future-generations.
24 of 1538 NOTE documents. Back to the full index.
| # | Title | Summary | Status |
|---|---|---|---|
| NOTE-1247 | The epistemic challenge to longtermism | A model of the epistemic objection to longtermism. A $1 million existential-risk intervention changes the chance that civilization exists in 1000 years by a conservative 2 x 10^-14, and its effect lasts until an “exogenous nullifying event” arrives at a constant rate r, which acts as an exponential discount. Against a global-health benchmark of 10,000 QALYs, it wins if r is below about 0.000135 per year with interstellar settlement, or 0.000000012 per year on Earth alone. Small credences in settlement, low r, fast travel or star-sized value make it win by orders of magnitude, but then the win can rest on probabilities as small as 10^-18. So the case for longtermism depends on non-obvious empirical claims or on Pascalian fanaticism, and its horizon may be thousands or millions of years, not billions. | Read |
| NOTE-1252 | A Perfect Moral Storm: Climate Change, Intergenerational Ethics and the Problem of Moral Corruption | Climate change is a “perfect moral storm” in which three storms converge. The global storm is a tragedy of the commons between states. The intergenerational storm is worse: carbon dioxide stays in the air for centuries, 17–33% of it for a thousand years by the estimate quoted, so the benefits of emitting go to the present and the costs to later generations. Every generation then has reason to overpollute, and since generations do not coexist, no institution or reciprocity can stop it. The theoretical storm is that our theories cope badly with such cases. Together they make moral corruption easy: each generation can seem to take the problem seriously while exploiting its position in time, and need not admit it, “either to others, or, perhaps more importantly, to itself”. | Read |
| NOTE-1259 | Future self-continuity: how conceptions of the future self transform intertemporal choice | A review, mostly of the author’s own work. Following Parfit, people who feel less connected to their future self treat it like a stranger and save less for it. Similarity to one’s self in ten years, chosen as overlapping circles adapted from Aron’s scale, predicted patient choices and assets; a future self represented neurally like another person predicted steeper discounting (r = .47); seeing one’s age-progressed face raised saving; respect for the elderly tracked and raised saving. Directly reported liking and caring for the future self did not predict choice, but were near ceiling. The future self may extend, at life’s end, to one’s offspring. | Read |
| NOTE-1260 | Psychometric validation of the Loyola Generativity Scale in Chilean teachers: evidence for a three factor structure | 430 Chilean elementary teachers completed a Spanish Loyola Generativity Scale, McAdams and de St. Aubin’s 20 items on a 4-point scale. Four items were dropped for poor sampling adequacy. Of one- to four-factor solutions, three factors were retained (RMSEA = 0.043, CFI = 0.989, 55.1% of variance); four fit slightly better but were judged over-extracted. Its factors: Generative Inefficacy, three negatively worded items; Generative Action, eight items; and Generative Legacy, three items on symbolic continuity. Action and Legacy correlate .45. The authors call the analysis confirmatory, but it chooses the number of factors from the data. Only 14 of the 16 retained items appear in the final solution, without comment. | Read |
| NOTE-1261 | Generativity as a Route to Active Ageing | Two Heidelberg gerontologists read generativity, Erikson’s “concern in establishing and guiding the next generation”, as the psychological form of an ethical “joint responsibility” that lasts into very old age. They follow McAdams: generativity is driven by cultural demand and an inner desire, which is a need to be needed and a need for symbolic immortality, “to invest resources and potentials into things that outlive the self”. It is not altruism, and its motives can be selfish or altruistic. In 3,308 older adults in Mexico, Spain and Germany, generativity predicted optimism and life satisfaction. An intervention of intergenerational projects in Belarus, Russia and Ukraine raised generativity scores in old and young. No statistics are given. | Read |
| NOTE-1263 | Future People, the Non-Identity Problem, and Person-Affecting Principles | Most of our choices change who will exist, so a policy that makes later lives much worse may be worse for no one who ever lives. Narrow person-affecting principles then imply that such a policy, or ending human history by having no children, is not wrong. Parfit rejects them, and retracts his own 1984 appeal to an impersonal principle. Being caused to exist with a life worth living is an intrinsic “existential benefit”. On the Wide Principle an outcome is worse if it benefits people less, which solves the non-identity problem in person-affecting terms and keeps “better than” transitive. A Wide Dual Principle, counting benefits to each and benefits to all, does not imply the Repugnant Conclusion. We may favour present people in what we do, but if every generation does, “we and our successors might together wreck the Earth”. | Read |
