Next week, I will finally try to give a lecture making general points about consequentialism, and the other ethical perspectives that compete with it. Now, though, I want to discuss one last major problem case for moral theory, and see what consequentialism says about it. The key case is famine relief, but before we talk about that, consider this: you’re walking along the sidewalk, past a shallow pond, and you see a small child drowning in it. No one else is around to help, but the pond is shallow enough that you could pull the child out with no risk to yourself. Do you have an obligation to do this? What if you’re wearing an expensive suit, say a $500 suit, and know wading into the pond to save the child would ruin it–is that good enough reason for not saving the child, if you would otherwise have an obligation to help? I think most people would say no.

In the article that provided the main inspiration for this lecture, Peter Singer suggested that this was illustrative of a general quasi-consequentialist moral principle:

…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. By “without sacrificing anything of comparable moral importance” I mean without causing anything else comparably bad to happen, or doing something that is wrong in itself, or failing to promote some moral good, comparable in significance to the bad thing that we can prevent.

I call this a quasi-consequentialist principle, because it’s just a slight variant on the consequentialist doctrine that you should always do whatever will have the best consequences, on balance. Singer words it to appeal to people who don’t agree with his brand of consequentialism, for example, the “something wrong in itself” clause goes against the grain of your standard form of consequentialism. Singer also offers up a weaker principle. He thinks his stronger principle is correct, but thinks the weaker principle is enough for his purposes: “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.”

Now the application: there are hundreds of millions of people suffering from malnutrition. According to one estimate, over half the deaths in 2006 were related to malnutrition–36 million people. And in the most desperate cases, it’s likely that you could prevent one of those deaths for rather less than the cost of replacing a $500 suit. Peter Unger, in his book /Living High and Letting Die/, suggested an estimate of $64 for greatly increasing the chance of a two year old in Africa living to six years old, at which point long-term survival prospects increase dramatically. At any rate, if an expensive suit isn’t morally significant compared to the life of a drowning child, then the $64 shouldn’t be significant compared to the life of a starving child in Africa. So, according to even the weaker of Singer’s two proposed principles, you are morally obligated to give that $64 dollars. But then we find something surprising: since there’s a lot more than one starving child in the world, it seems you’d be obligated to give until giving any more would require you to give up something of moral significance. Plausibly, that doesn’t mean giving enough to push you into poverty, but the average middle-class person in the United States and similar rich countries spends a fair bit of money on trivial luxuries that don’t seem to have any moral significance at all.

Before moving on to the important responses to this argument, I want to emphasize how if you accept Singer’s principle, there really isn’t any way to avoid his conclusion. For example, you might think that in the case of famine relief, there are too many people who need saving. This, though, doesn’t change the fact that you’re still in a position to do some good. Nor does the fact that there are other people who might possibly help change the fact that you’re in a position to do some good. You know that most of them won’t help, or at least won’t help enough to make your help superfluous. Distance might have made a practical difference once, but given how easy it is to get money to competent aid workers, it seems not to make much of a practical difference in the modern world. And if you’re worried about overpopulation, Singer would argue, you should support charities that fight overpopulation. This could be directly, as in international organizations for birth control, but from what I understand of the sociological data, it can be done indirectly: better education for women tends to substantially lower the birthrate.

So to reject Singer’s conclusion, you just have to reject his premise. And some philosophers–even philosophers sympathetic to his views like Peter Unger–have suggested that at first glance, that doesn’t look that hard to do. The principle may look like a good explanation for why we have an obligation to help in the drowning child case, but it produces counter-intuitive results in other cases–if you’re feeling bold, you can just say “it produces the wrong results in other cases, i.e. it tells us things that simply aren’t true, and it logically follows from that that the principle can’t be true.” The claim that we don’t have the sort of extraordinary moral obligations that Singer says we do, it can be argued, is as good a starting assumption for moral philosophy as anything—at least as good as the claim about needing to save the drowning child, certainly a better starting point than Singer’s broad principles. And there’s a very natural way of extending this, to turn this into an objection against consequentialism itself. Some people seem to regard the strict moral demands of consequentialism on famine relief as just as bad as what it allows in terms of harming some to benefit many. Consequentialism is both too lax and too strict, at least that’s what anti-consequentialists have said.

When you’re judging whether Singer’s principle has absurd consequences, it’s important to notice how large the difference between the two versions of the principle is. Plausibly, providing for the education of your own children is morally significant, but not as morally significant as the deaths of other children. So, the weak principle, while making a fairly strong demand in requiring you to give up frivolous luxuries to help the very poor, would not require you to give up paying for your children’s education for the same reason. But if we accept the strong principle, and the claim about the relative moral significance of education and starving to death, you would be required to give up paying for your children’s education to help the poor.

