Concept

Effective Altruism

philosophyethicseffective-altruism
Teach me this →

Effective Altruism

Effective altruism (EA) is the project of using evidence and careful reasoning to work out how to benefit others as much as possible, and then acting on it. It treats doing good as a question with better and worse answers — one open to the same scrutiny a scientist or an investor brings to their work — rather than a matter of taste or sentiment.

The movement grew out of moral philosophy, above all the utilitarian tradition that asks us to weigh everyone’s wellbeing impartially. Its practical claims are that the best charitable interventions can be hundreds of times more effective than the median, that a person in a rich country can do an extraordinary amount of good cheaply (the classic example: preventing deaths from malaria for a few thousand pounds), and that where you can help is no reason to prefer it — distance and nationality carry no moral weight.

The core moves

  • Impartiality. Each person’s suffering counts equally, wherever or whenever they live. Peter Singer’s drowning-child argument presses the point: if you would ruin your shoes to save a child in front of you, distance cannot explain why you would not give to save a stranger far away.
  • Marginal thinking. Ask not ‘is this cause good?’ but ‘where does the next pound do the most good?’ — the same logic of marginal returns an economist uses.
  • Cause neutrality. Start from the goal (do the most good) and let it select the cause, rather than starting from a cause and rationalising it. This has pushed EA from global health toward animal welfare and, more controversially, toward the long-run future.
  • Longtermism. William MacAskill’s extension: if future people matter morally and there could be astronomically many of them, then positively shaping the long-term future — reducing extinction risk, steering transformative technologies — may be among the most important things we can do.

Where mainstream views differ

Critics charge that the impartial calculus is too demanding, crowding out the local loyalties and personal projects that give life meaning; that quantifying good biases EA toward the measurable and against the systemic or political; and that longtermism’s vast expected-value sums can justify almost anything, licensing present harm for speculative future benefit. The 2022 collapse of FTX — whose founder professed EA motivations — sharpened the worry that the framework can rationalise reckless or self-serving conduct. Defenders reply that demandingness is a feature of taking ethics seriously, and that EA’s record in global health is concrete, audited, and large.

Peter Singer himself answers the demandingness charge on consequentialist rather than heroic grounds: the saintly standard — give until you are as badly off as those you help — is self-defeating, because an ask pitched that high makes people abandon the effort altogether. The right demand is the one that maximises total giving, which he takes to be a modest, progressive scale (a percent or two at ordinary incomes, rising towards a third for the very rich) rather than the impartial maximum. On this view demandingness is not a bullet to bite but a design problem — calibrate the ask to human nature, since an unmet ideal raises no one’s welfare. He grounds the underlying impartiality on the untrustworthiness of moral intuition: our partial intuitions ‘have arisen in various ways that suggest that they are not likely to be tracking moral truth’. See Peter Singer on Sentience, Effective Altruism, and the Ethics of Doing Good.

In the wiki