Print This Post Print This Post

Moral philosophers disagree about a lot of stuff.  They disagree, for example, on whether moral properties exist and, if so, what the heck they are and how we have knowledge of them; on whether one can derive an ‘ought’ from an ‘is’ and, if not, whether this really matters or not; on whether moral judgments are the deliverances of affective or purely cognitive faculties; on whether moral omissions have the same status as moral comissions; and a whole lot besides.

One particular claim, though, seems to have widespread endorsement—the claim that ordinary folk are objectivists when it comes to morality.  According to this view, ordinary folk believe moral issues admit of a single correct answer, and reject the idea that two people with conflicting positions on a moral issue may both be right.  This claim of  ‘folk objectivism’ enjoys a surprising degree of consensus, and can be found in the works of a diverse range of moral philosophers with disparate theoretical commitments (e.g. Blackburn 1984; Brink 1989; Gibbard 1992; Mackie 1977; Shafer-Landau 2003).  It is a datum that most metaethical theories try to vindicate or accommodate.  But is this claim correct?  The answer would seem to be important, as the claim of folk objectivism has played a significant role in theorizing about the nature of ethics.

For example, metaethical objectivists sometimes argue that their theories have a leg up on non-objectivists because their theories can better account for the objectivity in normal moral discourse (Brink 1989; Shafer-Landau 2003).   Non-cognitivists, by contrast, feel compelled to provide a story that would explain the objectivity in normal discourse while claiming, at a general level, that moral judgments are not beliefs but are expressions of emotions and desires (Blackburn 1984).  And error theroists claim that ordinary moral discourse is in massive error precisely because it refers to moral properties that are objective and not contingent upon any person’s desires or preferences (Mackie 1977).

But are the folk objectivists about morality?  I’ve been exploring this issue using experimental methods with the help of my colleagues John Park, David Tien, Jennifer Wright, and Joshua Knobe.  Our findings bring the claim into question.

For example, in one study, we asked subjects to interpret a disagreement on the moral status of the following action: “Dylan buys an expensive new knife and tests its sharpness by randomly stabbing a passerby on the street.”  Subjects were asked to imagine that one classmate thinks Dylan’s act was morally wrong, whereas another thinks Dylan’s act was morally permissible.  When asked whether these two individuals could both be correct in their judgments, the folk were predictably objectivist.   They rejected the notion that these individuals could both be correct.

Things began to shift, though, when we depicted the disagreeing individuals as belonging to different cultural groups.  When the disagreement was between an American classmate and a member of an Amazonian warrior culture, or a member of an extraterrestrial species called the Pentars, objectivity levels dropped in turn (as measured through subjects’ agreement with the claim that at least one of the disagreeing individuals must be wrong).

It seems as though subjects think that there could be objectively correct moral judgments within cultures, but not across them.  The greater the disparity of the cultural groups, the more the folk started to embrace a relativistic conception of morality.

The effect came up in studies done at Duke, the College of Charleston, Baruch College, and even the National University of Singapore.  It came up when the moral violations were described as occurring in the subjects’ own neighborhood (as opposed to some unspecified location).  It came up when every subject was given all three culture conditions in random order (instead of just one of them).  And it fits into a pattern with the results of other recent studies examining the same phenomenon.  So it may turn out that people aren’t actually objectivists about morality after all.

If so, should this affect how metaethicists go about their business?  For example, is ‘folk objectivism’ a datum that philosophers can safely put aside when theorizing about the nature and status of morality?  We’d love to hear how specialists in moral philosophy would interpret these results.

Full paper here.

Many thanks to Matthew Liao for the opportunity to share the results of this research project on this blog.


  1. 1. Posted by Clayton Littlejohn | October 16, 2010 8:51 pm

    Hey Hagop, very interesting paper. Wanted to raise a few generalish worries. One has to do with the contrast between objectivism and relativism. Are these clearly defined in a consistent way by the authors of this paper and the authors discussed who say the folk have a tendency towards objectivist views?

