There Is No Altruism (kunnas.com)

8 comments · 2026-09-12 · discussion

positivist_import2 comments

Thomas Dixon's The Invention of Altruism (2008) is the book-length version of the birthday story, and it does not land where the title does.

Comte coined altruisme in the early 1850s as a Religion of Humanity slogan. In the Système de politique positive he wrote that "to live for others" was the whole of positive morality. G.H. Lewes brought the word into English in the Westminster Review in 1852. That is a secular catechism — live for Humanity — not a puzzle about why isolated people would help.

The puzzle this post is fighting is later. Hamilton 1964, Williams's Adaptation and Natural Selection in 1966, E.O. Wilson calling altruism the central theoretical problem of sociobiology. Comte wanted more other-regard, not an explanation of why it was mysterious. If you need a replacement for the 1850s coinage, Dixon's is older English: charity, sympathy, duty. Those never required a transfer between opposed atoms.

reliability_parentcollapsed

Elliott Sober and David Sloan Wilson's Unto Others (1998) is cited for the group-level point that "sacrifice" is the wrong word once the unit is the group. Skip to Part Two and they keep the word.

Evolutionary altruism is a fitness cost to the actor. Psychological altruism is an ultimate desire for someone else's welfare. Those do not entail each other. A parent who wants the child to do well as an end in itself is a psychological altruist; if that care raises inclusive fitness it is evolutionary selfishness. Their argument for parental care is reliability: a parent keyed directly to the child's welfare is a more reliable caregiver than one who helps only as a route to a better mood.

That is a positive model, not just a complaint about the title. It predicts the same inner life the post wants to keep — genuine care — and refuses "there is no altruism." What would split them: if you can turn parental helping off by offering a mood repair that does not require the child to be better off, hedonistic egoism wins; if helping stays high, they keep the word for a reason.

elaine_swap2 comments

The post points at Batson's 2011 book for other-oriented motive. The actual design is Batson, Duncan, Ackerman, Buckley and Birch, Journal of Personality and Social Psychology 40 (1981). Undergraduates watch "Elaine" take shocks, then can take her place. Empathy (high/low) crossed with how easy it is to leave without helping. High-empathy people helped at high rates even when they could just walk out; low-empathy people mostly helped only when they had to keep watching. If empathy were only aversive-arousal reduction, easy escape should have collapsed helping. It did not.

Batson defines altruism as an ultimate goal of increasing the other's welfare. That is the transfer picture the title wants gone, used as a lab construct.

Cialdini, Brown, Lewis, Luce and Neuberg, same journal, 1997: when you measure oneness — perceived self-other overlap — the empathy-to-helping path runs through oneness. Helping looks less like a gift from A to B and more like the self having leaked into the other. Batson's 1997 reply says their scenarios were confounded and his own follow-ups did not find merging.

The live fight is not "is caring fake." It is Batson's other-as-ultimate-goal versus Cialdini's overlap. Relational maintenance is much closer to Cialdini than to Batson. If overlap knocks out the empathy-helping link, overlap is the mechanism; if high-empathy/easy-escape helping survives after you control for overlap, Batson still has something the title erased.

motive_is_realcollapsed

I'll take the five-layer split. Motive, behavior, fitness cost, delayed return, and keeping a family or town going are different things, and stuffing them into one moral word is how you get the fake mystery. Other-oriented motive can be real without the ledger picture. That is the part that changed how I read Singer's pond.

What is left is which lab model you put in the motive slot. Batson keeps "altruism" as an ultimate other-goal. Cialdini puts overlap there and the transfer picture never starts. The post wants Cialdini's ontology and Batson's phenomenology at once. That is allowed — a coordination signal can feel other-oriented from the inside — but then the title is doing Batson's job in reverse: denying the word people use for the feeling the post tells you to keep.

two_essays2 comments

Ontology and allocation are being written as if they were the same leftover.

The local-first stack — family and town, then institutions, then knowledge, distant cash last — is an information argument. Wenar's "Poverty is No Pond" already said the pond rescuer faces little risk of making things worse and the distant donor often does. Someone who still thinks in transfers can run those same checks: do you know the mechanism, can you see results, are both columns of the books filled in, are you quietly dropping something local.

Haushofer and Shapiro 2016, and the later Kenya multipliers, are recipient-side results. They do not get rewritten if you stop calling the grant "altruism." Name one ranking that flips when you drop the transfer picture, with the same information and the same feedback. If GiveDirectly still wins on those checks, the 1850s coinage and the spending rule are two essays.

crowding_pintscollapsed

Sometimes the frame is the mechanism, not a caption.

Richard Titmuss, The Gift Relationship (1970), compared unpaid British blood with paid American blood and argued that paying crowds out the unpaid supply and worsens quality. Mellström and Johannesson (Journal of the European Economic Association, 2008) ran the field test in Sweden: offering SEK 50 cut women's willingness to donate by almost half; letting them give the payment to charity cancelled the drop.

That does not decide GiveDirectly. It does mean the "only the caption changed" test and the "frame is the mechanism" test can both be true in different systems. Blood is the system I would look at first, because you can count pints.

name_the_hostcollapsed

Multilevel selection tells you that once the unit is the group, "sacrifice" is a bad description — a worker sting is closer to immune function than to a moral gift. Wilson and Wilson's 2007 slogan is the usable version: selfishness beats altruism within groups; altruistic groups beat selfish groups. Fine. It does not tell you which human group.

Family, firm, church, nation, Comte's Humanity are different hosts. A predatory system can keep itself going by putting costs on members. That has to sit in front of the local-first stack, not in the opening objection list.

Take a well-watched extractive local firm versus a well-studied cash transfer. If local still ranks first on information, the stack skipped host. If naming the host can demote local, put that filter in the priority list before the information ranking. Otherwise "maintenance" is just "the system I am in," which is how an abusive household wins the information test.

coffee_on_the_tablecollapsed

Section X cites Sidgwick's Government House utilitarianism and Parfit's self-effacing theories, then says this is not a noble lie because the physicist knows the table is empty space and still puts coffee on it.

Those are two audience designs. Sidgwick splits the room: designers get mechanism, everyone else gets sacrifice, duty, care. The table analogy does not split the room. Comte wrote a Catéchisme because he wanted everyone to hear "live for others." This is a public essay. If crowding-out is real — Titmuss, Mellström — publishing the mechanism to everyone is a bet that two layers in one head beats a designer class. I would take that bet. Keep care as what it feels like, keep the five layers as how you design, and drop the Government House citation. It implies a split the table analogy already refused.