Aging research and population ethics

This is the sec­ond of a se­ries of posts in which I’m try­ing to build a frame­work to eval­u­ate ag­ing re­search. Pre­vi­ous post: A gen­eral frame­work for eval­u­at­ing ag­ing re­search. Part 1: rea­son­ing with Longevity Es­cape Ve­loc­ity.


The first part of this post will ex­plore the po­ten­tial sen­si­tivity of the im­pact of ag­ing re­search on two differ­ent views of pop­u­la­tion ethics. Un­der the per­son-af­fect­ing view of pop­u­la­tion ethics, in which cre­at­ing new lives has neu­tral moral value, it seems that ag­ing re­search is re­ally valuable. Un­der the im­per­sonal view, in which cre­at­ing new lives has a pos­i­tive moral value, it could be less clear. By look­ing at de­mo­graphic trends, and an­a­lyz­ing the mo­ti­va­tions for why peo­ple have chil­dren, it turns out that sav­ing peo­ple by has­ten­ing the ar­rival of LEV wouldn’t pre­vent births and could ac­tu­ally in­crease the av­er­age fer­til­ity rate of the world. This leads to a coun­ter­in­tu­itive re­sult: Aging re­search could be even more valuable un­der the im­per­sonal view of pop­u­la­tion ethics.

In the sec­ond part of the post, I’ll ex­plore how to rea­son about moral weights, which could also in­crease the im­pact of mak­ing LEV come closer if longer lives are val­ued more than shorter lives for rea­sons other than QALYs. There are var­i­ous ar­gu­ments for why one should pre­fer some kind of age-dis­count­ing or its con­trary, but the an­swer ul­ti­mately de­pends on what a 1000-year-old life and mind looks like and how it is differ­ent from the life and mind of a shorter-lived per­son. There­fore, tak­ing a neu­tral stance is sug­gested un­less it is be­lieved that the fu­ture, if there will be one, is more likely to be bet­ter than the pre­sent. In that case, the lives of peo­ple saved through LEV should count for more.

Pop­u­la­tion ethics

At first glance, the im­pact of ag­ing re­search seems to greatly change de­pend­ing on if you adopt the im­per­sonal view of pop­u­la­tion ethics or the per­son-af­fect­ing per­spec­tive. In the im­per­sonal view, cre­at­ing new lives is re­garded as good. As­sum­ing that there isn’t suffer­ing at the end of life and peo­ple get re­placed im­me­di­ately, this view holds no eth­i­cal differ­ence be­tween mak­ing peo­ple live longer and re­plac­ing them with new peo­ple. Un­der the per­son-af­fect­ing per­spec­tive, how­ever, cre­at­ing new lives is not val­ued: only already ex­ist­ing peo­ple are val­ued, and thus how bad it is to die de­pends on the amount of well-be­ing lost.

MichaelPlant re­minded me of this point un­der my pre­vi­ous post. I gave an an­swer there but don’t think it is suffi­cient. There­fore, a more ac­cu­rate anal­y­sis of this con­sid­er­a­tion is war­ranted here.

It seems like if the per­son-af­fect­ing per­spec­tive is adopted, then ag­ing re­search has enor­mous value. That is the im­pact out­lined in the pre­vi­ous post.

It seems, though, that ag­ing re­search could have at least the same value un­der the im­per­sonal view if elon­gat­ing healthy life does not mean tak­ing the space of po­ten­tial new­borns. When try­ing to de­ter­mine if it would, it’s tempt­ing to think about the very far fu­ture and start with this ques­tion: will hu­man­ity use all the re­sources at its dis­posal at any given time? Even if hu­man­ity will not use all the re­sources at its dis­posal, will it still con­trol its pop­u­la­tion growth in or­der to max­i­mize well-be­ing? If the an­swer to one of these two ques­tions is “yes”, then it seems like elon­gat­ing life would pre­vent births.

