New Way Trolley Problem Shows We’re Awful
Tiffany Sun, a student at Rosyln High School in New York, was one of 40 finalists in the 2015 Intel Science Talent Search with an experimental philosophy project on the Trolley Problem. That’s the good news. The bad news? What she learned. From an article at Cogito.org:
Tiffany said the first step to conducting her research was coming up with experimental stimuli to apply to the traditional logical problem. “I manipulated the characteristics of the lone person on the track,” she says. “I used descriptions of variables—[physical] attractiveness, socio-economic status or wealth, and [having a physical] disability.”…
Tiffany’s results showed that the participants were more like to sacrifice (kill) the person on the alternative track if that person was described as poor (low SES), disabled, or unattractive. Attractive, well-off, and people without disabilities were less likely to be sacrificed to save the five people on the other track. The highest sacrifice rate was for the disabled person, at 74 percent. The next highest was for the plain woman, at 69 percent…
Tiffany notes that to her, one of the most interesting gender difference that emerged was that men chose to sacrifice plain women at a rate of 82 percent, while women sacrificed the plain woman only 52 percent of the time.
You can read more about Ms. Sun’s experiment, which involved 300 survey respondents, in the Cogito.org article, Student Science, and at The Island is Now.
Ms. Sun first learned about the trolley problem in a summer logic course she took from the Center for Talented Youth (CTY), between 9th and 10th grade. So an alternate version of this story could be headlined: “Early Exposure to Philosophy Influences Young Woman’s Choice of Science Project.” Not an unimportant point!
UPDATE: Technically, “the trolley problem” has come to refer to an apparent incongruity in most people’s judgments about what to do in a pair of cases, “bystander” and “footbridge.” Ms. Sun’s research seems to have employed only one of these cases, “bystander,” and so, on this more technical usage, it is a mistake to report her as having worked on “the trolley problem.” I did not point this out in the OP as I thought it was tangential to the main point of the post and a bit pedantic,but since a couple of people mentioned this on FB and email, and since this post seems to be getting shared a bit, I suppose I should not miss the opportunity to note this point. And while we’re at it, let me tell you what “begging the question” really means, kids…
There is published research showing that trolley judgments are sensitive to the race of the person on the track, though among politically liberal white participants, this leads to an increase in a willingness to sacrifice the white victim.
http://journal.sjdm.org/9616/jdm9616.pdf
I always manipulate the case in these kinds of ways when talking through trolley problems with my classes. Making an explicit show of the possibility that we’re more likely to sacrifice a poor, disabled, or unattractive person tends to get students’ justice hackles up in what seems like a valuable way.
We’re all awful…but men are more awful.
When I go through the “fat man” example, I always run into students who say that the fat man is less valuable, morally, because of his size. Even when I clarify that he lives a perfectly rewarding life, some students continue to consider him intrinsically less valuable than his thin counterparts. It’s depressing.
Why do you think the study supports the conclusion that men are more awful than women? (maybe on average men are more awful, e.g., violent crime statistics are an indication. But that’s a matter not related to the study).
There doesn’t seem to be enough info available to us, as far as I an tell, for the following reasons:
According to Sun’s account, the study says that the overall rate of sacrificing one person to save one was the same for both men an women. So, if it’s awful to be willing to sacrifice a person to save five, then as far as we know, the study does not support either men or women being more awful.
On the other hand, if the problem is not willingness to sacrifice people to save more people but unjustified biases, I still don’t see that the data supports that men are more awful.
On that note, let’s say the rate of sacrifice is R½ (it might be slightly different from men and women, but we do not know that).
On average, in the study, men were willing to sacrifice the “plain woman” in order to save 5 people in 82½ of cases, whereas women are willing to sacrifice her in 52½ of cases. That may well indicate a bias against the “plain woman” in the population of men if 82 is greater than R, because then, men in the study on average are more willing to sacrifice her to save 5 people than they are willing to sacrifice a person on average, in order to save 5 people (note: however, even this is problematic; the study should have asked another question in which no information about the single person on the track is specified).
On the other than, the results may similarly indicate a bias in favor of the “plain woman” in the population of women, if 52 is lower than R, because then, women in the study are less likely to sacrifice her in order to save 5 people than they’re willing to sacrifice a person on average (note: however, even this is problematic; the study should have asked another question in which no information about the single person on the track is specified).
Whether the degree of bias with respect to the “plain woman” is greater among men or among women in the study is unknown to us, given the info available in the articles the OP links to (we don’t have access to the study).
So, based on that, there is insufficient information to ascertain which group exhibits a greater bias in this case.
There may be other reasons for your assessment, but I’m not sure what it is, so I’d like to ask why you come to the conclusion that men are more awful.
That aside, I don’t think the conclusion that we (or even all of the participants in the study) are awful is supported. The study does indicate a number of awful biases, but it does not indicate that all of the participants have them (it’s not suitable to detect that).
Sorry, I had missed part of the info.
There is enough info to conclude that men in the study are on average a lot more likely to sacrifice the “plain woman” than they’re likely to sacrifice women on average, whereas women make an assessment very similar to their average in the case of the “plain woman” – a result that indicates a much stronger bias among men in the “plain woman” case than among women, though the amount of bias is still difficult to ascertain due to a lack of a scenario in which the person on the track is not described, which leaves us with the only option of comparing the results in specific scenarios with the average result considering all scenarios.
Arthur, I have similar issues in my classes regarding students and “the fat man,” although what happens in my experience is that they insist that anyone fat enough to stop a trolley is surely unhealthy and will probably die sooner than average, and use that to justify killing him.
I like to use a “man with a very heavy backpack” instead of the “fat man” when I teach this. It’s a good way to avoid triggering people’s biases and problematic associations regarding weight. I think that Peter Unger came up with this variant in the case for Living High and Letting Die.
Surely at least part of the issue is that some of those canvassed are thinking in a utilitarian way and utilitarian thinking done correctly (ie dispassionately) does take into account things like disability, wealth, attractiveness etc in deciding who to save and who to sacrifice. See for example Cora Diamond’s article ‘How many legs’