I sometimes give money to homeless people when I see them around London. Not always - sometimes I don’t have any change on me, and sometimes I do have change but don’t give it away for whatever reason (maybe because I’m in a hurry, or maybe because I want to use to change to buy something from the shop and I’m too selfish to forego my purchase, or maybe something else). But I probably do give money around 50% of the time I see someone begging next to a supermarket or a bus stop. Imagine that I made a decision: every time I saw a homeless person and felt the urge to give them money, I would refrain from doing so, and instead make a mental note of the amount I was going to give them, and gave it to the Against Malaria Foundation, or the Maximum Impact Fund, or any number of effective charities that use interventions that have been tested rigorously and are known to save lives and reduce suffering. Is this morally right? Wrong? Neutral? Note: To avoid irritating arguments in the comments about whether it’s right to give to homeless people at all because they may buy drugs, alcohol, and so on, let’s assume that all homeless people in this scenario spend their money on shelter and warm food. Let’s also assume that giving money to these charities really does save more lives and reduce more suffering than giving the money to homeless people in the UK/USA/Europe.
The reactions I’ve heard to this question when I’ve posed it to people are pretty interesting. Scott Alexander has a short story about someone who creates an AI that is able to generate controversial statements that seem obviously, trivially true to some people, and obviously, trivially false to others, causing them to argue for hours on end without even understanding how people on the other side can hold their view. The responses I get to this question reminds me of that story - some people say it’s obviously wrong to see a homeless person, deny them money, and then give that £3 or whatever it is to AMF. Why can’t someone give the £3 to both AMF and the homeless person?
Other people say that it’s obviously right - any amount of money that I intend to give to a homeless person would do more good given to people in Sub-Saharan Africa to buy malaria nets or provide Vitamin A supplements. Giving the money to both the homeless person and the charity doesn’t solve the problem, because as long as I am giving £X to the homeless person, that money would always be better spent at the margin by donating it to an effective charity. To those people, to say that I ought to forego giving some amount of money to an effective charity so that I get warm fuzzies by giving it to a homeless person isn’t morally right, it’s selfish!
At first I thought this was basically just a test to pick out who is a consequentialist. But I’m not sure that’s right - I think of myself as a consequentialist but still probably side more with the people who say you ought to give money to the homeless person, although I’m pretty sympathetic to both arguments. I sort of want to play a get out of jail free card and say that I think that not giving money to the homeless person has hidden negative consequences - to stop seeing people as human beings and start quantifying the good you can do in this way can might numb you to suffering in such a way that you become less likely to help people at all. But we can just stipulate away this problem - suppose that we know with certainty that redirecting the money away from donations to the homeless and towards people in Africa has better consequences - that the donor will continue to agonise over this decision and will become no less receptive to suffering than he/she was before? One way we might be able to know this is by asking about a past person - imagine we heard of a historical figure who decided to stop giving money to the homeless and to give that money to effective charities, and continued to do so all his life and never became less receptive to suffering - was that person’s decision morally right?
There’s also another interesting thing I’ve noticed here, which is that some people seem to make the claim that doing this mental calculation and giving money to an effective charity is worse than giving no money at all. People have said that it’s perfectly understandable to walk past a homeless person and not give them any money (as I do about 50% of the time), but it’s not acceptable to make a note of their presence and then give £3 to an effective charity, because it’s so dehumanising. I can understand this argument, but I don’t think I sympathise with it much, that’s the point where my consequentialist side really does come out.
What do you think? Does one view on this question seem obviously right to you, or are you (like me) more unsure? Leave a comment or send me a message on Twitter with your thoughts.
As a high-decoupler, I think it's only reasonable to extend the thought experiment to *taking* money and donating it to an effective charity. Also some transaction that involves complex numbers.
But seriously, when I think about these sorts of things I also factor in the pragmatic effect on propensity to donate anything. It's not exactly in the ethical realm that you're trying to write about, but I figure that (given that a donor's post-donation feelings will impact future donation amounts) trying to establish one right answer will leave a significant fraction of people feeling blocked from whichever option (tangible or abstract) would feel emotionally sustainable to them.
I don't think the thought experiment caveat about beggars never spending money on alcohol or drugs is reasonable. Warm food and shelter are available for free, for rough sleepers in any city in the UK. Most beggars are doing so specifically to feed a drug habit, and 80-90% of beggars are not homeless to begin with.