You are standing by a railroad track when you notice that a trolley, with no one aboard, is rolling down the track, heading for a group of five people. If the trolley continues on its present track, they will all be killed. The only thing you can do to prevent this tragedy is throw a switch that will divert the trolley onto a sidetrack. But there is one person on this sidetrack, and he will be killed. Should you throw the switch?
In another version of this dilemma, the trolley is again rolling down the track, heading for a group of five people. This time, however, there is no switch or sidetrack. Instead, you are on a footbridge above the track. You consider jumping off the bridge, in front of the trolley, thus sacrificing yourself to save the five people in danger, but you realize that you are far too light to stop the trolley. Standing next to you, however, is a very large stranger. The only way you can stop the trolley killing five people is by pushing this large stranger in front of the trolley. He will be killed, but you will save the other five. Should you push the stranger?
Philosophers have been pondering these dilemmas since the 1960s. Many — notably Judith Jarvis Thomson and Frances Kamm — have agreed that you may throw the switch, but must not push the large stranger. But why? In each case, we face a choice between bringing about the death of one person or allowing five to die. Where does the moral difference lie? The discussion has embraced a variety of moral theories, but it has not produced a truly satisfying answer.
What philosophers did not do, until recently, is take an interest in empirical research about our responses to these or other dilemmas. Now, as philosopher Kwame Anthony Appiah describes in his concise yet erudite and engagingly written new book, "Experiments in Ethics" (Harvard University Press, 288 pages, $22.95), this is changing.
Harvard psychology professor Joshua Greene, when still a graduate student, wanted to understand what was going on in people's brains when they made these moral judgments. His subjects underwent functional Magnetic Resonance Imaging, or fMRI, which gives real-time images of brain activity, while he asked them to make moral judgments about the dilemmas mentioned above. He found that when subjects were asked about situations involving "personal" violations like pushing the stranger off the footbridge, the areas of the brain associated with emotions were more active than when they were asked to make judgments about relatively "impersonal" violations like throwing a switch.
More significant still, however, is Mr. Greene's finding that the minority of subjects who did consider that it would be right to push the stranger off the footbridge took longer to reach their judgment, and had more activity in the parts of their brains associated with cognitive activity, than those who said that it would be wrong to push the stranger off the footbridge. Does this indicate that our intuitive rejection of pushing the stranger is less thoughtful and considered than the view that in these extraordinary circumstances this is, after all, permissible?
Marc Hauser, a professor of psychology and evolutionary biology at Harvard, has also explored our responses to the trolley dilemmas, putting them, and similar ones, on the Internet, in what he calls a "Moral Sense Test," available in several languages. After receiving tens of thousands of responses, he has found remarkable consistency across Internet users of different nationalities, ethnicities, religions, ages, and sexes. He is currently seeking to extend his survey to people living in societies still untouched by the Internet. Mr. Hauser sees his work as exploring our evolved universal "moral sense." It is not only in regard to moral dilemmas that the experimental method has cast light on the nature of ethics. As Mr. Appiah (who teaches, as I do, at Princeton) describes, research in psychology has cast doubt on the long-standing view that a person's moral character is the dominant factor in whether he or she does the right thing. Instead, quite minor aspects of situations are often more significant. In one amusing set of experiments, theology students were told to prepare a lecture on the parable of the good Samaritan. They were given a briefing about the lecture in one building before being told to walk to another, where they were to give the lecture. On leaving the first building, they passed a stranger who was groaning by the side of the path, apparently in need of assistance. Students who had been told they needed to hurry to give their lecture were six times less likely to offer help than those who were told they had plenty of time.
Some philosophers see such experiments as a clumsy takeover attempt by scientists who are ignorant of the philosophical issues raised by questions in ethics. Mr. Appiah is not of this view. He points out that it would be more accurate to see the late-20th-century attempt to build a wall of separation between science and philosophy as an aberration. In earlier periods, the study of ethics was seen as part of the "moral sciences," that is, an endeavor that drew on all of the relevant sciences, as well as philosophy, in order to understand ethics. Because he sees the quest for scientific knowledge as very much part of the philosophical tradition, Mr. Appiah warns not only against "baseless fears" of the damage that experiments in ethics will do to ethics, but also against "exaggerated hopes" that the rediscovery of such an approach will answer all our puzzles about ethics.
One reason for Mr. Appiah's modest expectations about the impact of experiments in ethics is that, as he rightly points out, experiments can tell us what is the case, but not what we ought to do. That is clearly correct. As David Hume pointed out in the 18th century, no "ought" can follow from a set of premises that consists only of sentences about what is the case. Nevertheless, given the extent to which we rely on our intuitions, both in everyday life and in philosophical discussions of ethics, some experiments may offer persuasive grounds for changing our moral views.
For example, distinct but converging lines of research by Jonathan Haidt and Mr. Greene suggest that many of our moral judgments derive from an affective system that operates at an unconscious level and that presumably evolved because it helped our ancestors to survive and reproduce. This has clear implications for our responses to the trolley dilemmas. For all of our evolutionary history, it was possible to kill people through physical violence, and so we may have evolved an unconscious aversion to it, suspended only in special circumstances like war. For almost all of this long period of time, it was impossible to kill someone remotely by throwing a switch, for there were no switches. That could be why the image of pushing a stranger off a high place arouses a strong negative reaction in us, whereas the image of throwing a switch does not — even though both actions bring about the death of one and save five. But if this is the correct explanation for why most people are willing to throw the switch but not to push the stranger, it surely must make us consider whether we should stand by those responses. After all, what is the moral relevance of the fact that one way of acting was possible when we lived in technologically simpler societies while the other was not?
The gap between "is" and "ought" may not be sufficient to prevent experiments in ethics from transforming the way philosophers think about many ethical quandaries. There is, however, another reason why Mr. Appiah thinks that experimental research can play only a limited role in ethics. In his view, the entire focus on problems like the runaway trolley — or even on more realistic dilemmas like end-of-life decisions in medical care — is misplaced. More important, he thinks, are the everyday questions that Aristotle raised, about how to live well. These are, of course, very important questions, and it is no doubt true that when we contemplate the nature of the good life, it is hard to see how experiments will replace philosophical reflection. But they are not the only important philosophical questions we can ask about ethics, and on some of these other issues, experimental research could, by debunking some of our moral intuitions, have a major positive impact on how we think about ethics, and how we live by them.
Mr. Singer is a professor of bioethics at Princeton University. His books include "Animal Liberation," "Practical Ethics," "The Expanding Circle," and "The Way We Eat: Why Our Food Choices Matter," with Jim Mason.