The Child in the Pond: How a Three-Page Essay by Peter Singer Launched a Global Movement — and Why the Question of How Far Our Duties Reach Echoes All the Way into the Boardrooms of Tech Billionaires

The Hook: Your Shoes or a Life

Imagine you are walking to work along your usual route, past a shallow pond. Suddenly you see that a small child has fallen in and is drowning. The pond is shallow enough that you could easily wade in and pull the child out. But you are wearing expensive shoes and freshly pressed trousers; if you step in, you ruin both, and you will be late for an appointment. The question: may you walk on?

Almost everyone who hears this story answers without hesitation: of course not. To weigh a pair of shoes against a child's life would be morally monstrous. Anyone who walks past while a child drowns before their eyes has done something deeply wrong — and no amount of tidy trousers can outweigh it.

It was precisely this seemingly harmless intuition that the Australian philosopher Peter Singer turned, in 1972, into the lever for one of the most influential and uncomfortable arguments in modern moral philosophy. In his essay Famine, Affluence, and Morality, written under the impression of the famine among Bengali civil-war refugees in 1971, he posed a simple but momentous follow-up question: if you are obligated to save the child in the pond even though it costs you your shoes — why are you not obligated to save the child in a distant country who is dying of hunger or malaria, whom you could save just as reliably with a donation, even though that too costs you barely more than your shoes?

Two generations later, a global movement grew from that question: effective altruism. It promises to free charitable action from warm feeling and subject it to the cool logic of effectiveness: how, with a given sum, do you save the most lives? This article takes you on the full journey — from Singer's precise argument through its radical consequences, the birth of a movement with billion-dollar budgets, the turn toward concern for humanity's distant future, the sharpest philosophical objections, all the way to the spectacular crash when one of its most prominent funders ended up in court as a fraudster. At the end stands no comfortable answer, but a question that can no longer be shaken off.

Part 1: The Anatomy of the Argument

The Three Premises

Singer's argument is so powerful precisely because it requires no exotic assumptions. It gets by with one empirical observation and two principles that almost no one seriously wants to dispute.

The empirical premise runs: suffering and death from lack of food, shelter, and medical care are bad. That is hardly controversial.

The strong principle, in Singer's own formulation, runs: "If it is in our power to prevent something bad from happening, without thereby sacrificing anything of comparable moral importance, we ought, morally, to do it." Note the calibration: not "without sacrifice," but "without a sacrifice of comparable moral importance." The ruined shoes are a sacrifice — but they are not a sacrifice morally equal to the death of a child.

The conclusion: if both premises hold, then affluent people are morally obligated to donate a substantial portion of their income to prevent avoidable suffering in the world — and to keep doing so up to the point at which further giving would cost them something of comparable moral importance. Not as praiseworthy generosity ("supererogatory," that is, going beyond duty), but as plain duty.

This last shift is the real charge of dynamite. Singer attacks the deeply anchored distinction between duty and charity. In ordinary moral common sense, giving counts as meritorious but voluntary: whoever gives is a good person; whoever does not give does nothing wrong. Singer disputes precisely this zone of the voluntary. If the child-in-the-pond principle is correct, then not giving in the face of avoidable death is not morally neutral but as wrong as walking past the pond.

The Two Screws: Proximity and Numbers

Why does the distant case feel so different from the near one? Singer identifies two factors that distort our intuition without being morally load-bearing.

The first is spatial and personal proximity. We feel a stronger duty toward the child we see than toward the one we merely read about. But Singer asks: why should mere distance matter morally? That I do not see the child in Bengal makes its suffering no less real and my capacity to help no smaller. Proximity produces a stronger feeling, but not a stronger claim. In a globalized world, where a bank transfer reaches an aid organization on the other side of the earth in seconds, geographical distance has become morally irrelevant.

The second factor is the number of others who could also help. At the pond I may be the only one — but with global hunger there are millions of other affluent people who could help just as well. Doesn't that distribute my responsibility? Singer exposes this thought as a comfortable excuse. First, the mere existence of other possible helpers changes nothing about what actually happens: if the others do not help, the child dies, no matter how many could theoretically have helped. Second, the same argument applies to everyone equally — if each points to the others, no one helps. The number of potential helpers does not diminish one's own duty.

The Strong and the Moderate Version

Singer himself was aware of the radicalism of his strong version. It demands giving up to the margin of marginal utility: up to the point at which, by giving more, I harm myself as much as I benefit others. Followed consistently to its end, this would push most affluent people to a level near subsistence — a life almost no one leads voluntarily.

For this reason Singer offered a moderate version that presupposes only a weaker principle: "If it is in our power to prevent something bad from happening, without thereby sacrificing anything morally significant, we ought to do it." Here it suffices that what is sacrificed is not itself morally important — say, the purchase of a second pair of fashionable shoes or an expensive dinner. Even this weaker version, Singer emphasizes, demands far more than most of us actually give. It does not require subsistence, but it draws the line of the morally permissible far below what counts as normal in affluent societies.

Key point. Singer's argument neither stands nor falls with utilitarianism. You need not be a utilitarian to admit that you must save the child in the pond. Therein lies its force: the principle sounds modest, yet it leads to a conclusion uncomfortable for common sense.

Part 2: From Argument to Movement — Effective Altruism

The Second Ingredient: Effectiveness

Singer's essay answers the question of whether we must help, but not the question of how we can best help. This second question became the core of a movement that formed around 2011 at the University of Oxford and gave itself the name effective altruism. Its guiding idea: if we are going to give, then we should direct the money to where it does the most good — not to where it feels best or where the advertising is most moving.

That sounds obvious, but it is not. Different charitable projects differ in effectiveness not by percentages but often by orders of magnitude. A guide dog for a person in a rich country costs tens of thousands of euros; for a fraction of that, blindness from trachoma can be prevented in numerous people in poorer countries. Effective altruism demands facing this uncomfortable arithmetic: equal compassion, but very unequal benefit per euro.

GiveWell and the Measurement of the Good

The methodological heart of the movement is the charity-evaluation body GiveWell, which assesses programs not by their administrative costs or their good reputation but by demonstrable impact per donated dollar. The most famous metric is the estimated sum required to save a human life.

According to GiveWell's estimates for the 2022–2024 period, the organization directed funds to the Against Malaria Foundation (which distributes insecticide-treated mosquito nets) at an average cost-effectiveness of roughly $5,500 per life saved. Across various programs and regions, GiveWell puts the cost of averting a death at roughly $3,000 to $8,000, depending on location and circumstances. These figures are estimates with substantial uncertainties, not accounting-precise prices — but they make Singer's abstraction tangible. The sum a Central European household casually spends on a weekend corresponds, on this reckoning, to a measurable fraction of a saved life.

The Institutions: Giving, Earning, Advising

An ecosystem of organizations grew up around this idea. Giving What We Can, founded in 2009 by the philosophers William MacAskill and Toby Ord, asks members for a public pledge to donate at least ten percent of their income to effective charity. The Centre for Effective Altruism (2012, likewise MacAskill and Ord) became the organizational umbrella. 80,000 Hours — named after the approximate length of a working career — advises young people on how to do the most good with their careers.

From 80,000 Hours also came the movement's most controversial concept: "earning to give." The reasoning: someone who, instead of a direct helping profession, takes a highly paid job in finance or technology and donates the bulk of the salary may under some circumstances do more good than through their own on-the-ground work — because the donated money can fund several well-paid specialists in aid projects. This logic is effective altruism thought through to its end — and, as we shall see, it would inflict severe damage on the movement's reputation.

Part 3: The Turn to the Distant Future — Longtermism

From Near Suffering to the Whole Future

During the 2010s, an influential part of the movement performed a remarkable shift. If morality is about doing the most good, and if spatial distance is morally irrelevant — why then should temporal distance count? A person who will live a thousand years from now is just as much a person as one living today. And since potentially astronomically many people could live in the future if humanity survives, the greatest lever for good might lie in securing that future at all.

This line of thought, given the term longtermism by MacAskill and Ord around 2017, MacAskill defines in two degrees of strength: longtermism is "the view that positively influencing the long-term future is a key moral priority of our time"; strong longtermism makes it the key priority. Toby Ord's book The Precipice (2020) supplied the dramatic foundation: Ord estimates the probability that humanity will suffer an existential catastrophe within the next century at roughly one in six — a moral game of Russian roulette. This estimate is dominated by human-made risks; the risk from an unaligned artificial general intelligence alone Ord puts at roughly one in ten, higher than all other sources combined.

The Reshuffling of Priorities

This shift was not merely theoretical. It redirected real funds. From about 2014 onward, the Open Philanthropy Project increasingly financed projects to reduce existential risk; by 2022 a substantial share of EA funding (estimates cite more than a third) flowed into longtermism and existential risk — AI safety, biosecurity, pandemic prevention — rather than into the classic anti-poverty programs with which the movement had begun.

Here opinions divide, even within the movement. The transition from the measurable child in the pond ($5,500, one life saved, verifiable) to hypothetical trillions of future people (barely quantifiable probabilities, centuries in the future) transforms an initially empirically disciplined movement into something far more speculative. I am of the opinion that precisely this tension — between the hard, verifiable effectiveness of poverty relief and the soft, assumption-dependent calculation of the distant future — marks the deepest internal fault line of effective altruism. Both invoke the same principle ("do the most good"), but they pull it into opposite epistemic worlds.

Part 4: The Major Objections

Singer's argument and the movement that grew from it have provoked an equally large countermovement. The most important objections do not strike peripheral questions but the core.

