Imagine you are walking by a shallow pond and see a drowning toddler. Do you have a moral obligation to save the child, even if it means ruining your clothes? It seems so.
Now consider that there are millions of people suffering and dying due to absolute poverty—the inability to maintain basic standards of living. In 2017, over 700 million people lived on less than USD $1.90 per day [1] , over 800 million lacked clean drinking water [2] and over 800 million people did not have enough to eat. [3] Over 5 million children died in 2019 from preventable and treatable diseases. [4]
Contemporary philosopher Peter Singer [5] famously argues that if you’re obligated to save the drowning child, you are equally obligated to help save people dying due to absolute poverty by donating to effective aid agencies. This essay explains his argument and considers some common objections to it.
Singer’s argument depends on a fairly straightforward moral principle: if we can prevent something very bad from happening without sacrificing anything of comparable moral importance, then we are morally obligated to do so. [6]
This principle explains why we should save the drowning child: her life is far more important than your outfit.
But millions of people are suffering or dying from absolute poverty and many of us could easily do something to prevent this by donating to effective aid agencies. Further, our doing so wouldn’t require that we sacrifice anything of comparable moral importance: we would just need to spend less money on things less important than human life: e.g., vanilla lattes, Netflix, and other luxuries. So, Singer concludes that it is wrong for many of us not to donate—it’s like letting the toddler drown in the pond to prevent our clothes from getting ruined.
Singer’s argument has radical implications for how many live, since most things routinely purchased in affluent countries are less important than a human life. It has thus given rise to many objections, several of which are discussed below.
Some interpret Singer’s argument as too demanding—implying that we must live simplistically, giving away any and all money we would otherwise spend on luxuries.
Singer responds that if his argument is sound then it doesn’t matter how radical it is—morality is sometimes challenging. [7]
Others suggest that we consider making a less radical commitment and donate ten percent of our income. [8] Research suggests this donation wouldn’t negatively affect the average donor (with about a 0.1 drop on a 10-point happiness scale for an average US citizen). [9] Indeed, many enjoy helping others and find it increases their happiness. [10] And it’s worth noting that meaningful good would still be done with an even smaller commitment. [11]
Others argue that we would be obligated to help the drowning child because she is part of our local community, arguing that our moral obligations derive from the relationships we have with others, [12] and so we are not obligated to help those in far-distant countries.
But suppose that there were a button in the pond which, if pressed, would save a toddler’s life in another country. If you think you’re still obligated to ruin your clothes to press it, then distance doesn’t change our obligations.
Others argue that we should focus on local poverty first. [13] But the costs of helping people in wealthy countries are significantly higher than aid in impoverished countries—making your donation less effective. [14]
Some argue that we don’t have an obligation to donate to aid agencies since there are many others who aren’t donating.
But suppose in the drowning toddler case, there were people standing at the shore watching and doing nothing. [15] Virtually no one would stand and watch a toddler die when they could easily wade in and save her, regardless of the inaction of others.
Some argue that issues related to absolute poverty are really the responsibility of the government. Others say that absolute poverty is the result of underlying structural problems and so we should solve those structural issues instead.
But suppose the reason the toddler is drowning is because there’s a playground right next to the pond and toddlers keep falling in. Surely, we should advocate to move the playground—but not before saving the toddler! Similarly, it seems like we have reasons to both help those who are currently suffering and advocate for larger scale solutions. [16]
Others point out that in the pond example, I personally save the child whereas if I donate, my donation could be stolen, embezzled, or otherwise fail to save lives.
To address this worry, several organizations have been created which identify aid agencies with a proven track record and monitor their success. [17] And suppose that when wading into the pond, you only had a 75% chance to save the toddler’s life. Wouldn’t you still try to save her?
Another objection is that Singer’s argument requires that we ignore our personal commitments.
For example, paying your friend’s exorbitant US medical costs is less effective than paying for medical costs in a poor country. But friendship ignores concerns of effectiveness in favor of loyalty and personal commitment. [18] Arguably, personal commitments and moral identities require that we at least consider donating to less effective causes which support our personal commitments. [19]
But, Singer might contend that you should do both—donate to effective causes in addition to (not in place of) your friend’s medical costs or your other moral identities. [20]
The response to global poverty is among the most practical ethical issues that exists. Many reading this could do something right now by making a donation online , of any amount, to effective aid organizations.
Singer’s work, along with that of others, [21] has inspired a movement called “effective altruism” which advocates for altruism—being concerned for others for their own sake [22] —in the most effective ways possible, generally by giving to aid agencies which do the most good. [23] If they’re right, then many of us will need to radically change our lives for the good of all.
[1] The World Bank (2020).
[2] World Health Organization (2019a).
[3] World Health Organization (2019b).
[4] World Health Organization (2020).
[5] To learn more about Peter Singer, see his webpage at www.petersinger.info.
[6] Singer’s 1972 essay “Famine, Affluence, and Morality” is the first and most famous exposition of the effective altruism movement. Singer offers two versions of his principle (see pages 231 and 235)—this formulation combines them.
[7] This is the first response Singer offers on page 236 of his (1972).
[8] MacAskill, Mogensen, and Ord make this argument in their 2018 essay. Singer (1972, 235) makes a similar move by considering a weaker version of his original principle: that our moral obligation is to prevent something very bad from happening without sacrificing anything of comparable moral importance. MacAskill, Mogensen, and Ord offer an even weaker (but still meaningful) suggestion that we draw the line at 10% of income, which is quite a bit over the average amount donated in the United States each year of about 2%.
[9] MacAskill, Mogensen, and Ord spend quite a bit of time outlining their evidence for their claim that giving is not demanding. See their (2018, 186).
[10] MacMillan (2017) discusses a study in which generosity was shown to lead to higher reported levels of happiness.
[11] For example, in 2016, only 44.3% of Americans gave more than $25 to charity (Michigan Institute for Social Research 2017). If those people were to give $50, that would amount to several billion dollars in additional donations.
[12] For example, Friedman (1991); Reader (2003); and Abelson (2005).
[13] Angus Deaton (2018), for example, argues that there is a sufficiently high number of people in absolute poverty in the US to merit focusing attention on eliminating that poverty.
[14] Singer (2018) makes this argument in response. He discusses this objection in conversation with Julia Taylor Kennedy in this 2011 “Extreme Poverty” video.
[15] Singer offers this modification in his original 1972 essay, page 233.
[16] Ashford (2018) argues that we have a primary duty to help resolve the structural problems, but a back-up duty to help those suffering in the meantime.
[18] This argument is made by Amia Srinivasan (2015) in her review of MacAskill’s (2015) book.
[19] I make this argument in Boesch (2018).
[20] This response works particularly well if we are working with a weaker proposal, like that given by MacAskill, Mogensen, and Ord (2018). It is less effective if you take the stronger version of Singer’s argument.
[21] See, as important examples: Unger (1996); Singer (2015); and MacAskill (2015).
[22] Contrast altruism with ethical egoism, an exclusive concern for one’s own self, potentially at the expense of others. See Ethical Egoism by Nathan Nobis for an introduction to egoism.
[23] “Most good” is typically understood as the greatest amount of reduction in death or suffering due to lack of basic necessities (adequate food, water, medical care, shelter, etc.)