Marek Kohn—

Amnesia inclines us to assume that entry into language is painless, as does the apparent ease with which children typically become speakers. But many if not most skills require struggle to acquire, even if they seem effortless once mastered. We take language to be one of the most fundamental characteristics of being human. Why should we assume that entering into humanity is easy, any more than being born into the world is easy? Perhaps babies cry over words as well as milk and teeth.

Like the adult, the infant finds itself in a world that is not entirely unfamiliar to it. Both already recognise some regularities. If the adult did not already have some knowledge of the structure of the language, the din would just be noise. Unless the hearer is able to discern elements within it such as sounds, words or phrases, their repetition will serve no constructive purpose. Typically developing infants emerge with two or three months’ experience of hearing sounds, albeit muffled by maternal tissue and amniotic fluid. The most significant sounds they hear are their mothers’ voices. Changes in foetal heartbeat indicate that, even in the womb, they come to distinguish between the voices of mothers and others. Although the insulation surrounding them filters out higher frequencies, the rhythms of speech get through. So do sufficient vowel sounds for newborns to notice a difference between native and non-native ones. They also have a sense of the melody of a language, which they reproduce in their crying. When a group of French newborns and a group of German ones were compared, the French babies’ cries typically followed a rising melodic contour, increasing gradually in pitch and then falling away; the German cries traced a falling contour, rising at the beginning and then lowering gently. (German cries were also at their loudest soon after they began; French ones took longer to reach peak volume.) The differences followed the characteristic intonation of each language: in French, pitch tends to rise towards the end of words or phrases, before dropping right at the end; in German, pitch is more likely to peak on an accented syllable before gradually declining as the phrase is completed.

Babies bring a preference for their mothers’ voices out of the womb with them. Scientists gave newborn infants the means to exercise choices, in the form of an audio system connected to an artificial nipple that registered when it was sucked. The babies heard recordings of their mothers’ voices and of other women, reading And To Think That I Saw It On Mulberry Street, the first Dr. Seuss book. They found that by spacing how frequently they sucked the nipple they could select which voices they heard, and they chose their mothers’. A newborn baby will lie more still, its heart beating more slowly, if it hears its mother’s voice instead of an unfamiliar woman speaking. Newborns prefer to hear their mothers’ voices in the form familiar from the womb, with the higher frequencies filtered out.

These preferences are aspects of a fundamental bias for speech, as opposed to other kinds of sound. Given a choice between syllables and sounds that are not speech but have similar acoustic forms, newborn babies incline to the syllables. They also prefer the rhythms of speech to which they became accustomed before they were born. Some languages, such as French and Spanish, give roughly equal time to each syllable; others, such as German and Dutch, make the intervals between stressed syllables the same length, and shorten the ones in between to fit. Newborn babies pay more attention to speech in their mother tongue than to speech in a language that has a different basic rhythm pattern. Babies who had two mother tongues, because they heard their mothers speaking both English and Tagalog during the last trimester of pregnancy, showed no preference for one over the other. They were, however, able to distinguish between the two. (Tagalog, a major language of the Philippines, is syllable-timed; English, like its German and Dutch cousins, is stress-timed.)

The benefits that babies gain from the bias towards speech and the ability to distinguish its sounds are not confined to language learning. Preferences for the mother’s voice and the mother tongue are aspects of the preference in which the infant’s early life is enveloped, for the mother herself. By paying attention to the sound of her voice, and responding warmly to it, the infant will intensify the bond with her. That is probably the most effective way in which babies can help themselves to survive and flourish. An ear for differences in speech will also help a baby tune in to other caregivers. Language has a privileged place in development not because it is uniquely human, but because it is part of a complex through which infants secure their basic needs.

From Four Words for Friend by Marek Kohn. Published by Yale University Press in 2020. Reproduced with permission.

Marek Kohn is the author of The Race Gallery, As We Know It, A Reason for Everything, and Trust.

Further Reading: