by Tim Sommers
Honestly, in academic philosophy a great, or even a good, title is not really necessary. Perhaps, the most influential philosophy article of the twentieth century is called “On Denoting.” Really, Russell? That’s all you got? Compare Donald Davidson’s “A Nice Derangement of Epitaphs.” Now, that’s a title. Unfortunately, given that he argues there is no shared language, it has faded from view a bit. Thomas Nagel’s “What is it Like to be a Bat?” has got to be near the top of any “best of” list, especially considering its influence. But my favorite title, which I have borrowed for this article, is “Sometimes There Is Nothing Wrong with Letting a Child Drown” by Travis Timmerman.
I met Professor Timmerman a couple of years ago, a nice guy with no documented history of drownings, and asked him about it. (You don’t meet George Lucas and not ask him about Star Wars, right?) He told me that when the article is reprinted in textbooks and anthologies, it is consistently retitled “Reply to Singer.” Why?
Well, it is a reply to Singer. But really, I suppose, the point is to deemphasize the whole “children drowning” thing – which is too bad because it’s a real attention grabber. The article to which Timmerman is replying is the one in which Singer introduced a thought experiment, called “The Pond,” second nowadays only to the trolley problem in terms of how well-known it has become.
You are walking to work and pass a shallow pond. A small child has fallen in and is drowning. No one else is around. You could wade in and save her easily, but doing so will ruin your shoes and make you late to give a lecture. Almost everyone agrees, of course, that you must save the child; ruined shoes and a missed (or late) lecture are trivial next to a child’s life.
Singer builds a much larger argument on this shared intuition. If it is wrong to let the child drown to protect your shoes, it is equally wrong to spend money on luxuries like new clothes, restaurant meals, vacations, etc. while children elsewhere die preventable deaths that the same money, spent differently, could have saved them. Distance – spatial, temporal, or relational – he insists, is morally irrelevant. A drowning child ten feet away and a starving child ten thousand miles away place the same demand on us, provided our help would work equally well in both cases.
Singer’s actual principle is narrower than the pond alone suggests. If it is in our power to prevent something bad without sacrificing anything of comparable moral importance, we ought to do it. This is not full-blown utilitarianism (anything and everything we do should always maximize the overall, aggregate welfare of everyone with no threshold and no protected space for ourselves). The clause “without sacrificing anything of comparable moral importance” maybe builds in room for a career, a family, a modest life as long as the cost of keeping them isn’t trivial next to the good given up.
Timmerman’s central move in response is this. Forget one pond. Imagine thousands, or millions, of ponds scattered along your route, each with a drowning child in it which you could easily save if there weren’t so many of them, one after another, forever. This is not a fanciful exaggeration so much as a description of the world we actually live in. Nine million people die of malnutrition-related causes every year. (A quarter of a million people drown every year, but that’s not particularly relevant. Which is weird because all we keep talking about is drowning, but that’s philosophy for you.)
I am oversimplifying Timmerman’s argument quite a bit, but the point is that the single pond triggers our intuitions the way it does because it is singular. The many-ponds case asks what happens to those same intuitions once the rescue never stops.
At pond one, saving the child costs you wet shoes, which is trivial next to a life, so you must save her. At pond two, the same reasoning applies, and the same at pond three, and so on. At no single pond does the cost ever rise to “comparable moral importance,” because each rescue, evaluated on its own, still looks cheap next to a child’s life. Yet if you actually try to save every child at every pond, the cumulative toll is your career, your health, your family, your entire life redirected into an endless chain of rescues. Singer’s qualification is applied one rescue at a time, but the real cost of the policy only shows up in aggregate, and by the time it shows up, no single instance was ever the one that crossed the line.
That is Timmerman’s real target as I read him. A principle that is individually plausible at every step can be collectively unlivable, and nothing in the “no comparable importance” test, evaluated instance-by-instance, ever tells you where to stop.
