The Drowning Child Argument Is Simply Correct

Failure to donate to effective charities is like walking past drowning children and doing nothing

(I’d appreciate if you could share this article—I think, for obvious reasons, it’s important that people hear the ideas. Occassionally my articles convince people of things, and convincing people to take the Giving What We Can pledge saves like a hundred lives, so it’s worth trying to get this article out there. Restacking it is an easy, low-cost way of doing this!)

Imagine you were walking past a drowning child. The child kicks, screams, and cries, as they drown, and are about to be resigned to a watery grave, when you walk by. You can save them if you jump into the pool and pull them out. But doing so would come at a cost. You’re currently wearing a very expensive suit—about 5,000 dollars—or perhaps your suit is cheap but has a 5,000 dollar bill in your pocket that would be ruined if you save the child (it’s a very deep pocket—you can’t pull it out in time). Clearly, in such a case, even though it would cost significant money, you’d be obligated to jump into the pond to save the child.

Long before I was born, Peter Singer argued that this case shows that we have an oblgliation to donate to effective charities. The best charities—which you can, at any time, donate to—save lives for a few thousand dollars each. Just as you’re obligated to sacrifice a few thousand dollars to pull the child out of the pond and save them, you’re obligated to sacrifice a few thousand dollars to save a far-away child who would otherwise die of malaria.

In Singer’s original formulation, he used the drowning child case to support the following principle: if you can prevent something very bad from happening without sacrificing anything of comparable moral value, you should do so. For example, if you can prevent a woman from being raped or a person from being murdered at the cost of 700 dollars, you should do so, because averting rape and murder is more valuable than 700 dollars. From this, he deduces that you should give your discretionary spending to effective charities. You shouldn’t spend 5,000 dollars going to Hawaii when you could instead save a person’s life.

But I think we can make a much more direct argument: failure to give to effective charities is morally equivalent to walking past drowning children. Therefore, you have an obligation to give to effective charities, just as you would have an obligation to pull drowning children out of ponds (it seems this is how everyone has, in the intervening years, interpreted Singer’s argument, even though it’s not what was originally intended).

In both the case where you pull the child out of a pond and the case where you donate to effective charities, you can avert a death at the cost of just a few thousand dollars. This seems to be the salient feature of the situation—the reason to wade in and save the child seems to be that you can save a life at small cost. The alternative is to come up with some gerrymandered explanation of why you should save the child from the pond but not from malaria, but that’s less plausible than the simple account that you should prevent terrible things from happening if you can do so at comparatively minor cost.

Still, lots of people argue that there are important differences between pulling kids from ponds and donating to, say, the Against Malaria Foundation. Let’s address them.

The most common claim is that there’s a difference in terms of proximity. You are only obligated to save the child because they’re near you—if they were far away, you wouldn’t be obligated to save them. This account suffers from two problems: it’s false in a first way, and it’s false in a second way.

First of all, the idea that you’re only obligated to save people who are near you is crazy. Imagine that you could wade into the pond to press a button that would save a child from drowning who was far away. Clearly, you should still do that. But in that case, there’s as much lack of proximity as there is when you donate to effective charities.

Second of all, proximity—at least in the sense of someone being physically close to you in space—is obviously not morally important. Suppose that a child is drowning in a plane and it costs money to press a button that would save them from drowning. Would your reason to save them decrease as they recede into the distance—as they get farther away? Is your obligation to save aliens within one galaxy of you much stronger than your reason to save aliens within two galaxies of you? No, that’s crazy! It doesn’t get less important to save people simply because they’ve taken planes far away.

When claiming that proximity refutes the drowning child argument, lots of people like to say is that you have a great obligation to your friends and family. I don’t know what prompts them to say this in response to the drowning child argument, as it has nothing to do with the argument! Even if you have special obligations to your friends and family, your reasons to save drowning children that you don’t know are still equal to your reasons to save kids you don’t know who might get malaria. The drowning child is not your child—they’re a child that you don’t know personally.

People often claim that you have a greater obligation to those in your own country than to foreigners. I’m doubtful of this, but let’s grant it. Now imagine that you’re on the Mexican border and see a drowning child. They’re not a member of your country. Nevertheless, you should wade in the pond and save them, even at the cost of an expensive suit. Failing to give to effective charities, I claim, is like ignoring the drowning Mexican child—even though they’re not part of your country, you still have an obligation to save them.

Additionally, it’s often claimed that there’s an important difference in that in the drowning child scenario, you’re the only person who can save them, while when giving to charity, others can save them to. I’ve always found this idea super weird: your reason to save people doesn’t evaporate just because other people aren’t following their duty to save people. We can see this by imagining in the drowning children that there are a bunch of nearby assholes ignoring the child as he drowns. Does that eliminate your reason to save the child? No, obviously not. But this case is, in terms of other people not acting to save the child, analogous real world charitable donations.

The final consideration—and this one is the only that bears any weight—is that there are many drowning children. Imagine that there wasn’t just one drowning child, but hundreds of thousands—you could never save them all. It’s plausible that you wouldn’t be obligated to spend your entire life saving children, never enjoying things.

The main thing to note about this is that even if it’s right, maybe it means we’re not all required to spend all of our time saving children, but it still means we’re required to do a lot. A person who never saved even a single drowning child, who ignored the cries of every child who drowned, would be monstrous. So while perhaps you don’t have to give all your money to effective charities, accepting this reasoning would still mean you have an obligation to make charitable giving a main part of your life—say by giving a significant share of your income to effective charities.

I’m also dubious that this justifies spending money on luxuries. In a world where kids were constantly drowning, it doesn’t seem justified to, say, spend thousands of dollars on vacation when you could instead save a child. A child’s life is just so much more important than a trip to Europe. Your reason to save a child doesn’t depend on how many previous children you’ve saved—or so it seems. If I can’t remember whether I lost 10,000 dollars yesterday saving drowning children or gambling, it doesn’t seem I need to figure out which of these I did to decide whether I should save a drowning child. But if we accept this principle, that whether you previously spent your money on saving children or doing other stuff doesn’t affect whether you should currently spend your money on saving children, then your reason to save children is the same as it would be if you hadn’t saved any children. But clearly, if you were choosing between saving a child from a pond and going on vacation, and you hadn’t saved any children, you’d be obligated to save the child. It follows then that you have an obligation to save a child if the alternative is going on vacation.

This argument has, since I’ve heard it, struck me as obviously, irrefutably correct. We certainly have an obligation to make saving children—when we can save hundreds at comparatively minor cost—a significant life project. If a person can save a life a year, without majorly jeopardizing their welfare, just by tithing to effective charities, failing to do so seems clearly immoral.

If you’re convinced by this, I’d encourage you to take a Giving What We Can pledge (if you do this in response to this article, I’ll provide a free paid subscription) or give to GiveWell charities. Most people are, inadvertently, doing things as bad as walking past drowning children. We have significant reason to stop doing this.