    Another worry has to do with using the respondents results to classify them relativists rather than, say, pluralist objectivists. Whether an act is wrong (all things considered) has to do with the balance of reasons, let’s suppose. It’s striking that the respondents respond in the way that they do, but one objectivist friendly explanation is that the kinds of reasons that apply to agents in these different settings differ. If they did, it might be that the same principles determine which actions are right, but different principles apply in these three different societies you ask the respondents to imagine. Is there a way to control for this so you know the respondents aren’t assuming that these different societies came to have such radically different norms as a result to the fact that they live in significantly different circumstances? (Just to make this concrete, think of the discussion of relativism by James Rachels where he argues that much of the alleged moral differences between cultures is superficial and reveals not principled differences concerning the value of life, but differences in the way these principles instruct us to act depending upon differences in our circumstances. There’s the classic example from his article (which I doubt is historically accurate) of eskimos who are thought to practice euthanasia and infanticide. We don’t allow that around here, and so the relativist says that there’s a deep moral disagreement. The problem is that if we faced the kinds of situations the eskimos were thought to in these (likely fictional) stories, we would start to act in the ways that they would even if we stuck to our principles. If they faced the kinds of situations we would, the rates of euthanasia and infanticide might decrease.)

    If I ask my students whether infanticide or euthanasia is wrong, they often say that it is. If I then ask them what they would do if they were in the kinds of situations where people practiced infanticide, they admit that they would think that this was an acceptable but regrettable thing to do. If I ask them to find a single principle that would make sense of this, they are pretty good at doing that. This is a clear case of superficial relativism obscuring a deeper sort of objectivism, and I wonder how we could know that this isn’t happening again. Granted, I can’t think of particularly good reasons for the subjects to engage in the sorts of behavior described in the vignettes, but to make the stories realistic, I wonder if the respondents are assuming that they must have some reasons or other and this pollutes their judgment.

    Final worry, “wrong” and allied moral terms are notoriously hard to interpret. Some ordinary folk think that there’s a blame component and some do not. Suppose the objectivist view is about permissibility, rightness, etc…, and suppose that these don’t have a blame component. But, suppose the folk being polled aren’t terribly sensitive to this. They might be reticent to blame people who are merely conforming to social norms and this might affect the way that they use moral vocabulary. But, they still might have moral outlooks similar to that of the objectivist even if they don’t have facility with the moral language. The objectivist view is about deontic concepts, not the use of moral language.

  2. 2. Posted by Joshua Knobe | October 17, 2010 12:54 am

    Hi Clayton,

    These are very nice points, and I’m sure Hagop will have more to say about them, but I wanted to start off by just making a quick clarification.

    Participants in the study were not asked to imagine that different people, inhabiting different cultures, were each performing an action of broadly the same type. Rather, they were just told about a single person, from one particular culture, performing one particular action. Then participants are asked to imagine that different judges are arriving at opposing judgments of this one person’s action.

    So, for example, participants could be told that a man named Dylan (who lives in New Jersey) kills an innocent person. Then they would be told to imagine that one of their classmates thinks that what Dylan did was morally bad but that a person from an Amazonian tribe thinks that what Dylan did was morally permissible. After it is explained that these two people have opposite opinions in the case, participants are asked whether one of them has to be wrong or whether they can actually both be right.

    So, given that both judges are making judgments about the very same agent, we thought we might be getting at genuine relativism rather than just the idea that different agents have different reasons. Does that sound right to you?

  3. 3. Posted by Tina Shoebb | October 17, 2010 2:26 am

    A great many people these days have been infected with relativistic thinking. Which means that it is pretty difficult to find people who count as ‘folk’: people who have not been exposed to cod philosophy of one sort or another.

    So if you want to examine what ordinary people’s philosophical intuitions would be prior to being corrupted, you are going to have to search pretty hard to find such people.

  4. 4. Posted by Hagop Sarkissian | October 17, 2010 5:06 pm

    Hi Clayton,

    Thanks for posting your thoughts and reactions! I agree with Joshua’s response to the pluralism worry. We designed the experiments with the intent of trying to avoid this worry by having the two individuals from different cultures judge one token act by one specific person. The subjects weren’t being asked whether, for example, one kind of act may be acceptable in one cultural context and not in another, but rather they were given a case of disagreement about only one action by one person in one particular place. Does that help with your worry?

    We also did a follow-up study where half the subjects were asked the usual question about whether the differing individuals could both be correct in their moral judgments, and half were asked a different question about whether the differing individuals both had “good reason” to believe judge as they did. Here, subjects tended to say that the individuals from the other cultures had no good reason to judge as they did in the moral case, but were nonetheless correct in their judgments. More research is needed to clarify what’s going on, but it seems as though the folk are willing to grant that people from different cultures may hold very different moral beliefs that are not illicit or false, and that they don’t necessarily make such judgments because they think there are good reasons supporting the different moral beliefs. (This is study 6 in the paper attached to the original post.)