Start­ing with these ques­tions and think­ing about the far fu­ture is wrong. Re­minder: most of the im­pact of ag­ing re­search comes from mak­ing the date of LEV come closer and sav­ing the peo­ple who wouldn’t oth­er­wise have hit LEV. If LEV will hap­pen, then it’s very prob­a­ble that it will hap­pen in this cen­tury or the next. There­fore, to an­swer the ques­tion “will elon­gat­ing life pre­vent births?” we need to ac­count for how so­ciety cur­rently works and the cur­rent de­mo­graphic trends.

It seems that the choice of mak­ing chil­dren is not cur­rently mo­ti­vated by lack of re­sources (poverty); on the con­trary, the num­ber of chil­dren is go­ing down sharply with in­creased stan­dards of liv­ing. This is a trend that is prov­ing true in ev­ery part of the world, un­der­de­vel­oped na­tions in­cluded.

This means that mak­ing peo­ple live longer in this or the next cen­tury is not go­ing to pre­vent po­ten­tial births. They will prob­a­bly hap­pen less and less re­gard­less, and mak­ing old peo­ple healthy and pro­duc­tive is go­ing to pre­vent the eco­nomic dis­aster that is loom­ing due to an in­creas­ingly aged pop­u­la­tion, even in un­der­de­vel­oped coun­tries.

Quite the in­verse could prove to be true, though: peo­ple with longer lifes­pans could have more chil­dren, sim­ply be­cause they will have much more time to pro­cre­ate via the child­bear­ing win­dow be­ing ex­tended. There­fore, the fer­til­ity rate will prob­a­bly in­crease. This con­sid­er­a­tion could even be the rea­son why a sce­nario in which pop­u­la­tion con­trol will be needed could prove true, al­though I tend to think that ceiling is very far away in the fu­ture, due to tech­nol­ogy still hav­ing an am­ple mar­gin of im­prove­ment. In case longer lifes­pans ac­tu­ally in­crease the world fer­til­ity rate, then the im­pact un­der the im­per­sonal view of pop­u­la­tion ethics is the sum of the QALYs saved due to mak­ing LEV come closer, plus the QALYs of the new­borns of the peo­ple saved, who wouldn’t oth­er­wise have been born.

Ad­di­tion­ally, if longer lives are more valuable than shorter ones for rea­sons differ­ent than the num­ber of QALYs, the neu­tral view could still value longer lives over perfectly re­plac­ing them with shorter lives. This brings us to how to choose moral weights.

Mo­ral weights

An im­por­tant ques­tion that could sub­stan­tially af­fect the mea­sure of im­pact is how to choose moral weights. I think that it’s prac­ti­cally im­pos­si­ble to come to a defini­tive an­swer due to a lack of em­piri­cal in­for­ma­tion, but I can out­line pos­si­ble ways to rea­son about the prob­lem.

The cen­tral ques­tion seems to be: is a 1000-year life in­trin­si­cally more, less, or equally as valuable as many shorter lives that sum up to 1000 years?

One ar­gu­ment for why it could be less valuable could be this sim­ple one: it’s only one life. I wouldn’t find strange if many peo­ple would find differ­ent shorter lives more im­por­tant than a sin­gle long one be­cause of some kind of in­tu­ition re­gard­ing a prefer­ence for va­ri­ety or even fair­ness. This is also sup­ported by the in­tu­ition that many peo­ple would choose to live a 90% chance of liv­ing for a nor­mal hu­man lifes­pan than a 10% chance of liv­ing a 900-year lifes­pan.

One life, in­tu­itively, is “fresh” only once. Some­one may value shorter lives more be­cause they could be, in­tu­itively, more im­bued with fresh ex­pe­riences. Each one goes through in­fancy, ado­les­cence, and all the other phases of life.

At first glance, the first ar­gu­ment seems weaker: af­ter all, one per­son is never re­ally the same. The mind changes con­tin­u­ously, and some­one could re­tain very lit­tle of them­selves liv­ing cen­tury af­ter cen­tury. Would this per­son ex­pe­rience less nov­elty? It’s pos­si­ble, un­less the fu­ture re­serves re­ally in­cred­ible new ex­pe­riences and sur­prises. How­ever, is nov­elty all there is to con­sider?

Many lives are, prob­a­bly, im­bued with more nov­elty, but one long life could mean in­sight and ac­crual of knowl­edge that would be im­pos­si­ble for a sin­gle lifes­pan. Anec­dotes of old sci­en­tists and lu­mi­nar­ies with vast vi­sions of their fields but lack­ing the sharp­ness of mind to con­tribute, es­pe­cially in hard sci­ences or math­e­mat­ics, are com­mon. Each one of them dy­ing is a burnt library of in­sight and knowl­edge. Sev­er­ing their lives at that point means also pre­vent­ing any fu­ture ex­pe­rience re­sult­ing from that knowl­edge. In some sense, it feels like stop­ping to play when the fun be­gins, and this could also say some­thing about nov­elty, which may not be ex­tin­guish­able very soon. Much longer lifes­pans could also pos­si­bly mean deeper and oth­er­wise im­pos­si­ble-to-ex­pe­rience emo­tions and states of mind, mak­ing longer lives more valuable. This seems ob­vi­ous if we take, again, the ex­am­ple of lu­mi­nar­ies: for a com­mon in­di­vi­d­ual, a nor­mal hu­man lifes­pan may be not enough to ac­quire the knowl­edge of a lu­mi­nary. Thus, a short life may con­sti­tute a hard wall against what can be ex­pe­rienced by most peo­ple.

Another in­tu­ition that would make one con­sider a longer life more valuable is this: I think there is a pretty strong case for prefer­ring to have one gen­er­a­tion of peo­ple liv­ing 80 years than mul­ti­ple gen­er­a­tions of chil­dren liv­ing till the fifth year of age. There­fore, maybe the same in­tu­ition could ap­ply for longer lifes­pans. Are peo­ple liv­ing till 100 like chil­dren if com­pared to some­one liv­ing to 1000 years old? The an­swer to this can’t be defini­tive. I think the an­swer de­pends on in­for­ma­tion we cur­rently don’t have: how a 1000-year life and mind looks like and how it is differ­ent from the life and mind of a shorter-lived per­son.

In­tu­ition on how to as­sign moral weights sug­gests both is­sues: If we lean to­wards valu­ing longer lives more, we could be over­es­ti­mat­ing how much more “en­light­ened” a hu­man mind can be­come. If we lean to­wards valu­ing shorter lives more, we may un­der­es­ti­mate the same vari­able or even com­mit a mis­take akin to scope in­sen­si­tivity if we don’t think about the prob­lem deeply enough.

One con­sid­er­a­tion that could shift the nee­dle con­sid­er­ably on this is if you deem it prob­a­ble that the fu­ture will be bet­ter than the pre­sent or if you think, in­stead, that the far fu­ture will be worse. I think that the fu­ture is more likely to be of a utopian kind or sim­ply de­void of life than worse than the pre­sent, and the prob­a­bil­ity of fu­ture ex­is­ten­tial risks has to be fac­tored in as a dis­count of im­pact, but it’s not part of moral weights, so I would tend to eth­i­cally value longer lives more than shorter ones for this rea­son.

How­ever, if you think that the prob­a­bil­ities of the fu­ture be­ing bet­ter or worse than the pre­sent offset each other, then there are good ar­gu­ments for both meth­ods of ap­ply­ing moral weights, and I would ar­gue to ap­ply nei­ther age-dis­count­ing nor the con­trary. A neu­tral stance is prob­a­bly prefer­able. That said, differ­ent an­a­lysts should feel free to think about the prob­lem them­selves, and if they be­lieve that one out­come is more likely than the other, they may want to cor­rect these crude es­ti­mates.


Cross­posted from the EA Forum