Singer is trying to prove that giving famine aid is morally mandatory. Here is his argument for that conclusion.
Suffering and death are bad. [Moral Assumption]
If (a) it is in our power to prevent something bad from happening, (b) without thereby sacrificing anything of comparable moral importance, (c) we ought, morally to do it. [Moral Principle]
We could prevent suffering and death by giving aid without thereby sacrificing anything of comparable moral importance. [Factual Assumption]
Therefore, we ought, morally, to give aid. [Conclusion]
The balance of the article involves taking up and answering objections to points 2 and 3: questions about the moral principle are discussed on pp. 236-39; questions about whether famine aid really would relieve suffering are taken up on pp. 239-43.
Our focus will be on the principle: point 2. We want to know how he argues for it and how he defends it against objections.
Singer’s argument for this principle involves an example of a drowning child. His idea is that it is obvious that you should save the drowning child and that the principle is the best explanation of why you should do so. This is evidence that the principle is correct. If the principle is correct, in turn, he can use it in his argument about famine aid.
These are the major parts of Singer’s article. I find flat outlines like this helpful for following the course of an argument. Even though they aren’t very detailed, they keep me oriented. And because they aren’t very detailed, they are not hard to make.
Introduction to the problem (229-31).
Two versions of the principle he wants to defend and the basic argument in its favor (231–32). (This is the part that we will focus on in our discussion.)
Objections to Singer’s basic argument for his principle: the famine case is unlike the drowning child case (232–35).
Radical consequence of the argument: many acts that, according to commonsense ideas of morality, are merely matters of charity are in fact moral duties (235–36)
Objections to the principle itself. The objections maintain that an argument that conflicts too much with our commonsense understanding of charity and duty must be mistaken. Singer argues that we should admit that our commonsense understanding of these matters is indefensible (236–39).
Objections to the practical steps Singer recommends. These objections accept the idea that we’re required to give to those in need but they express doubts about whether privately provided famine relief will meet this goal. The third point isn’t really an objection but more of a question: how much should we give? (239–43)
Our focus is on Singer’s principle. He argues for it using the example of the drowning child. The idea is that if you think you must save the drowning child consistency requires that you agree that you must give a lot to famine aid.
We have to ask whether that is so. Could you agree that you must save the drowning child without agreeing that you must give a lot to famine aid? Is there a different principle that would tell you that you must save the drowning child without also telling you that you must give to famine aid?
Here is another way to put it. Look for ways that the two cases are different. Do those differences matter? Also, has Singer shown that they do not matter? (That is a slightly different question.) Those are the questions that we are going to talk about.
You may notice that there are two versions of Singer’s principle: a strong and a moderate version.
Here is what that means. Singer’s moral principle has two parts:
A part explaining why we have duties: we have to prevent suffering and death. (Roughly 2a in the argument I laid out in the first section.)
A part that makes exceptions to those duties: we do not have to prevent suffering and death if doing so means making a specified kind of sacrifice. (2b in the argument laid out in the first section.)
The strong version is: if it is in our power to prevent something bad from happening, without thereby sacrificing anything of comparable moral importance, we ought, morally to do it (Singer 1972, 231). It is described as strong because it requires a lot. The exceptions are narrow and the sacrifices it imposes can be severe.
The exceptions are for cases where preventing something bad would come at the expense of something of “comparable moral importance,” such as sacrificing another life. When that is so, we are not required to prevent the bad thing from happening.
The moderate version is: if it is in our power to prevent something very bad from happening, without thereby sacrificing anything morally significant, we ought, morally, to do it (Singer 1972, 241).
Here the exception to the duty is triggered at a much lower level. If saving the drowning child meant violating a promise to meet someone for lunch, that would be enough to show that you do not have a duty to do so. Breaking a promise is morally significant, after all.
Each version has problems. The strong version appears to claim more than the drowning child case can establish. If readers thought that it was morally mandatory to save the drowning child even at significant risk to the rescuer’s life, then the example might support something close to the strong version. But that is not how the example is presented so I don’t see that Singer has made his case for the strong version.
The moderate version has the opposite problem. Would you really say that the duty to save the child could be trumped by a lunch date? I hope not!
So it seems to me that more work remains to be done to formulate and argue for a persuasive version of Singer’s principle.
These are the points that you should know.