The Demandingness Objection

The weightiest objection is the demandingness objection: a morality that demands giving up to the margin of marginal utility overtaxes the human being and leaves no room for a life of one's own. Its classic formulation comes from Bernard Williams in his Critique of Utilitarianism (1973). Williams argued that the relentless consequentialist demand for the optimal outcome grinds down the personal "ground projects" — the bonds, undertakings, and passions that give a life meaning in the first place. Such a morality forces the human being to regard his own life only instrumentally, as a mere vehicle for the promotion of impersonal value.

Williams coined the famous image of the "one thought too many": if a man standing before his drowning wife and a drowning stranger must first consult his moral theory as to whether it is permitted to favor his own wife, then he has had one thought too many. The immediate affection for the wife is not a justification that must be checked for its compatibility with impartial morality — it is a reason in itself. A morality that summons even the love for one's own person before the tribunal of impartiality misses, on this view, something essential to human life.

Susan Wolf's "Moral Saints"

Related but independent is the critique of the philosopher Susan Wolf in her essay Moral Saints (1982). Wolf asks provocatively whether a person who devotes their life entirely to maximizing the moral good — a "moral saint" — is even a desirable ideal at all. Her answer is a clear no. A world in which there were only moral saints would be a world without master musicians, without top chefs, without Wimbledon athletes, without great novelists — for all these forms of excellence demand time, devotion, and resources that the saint would inevitably withdraw from charity. No one really wants this world, says Wolf, and it would be downright crazy to want it. A fully flourishing human life comprises more than moral perfection; the non-moral goods — beauty, skill, knowledge for its own sake — belong to it inalienably.

The Critique of the Utilitarian Foundation

A third strand doubts the calculative foundation. If the sole criterion is maximized aggregate benefit, then the logic can lead into uncanny regions: a system that yields excellent results on average can turn out catastrophically for a small group — the classic concern that separates utilitarianism from rights-based thinking and that also runs through the trolley problem. Effective altruism's aggregating reckoning stands squarely in this tradition. Its defenders reply that in practice the movement precisely does not violate human rights but simply directs donations; its critics counter that the mindset — people as interchangeable bearers of utility, actions measured only by outcomes — opens a dangerous slope, at the foot of which stands the argument that the end justifies the means.

The Movement's Reply

The defenders of effective altruism take these objections more seriously than caricatures suggest. To the demandingness objection they respond with the moderate version and Giving What We Can's ten-percent norm: a fixed, livable threshold instead of the crushing margin of marginal utility. They emphasize that a morality does not become false by being demanding — after all, earlier generations regarded the abolition of slavery as "unrealistically demanding." And they point out that the movement has saved real, verifiable lives while the philosophical subtleties were being debated. The dispute remains open; but it no longer turns on whether we should help, but on how much and in what way.

Part 5: The Fall — the FTX Scandal

No event has shaken effective altruism as much as the rise and fall of Sam Bankman-Fried. The founder of the crypto exchange FTX was the poster child of "earning to give": he had publicly declared that he wanted to acquire his fortune in order to donate almost all of it to longtermist causes, and to parts of the movement he counted as a model of how wealth and charity could be combined strategically. The FTX Future Fund that he financed became one of the largest funders of longtermist projects.

At the end of 2022, FTX collapsed. It came to light that Bankman-Fried had, in all probability, committed fraud on a gigantic scale — customer funds had been misappropriated. The EA community reacted with shock and anger. Dustin Moskovitz of Open Philanthropy conceded that the movement needed a clear story that rejects the principle "the end justifies the means." For precisely this suspicion now hung in the air: had the utilitarian logic — maximize the good, cost what it may — given a clever young man the intellectual permission to override the rules, because he intended to use the proceeds for a higher purpose after all?

Peter Singer himself suggested the scandal might lead the movement to reconsider its relationship with billionaires and with the crypto world. The episode became a real-life object lesson in the demandingness and the utilitarianism objections at once: an ethics that looks only at outcomes and slights the integrity of the action — Williams's "ground projects," the attachment to rules and promises — is vulnerable to being hijacked by someone who massages the calculation in their own favor. Whether this was a flaw of the idea or of its abuse by a single person is disputed to this day. But the movement that had set out to calculate the good had to learn painfully that trust and integrity cannot be entered on the same balance sheet as saved lives.

Part 6: The Enduring Challenge

What remains once you set aside the movement, its billion-dollar budgets, and its scandal? Singer's argument remains — and it is astonishingly hard to refute. Most people who encounter it for the first time instinctively try to find a difference between the child in the pond and the child in the distance that would justify their own inaction. Almost all of these differences — the distance, the number of other helpers, the uncertainty about impact, the absence of a direct relationship — crumble on closer inspection into psychological factors that explain why we feel differently but do not justify why we would be permitted to act differently.

