In her memoir, Eva Hoffman, a writer and academic, reflects on being a bilingual and bicultural immigrant to North America from Poland. She describes two languages issuing competing commands in her head:
—Should you become a pianist?
the question comes in English.
—No, you mustn't. You can't.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
—Should you become a pianist?
the question echoes in Polish.
—Yes, you must. At all costs.
Typically we regard language as conveying information, not changing it. Yet numerous bilinguals have noted anecdotally that their first language feels more emotional to them. As Hoffman—and a growing body of research—illustrates, the tenor of our thoughts can depend on the tongue in which they are spoken. For example, in psychotherapy with a dual-language therapist, bilinguals sometimes use their mother tongue when they want to feel the full impact of a topic but then switch to another language to achieve emotional distance. Now evidence suggests that this “foreign language effect” even applies to our moral judgments.
The idea that different languages can lead us to distinct conclusions has wide-ranging implications. For couples from different linguistic backgrounds, the effect could sway the emotional give-and-take between two lovers. In foreign policy it might nudge certain leaders toward a more rational decision and others toward a more intuitive one. At a more fundamental level, the foreign language effect raises questions about human moral integrity. Your opinions of what is right and what is wrong are not absolute. They can be swayed by seemingly trivial conditions—including the language in which you frame your belief.
At the Whim of Language
The foreign language effect draws some of its influence from the ethos of its culture. For example, when Chinese-English bilinguals were randomly assigned to answer a self-esteem questionnaire in either Chinese or English, those who responded in Chinese received lower self-esteem scores than those who replied in English. Reading self-esteem questions in English cues bicultural respondents to adopt an American self-enhancing bias. Considering the same questions in Chinese might lead respondents to draw on the traditional Chinese virtue of modesty.
Yet culture is not the only reason. Presented with classic scenarios from behavioral economics in which people routinely make slightly irrational choices, bilinguals behave somewhat more logically when evaluating vignettes written in their nonnative language. Speaking a language with imperfect fluency, it seems, reduces the engagement of the brain's emotional circuitry.
New research on the subject, published in 2014, ventures into the realm of morality. Cognitive scientist Albert Costa of Pompeu Fabra University in Barcelona led a team that investigated how language might influence decisions about the “trolley problem,” a well-known scenario used by philosophers and psychologists to gauge people's moral reasoning. It goes like this: a trolley is speeding toward five oblivious track workers. You are standing nearby, close to a fork in the tracks. You have time to flip a switch to divert the trolley onto a different track, where only one worker is stationed. Should you flip the switch, thus killing one person and saving the other five lives?
According to a utilitarian view of the situation, saving more workers amounts to the greater good, so it is morally acceptable to kill one to save five. Most people—close to 70 percent, in some studies—consider it permissible to flip the switch. When psychologist Joshua Greene of Harvard University presented this and similar dilemmas to people while they underwent brain scans, he found activation in the executive function areas of the prefrontal cortex, located right behind the forehead, which suggests that people were consciously reasoning through this decision.
A related dilemma requires doing more than flipping a switch—now you must kill someone with your own hands. You are positioned on a footbridge over a train track with the trolley speeding toward five workers. Next to you stands a large man wearing a heavy backpack. If you shove this man off the bridge onto the tracks below, he will die, but he and his heavy backpack will stop the trolley, thus saving the five workers. Respondents usually report aversive feelings such as a sense of horror at the request to push someone to certain death. Typically permissibility ratings for this dilemma are in the range of 12 to 20 percent—far below the 70 percent for the switch scenario. In Greene's neuroimaging study, the footbridge scenario—with its requirement to personally kill someone—elicits strong activity in areas of the brain that mediate emotional responsiveness.
Now imagine evaluating the footbridge dilemma in either your native tongue or a foreign language, in which emotions are expected to be less active. As you might expect, Costa and his colleagues found that only 20 percent of respondents using their native language said that personally spilling blood to save five people was permissible. When evaluating in a foreign language, 33 percent of respondents considered bloodshed permissible. The effect held across several language pairings, including Spanish/English (recruited from the U.S.), Korean/English (Korea), English/French (France), and Spanish/Hebrew and English/Hebrew (both in Israel).
To rule out any effects from the cultural norms linked with a language, the researchers next compared responses to the footbridge and flip-switch dilemmas from native Spanish speakers who had learned English and native English speakers who had picked up Spanish. No matter the language, more respondents favored the utilitarian option in the foreign tongue (44 percent) than in the native language (18 percent, on average). No such difference was found for the switch-flipping scenario. This finding suggests that the foreign language effect is specific to moral dilemmas with a strong emotional component.
Tribal Thinking
Although Costa and his colleagues sampled from a variety of native and foreign language pairings—an admirable accomplishment—a different cultural phenomenon might also explain the results. Using a native language could induce the feeling of reasoning about your own people—your so-called in-group. Conversely, a foreign language could signal that the scenario is more relevant to strangers and outsiders. Indeed, studies have shown that assigning specific ethnicities to the victims in the trolley problem can change the pattern of responses in complex ways.
The research also raises a question: What about reasoning in a second language that is not foreign? People who reside for years in another country may ultimately feel as proficient and emotionally engaged when speaking its language as when using the language of their homeland. Or consider the children of immigrants who grow up to become “balanced” bilinguals. They commonly identify more strongly with the language of their parents but achieve similarly high proficiency in both that language and the language of the dominant culture. Would these individuals also think in a more utilitarian way when using their second language?
Although these groups have not yet been studied in moral-reasoning dilemmas, my bet is no. I make this prediction based on my own research on the emotional charge of different languages. In one study of Spanish/English bilinguals, for example, I recorded the electrical conductance of their skin while they listened to emotional phrases in their first or second language. Even a slight change in sweat levels can indicate that feelings are stirring, so we used this technique to measure our participants' reactions.
People without extensive immersion in their second language had lower skin conductance in the foreign language. The skin readouts of balanced bilinguals, however, were about the same. Thus, my conclusion is that a language elicits strong feelings when it is used routinely in emotional contexts—as in conversations between a parent and child or two old friends.
Whether or not you are multilingual, these findings reinforce what psychologists have said for decades: we humans are chameleons. We can be different people in different situations, bristling when an intimate partner criticizes us but apologetic when a co-worker does so. Some philosophers claim that belief in a unitary self is an illusion, albeit a helpful one. So, praemonitus praemunitus! Or put in another language: by understanding the many forces shaping our thoughts and behavior, you are forewarned and forearmed.