Five-month-old babies prefer their own languages and shun foreign accents


Blogging on Peer-Reviewed ResearchDiscriminating against people who do not speak your language is a big problem. A new study suggests that the preferences that lead to these problems are hard-wired at a very young age. Even five-month-old infants, who can't speak themselves, have preferences for native speakers and native accents.

The Tower of Babel story highlights the conflicts that can arise when people don't speak the same language.The human talent for language is one of our crowning evolutionary achievements, allowing us to easily and accurately communicate with our fellows. But as the Biblical story of the Tower of Babel relates, linguistic differences can serve to drive us apart and act as massive barriers between different social groups.

These barriers can give rise to linguistic discrimination, a far more insidious problem that it seems at first. Language-based prejudices have led to horrific acts of human abuse, and even civil wars. Genocide often finds itself paired with linguicide, since a race can be killed off more thoroughly if their language follows them.

Even today, people in a linguistic minority can find themselves denied access to healthcare, or at a disadvantage when looking for jobs. The issue cuts to the heart of several ongoing debates, from the role of second languages in education to whether immigrants must become fluent in the tongue of their host country.

It should therefore be unsurprising to learn that we have strong preferences for our own language and for those who speak it. But Katherine Kinzler and colleagues from Harvard University, have found that we develop these preferences from an incredibly young age, before we can speak ourselves, and well before we can even hope to understand the social issues at stake.

Kinzler tested 24 infants, aged 5 to 6 months, from households that only spoke English, to see if they had any linguistic preferences. Each toddler watched videos of two women, one speaking English and the other, Spanish. The women were all bilinguals and swapped the language they used in different trials to make sure that the babies weren't showing preferences for physical traits like skin colour.

The babies were then shown the two women side by side, but no longer speaking. They strongly expressed their preference for the English speakers by gazing at their screen for a longer time (measuring gaze time like this is a standard test used by child psychologists).

Once developed in early infancy, these preferences stick around into childhood, and most probably well beyond that. In very similar experiments, Kinzler found that older infants (10 months or so) prefer to accept toys from a woman who spoke their native language.

Even young infants can discriminate between their language and others.The babies, from either Boston or Paris, were shown alternating films of an English or French-speaking woman, who spoke for a while and then silently offered the child a toy. Two real toys then appeared on the table in front of the infant, and they were twice as likely to pick the one in front of the native speaker.

So even though the offering of the toy involved no spoken words, the infants still gravitated towards the woman who had spoken earlier in their familiar tongue.

Infants can even pick up on subtle differences in dialect. Even when two speakers are talking in the same language, 5-month old infants will prefer someone who speaks with a native accent to someone who speaks with a foreign twang. Older children (5 years or so) will similarly prefer to befriend another child who speaks with the same accent.

At that age, children will have barely any understanding of the social circumstances that leads to different groups of people speaking the same language in different ways. And it's unlikely that their parents had much influences, since even the 5-month-old toddlers had these preferences.

These early preferences can act as the foundations for more destructive behaviours and conflicts later on in life. But we must be very careful - an instinctive basis for a behaviour does not in any way justify it.

Instead, by telling us about the basis of linguistic prejudices, these results suggest that we must work even harder to overcome them. If they are hard-wired from an early age, then education from an early age seems like a sensible first step.

Perhaps, exposure to multiple languages early in life can soften these preferences, and it would be fascinating to see if the same results hold for babies from bilingual households.

Reference: Kinzler, Dupoux & Spelke. 2007. The native language of social cognition. PNAS 104: 12577-12580

More on child development: 


More like this

Interesting... an anecdote: I used to work for a large international circus, one of the neat parts about being with them was they travelled with a school for the performer's children. these are children of parents from all around the world. They were taught in english and french, but were surrounded by russian, chinese, korean, spanish etc.

It was actually quite irritating. (in a jealousy sort of way...) to be around some of these children as every time they came to a point in a sentence that they couldn't remember the word in a particular language they were speaking at the moment, they would switch to another language and the conversation would continue on in the new language.

Why would the observer infer anything from this experiment, other than the obvious: the infant has heard only one language from within the womb and without, and is simply responding to the one that sounds most like mom. Infants' eyesight develops in ways to bond infant and mother(at birth, the typical infant can focus in a vague way on faces, and focuses best at a distance that approximates the distance from mother's arms to her face), hormones in mother develop to bond infant and mother, why would language not also develop in such a way to bond infant and mother? I hardly think that this can be said to determine language preference...children have evolved to prefer mom, period, and to be able to identify her as a matter of survival.

It would be interesting to do this experiment on older children, perhaps with someone reading in their native language word a child doesn't understand (a technical manual or medical article?), and see if the same prejudices occur.

I am definitely not a linguist. I'm just an interested layman.

@coryy: Two speakers of a particular language only "sound the same" on a very abstract level. A word spoken by two different people, or even the same person at different times, can have very different waveforms. It takes an enormous amount of mental processing to parse an utterence into phonemes, or otherwise determine whether a speaker is using a familiar language. The research shows that quite young infants* have emotional reactions based on this high-level processing. It is not at all obvious that this should be the case.

Ed Young said "Perhaps, exposure to multiple languages early in life can soften these preferences, and it would be fascinating to see if the same results hold for babies from bilingual households."

I would be very surprised if that turned out to be true. I suspect that children in bi-lingual households have a strong preference for the two languages (and accents) their care-givers use and a bias against everything else. A child exposed to enough linguistic diversity (thirty different languages, say) might be less discriminating, but such a child would have a very hard time learning to talk!

*A five-month-old is not a toddler.

By Grep Agni (not verified) on 17 Jun 2009 #permalink

Sorry about miss-spelling your name above, Mr. Yong. I've been reading it incorrectly for a few weeks, now.

By Grep Agni (not verified) on 17 Jun 2009 #permalink

@Grep Agni:

"Two speakers of a particular language only "sound the same" on a very abstract level." Layman you may be, but I doubt an expert could cut to the heart of this matter any more directly. It is indeed a stunning claim that five month olds seem to have the ability to classify speakers according to language.

I am bilingual myself and I still remember the intense shock my infant seemed to register the first time I happened to speak to her in my 'other' language, complete with very different intonation. She was a few weeks old at the time. She herself was not a happy bilingual in her early years.