concern for the far future’s existence rests on what the living and the dead value, not on future people’s claims. That is the entry’s structure, not an inference.
• The owner’s thesis (CLAIM-491). These are inferences; the entry does not discuss the self.
  • Neither an extended self nor an impersonal valuation is the entry’s ground, but the gap it leaves is where the thesis would sit. A future that is merely good is no one’s claim on this view. What makes it matter is that the living’s projects and attachments presuppose it. If those projects count as part of an extended self, the thesis has a way to read the strongest concern for the far future as self-interest. If they do not, Scheffler’s case stands against it.
  • Meyer’s obligation is owed to the living and the dead, with future people only as beneficiaries. That fits the thesis’s shape: concern whose object is the future, whose ground is people who exist or existed.
  • Rawls’s first device was the thesis’s shape, and he dropped it. In A Theory of Justice, savings rested on the contractors’ care for their descendants, a self extended along a family line. In Political Liberalism he replaced it with mutual disinterest plus a consistency condition. A theory of justice that once rested on extended concern now avoids relying on it.
• A dated note to CLAIM-491 is not needed. The entry bears on the thesis only through Scheffler, who is unread. The note belongs on QUESTION-299 (section 3 below).
• No THEORY filed.
• No ML instruction.

Limitations

• An encyclopedia entry that argues. The threshold and disjunctive notions are the author’s own position (Meyer 2003, 2009a), presented alongside the survey. Impersonal views are reported, not examined.
• Scheffler at second hand. The afterlife conjecture is given in McKinnon’s gloss.
• The 2026 revision’s additions are thin. Several 2025 works (Kallhoff, Nolt, Watene, Whyte, Ochi) are summarized in a sentence each.
• Typos and slips.
  • “Parfit 1984: 363, 295–396” (§1), presumably 395–396.
  • “therelevant” (§2.2). In the same section, “bringing into existence a person whose life is worth living still falls below the threshold” lacks a word.
  • “most of who” (§1); “as maximizes” (§4.2); “weak sufficientiarianism” (§4.3); “threshold Notion” (§3.1).
  • “in subjective rather than objective subjective terms” (§4.4).
  • “allows ist content” (§5).
  • An unmatched parenthesis in §1’s first paragraph, and a stray quotation mark after the link to the Ramsey entry (§4.5).
  • “Waligore emphasizes” (§5) with no year; the bibliography has Waligore 2026.
  • Note 2: “whether future people can be said to and have a right”.
  • Bibliography: Narveson 1973’s title is followed by ””:,”.

Open questions

• Does the threshold view’s admission (§4.6) leave the continuation of humanity to non-correlative duties, or to valuing alone? Scheffler’s own text, unread, would say which he means.
• Is the obligation not to destroy what future-oriented projects depend on owed to the dead in a sense that would make it a legacy interest (QUESTION-301)? The entry says it is owed to past people without saying how they can be wronged.