A colleague recently sent me a link to a piece in Scientific American Mind called “Accents Trump Skin Color.” Reporter Agata Gluszek reviews work by Katherine Kinzler and colleagues suggesting that, for young children, accent is as important as visual cues to race, gender, and age in selecting friends. The magazine article was interesting, and led me to look for the research paper it was based on.

That publication is K. Kinzler, K. Shutts, J. DeJesus, and E. Spelke, “Accent trumps race in guiding children’s social preferences,” Social Cognition 27(4), 2009, pp. 623–634. (It is available here, but only if you or your institution subscribes to the journal.)

The authors, social psychologists from Harvard University, suggest that their subjects, white English-monolingual 5 and 6 year olds in the Boston area, relied as strongly on language cues in choosing friends as they did on visual cues. This supports work in social psychology suggesting that people use non-visual cues, including ethnic categories or language choice, in various types of identity work.

In a series of experiments Kinzler and her colleagues showed 5- and 6-year-old white children pairs of photographs of children’s faces, matched with audio clips of children speaking. Subjects were asked to put a sticker on the picture of the child they would like to be friends with. In one experiment, when pictures were shown without any audio children chose to be friends with a child of their own race about 75% of the time. When audio clips showed the white pictures “speaking” French-accented English, however, the subjects chose the black, native-English speaking pictures more than 70% of the time. One way to interpret these findings, according to the authors, is “that children privilege information about how individuals sound over how they look in guiding their social preferences” (p. 629).

As I say, the findings are interesting and fairly persuasive (though of course I hardly need to be persuaded that language use is an important constituent of cultural identity). I do have some minor criticisms of the work as presented. For example, they find that white, English-speaking subjects prefer black children who speak with a native English accent over white children who speak French-accented English, but do not compare this to other possible combinations (black and French-accent, white and American accented). Its possible that including such comparisons would not have significantly changed the outcome, but it would be interesting to compare.

It was also slightly odd to see Mark Baker’s The Atoms of Language cited to support the assertion that language may have served as a better indicator of group membership than visual cues throughout evolutionary history. It has been some time since I read Baker’s book, but I recall him making almost the opposite argument – that languages, especially their syntactic parameters, are not strongly correlated with other aspects of culture.

Still, the paper is well done, generally persuasive in its conclusions, and relatively accessible even for non-psychologists with an interest in the social psychology of identity and language. The questions for future research raised in the paper might be profitably addressed by linguistic anthropologists as well as social psychologists: Do similar accent preferences hold for children from multilingual settings? How do language or accent preferences compare to gender or age preferences? Does in-group preference trump the social prestige of particular languages? Do these patterns change as children get older? Can children be led to overcome these prejudices?