| NOTE-1267 | Population ethical intuitions | Nine experiments with US online samples. People require more happiness than suffering for a population to be better existing than not (about 75% happy; about 61% when the two are stated to be equally intense). Yet they judge adding one extremely happy person as good as adding one extremely unhappy person is bad: they reject the intuition of neutrality and the procreation asymmetry. With averages fixed they prefer larger happy and smaller unhappy populations, but the preference for more happy people fades and, at a billion against ten billion, reverses. With totals fixed they prefer higher averages. In conflicts averagism tends to win, to the point of preferring to add near-maximally unhappy people who raise the average. Prompts to reflect move judgements toward totalism. | Read |
| NOTE-1269 | Shaping Humanity’s Longterm Trajectory | Humanity’s value per unit time is a curve, the future’s value the area under it, and an intervention’s value the area between two curves. Value must be summable and separable across time, and there is no pure time preference. Four idealised changes have values independent of the curve’s shape. An advancement is worth about (v_τ − v_0)δt, which does not grow with the future’s duration, and is negative if the end comes sooner too. A speed-up scales with (v_τ − v̄)τ, a gain with τ, an enhancement with v̄τ. Raising near-term survival by a factor γ_p is worth (γ_p − 1)v̄τ, like an enhancement, and is more surely contingent. So advancing progress rarely rivals existential risk reduction: on a million-year future a year’s advancement equals a one-in-a-million gain in survival. | Read |
| NOTE-1270 | The Psychology of Existential Risk: Moral Judgments about Human Extinction | Parfit’s three outcomes, put to laypeople: no catastrophe, 80% of humanity killed, 100% killed. Most judge extinction bad and to be prevented (78% on a binary item), but of those who rank the outcomes as expected only 23% (UK), 19% (US) and 37% (Oxford students) judge the step from 80% to 100% the greater. More do when the victims are zebras (44%, 35%), when the catastrophe sterilizes rather than kills (47%, 29%), or when reminded that only extinction ends humanity’s future (50%, 39%). A majority do only when the surviving future is described as long and utopian (77%, 67%, 77%). Existential-risk mitigators do so unprompted (86%). The authors read this as a focus on immediate harm to fellow humans and neglect of the long term. | Read |
| NOTE-1273 | Leaving a Legacy: Intergenerational Allocations of Benefits and Burdens | Six experiments on intergenerational allocation. Framing a climate decision as leaving burdens rather than using up benefits raised concern with one’s legacy, felt responsibility and affinity toward future generations, ethical concern and moral emotions. In allocation tasks people favoured themselves more with benefits than with burdens, but only when the other person would receive the resource in the future, and only when they themselves were a party to the trade-off. Perceived ethical considerations accounted for the first condition. The authors read this as legacy concern, the wish to extend the self beyond death. Legacy concern was measured only in the attitude studies, never alongside an allocation. | Read |
| NOTE-1279 | Activating the legacy motive mitigates intergenerational discounting in the climate game | A climate collective-risk game with 30 groups of six. When saving the group’s endowments paid the players the next day, 9 of 10 groups reached the target; when it paid for trees for future generations, 2 of 10 did; with three messages added to activate the legacy motive (death awareness, the powerlessness of future generations, and paying forward a past generation’s sacrifice), 7 of 10 did. Each message raised investment. The messages restored a fair-share norm rather than extra giving. No manipulation checks and no control message, so which mechanism worked is unknown. The authors call the legacy motive a way of leveraging self-interest. | Read |
| NOTE-1287 | A Mathematical Theory of Saving | How much should a nation save? For a community that lives for ever, with fixed numbers, tastes and motives, and that does not discount later enjoyments (“ethically indefensible”, from “the weakness of the imagination”), saving times the marginal utility of consumption must equal the shortfall of present net enjoyment from Bliss, the maximum obtainable. The rule does not depend on the interest rate, and asks much more saving than usual: £300 out of £500 on an illustrative scale. A constant discount rate below the interest rate gives the same rule for a “modified utility”; one above it runs consumption down to subsistence. The paper then takes discounting as a fact of saving. People care less for heirs than for themselves, and families with different rates end divided between bliss and subsistence. | Read |
