Why behaving morally might make people like you less
I've been vegan for a while now. Although usually I don't bring it up too often, occasionally it has to be shared to explain why I say no to certain foods. Something that has happened quite often is that people start explaining why they eat meat, or why being vegan does not make you a morally superior person. In fact, the last time someone was telling me why they could never be vegan, they mentioned that people only do it to make themselves seem like better people.
Online and in real life there's quite a large dislike for vegans and other people that try to do something they belief is morally good. For instance, a common attack against progressive-green parties in the Netherlands is that they try to be the "best-behaved kid in class". Why do people have negative reactions to others who try to do good?
Do-gooder derogation
Minson and Monin (2012) coined the term "do-gooder derogation" to describe the social backlash that morally motivated people tend to attract. The paper studied reactions of meat-eaters to vegetarians as an example. In the first study, 47% of participants freely associated negative terms with vegetarians, words like "judgmental" or "self-righteous," and the negativity of those words correlated with how much participants expected vegetarians to see themselves as morally superior. In a second study, if they asked participants to imagine how a vegetarian would view their diet, they would give lower ratings to the vegetarian. To be clear: the vegetarians in question didn't say anything, it was perceived judgement where there was none.
The core finding is that the derogation is not really about the do-gooder's actual behavior. Rather, it seems to be about the contrast between the behavior of the morally-motivated person and the subject. People seem to be defending themselves against a judgement that hasn't been given.
Moral rebels and anticipated judgment
Monin, Sawyer, and Marquez (2008) documented this across four studies. The setup was that participants either went along with a morally uncomfortable task, or watched as an observer. They then rated someone who had refused to do it on principle. In one study that task was recording a speech against a popular university policy. In another it was a suspect lineup where the obvious guilty party, based on circumstantial details, was the only Black suspect. In both cases, active participant who had already complied with the task disliked the refuser; observers liked them.
The mechanism is thought to be imagined rejection. Active participants expected that the rebel would think poorly of them for going along, leading to them disliking the rebel. Even though the rebel wasn't present anymore and did not have interactions with the participant. Interestingly, in the same paper they showed that participants who wrote about an important personal value beforehand showed no derogation at all, and even questioned their own choice afterward. The threat was to self-image, and shoring that up elsewhere helped defuse the situation.
It even appears in children: Tasimi, Dominguez, and Wynn (2015) tested children aged 8 to 10 in a donation task. Without a comparison condition, 94% of children preferred the more generous character. When children were placed in a condition where another child had donated more than they had, that preference dropped by about 20%. The effect did not appear when the generous individual was an adult. It looks like social comparison with a peer who has outperformed you is what triggers the response.
The expulsion test
The effect can go as far as to work against self-interest. Parks and Stone (2010) ran public goods experiments where participants could vote to expel other players from the group. Players that shared too much, those who gave far more than the group average, were expelled about as frequently as people who shared too little. The most common justification given for expelling a highly generous player was that their contributions made everyone else's look comparatively unimpressive. Paradoxically, contributing a lot to a shared resource can make you look bad!
Herrmann, Thöni, and Gächter (2008) ran public goods games across 16 different participant pools from 15 countries. The antisocial punishment of high contributors showed up across cultures, not just in what we typically view as "individualistic" societies like Western Europe or North America. The intensity varied, and the paper noted some differences between regions, but the basic pattern of punishing over-contributors was widespread enough.
Why does this happen?
The first mechanism identified by research is anticipated moral reproach. When someone around you makes a visibly moral choice, you start running a simulation of what they think about you. A convincing piece of evidence comes from the Monin et al. paper: when the threat to self-image is removed by having a person reaffirm their own values beforehand, the negative response to the do-gooder disappears entirely.
The second is social comparison. The do-gooder raises the bar, and everyone else now looks less generous by proximity. The person who gives more is not just doing a good thing, they are also making you look worse by contrast. This is what the children's study and the public goods expulsion results point to most directly.
These two mechanisms are not mutually exclusive. In most real situations they probably compound each other.
It's a frustrating effect to learn about, because it means that trying to do good comes with a social tax. The research points some practical points: framing your choices as personal rather than universal reduces the friction, since there is less room for someone to read your choice as a verdict on their own. Secondly, publicly visible moral choices also attract more suspicion than quiet ones, since people can infer that you want credit for it. These points have really bad implications for us all. If the path of least resistance for a do-gooder is to keep quiet and frame good behaviour as personal choices, good behaviour becomes harder to observe, normalize, and spread.
