Using a foreign language changes moral decisions

Would you sacrifice one person to save five? Such moral choices could depend on whether you are using a foreign language or your native tongue. A new study from psychologists at the University of Chicago and Pompeu Fabra University in Barcelona finds that people using a foreign language take a relatively utilitarian approach to moral dilemmas, making decisions based on assessments of what's best for the common good. That pattern holds even when the utilitarian choice would produce an emotionally difficult outcome, such as sacrificing one life so others could live.

"This discovery has important consequences for our globalized world, as many individuals make moral judgments in both native and foreign languages," says Boaz Keysar, Professor of Psychology at UChicago. "The real world implications could include an immigrant serving as a jury member in a trial, who may approach decision-making differently than a native-English speaker." Leading author Albert Costa, UPF psychologist adds that "deliberations at places like the United Nations, the European Union, large international corporations or investment firms can be better explained or made more predictable by this discovery."

The researchers propose that the foreign language elicits a reduced emotional response. That provides a psychological distance from emotional concerns when making moral decisions. Previous studies from both research groups independently found a similar effect for making economic decisions.

In the new study, two experiments using the well-known "trolley dilemma" tested the hypothesis that when faced with moral choices in a foreign language, people are more likely to respond with a utilitarian approach that is less emotional.

The first experiment presented with the "footbridge" scenario of the trolley dilemma. Study participants are asked to imagine they are standing on a footbridge overlooking a train track when they see that an on-coming train is about to kill five people. The only way to stop it is to push a heavy man off the footbridge in front of the train. That action will kill the man, but save the five people. In other words, were faced with the dilemma of choosing between actively sacrificing one person, which violates the moral prohibition against killing, or by inaction allowing five people to die.

The researchers collected data from people in the U.S., Spain, Korea, France and Israel. Across all populations, more participants selected the utilitarian choice—to save five by killing one—when the dilemmas were presented in the foreign language than when they did the problem in their native tongue.

Even with randomizing the participants' language groups, "those using a foreign language were twice as likely to respond with the utilitarian approach that is more in the service of the common good of saving more people," said lead author Albert Costa from the Center of Brain and Cognition, Universitat Pompeu Fabra, Barcelona. Costa is currently a visiting professor at UChicago.

The second experiment included a version of the dilemma that is less emotional. In this dilemma, the trolley is headed towards the five men, but you can switch it to another track where it would kill only one man. People tend to be more willing to sacrifice the one man by pulling a switch than by pushing him off the footbridge because the action is less emotionally intense, the researchers note. The language of presentation did not affect participants' decisions in this dilemma; with either language, the vast majority of people prefer the utilitarian option in this less emotional scenario.

The team evaluated data from 725 participants, including 397 native speakers of Spanish with English as a foreign language, and 328 native speakers of English with Spanish as a foreign language. Each participant received the two dilemmas either in their native language or a foreign one. When presented with the less emotional scenario, more than 80 percent of participants preferred to divert the train and that percentage remained high in their native and foreign language. On the other hand, when presented with the more emotional scenario, people are once again significantly more likely to sacrifice one to save five when making the choice in a foreign language.

Keysar says decisions appear to be made differently when processed in a foreign language. "People are less afraid of losses, more willing to take risks and much less emotionally-connected when thinking in a foreign language."

Co-author Sayuri Hayakawa, a UChicago doctoral student in psychology, says the way we learn the language is key. "You learn your native language as a child and it is part of your family and your culture," she said. "You probably learn in less emotional settings like a classroom and it takes extra effort. The emotional content of the language is often lost in translation."

"What this study tells us is that can be affected depending on whether the language in which it is presented is a native or foreign one," Costa said. "Awareness of this impact of languages on is fundamental to making more informed choices."

The study is published online by PLOS ONE.

Explore further

Study says empathy plays a key role in moral judgments

Journal information: PLoS ONE

Citation: Using a foreign language changes moral decisions (2014, April 28) retrieved 23 October 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Apr 29, 2014
These so called 'moral dilemma' games are riddled with flaws and should not be taken seriously.

How many people would trust a source of information that says that if they take a certain action, murdering an individual, that other people might be saved??

If the person in the real situation does nothing they will be free of blame. If they take responsibility for the situation, an enormous leap that few people have the confidence to make, especially when the murder of one individual is involved, then they will be charged with murder of that one individual.

It is up to a jury to decide if that person really was in possession of information reliable enough that five people would be killed if one person was not murdered and therefore the murder was justified.

Some jurisdictions would still charge and convict the person of manslaughter even if it was a choice between 5 or 1. (cont below)

Apr 29, 2014
Does this still apply if you're talking programming languages rather than 'people' languages ??

Apr 30, 2014
Non compos mentis.

The rational content of dilemma is zero. The researchers are welcome to infer more than this.
And they do.

May 02, 2014
One scenario that would cause far less confusion is a scenario where you are driving a car and the breaks fail. You can continue on the road, killing five, or swing into a side road killing one.

This removes the issues of taking responsibility and does not allow for the "no response" option (do nothing and avoid responsibility) and it removes the culpability dimension because you are going to be responsible for your actions either way, but not responsible for the resulting accident (unavoidable brake failure). It also removes the surety dimension ie you don't have to trust the source of the information.

There is also the issue of what else you could do, like sounding the horn and flashing the lights. But this is true for either outcome (side road or main road) and so it cancels out.

I bet that my scenario would get close to 100% swinging into the side road...

May 08, 2014
Russell, the researchers certainly can consider the rational content as zero, they are experienced intelligent researchers. But subjects do not behave in this way.

There is no explanation, given the 5:1 choice, why anyone, given the researcher's constraint on rational consideration, would choose 5 and the researchers never offer any rational explanation for this.

I have offered several rational explanations, mainly that 'noise' in the form of real life experience colours ordinary people's choice. I have also pointed out a similar model that does not have the noise component.

I had a similar argument with Prof. Herbert Gintis on game theory telling him that tribal people would respond differently because they treat games as real. When the experiment was finally performed on tribal people they behaved just as I predicted. The discussion in the Yahoo! groups' archive (Evolutionary Psychology).

May 15, 2014
The researchers never offered any rational explanation for content having no rational content.
You do. Too bad.

If you want to infer predictability from randomness no one will stand in your way.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more