| NOTE-1293 | Responsibility to Future Generations: A Moral Reframing Strategy for Combatting Climate Change | Felt responsibility to future generations is held by most people in 33 of 34 European countries and is nearly unrelated to ideology, age and gender. In US samples it is endorsed more than responsibility to cut one’s own climate contribution, is half as tied to conservatism (r = −.21 against −.45), and predicts self-reported proenvironmentalism. A preregistered experiment (N = 3,175) found five brief interventions each raised it a little; a letter to a person born today did most, a longtermist text least, and reflecting on past generations’ sacrifices raised it without raising responsibility for climate. Nothing measures how far the self extends, and responsibility is not separated from caring. | Read |
| NOTE-1298 | The Afterlife (Tanner Lectures I and II) | Scheffler’s 2012 Tanner Lectures. “The afterlife” is the survival of other people after one’s death. Few would be indifferent to the earth’s destruction thirty days after their death, or weigh its goods against its evils. Many projects would lose their point, and in an infertile world, where no one dies early but no one more is born, even enjoyments might. Humanity’s survival matters to us in its own right and as a condition of other things mattering. In that respect it matters more than our own survival or our loved ones’. Scheffler reads this as our dependence on future people, not altruism toward them, and as showing that individual valuing presupposes humanity as an ongoing project. | Read |
| NOTE-1300 | The Significance, Persistence, Contingency Framework | The value an event E contributes through a feature F of the world is significance × persistence × contingency: value per unit of the time E adds (ΔV/ΔT), how long F lasts (T), and the share of that duration due to E (ΔT/T). Under uncertainty the factors are taken ex ante (ΔEV/ΔET, ET, ΔET/ET), which keeps the product but is not the product of expected factors. An alternative takes significance as V/T and makes contingency the sum of the semi-elasticities of significance, persistence and the probability that F exists. The result estimates the “importance” factor of the ITN framework. It is an accounting device that assumes expected-value maximisation; it says nothing about who is moved by the value it counts. | Read |
| NOTE-1302 | The Case for Strong Longtermism | Axiological strong longtermism: in the most important decisions facing agents today, every near-best option is near-best for the far future and delivers much larger benefits there than in the near future. The case runs through a benefit ratio. Bednets save 0.025 lives per 100 in expectation; AI safety beats bednets even on an Earth-only 10^14. A stakes-sensitivity argument carries this to what one ought to do. Discounting would defeat it: a zero pure rate of time preference is “near-essential”. Fanaticism is the weakest point the authors admit. They give an impersonal good as the ground and say nothing about motivation. | Read |
| NOTE-1304 | Intergenerational Justice | The entry argues as well as surveys. Future people’s existence, number and identity depend on our choices, so harm understood as making someone worse off than they were, or would otherwise have been, cannot reach them: the non-identity problem. A threshold notion of harm can: we harm someone by causing her to exist, or to be, below a normatively set level. Read as sufficientarian, in basic needs or capabilities, this grounds future people’s rights, and Rawls’s just savings principle is one version of it. Being weakly person-affecting, it owes merely possible people nothing, and gives no reason to prefer a flourishing future to none. Meyer fills that gap with concerns that are not rights: Scheffler’s afterlife conjecture, and a duty owed to present and past people not to destroy what future-oriented projects depend on. | Read |
| NOTE-1306 | Review of Samuel Scheffler, Death and the Afterlife | Cottingham reviews Scheffler’s book: two 2012 Tanner Lectures and a third paper given at Chicago the year before, four commentaries and replies. He reports the afterlife conjecture and its upshot, that in one respect our own survival matters less to us than humanity’s. The third lecture argues that a life must be bounded, so immortality is no remedy, while confidence in our values needs others to live on. Frankfurt doubts the concern is altruistic; Wolf holds we could and should “snap out of it” and go on. Cottingham regrets that religious reflection on finitude is left out. | Read |