The most honest way of dealing with the argument perhaps lies not in refuting it but in negotiating its reach. One can reject Singer's strong version without disputing the moderate one. One can grant Williams that a life without projects of one's own is no longer a human life, and grant Wolf that the moral saint is not an ideal — and still admit that most of us remain far below what even the weakest version demands. The interesting question is then no longer "Must I give everything?" but "Why do I give so much less than I easily could?"

The Central Takeaway

If you want to take one single practical lesson from Singer's essay, let it be this: test your excuses by bringing the case closer. Whenever you catch yourself justifying a failure to help by appeal to distance, uncertainty, or the responsibility of others, ask whether the same excuse would hold at the shallow pond. Would you walk on because the child is far away? Because someone else might come along? Because you are not entirely sure your intervention really helps? If the excuse would sound ridiculous at the pond, it is ridiculous in the distant case too.

From this follows no compulsion to sainthood, but a modest yet concrete shift. You need not give down to subsistence. But you can choose a fixed, livable threshold — the ten percent popularized by the movement is one possible anchor — and you can take care that your giving is effective, not merely well-meant: that it flows to where it prevents the most suffering per euro. Effective altruism may be assailable in its speculative offshoots and in its billionaire problem. But its sober core — that equal suffering counts equally, no matter where it occurs, and that compassion without effectiveness remains half-hearted — has outlasted the fall of its most prominent funders.

Reflection Question

Singer claims that a sufferer's distance does not diminish our duty. Almost everyone assents to the abstract thesis — and simultaneously lives as though it did not hold. Suppose you accept the argument with your head but continue to feel the stronger obligation toward those near you: is this gap between insight and action a moral failure you ought to remedy — or is partiality toward the near, as Bernard Williams holds, itself a legitimate, inalienable part of a good human life? And if both are true at once: where exactly would you draw the line between the duty you acknowledge and the life you allow yourself?

Cross-References in the Vault

• The Switch in the Machine: The Trolley Problem and the Ethics of Autonomous Vehicles – the fundamental opposition between utilitarian aggregation and rights-based inviolability that also shapes Singer's aggregating logic.
• The Billions Barely Alive: Parfit's Repugnant Conclusion and the Ethics of Future Generations – the distributive ethics of future generations, philosophical foundation of longtermism.
• The Child in the Road: Moral Luck and the Limits of Responsibility – how much our moral judgment depends on chance: the flip side of Singer's question about the reach of duty.
• Behind the Veil of Ignorance: How Rawls Reinvented Justice – and Why His Thought Experiment Still Echoes in the Ethics of Artificial Intelligence – Rawls's inviolability of the person as the sharpest counterposition to utilitarian aggregation.
• Meant or Merely Foreseen: The Doctrine of Double Effect and the Power of Intention – another building block of an ethics beyond pure consequentialism, in which intention stands against mere outcome.
• Losses Loom Larger: How Prospect Theory Dethroned Rational Man – and Why Its Most Famous Principle Is Now on Trial Itself – why we decide, psychologically, differently from what the cool calculus of utility suggests.

Sources

1. Singer, P. (1972). Famine, Affluence, and Morality. Philosophy & Public Affairs 1(3). Overview, argument structure, and historical context (Bangladesh 1971): Famine, Affluence, and Morality (Wikipedia); study guide with premises and the strong/moderate versions: Study Guide (Utilitarianism.net); full-text framing: Giving What We Can
2. On the effective altruism movement — history, principles, criticism: Effective altruism (Encyclopædia Britannica); on its emergence: History of effective altruism (EA Forum)
3. GiveWell on the cost per life saved (February 2024 version, Against Malaria Foundation ~$5,500, range ~$3,000–8,000): How Much Does It Cost to Save a Life? (GiveWell); Against Malaria Foundation (GiveWell); 2024 metrics: GiveWell's 2024 Metrics and Impact (GiveWell Blog)
4. On longtermism — definition, MacAskill/Ord, allocation of funds: Longtermism (Wikipedia); Longtermism (Centre for Effective Altruism); on MacAskill's program: What We Owe the Future (Wikipedia)
5. Ord, T. (2020). The Precipice: Existential Risk and the Future of Humanity. Bloomsbury. (Existential risk ~1 in 6 this century; AI risk ~1 in 10): The Precipice (Wikipedia)
6. On the demandingness objection and Bernard Williams's "one thought too many": Demandingness objection (Wikipedia); The Demandingness Objection (Utilitarianism.net)
7. Wolf, S. (1982). Moral Saints. The Journal of Philosophy 79(8). Summary and framing: Course Notes – Susan Wolf, "Moral Saints" (The Electric Agora)
8. On the FTX scandal and its consequences for effective altruism: Sam Bankman-Fried (EA Forum); Effective Altruism After Sam Bankman-Fried (Seven Pillars Institute)

← All articles