    Regarding your final worry, I take it the idea is something like this: some people might think an action is wrong only if the person committing the action is blameworthy in some way (for example, by focusing on the person’s intentions or desires or culpability), whereas other people might think an action is wrong in and of itself without any mention of blameworthiness. That seems right to me. We did a study to see whether folk were reticent to blame people because they were simply conforming to the prevailing beliefs in their society. In this study, we found a sharp asymmetry depending on the type of belief in question. When it came to factual beliefs (say, about whether Napolean rode into battle on a horse or in a helicopter), folk were perfectly willing to say that people with differing judgments couldn’t both be correct, even when they were given some plausible story why one of the persons had a false belief. By contrast, in the moral cases, subjects again showed the tendency to say that the differing moral judgments could both be correct. (Study 6 is again relevant, though Study 5 touches on this issue as well.)

    Don’t know if that addresses all your worries, but hopefully its enough to get the discussion going.

  5. 5. Posted by Hagop Sarkissian | October 17, 2010 5:19 pm

    Hi Tina,

    An interesting comment, though I think we may risk begging some central questions by assuming which beliefs are ‘true’ or ‘authentic’ and which reflect a process of poisoning. Why can’t objectivist thinking about morality be something that ‘infects’ people, for example? Here’s a testable hypothesis: do students who take philosophy courses end up becoming more objectivist in their views about morality? If so, which is the real ‘folk’ view?

    But let me voice my agreement with the claim that it is difficult to come to some definite notion of who count as ‘the folk’. For purposes of our studies (and many others which examine folk views), the folk are taken to be any people who haven’t yet undergone philosophical training by taking a course in philosophy.

    But I don’t think this addresses your worry. I take it you mean that folk might be infected not by philosophical conceptions of relativism but rather by a relativistic tendency in public discourse, or a relativistic cultural context. Is that right? If so then I would have to say that my own experience here in the U.S. doesn’t really conform to that. I don’t hear many talk show hosts or read many op-ed pieces claiming that there could be multiple correct perspectives on prevailing social and political issues, for example. Rather, it seems as though we are in the grips of a divisive ‘us’ vs. ‘them’ mentality. But discussing this would take us too far afield, I’m afraid.

  6. 6. Posted by Clayton Littlejohn | October 18, 2010 7:47 pm

    Dear Hagop and Joshua,

    On the pluralism vs. relativism issue, I think it helps address the worry that differently situated judges are focused on the same token action.

  7. 7. Posted by Laurence Thomas | October 19, 2010 1:58 am

    Greetings Sarkissian–

    The thought that occurs to me is that human beings are remarkably malleable. It does not take much for that which seems utterly natural to one group people to be viewed as utterly disgusting by another. I am always amused by the fact that in France I can kiss some very close ever so heterosexual males on each cheek (they would be offended if I did not do so) whereas this very behavior could not be more out of the question in North America.

    Change in significant ways the experiences that people have from the very start of their lives and one changes substantially the moral intuitions that they have. Indeed, just look at how different our views of women are nowadays compared to 100 years ago.

    So it hardly surprises me that we get substantial differences across cultures, where there is a marked difference between the cultures in question.

  8. 8. Posted by Jean Kazez | October 21, 2010 4:37 pm

    When subjects say the American and the Amazonian could both be correct, could it be that they’re actually obeying a moral principle that forbids judging the Amazonian? If so, then you’d expect them to say both could be correct even if the disagreement was about the number of planets, or math, or some such. Or maybe it makes sense to think they would be especially wary of judging the Amazonian’s moral beliefs, since those kinds of judgments seem like a prelude to interfering with other cultures–something most people consider wrong. It seems like the role of tolerance would have to be teased out before you could draw the conclusion that a lot of people are non-realists about morality. I’m wondering if you have data about non-moral disagreements as well.

  9. 9. Posted by Simon Rippon | October 21, 2010 4:44 pm

    Hello Hagop,
    Interesting stuff, but I strongly doubt that study 5 in the paper is well-designed for ruling out the following competing non-relativist hypothesis: Your subjects may be confused about whether the aliens and tribal peoples are said to have the thought that the action is or is not morally wrong (call this the “same thought” interpretation), or whether they are said to have a thought expressed with the words “morally wrong” (in their respective languages; call this the “same words, different thought” interpretation). I think neither of your non-moral questions in study 5 is well-designed to test for this (though perhaps it was not intended to test for this at all?)

    Arguably the pasta question could be helpful for answering my worry, but the belief attributed to the tribesman by that question is so improbable even under the “same words, different thought” interpretation that I think it is probably not a good test (that is: It’s hard to imagine an Amazonion warrior culture society that has anything called “pasta” which grows on trees AND is harvested by special farmers called “Pastafarians” every five years – why did you feel it necessary to add the second conjunct?)

