When California reinstated bilingual education in a 2016 referendum, the text of the new law declared: “A large body of research has demonstrated the cognitive, economic, and long-term academic benefits of multilingualism and multiliteracy.” This argument that bilingual education is useful because it confers general mental benefits — such as improved “executive function” and “cognitive flexibility” — has proven irresistible for bilingual advocates. But it’s probably not true. As I detailed in an essay for The American Conservative last year, researchers have struggled to replicate the positive results from previous studies of bilingualism. Many are now skeptical that a “bilingual advantage” exists at all.
A new meta-analysis published in Psychological Bulletin provides more reason for doubt. The authors averaged the results of 152 studies and found little evidence that bilingualism confers any general cognitive benefit. So what happened to all of the exciting “bilingual advantage” research that we hear so much about? When the authors of the meta-analysis compared effect sizes across the various studies, they found that the most imprecise studies — imprecise in the sense that their small sample sizes introduced a lot of random error — tended to report the most positive results. This is classic evidence of publication bias, whereby shaky studies are more likely to be published if they produce results that go in a favored direction. After correction for publication bias, the effects of bilingualism on cognitive inhibition, monitoring, shifting, and working memory were all effectively zero. Effects on attention and verbal fluency were actually negative, albeit small.
“The idea that bilinguals outperform monolinguals in cognitive control functions seems to have already been accepted by the popular media and educators, because of a number of influential studies reporting a bilingual advantage,” the authors write. “Our thorough meta-analysis, however, suggests that healthy bilingual adults do not have such a cognitive control advantage.”
The broader lesson here is to beware of the “objective” answer to a fundamentally political question. For years the debate over bilingual education hinged on our subjective preferences for assimilation versus multiculturalism. Bilingual advocates short-circuited that debate, arguing in effect: “You may be worried about assimilation, but forget about it because Science says bilingual education makes people smarter. You’re not against Science, are you?” As we now know, science says no such thing. But the damage is already done, as English immersion is no longer the default pedagogy in the Golden State.