What To Do?

We’ve been giving morality, and the universality of moral intuitions, a good going over lately (particularly in this discussion, which now has over 100 comments). Readers with an interest in this topic might like to have a look at Harvard University’s Moral Sense Test. Feel free to share your thoughts here.

Note: Don’t read the comments below before you take the test!!

Related content from Sphere


  1. Charles says

    I suppose I should preface this by saying that this comment contains SPOILERS!

    These things never test exactly what they say they are going to test. I suppose this is only natural, since you want to experiment with clean slates, so to speak, but I always feel a little deceived when I take one of these tests. Trying to figure out what they are *really* testing for is the challenge.

    (The first part seemed to be at least part memory/recall, and I’m pretty sure I answered most of the questions at the end correctly. There was only one that I wasn’t completely sure of, although I think I probably got the right answer.)

    The second part was a bit confusing. I’m guessing that they were trying to see if we would be more likely to find certain acts more reprehensible if we are first exposed to offensive images. Or maybe not, who knows? For me, one of the acts was clearly more reprehensible than the other, regardless of the image I saw beforehand.

    To address the series of questions in the first part, though, I have to say that I hate this sort of thing. For one, things are never this black and white. Surely you would be able to warn at least one of the parties involved, no? Yes, yes, I know. This defeats the purpose of the test, which is to ask the question: is it OK to sacrifice a few for the good of the many? Here the ratio is only 5:1. But what if it were 100:1, or 1,000,000:1. Would the life of that single person be worth saving the rest? Mathematically speaking, I suppose so, but for every question that asked me if I would trade one for five, I answer with a 2 (that is, just one above “forbidden”). If it were my life in the balance, perhaps I could be justified in making the choice, but if it is not, how can I play God? Who am I to decide if someone should live and someone else should die?

    The only exception for me was the question with the psycho. The five people weren’t normally in danger, but the psycho’s actions would result in their deaths. I answered that with a 6 (one below “obligatory”), which probably nullifies everything I said above, but in my mind there is a difference.

    Truth be told, though, if this happened to me enough times in real life, I’d probably just invest in a high-powered rifle and shoot anyone stupid enough to walk on train tracks that are still in use.


    Posted July 22, 2008 at 1:51 am | Permalink
  2. Addofio says

    Forced-choice tests distort the phenomena they are intended to study. In some cases the distorion is within legitimate limits, on the order of the simplifying assumptions made by pretty much any scientific model. In this case, if they are interested in cultural (or, for that matter, gender) effects, this distortion will be so severe as to invalidate the study, or at least that aspect of it. Many cultures simply reject the kinds of assumptions presented, and a fundamental aspect of the test is to accept them and reason only in terms of the info presented–a very modern, Western, academic thing to do, that even modern academic Western men are more likely to comply with than many modern Western women. The conditions of the test make it impossible to obtain valid data for such people. Although, this is Harvard, and they must know the above–maybe they have accommodated the test to these facts in some way not obvious to me. If not–shame on them.

    I ended up going through the test twice (was that morally forbidden, permissable, or obligatory?) The first time I was presented with only one scenario and a couple of questions about it. That seemed so odd that I went back. The second time, I seem to have gotten the same test as Charles. I sure hope these people aren’t betting the farm on the validity of their results.

    Posted July 22, 2008 at 11:20 am | Permalink
  3. Malcolm says

    Interesting comment, Addofio. What assumptions in this test would you imagine might be rejected?

    Posted July 22, 2008 at 11:54 am | Permalink
  4. Addofio says

    For starters, that there are only two courses of action available to the person making the decision. Or, people often simply reject some the the “facts” of the scenario, or they bring in other facts from outside the scenario–hence the plea in the intro to accept the facts and only the facts in the scenario. Basically, scenarios such as these put us in a box and ask us to reason within that box, which us academically trained people are usually very willing to do–but many people will simply reject the box altogether, in one way or another. It’s a way of thinking highly influenced by Western schooling.

    All this isn’t just my opinion or based only on my own experience–there’s ample research demonstrating these effects, over decades. If you push, I can give you references, but it *is* summer, and I *am* off duty. :-)

    Another thing that was interesting to me is that they told us virtually nothing about what the person confronted with the moral dilemma actually knew–only what the supposedly objective facts of the situation were. Since we nearly always take the mental state of people into account in making our judgements about their judgements or actions, I thought this was an interesting omission. And of course, by “we” I mean “people like me” :-).

    Posted July 22, 2008 at 3:53 pm | Permalink
  5. Malcolm says

    Yes, the testers are certainly trying to zero very specifically in on some particular distinctions among the various ways one might cause one death to save several lives. They aren’t rejecting the possibility of other actions, though, it seemed to me, but just asking whether the suggested action is forbidden, permissible, etc.

    I know that they have been running this test for people from a variety of cultural backgrounds, but I don’t know how consistent the results have been.

    Wish I were off duty…

    Posted July 22, 2008 at 4:37 pm | Permalink
  6. Malcolm says

    Hi Charles,

    I can understand your point about being reluctant to play God, but in the given scenarios either decision you make affects lives; indeed the decision not to act affects more of them.

    Posted July 22, 2008 at 4:43 pm | Permalink
  7. the one eyed man says

    I send a note to Harvard asking about the hypotheses they are testing & received the following reply:

    Thank you for your interest in our research! When you logged on to the Moral Sense Test (MST) you participated in a series of three experiments.

    The first is an experiment designed to examine the role of memory in our moral psychology. Specifically, it is an experiment designed to see whether immoral actions (e.g., pushing someone’s car off of a cliff because you don’t like them) are more memorable than immoral omissions (e.g., standing by and watching someone’s car roll off a cliff when you could easily stop it because you don’t like them). In this study, participants read one of two short stories at the beginning of the MST. The stories were identical in their consequences and in the beliefs and desires of the person who does something immoral. However, in one case the person’s *action* causes harm and in the other the person’s *omission* causes harm. At the end of the MST, after they have been thinking about other sorts of moral dilemmas, all of the participants are asked to remember some details about the story. Data from previous research have suggested that immoral actions are worse than immoral omissions. We are investigating the hypothesis that this difference in moral salience will yield a difference in the amount of detail that participants will be likely to recall.

    The second is an experiment designed to see whether there is any effect of the order in which moral dilemmas are presented to participants. In this experiment, some participants read a series of trolley car scenarios that began with a scenario that participants typically see as perfectly permissible and then progressed through scenarios that participants typically see as impermissible. Other participants began with a scenario that most participants typically see as impermissible and then progressed through scenarios that participants typically see as permissible. Previous data have suggested that there are a number of factors that play an important role in determining the permissibility of an action. We are investigating the hypothesis that these factors will continue to play the same role regardless of the order in which scenarios are presented.

    The final experiment investigates the role of emotion in making moral judgment. Specifically, this experiment tests whether induced disgust can affect the permissibility of harmless, but offensive social norm violations. In this study, participants are asked to read and then make judgments about emotionally charged scenarios after being shown either a disgusting or a neutral image. We apologize for any discomfort you may have experienced due to the disgusting image. It was necessary to provide the appropriate conditions to test your behavior. Data from previous research have been interpreted to suggest emotional responses drive moral judgment. According to an alternative theory, we experience emotions like disgust in response to our moral judgments. If that’s correct, a disgusting image will manipulate reports of feeling disgust, but not judgments of moralness or permissibility. We are investigating this hypothesis.

    Moral judgments play a critical role in our daily lives. However, it is unclear what sorts of non-moral factors play a role in structuring our moral judgments. A growing body of research suggests that many of the moral judgments we make operate intuitively, that is, without conscious access or understanding of the underlying principles. Understanding these intuitive moral principles opens a valuable window to our psychology, and also can help us make more informed moral choices. If you are interested in learning more about moral psychology, especially the way in which we are investigating moral psychology at the Cognitive Evolution Laboratory, you may wish to consult:

    Hauser, M.D. (2006). Moral minds: how nature designed our universal sense of right and wrong. New York: Harper Collins.

    Please do not disclose the research procedures and hypotheses that I have describe here to anyone who might participate in this study between now and the end of the data collection (May, 2009) as doing so could affect the results of our study.

    Thank you very much for your participation!

    Posted July 22, 2008 at 9:27 pm | Permalink
  8. Malcolm says

    Please do not disclose the research procedures and hypotheses that I have describe here to anyone who might participate in this study between now and the end of the data collection (May, 2009) as doing so could affect the results of our study.

    Nice going!

    Posted July 23, 2008 at 12:07 am | Permalink
  9. JO says

    Hi there,
    This is one time that I should not have read the comments first. I decided not to take the test after reading all of them as I would not want to jeopardize the results, but curiosity is killing me! Just thought that I would let you know that I haven’t abandoned the waka train, but have been in Philadelphia and then Brooklyn to see my son. After reading one of your earlier posts, I see that I was very close to where you live. Am now home but am very busy reading Kevin Kim’s book. I picked up “Freedom Evolves” by Daniel Dennett at Barnes and Noble so it will be next. It was the only one they had by him so I thought that I would go ahead and, at least, get a feel for his thinking.
    Salute from the VERY hot and humid Arkansas Ozarks,

    Posted July 23, 2008 at 4:13 pm | Permalink
  10. the one eyed man says

    You know me: the skunk at the garden party.

    I thought about the warning before I posted the note above, but figured that the likelihood that someone would have taken the time to go through all of the comments without having taken the test — and subsequently took the test and despoiled its results — was nil.

    Posted July 23, 2008 at 7:10 pm | Permalink