Why it is a question
The owner’s thesis (CLAIM-491) says apparently selfless action serves a self enlarged to include others. Legacy is the same move in time. Johnston’s Surviving Death (LIT-1759) made it explicit: the good person, who identifies with humanity, lives on in its onward rush. Johnston has since given that account of the self up (NOTE-1257). The record already holds the owner’s own version, “immortality” through influence that reinforces itself across generations, as CLAIM-420.
Four bodies of work bear on it.
- The self over time, measured. Hershfield (LIT-1702) measures continuity with one’s future self with overlapping circles adapted from Aron’s scale, and finds that people who feel less continuous treat their future self like another person. That makes prudence an instance of the thesis’s shape, and links it to Parfit (THEORY-464).
- Legacy motivation.
- Death and the group. Terror management theory (Burke et al., LIT-1687) draws on Lifton’s symbolic immortality (LIT-1714). On it, reminders of death make people identify more with a group they see as lasting (Castano et al., LIT-1719). The temporal and social extensions of the self then meet.
- Three philosophical readings of concern for what comes after us.
- An extended self. The thesis’s reading: what one cares about after death is part of oneself.
- Interests that outlast death. On Pitcher (LIT-1740) the living person can be harmed by what happens after their death. So acting for one’s legacy can be self-interested in the ordinary sense, with no extended self.
- Dependence on a future we will not see. On Scheffler (LIT-1754) much of what we value depends on our confidence that humanity will go on after we die. He reads this as dependence on future people, not altruism toward them: their going on keeps our valuing secure (NOTE-1298). It was filed as the hardest case for the thesis in time, and the reading makes it less hard.
What would count as an answer
- A test that separates the readings. Measure felt continuity with one’s legacy or with future generations, and test whether concern for them follows that measure or follows caring alone. That is CLAIM-492’s test applied to time.
- A case for one reading. An argument that concern for people one will never know cannot be self-interest on any self, or a model showing that it can be without triviality.
- Evidence from the legacy experiments. If legacy priming raises beneficence only through the wish to be remembered, the thesis’s reading, or Pitcher’s, gains. If it raises beneficence also where no one will know, Scheffler’s gains.
The wider question
What the living owe the future, as distinct from why they care about their own legacy, is QUESTION-299: longtermism, person-affecting views, the non-identity problem and Scheffler’s grounding of concern in what the living value.
Note of 2026-10-29: the legacy works, read
The originals are Unavailable. Works read in their place, and the two read themselves, change the question in five ways.
- Scheffler is less hard a case than filed (NOTE-1298; NOTE-1306). The Tanner Lectures claim that our valuing depends on confidence that others will live after us. Scheffler says the surprise shows “less that we have been blinded to our own altruism than … that we have overestimated our independence and self-sufficiency”. He explains even his preference to die sooner so that humanity survives by the limits of individualism, not of egoism. For the thesis, the self served would be one’s standing as a valuer: a third sense of “self-interest”, after welfare and after a self-concept kept true. The triviality objection applies to it.
- Pitcher’s reading depends on the theory of welfare (Luper, SEP, NOTE-1294). Posthumous harm is the thwarting of desires held while alive, so legacy-seeking is self-interest with no extended self only on a desire-based theory of welfare. If any satisfied desire counts as one’s good, it is self-interest by definition.
- Terror management is weaker support (Many Labs 4, NOTE-1322; Chen et al., NOTE-1271). Seventeen labs (N = 1,550) found no mortality-salience effect: g = 0.07, 95% CI [−0.03, 0.17]. Corrected for publication bias, the meta-analytic effect falls from about r = .35 to about .18, or to nothing. Castano et al. (NOTE-1297) state the join of the social and temporal self as a conjecture: a lasting group lets the self “project themselves in space and time, beyond their personal death”. Their one experiment (N = 48) builds on that contested effect and measures no self.
- Legacy motivation is real but not yet separated from caring.
- Generativity includes a need for symbolic immortality (Kruse and Schmitt, NOTE-1261). A recent validation separates a three-item “Legacy” factor from generative action (Sandoval-Obando et al., NOTE-1260), but its items are not printed.
- Wade-Benzoni’s experiments (NOTE-1273) never measure legacy concern and generosity in the same study. Two of their p-values do not match their statistics.
- Hurlstone et al.’s legacy messages (NOTE-1279) bundle identification with generations, a moral appeal and one’s own death, so they cannot tell the thesis’s reading from Pitcher’s.
- Symbolic immortality, as Sturesteps measured it (skimmed, NOTE-1312), is consolation that goes with accepting death as final, not a motive to act for the future.
- The self over time is measured as the thesis needs (Hershfield, NOTE-1259). Felt similarity to one’s future self predicts patience where directly reported caring did not, but that caring was near ceiling. Hershfield guesses that the future self may continue in one’s offspring, which is the untested bridge from prudence to legacy.
The question’s third philosophical reading is reworded to match Scheffler: dependence on a future we will not see, not care for a future that does nothing for us.
Note of 2026-10-29: Zaval’s legacy measures, read at second hand
Wickersham et al. (NOTE-1316), read in place of Zaval et al. (LIT-1724), reuse its legacy writing task and Legacy Motivation Scale with 988 adults.
- The effect replicates. The task raised pro-environmental intentions and hypothetical donations at every age, and the scale partly mediated it.
- The scale does not isolate the wish to be remembered. Of its eight items, two are about being remembered, four about self-regard, and two about responsibility and connection to future generations. Only the total is analysed.
- A parallel route. The task also raised love of nature, which carried more of the donation effect than legacy motivation did (0.138 against 0.097, in the supplement’s two-mediator model).
- The question’s test is still untried. No condition varies whether anyone will know what one did.