The Crying Song
A new study finds the roots of language in newborns' first sounds
"Crying baby" and "music" don't usually go together, but, hey, interesting science starts with questions no one else thought to ask—in this case, what are the musical contours of a newborn's cry? There had already been provocative research on what sounds a fetus can hear in the womb and what effect that has right after birth, with several research teams finding that newborns prefer their mothers' voices over those of other people, as in studies such as this and this. That makes sense, since Mom's voice is what a baby heard most for nine months. Newborns also prefer their native tongue to other languages for the same reason.
Now an intrepid team of scientists, three from Germany and one from France, has gone an intriguing step further: they have found that newborns cry in their native language. "We have provided evidence that language begins with the very first cry melodies," says Kathleen Wermke of the University of Würzburg, Germany, who led the research.
The reason I call her team "intrepid" is that they recorded 2,500 cries of newborn babies, 30 French and 30 German, between 2 and 5 days old. So a tip of the hat for sacrificing eardrums in pursuit of knowledge.
The idea was to extend the existing findings about what sounds babies can perceive—their native language, their mother's voice—to test what sounds they can create. Once the researchers had their recordings (no babies were harmed in the course of this research! All crying was spontaneous, due to hunger or thirst or general unhappiness rather than pain, as from having blood drawn), they set to work analyzing the cries' melodic qualities.
As anyone who has had the pleasure of hearing a crying baby for any length of time knows, the little tykes produce wails that vary in pitch. Since most of us have heard only the cries of babies whose parents speak our language (the exception being on international flights, where you can hear a veritable United Nations of crying babies), we tend not to think twice about the pattern of pitch changes. A crying baby is a crying baby.
But that is not what the scientists found. French babies tended to cry "with a rising melody contour," they will report in the December issue of the journal Current Biology, posted online Thursday. The cries sounded French: the pitch changed from low to high, rising toward the end of words as well as phrases within a sentence (though the final sound of a sentence has a lower pitch). In contrast, the German babies' cries had falling melodic contours. They sounded German: the pitch fell from high to low, which is consistent with the sound of German's falling melody contour, from the accented high-pitch syllable at the start of a phrase or word to the lower pitch at the end of a phrase. A French child says "papa," while a German one says "papa." There is, in short, "a tendency for infants to utter melody contours similar to those perceived prenatally," write the scientists.
"The dramatic finding of this study is that not only are [newborns] capable of producing different cry melodies, but they prefer to produce those melody patterns that are typical for the ambient language they have heard during their fetal life, within the last trimester," said Wermke. "Contrary to orthodox interpretations, these data support the importance of human infants' crying for seeding language development."
It had been thought that babies' cries are constrained by their breathing patterns and respiratory apparatus, in which case a crying baby would sound like a crying baby no matter what the culture, since babies are anatomically identical. "The prevailing opinion used to be that newborns could not actively influence their production of sound," says Wermke. This study refutes that claim: since babies cry in different languages, they must have some control (presumably unconscious) over what they sound like rather than being constrained by the acoustical properties of their lungs, throat, mouth, and larynx. If respiration alone dictated what a cry sounded like, all babies would cry with a falling-pitch pattern, since that's what happens as you run out of breath and air pressure on the throat's sound-making machinery decreases. French babies apparently didn't get that memo. "German and French infants produce different types of cries, even though they share the same physiology," the scientists point out. "The French newborns produce 'nonphysiological' rising patterns," showing that the sound of their cries is under their control.
Although phonemes—speech sounds such as "ki" or "sh"—don't cross the abdominal barrier and reach the fetus, so-called prosodic characteristics of speech do. These are the variations in pitch, rhythm, and intensity that characterize each language. Just as newborns remember and prefer actual songs that they heard in utero, it seems, so they remember and prefer both the sound of Mom's voice and the melodic signature of her language.
The idea of the study wasn't to make the sound of a screaming baby more interesting to listeners—good luck with that—but to explore how babies acquire speech. That acquisition, it is now clear, begins months before birth, probably in the third trimester. Newborns "not only have memorized the main intonation patterns of their respective surrounding language but are also able to reproduce these patterns in their own [sound] production," conclude the scientists. Newborns' "cries are already tuned toward their native language," giving them a head start on sounding French or German (or, presumably, English or American or Chinese or anything else: the scientists are collecting cries from more languages). This is likely part of the explanation for how babies develop spoken language quickly and seemingly without effort. Sure, we may come into the world wired for language (thank you, Noam Chomsky), but we also benefit from the environmental exposure that tells us which language.
Until this study, scientists thought that babies became capable of vocal imitation no earlier than 12 weeks of age. That's when infants listening to an adult speaker producing vowels can parrot the sound. But that's the beginning of true speech. It's sort of amazing that it took this long for scientists to realize that if they want to see what sounds babies can perceive, remember, and play back, they should look at the sound babies produce best. So let the little angel cry: she's practicing to acquire language. Question to readers: did your cranky, squalling newborn speak earlier than your placid, quiet one?