Two Concepts of EA

By Carolanne Jiang @ 2026-05-22T23:11 (+7)

This is a linkpost to https://carolannejiang.substack.com/p/two-concepts-of-ea

A lot of conversations evaluating EA — both internal and external — run on a slippage between two things that sometimes go by the same name. It seems quite important to me to separate the philosophical core of EA—to the extent that one can be identified and defined, and the social fact—as it exists as an empirical phenomenon. 

The points I want to make in this essay are pretty banal, but I think keeping them straight clears up a fair amount of confusion.

Consider the example of disputes about to what extent is EA utilitarian. Some arguments for and against it often go as such:

For:

• Citing someone that self identifies as an EA and also as an utilitarian
• The fact that Peter Singer self-identifies as a utilitarian
• Phrases such as “doing the most good”
• All of these discussions about aggregation, maximization, interpersonal cardinal comparability, et cetera.
• Caring a lot about the effectiveness of procedures via measuring outcome ≈ outcome based ethics ≈ consequentialism ≈ …
• Bentham’s Bulldog’s first name is “Bentham”…? Something like that
• et cetera.

Against:

• Citing someone that self identifies as an EA but does not self-identify as an utilitarian
• Will MacAskill does not self-identify as a utilitarian

• There exist paragraphs as such:

  Off the top of my head, here are some people who are not utilitarians and who might be interested in the question of effective altruism: A virtue ethicist cultivating the virtue of compassion. A deontologist doing supererogatory good deeds. An ethical egoist who knows that the warmfuzzies of truly helping someone is the best way to improve her own personal happiness. A Christian who knows that what we do unto the least of these we do unto him. A Jewish person who is performing tikkun olam. A Buddhist practicing loving-kindness. Someone who cares about fairness and doesn’t think it’s fair that they have so much when others have so little. A basically normal person who feels sad about how much suffering there is in the world and wants to help.

  —Ozy Brennan, You Don’t Have To Be A Utilitarian To Be An EA

• Wanting to do something and wanting to do so effectively is not necessarily utilitarian. ie. instrumental rationality does not seem to emit a clear moral orientation.
• In the 2015 and 2017 Rethink Priorities surveys, ~40-60% of respondents opted to answer the question “what moral philosophy, if any, do you lean towards?”; of which, ~50-60% selected Consequentialism (Utilitarian), which means, under rough estimations, only a quarter of EAs self identify as leaning towards utilitarianism. ¹
  • Of course self-identification with a term is meaningfully different from holding the view, of course, selection effect, and (insert social science words here)
• et cetera.

The list on both sides can go on. At this point, one can say some words about definitions, about social science, about practicality versus theory, about how lists end up hiding weighting, and about how you’re already pre-committing to a philosophical framework when you accept the existence of mere arguments, et cetera. We can dispute about any of these statements, but I hope you agree that something seems at least a little bit off here.

I acknowledge here that conversations about utilitarianism and EA do not always devolve into statements about definitions, or Wittgenstein, or infinite hypotheticals, or philosophical litigation over the metaphysics of hypotheticals, or something about something being peculiarly Western. But frequently enough they do, and they leave me pretty confused.

This phenomenon — disputes that loosely take this shape — relate loosely to a few other questions I sometimes encounter. Among them:

• is EA an opportunity or obligation?
• Is EA a normative framework, or no more than instrumental rationality?
  • Is EA a tool that helps you achieve what you want (maybe after extended, sincere reflection, eg. Coherent Extrapolated Volition or Idealized Reflection), or something that tells you what you should want?
  • Sometimes I want to describe some part of this with some term like a “wish fulfillment model of EA”; or better yet, through the piece whole hearted choices and morality as taxes

Some part of this kind of confusion, as it seems to me, comes from picking out one element of EA and treating a verdict on that element as a verdict on the whole. Another part seems to come from confusing the several ways we end up using the same term. On one hand, there is a philosophical core, to the extent it admits of a formulation. On the other hand, we have a social phenomenon (how EA manifests as a community, a field, a movement, or whatever definition strikes your fancy); the sort of thing you can do sociology on.² These are different kinds of objects, and they differ functionally. The philosophical core could function as a normative framework. The social phenomenon cannot, for it is not a set of claims.³

I don’t have entirely developed views on the philosophical core of EA as a normative framework: I'm uncertain whether a normative framework can reasonably be distilled from it, if so what it would look like, and whether I'd take it as my own guiding principle s— and of course there's plenty to say about the notion of "personal philosophies" in general. But I do think it's important that when we talk about EA, or argue about how utilitarian it is, we clarify which subject we're discussing.

1 I can supply the exact numbers later

2 There are also rhetorical moves such as ‘it’s really fundamentally a question’

3 There is also the ‘research field’ definition, which is probably a first candidate to be ruled out for potentially being a supplier of moral prescriptions, for everything you can say about facts vs. values in science.