Unger, as I’ve said, endorses the criticism that it’s pretty easy, at first, to reject Singer’s premise. But Unger still comes out endorsing Singer’s conclusion. How does he do it? The strategy explicitly adopted by Peter Unger to defend Singer’s conclusion–and which you see hints of in Singer’s own paper–is to systematically examine possible distinctions you might make between the two cases, and show that none of them work.

Consider distance. Singer suggests that taking distance into account violates a general principle of impartiality. You could respond, though, that it does no such thing, as the distance principle would apply to everyone equally–I’ll leave it up to you to think about whether that’s too technical a response. Perhaps a better argument against the difference principle comes from considering cases like the pond and the famine relief cases, but where the distance element is flipped. So maybe you get a distress call from a few miles away on a short wave radio, and you have a vehicle that can get there in time and be used for a rescue. And compare that to a case where you’re vacationing in a foreign country, and pass within a block of some starving children. It seems like the distress call case is morally like the pond case, and the vacation case is morally like the charity case, in spite of the flipped distance relation.

Now you should notice that this sort of strategy can be extended for all kinds of possible distinctions. For any proposed distinction, invent a pair of cases like our initial pond / charity case, but with that one element altered. For example: if you think perhaps it is not physical distance, but some sort of “social distance” that matters, such as being fellow countrymen or some such, then consider a case where you walk by a child drowning in a shallow pond while staying in South America. When you look at it that way, the “social distance” distinction seems implausible. If you think the key thing is that the drowning child case is an emergency, or more urgent, or whatever you want to call it, you can imagine receiving an appeal to give to charity with just that feature. This strategy works for an awful lot of possible replies.

You should be able to see that Unger’s basic strategy here is what we saw him applying in to the trolley problem in the first lecture on ethics, and what was applied to medical ethics in the last lecture. The strategy is taking ethical distinctions that look at first sound, and questioning whether they really make sense, especially by coming up with cases that run counter to the distinctions, and coming up with cases where the apparently clear distinctions turn out not to be so clear.

A problem that arises with this approach is you can start talking about cases very far removed from ordinary experience, F. M. Kamm, in trying to show that distance actually can make a moral difference, in spite of what Unger says, once proposed a case where you become aware of some emergency very far away thanks to super-vision, and are in a position to help somehow. Kamm has also proposed cases where by putting a fair bit of money into a machine, you can cause another machine to scoop some drowning children out of a pond. Kamm claims to have moral intuitions about these cases that back up his larger claims that distance is, in fact, morally significant. And when our intuitions are ambiguous, what are we supposed to do? We get into a debate about who has the burden of proof—does it need to be obvious that the cases support the proposed distinction? Does it need to be obvious that the cases undermine the proposed distinction?

These problems become even worse when you consider what could be called “combination responses”—responses admit that, say, distance alone can’t account for why you must save the drowning child but not give great amounts to charity, and the emergency status of the situation alone can’t account for why you must save the drowning child but not give great amounts to charity, but maybe the two factors combined account for this difference. Maybe it’s some other combination, maybe a more complicated combination. And when we try to come up with complicated hypothetical cases to evaluate the more complicated proposed distinctions, it gets harder to make obvious judgments.

In the last bit of this lecture, I’m going to stray into foreshadowing things that I’ve long been holding off on for future lectures—some of it I’ll launch into in full detail next lecture, but some of it will be saved for even longer than that. The first point is this: one alternative approach to the sort of consequentialism I’ve kept in the background in discussing all these moral problems says that we should focus not so much on action, but rather on rules. We want to formulate a set of moral rules that meet some criterion of goodness, whatever it is. And that suggests a couple of accounts of the problem we’ve been discussing. For example, it seems like there are some definite advantages to a rule requiring us to help others when we happen upon emergencies. Such a rule would give everyone a certain protection against emergencies. Its effects would be very different than the effects of a rule requiring us to help in all cases of significant need whatever. So maybe on that grounds, we can recommend a rule requiring us to help in emergencies, but not in any cases whatever. And if we are concerned with global malnutrition, and think it important to make some account of it in our system of rules, we can try to formulate a rule that, if universally followed would solve the problem, say by requiring everyone to donate $50 to the cause. I’ll have a lot more to say about the merits of this approach next lecture, but keep it in your heads for now.

The second idea is, to some extent, a stronger version of the first, and suggests that morality isn’t a set of absolute demands imposed on us by some extra-human moral reality, but rather that morality is a human invention, a set of rules to help society function. And it seems that a policy of helping members of our own society in emergencies has a value, from a self-interested point of view, that a strong requirement to alleviate ongoing problems wouldn’t have. To some people this approach will make a lot of sense; to some people it may sound like “fake morality,” and we’ll discuss it in depth much later in the course, when we get around to discussing the nature of truth in various subject matters.