Syndicate content

Excuse me Mr. Can’t you see the children dying?

Berk Ozler's picture

In 1997, Peter Singer wrote about a dilemma he’d pose to his students about a drowning child in a pond on their way to class: “would they be willing to save the child at the cost of getting all wet, having to go back home to change, and missing the first period?” After getting the expected answer that they all would, he’d ask about a hypothetical child far away, and ways that the students could save lives elsewhere at “no great cost – and absolutely no danger – to themselves. This would lead to a discussion of a version of effective altruism two decades ago.

Two versions of this classroom discussion practiced by our contemporaries in 2016 were written about recently. In January, Bruce Wydick, a respected colleague and a guest-blogger here previously, wrote about his adaptation, where the cost of saving the drowning child is the loss of your phone, which you don’t have time out of your pocket before jumping in. Then, however, he veered significantly from the original thought experiment to a decidedly more uncomfortable set-up: talking to the students about things they could do to reduce global poverty, he focused on GiveDirectly, talked to them about some of the evidence in this policy brief, and went for the punchline: for each of you who will give me your phones for two weeks (to be locked in a drawer), a donor will donate $50 to give directly. You can read the rest of it yourselves…Tim Ogden, at a seminar at his alma mater, upped the ante further by promising to give GiveDirectly $1,000 if 20 of the 60 students would give up their phones for two weeks – making it a all or nothing proposition and introducing a collective action problem in the mix.

As I was reading these (I read in reverse chronological order, as I saw Ogden’s post first, leading me to Wydick’s, and to the original from Singer), I (a) felt really uncomfortable, and (b) wasn’t even sure what the point was anymore. Peter Singer was challenging his students on the salience of distance, nationality, etc. and opening a debate about what one could do. Maybe I am missing something, but these recent examples seem more about shaming and peer pressure to have students give to a specific organization than causing a substantive discussion about effective giving.

Not sure whether I was alone in my discomfort with the whole thing, I asked a couple of friends for their opinions. I summarize their reactions below, but you’re probably wondering what my main worries were. I think there were three:

  1. The casual connection between the certainty of saving a child’s life and donating to an organization – even if it is to a very good one.
  2. The manipulation and the peer pressure: these are important topics that all of us, let alone unsuspecting students, need to think carefully about: what’s the point of having to decide on the spot, with little information (having to trust your professor’s or an alum’s word), with possible consequences of your public actions in your peer group?
  3. Finally, I worried about how these experiments would affect the students’ future actions with respect to effective altruism? The protagonists seem optimistic on this count, but I could easily see some people walking out of there, not wanting to donate to GiveDirectly anytime in the near future nor willing to listen to someone talking about hungry children in Africa. Without evidence, there’s no way to know…

My friends raised similar concerns, but also had some good suggestions:

"I think there are some interesting issues: jump in lake to save drowning child = reasonable certainty of saving a life. Donate to Give Directly = no evidence on mortality. Even the evidence we have is from a single impact evaluation. I think the original Singer thought experiment is interesting but it's a big jump to this implementation, especially when students don't have a chance to think through these arguments."

“...The problem in part is that there is no reason to take the phones except for shame/disutility. What's next, telling kids that I will donate money for everyone who agrees to get punched or agrees to eat food off the floor? After all, wouldn't you take a punch to save a child's life?”

“If they said, “Hey, your meal plan costs $10 for lunch. If you skip lunch today the university will buy a malaria net for a pregnant woman or buy 10 deworming pills or something similar then this would be a much more effective exercise.”

I found myself generally agreeing with these sentiments. Through our discussions, however, another useful thought emerged: at least if you’re conducting this experiment in class, you could put some more thought into its design and collect some follow-up data to teach us something useful. Who knows, maybe I am the one who is completely wrong: maybe, these students will go on to becoming much more effective altruists than they would have otherwise been or at least donate much more money to GiveDirectly or similar causes in the short run. Perhaps, however, there’ll be some negative and heterogeneous effects. Hold your nose, some might even end up sending unused t-shirts to a poor country or do voluntourism: are you willing to take that responsibility? I am not encouraging anyone to conduct this type of experiment, but, if you do, at least do us the favor of showing its impact…

And, if you’re an unsuspecting economics student and see one of these experiments coming during class, feel free to walk away: we at Development Impact have your hall pass right here…

P.S. A tangent here: Peter Singer's ideas of nationality and distance being morally irrelevant in your decision to help can and should extend to species. If we are to think about doing the most we can to reduce the total amount of suffering on earth, it does not make sense to limit ourselves to thinking about how to help humans only. Unless, that is, you subscribe to human exceptionalism: billions of farm animals suffer needlessly every day. Effective Altruism lists Animal Charity Evaluators as one of the organizations “founded by people aiming to have the great possible impact and who identify with the EA approach.” ACE, in turn, highly rates organizations like Animal Equality, FARM, The Humane League, and Vegan Outreach, and a few more. Next time you’re thinking about the most effective way to reduce suffering, I suggest considering these organizations as well…



Submitted by Timothy Ogden on

As I noted when Berk tweeted about his discomfort in response to my write-up of my version of the "experiment", I think it is totally fair to be uncomfortable. I'm still not sure how I feel about it.

Well I know that I'm uncomfortable. I'm uncomfortable when I think about the consumption choices I make and don't make in the context of global inequality. When I originally read Singer's set-up--which he does use to advocate giving to specific organizations by the way--my reaction was to reject the argument because I had no confidence that the organizations he was recommending money be given to (such as OxFam) would have the desired effect. GiveDirectly makes that knee-jerk rejection of the philosophical argument much more difficult. The intent of what I did was to get students to wrestle with the discomfort in a tangible way.

I'll write up a more complete response on Medium and share the link. Note that I am working with GiveDirectly to anonymously track future donations.

Submitted by Frances Kissling on

yes, what people were asked to give up -phones - was not an appropriate ask. The purpose of the drowning child exercise is that one gives up something of financial value, as that $$ not spent on a non-necessity can be spent on charity. It is the "live lightly on the planet" part of Singer's message and is central to the theory. Second, the major flaw in the thought experiment from an effective altruism view - and Singer's - is that neither Singer nor EA are specifically interested in saving children's lives. They are interested in ending extreme poverty and suffering. The number of lives you save is only part of the measurement of ending poverty and sometimes does not end poverty and actually increases suffering. (Not suggesting we let anyone die for that reason, just that we don't overstate it as the major impact on poverty and suffering. )

Submitted by Garth Luke on

Also isn't giving up your phone for a period not just the loss of a recreational device but also an essential tool for a whole range of social responsibilities such as contact with work, family etc. I for one would rather give the cash directly than be without a phone for two weeks.

Submitted by Rick on

I think it's also interesting to consider the implications of this experiment for students who are having trouble in their personal life. Imagine you are suffering from anxiety or depression - your phone helps keep you in touch with others, it has real utility in your life. Then your class does this experiment, and you see your peers start handing in their phones. You need your phone more than they do, but now you feel like a terrible person (keep in mind that in this example the person is suffering from depression and/or anxiety), and your anxiety and/or depression kicks in.

By its nature, this experiment was designed for students with privilege, but what happens when less-privileged students are asked to participate?

Add new comment