    The Napoleon question I believe is completely unhelpful for these purposes – and when asked first, would probably vitiate the results for the pasta question by encouraging the “same thought” interpretation to be held for the pasta case that followed it. It’s almost impossible to be confused about whether we’re talking about *our* Napoleon Bonaparte or *their* “Napoleon Bonaparte”.

    Moreover, each of these non-moral questions were asked in a *completely* different form from the moral ones, which further exacerbates the difficulty of getting the “same words, different thought” interpretation compared to the moral cases you used. Why was this? In these non-moral cases, you built two extra beliefs into the scenarios: the beliefs of the two named individuals who the tribesman and classmate then are described as agreeing or disagreeing with. Since these two named individuals are most naturally assumed (absent further information) to belong to the same culture, this has the effect of generating the “same thought” rather than “same words, different thoughts” reading.

    So I worry that what is presented as specifically moral relativism in your paper may be just the “same words, different thought” interpretation coming to the surface wherever it is available – which it is not in your non-moral cases.

    I raised this worry in comments about a year ago when you posted an early account of this work at Experimental Philosophy; Josh responded positively to the point then, hence my thinking that your study 5 may have been designed to rule it out.


  10. 10. Posted by Hagop Sarkissian | October 24, 2010 9:23 pm


    Thanks for your comment. I do think that what you say is possible. Let me mention a couple of studies that might mitigate the worry (without, of course, vitiating it).

    We did ask about non-moral cases, and here subjects were perfectly willing to say that the Mamilon was wrong in his beliefs (even when we gave an account of why the Mamilon would understandably have false beliefs about the non-moral facts, owing to limited or spurious sources for the beliefs). So it seems that subjects are willing to pass judgments on others in the non-moral realm, but not the moral realm. Maybe they hold a principle that it is wrong to judge others’ moral beliefs (as you say), but the question is why? One possibility is that subjects truly believe these foreign moral beliefs are wrong, but also think it would be intolerant to say so. Another possibility (and one that we think may be true) is that folk really are relativists about morality, that they believe moral claims are only true relative to certain moral frameworks.

    But I think that what you say here is an interesting possibility to explore in future research:

    “maybe it makes sense to think they would be especially wary of judging the Amazonian’s moral beliefs, since those kinds of judgments seem like a prelude to interfering with other cultures–something most people consider wrong”

    That is something I hadn’t thought of, and think is a plausible hypothesis to test out further.

  11. 11. Posted by Hagop Sarkissian | October 24, 2010 11:45 pm

    Dear Simon,

    Thanks for your continued interest in this project, and for your detailed comments and criticisms. These are very helpful indeed.

    In particular, I think you’re right that perhaps the added levels of belief in that study (5) may have introduced unnecessary additional details. So we ran a follow-up (Study 6), where the non-moral cases paralleled the moral cases much more directly. You can see a version of this study at the links below:



    What’s more, we added some extra bit of info to the description of the Mamilons to provide them some good reasons for having the strange beliefs in the non-moral cases (they have bad sources of knowledge about the outside world). Even with these changes in place, we got the same pattern of results.

    Looking at Study 6, do you think this partly addresses your worries? I’d love to hear what you think.

    In general, I think that the ‘same words / different thought’ interpretation would be encouraged in cases where we have the disagreeing individuals actually speaking directly and stating their opinions in their own words–for example, if they are quoted. Then it does seem possible that the thought/meaning may be different than the spoken word. But in reporting to the subjects what the classmate and the Mamilon believe, I think subjects are encouraged to think that we are reporting in our language the actual thought of the Mamilon, and not reporting his words. Does that make sense (even if it’s not convincing?). We’re telling subjects what the Mamilon thinks. (Of course, we–the experimenters–could be wrong, but why would the subject think so?)

    Of course, it’s always possible that subjects interpret things differently, and there’s no way to rule out your hypothesis given the studies we’ve run so far. I do think, though, that reporting their beliefs tends to ‘fix’ or ‘hold’ them in a way that is better than having them directly expressing their beliefs.

  12. 12. Posted by Jean Kazez | October 26, 2010 2:12 am

    Hagop, Thanks–now that I’ve noticed the link to the paper (blush-I didn’t before) I look forward to reading the whole thing carefully.

  13. 13. Posted by Jean Kazez | October 26, 2010 2:58 pm

    Isn’t it going a little too far to say the folk are relativists, given the mean score of 4 in study 1 (and the others)? Shouldn’t relativism yield scores of 3 or lower? The more cautious interpretation would be “not objectivism”-wouldn’t it?

    Also–I note that in all these studies you’ve got the exotic judge making the intuitively wrong moral judgment. What would happen if the exotic judge made the right judgment, and a home judge made the wrong judgment (maybe because of membership in some home subculture). Looking at this might help you eliminate moral tolerance for exotic cultures as a confounding variable.

    Finally, I’m puzzled that you think surveys conducted at several universities are enough to reveal how people innately think about morality, as opposed to just the elements of a college education. I distinctly remember the day in my own college education when I first encountered relativism–it was explicitly taught to students.

  14. 14. Posted by Jean Kazez | October 26, 2010 3:00 pm

    Sorry–as to the “4”–I meant in the “other culture” condition.

  15. 15. Posted by Simon Rippon | October 26, 2010 8:39 pm

    Hi Hagop – no, the new study you pointed me to does not address my concern at all! Although you got rid of the additional beliefs that complicated things in study 5, you basically once again ruled out the “same words, different thought” interpretation in the non-moral cases, and encouraged it in the moral cases, with the preamble you provided. You told us that the tribe learned their beliefs about “other cultures” through “history” books with “entirely fictional” contents. Then you asked us precisely, in the non-moral cases, about what can only be reasonably interpreted as their beliefs about our culture. That looks like stacking the deck against non-moral relativistic responses to me!

    Of course I agree with you that the “same words, different thought” interpretation would be the *wrong* interpretation of any of your questions, since you are not reporting, e.g. the direct speech of the characters in your scenarios. I’m just very skeptical that your subjects are aware of this distinction, and skeptical that they are avoiding thar very interpretation of your moral (“moral”) scenarios.

  16. 16. Posted by Hagop Sarkissian | October 29, 2010 2:15 am

    Hi Jean,

    Thanks for these further comments. I would avoid drawing the hard conclusion that ‘folk are relativists’. As you say, the data don’t seem to support such a strong conclusion. In the paper, we end up making a more conditional claim: to the extent that individuals end up entertaining different perspectives or points of view, and to the extent they are open to new experiences, then the folk will tend to adopt a more relativistic stance toward morality. This is not as strong a conclusion, but it’s one we think has some plausibility given the other studies we cite toward the end.

    I think your proposed twist on the vignettes are interesting. My hunch would be that a lot would hinge on how we describe the ‘home subculture’ (as you put it) within the vignette. If the subculture is described in sufficient detail and seems plausible, then perhaps we’ll get the same response pattern. But I don’t have any strong intuitions on that. It would be a worthwhile study!

    As for your last point–yes, I think that’s a valid one. Indeed, we report (although very briefly and only in passing) the results of another study done by Beebe & Sackris, where they surveyed individuals across a range of age groups using the same basic method (i.e. interpreting disagreement). What they found fits your suspicions: individuals roughly between the ages of 17-30 tend to be most relativist, whereas individuals in other age groups tend to be more objectivist. This can be interpreted in a number of ways, of course. One is that it is during this period in a person’s life when they are most open to possibilities and different perspectives. Another is that people really are objectivists after all, but abandon this belief for a short period in their lives. I look forward to seeing what others find in this regard.

  17. 17. Posted by Hagop Sarkissian | October 29, 2010 2:24 am


    This looks to be something to explore further. It seems we agree what the natural reading should be (given how the vignettes are worded), but you seem far more sceptical about whether the folk are actually reading it that way.

    I’m wondering how we could disambiguate these two interpretations for subjects. Perhaps the idea would be to make it clear to subjects what the two possible interpretations are, and then see which one they think most accurately describes how they feel about the case? Does that sound like a promising method to you? Or, instead, we might give half the subjects a prompt that reflects one interpretation, and half the subjects another prompt which reflects the other interpretation, and see if we get a difference between the responses?

    At any rate, we obviously can’t resolve how subjects are interpreting the disagreement in the vignettes without running more experiments.

  18. 18. Posted by Simon Rippon | October 29, 2010 10:38 am

    Hagop – thanks. I don’t have an idea for how you can get subjects to disambiguate these interpretations; I’m not sure whether it’s even possible to get them to do so. I think philosophers get very confused about things Kripke says because of a similar issue, even though philosophers should be fully aware of the difference. So all I would suggest is that you run non-moral scenarios/prompts that parallel the moral ones you use rather more closely. I proposed one such prompt involving rocks/”rocks” in my comments on your older blog post. If you can show that the folk do not give relativistic-sounding responses to prompts like that, your moral results I think would be much more telling.

Post a comment

Name: (required)

Email Address: (required) (will not be published)