| NOTE-1308 | High Risk, Low Reward: A Challenge to the Astronomical Value of Existential Risk Mitigation | On Ord’s simple model, with constant per-century risk r and constant value v per century, cutting this century’s risk by a fraction f is worth fv, whatever r is: halving 20% risk is worth the same as halving 2%. Absolute cuts are worth fv/r, linear and quadratic value growth give fv/r and fv(2 - r)/r^2, and cutting risk in every century gives (f/(1 - f))(v/r). In each, higher risk lowers the value of mitigation. At 20% risk a 10% cut is worth 0.5v to 4.5v. Astronomical value returns only if risk falls soon and stays very low (a time of perils of two centuries and 0.01% afterwards gives 160v). Thorstad finds the arguments for that hypothesis (wisdom, an existential risk Kuznets curve, space settlement) weak. | Read |
| NOTE-1314 | Astronomical Waste: The Opportunity Cost of Delayed Technological Development | Each second’s delay in settling the Virgo Supercluster loses the potential for about 10^29 lives (10^13 if all must be biological). On a total, undiscounted view that makes delay an enormous loss. But a one-percentage-point cut in existential risk is worth over 10 million years of delay, so the total utilitarian’s priority is safety: “Minimize existential risk!”. On a person-affecting view only people now alive or who will exist count, and extinction is bad for what it does to them. Given a chance that some now alive live to share the cosmic benefits, safety is then very important but must be balanced against speed. The grounding is impersonal on the first view and the present generation’s own good on the second. Nothing is said about motivation. | Read |
| NOTE-1316 | The impact of place and legacy framing on climate action: A lifespan approach | 988 American adults, aged 18 to 87 and sampled evenly by age, wrote for 5–7 minutes about what they wanted future generations to remember them for (Legacy), about a place they loved and climate change (Place), or how to scramble eggs (Control). Both tasks raised pro-environmental intentions and hypothetical donations, small effects equal across age groups. The Legacy task raised an eight-item Legacy Motivation Scale, which partly mediated its effect. But the Place task raised legacy motivation too, the Legacy task raised love of nature, and with both mediators in one model love of nature carried at least as much of the donation effect. The scale’s items mix wanting to be remembered with self-regard and felt responsibility to future generations. | Read |
| NOTE-1317 | The Precipice Revisited | Ord updates The Precipice’s four largest risks after about five years. Climate risk is down: emissions track between the two lower RCPs, and the IPCC’s likely climate sensitivity narrowed from 1.5–4.5°C to 2.5–4°C, partly by our action and partly by discovery. Nuclear risk is up: the war in Ukraine, New START’s coming expiry, and funding for the field halved. Pandemics and AI are mixed: Covid showed weak institutions and a brief unity, but vaccines came within a year; language models are not inherently agents and have read human values, while big-tech racing is bad and governance has improved. The book’s numbers (climate and nuclear 1 in 1,000, pandemics 1 in 30, AI 1 in 10 per century) have not moved at their rounding. | Read |
| NOTE-1318 | Population and Having Children Now | Narveson’s late restatement of the person-affecting view, after giving up utilitarianism. Creating a person neither benefits nor harms them, since before conception there is no one to compare. But the existence of people with good lives is normally good, and a reason to create them, “contrary, perhaps” to how his 1967 and 1973 papers were read. It matters because parents and the benevolent like it and because more people benefit others. Only harm to actual persons, “alive, breathing, and acting now (or at any rate in their lifetimes)”, can justify coercing procreation; possible people are owed nothing. Not aiding the poor is not harming them. The second half claims, with little evidence, that resources are created and that present and future people do not compete for them. | Read |
| NOTE-1321 | Existential Risk Prevention as Global Priority | An existential risk threatens premature extinction of Earth-originating intelligent life or the permanent and drastic destruction of its potential. Four classes: extinction, permanent stagnation, flawed realisation, subsequent ruination. On an aggregative, time-neutral view Earth alone could hold 10^16 future lives, so a millionth of a percentage point of risk is worth a hundred million lives. Maxipok, a rule of thumb for impersonal moral action, follows. Adams’ shared project of humanity, preference satisfaction and custodial duties concur. Bostrom finds a “deficit of motivation”: the good is pan-generational and public, its beneficiaries cannot pay, and moral sentiment is insensitive to scope. | Read |