Cross Post: Want to be popular? You’d better follow some simple moral rules

Imagine that an out of control trolley is speeding towards a group of five people. You are standing on a footbridge above, next to a large man. If you push him off the bridge onto the track below, his body will stop the trolley before it hits the five people. He will die, but the five others will be saved. Should you push the man off the bridge?

Before you make your decision, you should know that your popularity could depend on it. According to a new study of more than 2,400 participants, which we carried out with David Pizarro from Cornell University, the way you answer the “trolley problem” can have a big impact on how much people trust you. So let’s have a look at your options.

You might say yes; saving five lives outweighs the harm of killing one person. And you wouldn’t be alone: you’d be making a moral decision in line with “consequentialist” theories of morality. Consequentialists believe that we should aim to maximise the greatest good for the greatest number of people, even if this means causing some harm – for example, by killing one person to save five.

On the other hand, you might say no; killing someone is just wrong, regardless of any positive consequences there might be. Here, you’d be making a moral decision in line with “deontological” moral theories, which focus on moral rules, rights and duties. Maxims such as “thou shalt not kill” and “treat others as you would like to be treated” (otherwise known asthe golden rule) fit into this category.


Which do you choose?

Statistically, more people think that it’s wrong to push the man off the bridge to save the five others. On one level, this makes sense – we shudder at the thought of a friend or partner doing a cost-benefit analysis of whether you should be sacrificed for the greater good. So why do more people prefer this rule-based approach to morality?

Some scholars have argued that deontological intuitions arise from “irrational” emotional responses. But we thought there might be another explanation: namely, the power of popularity. We proposed that if people who stick to moral rules are considered to be better social partners, that might explain why more people take a deontological view.

Toeing the line may come naturally.

Over the course of human evolution, this could favour one type of moral thinking over another among the entire population. So, rather than reflecting irrational or emotional thinking, making moral judgements based on rules could be an adaptive feature of our minds.

To the test

We tested this hypothesis using several variations of the “trolley problem” and asked whether people who made deontological or consequentialist moral judgements were preferred as social partners.

Over the course of nine experiments, we found that people who took a deontological approach to the dilemmas (refusing to kill an innocent person, even when this maximised the greater good) were seen as more trustworthy than those who advocated a more flexible, consequentialist approach.

And not only did most people say they would rather trust a deontologist than a consequentialist – they also put their money where their mouths were. When asked to entrust another person with a sum of money, participants handed over more money, and were more confident of getting it back, when dealing with someone who refused to sacrifice one to save many, compared with someone who chose to maximise the overall number of lives saved.

Not so simple

But this wasn’t the whole story: simply deciding whether or not to sacrifice an innocent person was not the only thing that mattered. We also found that how the choice was made was crucial. Someone who had decided to sacrifice one life to save five – but had found that decision difficult – was trusted more than someone who had found the decision easy.

And it wasn’t always the case that those who refused to kill an innocent person were trusted more. Where the person who might be sacrificed indicated a specific desire to live or a willingness to die, people favoured individuals who respected those wishes – even if that involved killing.

These findings don’t just help explain how we came to have the moral intuitions that we do, but also how moral judgements play out on the world stage. Our results could help shed light on why we are often attracted to political leaders who communicate simple messages based on moral rules.

Consider, for example, a politician who says that gay marriage should be legal because marriage is a fundamental right in a fair and democratic society (a deontological perspective). This person is likely be to seen as more moral and trustworthy than one who says that gay marriage should be legal because it has positive economic or social consequences (a more consequentialist perspective).

So next time you have to make a tough call, remember – people like people who follow moral rules.


Jim A.C. Everett receives funding from the US-UK Fulbright Commission, and the Economic and Social Research Council.

Molly Crockett receives funding from The Wellcome Trust, Oxford University Press and The John Templeton Foundation.
  • Facebook
  • Twitter
  • Reddit

10 Responses to Cross Post: Want to be popular? You’d better follow some simple moral rules

  • Keith Tayler says:

    Thought experiments like the “trolley problem” are amusing but should never be taken too seriously. Much the same goes for evolutionary psychology which can provide some entertaining speculations. Put the two together and you can have riot.
    I do not accept that ‘[t]hese findings don’t just help explain how we came to have the moral intuitions that we do, but also how moral judgements play out on the world stage.’ There are numerous other explanations for why people in thought experiment might claim to prefer people that do not deliberately kill another to prevent an accident. Not sure why your explanation does any more work than the others .

  • Nick says:

    “…they would rather trust a deontologist than a consequentialist”

    Though I can’t access the study, it sounds very much like an engagement with Guy Kahane’s work on this topic would help this project move forward (he’s just down the hall from you at Oxford). One big issue is that Trolley cases don’t actually distinguish between Deontology and Consequentialism. After all, people who think that we shouldn’t push the man off the bridge might be tacit rule-consequentialists. Moreover, W.D. Ross’s deontological theory says that there are duties of beneficence, and someone who thinks we should push the guy off the bridge might be thinking along those lines.

    I’d also be interested to know whether you tested the following hypothesis: people’s trust-judgments are made, not on the basis of ascribing some tacit moral theory to their targets, but on the basis of perceptions of motive. That is to say, people who make an easy decision to push the guy off the bridge are seen as heartless, people who agonize over it are seen as having the correct balance of motives. This seems far more plausible, as a psychological matter, and it highlights certain theoretical alternatives (neo-Humean or sentimentalist theories) which are sadly neglected in this literature.

  • Eyeore says:

    The “trolley problem” is framed somewhat dishonestly. The moral dilemma should be: “would you jump off the bridge to save 5 others” not whether you would, essentially, commit murder. If one’s motivation is to save other people, why would such a person not consider themselves as much of a potential “resource” as they would the person standing next to them? Interesting that those who pose this dilemma haven’t seemed to think of that.

    • Keith Tayler says:

      This is why this thought experiment is pretty much useless. There variants that prevent the decision making from throwing themselves off the bridge. Nonetheless, it cannot be treated as anything more than amusement or as an example of why thought experiments are of little or no help to ethics.

    • tmc says:

      Wouldn’t the fact the decider would be dead defeat the question of trusting that person? Who doesn’t trust Mohammad Atta…now, for example?

    • alex says:

      because it doesn’t test the willingness to sacrifice oneself or to save other people. It tests our moral beliefs about killing and letting die (Thomas Aquinas’ double effect)

  • Tyle says:

    Nick took the words right out of my mouth, and then some. But I also wanted to bring up a much narrower point. In passing, you assert that consequentialist views are more flexible than deontological views. But it seems to me that these two axes are independent. A consequentialist theory could be very demanding, or not, and a deontological theory could be very demanding, or not. For example, a consequentialist might think that the fat man *must* be pushed off the bridge, or might think that while it would be good to do so, this is not morally required. Similarly, a deontologist might think that you *must* allow the five to die, or might think that while it would be good to do so, this is not morally required.

    It is true that under a consequentialist view, moral prescriptions can’t be summarized by universal rules. But I think that it is misleading to describe this as flexibility, because there may still be a definite and non-negotiable moral prescription in every situation.

    • Nick says:

      Doesn’t look like the authors are planning to respond to us. This is a shame, as their study will now just contribute to the mounting confusion that Kahane is trying so hard to clear up.

  • Paul Treanor says:

    Note that the trolley problem has been discussed here earlier in a real-world case, namely that of driverless cars (AVs). There is an emerging lobby by car manufacturers, to allow the vehicles to be programmed to drive over pedestrians and cyclists, when necessary to avoid harm to the passengers. If you surveyed public opinion, I suspect that the majority would agree with that, certainly in countries like Britain and Germany where there is widespread hostility to cyclists.

    That real-world case does cast doubt on the validity of the research by Jim Everett and Molly Crockett. I suspect that they have manipulated their results, by failing to specify the religion, politics, and ethnicity of the individuals in the trolley case. Would you push a large rabbi off a bridge, to save the lives of five Hamas fighters? You will get different answers to that question in Tel Aviv and Gaza. Would you push an ISIS fighter off a bridge, to save the life of David Cameron? Would you push a dog off a bridge, to save the life of Donald Trump? And would *not* pushing the dog make you popular with Trump supporters?

    I understand that abstraction from the real world is necessary for this kind of experiment. However, when any kind of political, religious or ethnic preference can affect the results, then the experiment must be designed to take account of that. On the basis of the summary here, that does not appear to be the case.

  • Papadoc says:

    I’m seeing no correlation of this trolley concept with societal moral issues. Humanitarian morality does not exist as much as the humanitarian wants to believe. Morality is discerning right from wrong, as we all agree. Humans did not invent morality. You can recognize right from wrong only if you know what they are. Morality comes from God, not man.


Subscribe Via Email