This maps directly onto the case Singer actually wants us to accept. It’s not that any single donation to save a life would ruin your life. It’s that, if the argument holds for one dollar, it seems to hold again tomorrow, and the day after, and every day thereafter, until you have given away nearly everything you own that isn’t strictly necessary for survival. Global poverty isn’t one pond; it is millions of ponds, refilling every day. Timmerman’s point is that a principle built to handle the single, dramatic rescue may not have the resources to say when, or whether, an agent facing an endless procession of equally cheap, equally urgent rescues is permitted to stop.
Timmerman supplements the many-ponds argument with a second, related case. Suppose you pass the pond and see the child drowning, exactly as before, but this time you know with certainty that if you don’t jump in, a lifeguard arriving moments later definitely will. Your intervention isn’t necessary for her survival; it only determines who gets wet. Most people’s sense of obligation weakens sharply here. Saving her starts to feel pointless rather than heroic, since the outcome doesn’t depend on you.
Timmerman uses this to argue that part of what makes the original pond case compelling isn’t merely a bad outcome preventable at low cost. It’s that your particular action is the only thing standing between the child and death. Global poverty, he suggests, is sometimes structurally more like the redundant lifeguard than the original pond. Aid agencies pool resources from millions of donors, and no single contribution is strictly necessary for any particular child’s survival. But this case is best read as a companion to the many-ponds argument, not a replacement for it. It targets necessity, while many-ponds targets aggregation, and Timmerman needs both to unsettle Singer’s principle fully.
It’s worth remembering that Singer’s principle, even before Timmerman’s challenge, was never meant to collapse into utilitarianism. A doctor who could either fund a comfortable retirement or donate to prevent overseas blindness is, on strict utilitarian grounds, obligated to give until her own marginal loss equals others’ marginal gain, no exceptions. Singer’s principle instead lets her keep what protects things of comparable importance, things like her family’s basic security or her ability to keep practicing medicine at all. A teacher who tutors struggling students for free is, similarly, permitted to stop once further tutoring would cost her own health or her relationship with her own children, rather than being conscripted indefinitely the way utilitarianism would require.
None of this makes Singer’s qualification that wide a loophole. Taken seriously, “comparable moral importance” rules out most ordinary luxury spending: a $3,000 kitchen renovation, a car upgraded for looks rather than need, routine expensive dining, none of it comes close to the weight of a preventable death. This is why Singer’s own conclusion, even in its qualified form, is that most people in wealthy countries should be giving far more than a token amount.
And this is where the many-ponds argument, for all its force, has a real limit. It shows that an unbounded, exceptionless duty to rescue is unlivable in principle. It does not show that any particular act of giving you’re actually facing this year is one rescue too many. Most people in wealthy countries are nowhere near the point of exhaustion the many-ponds case describes. You could say they are, at most, at the very first pond, deciding whether to skip the kitchen renovation. The lifeguard case, likewise, shows that redundancy matters, but most real donations, to bed-net programs, direct cash transfers, targeted interventions, are not guaranteed to be replaced by someone else’s dollar if you don’t give; in most cases that specific good doesn’t happen.
Put together, Singer’s principle sits in an uncomfortable middle position. It is weaker than utilitarianism because it protects some space for the agent’s own comparably important interests. It is stronger than ordinary moral common sense because that protected space is far smaller than comfortable people assume.
And Timmerman’s many-ponds argument is the sharpest challenge to it, not because it shows the principle is false, but because it shows the principle, applied literally and endlessly, cannot say where the protected space ends before it swallows an entire life. Whether that shows the principle is wrong, or only that it needs a rule for aggregating cost across repeated rescues rather than assessing each one alone, is the real unresolved question Timmerman leaves us with.
The pond is a device for stripping away the excuses of distance and diffusion that let people ignore preventable suffering. Timmerman’s contribution is to show that some excuses, redundancy, and above all the sheer scale of need multiplying the same low-cost rescue without limit, are not always merely excuses, but real objections. We are nowhere near the exhausted end of Timmerman’s many ponds, and Singer’s weaker principle still asks more of us than most of us are currently willing to give.