In the past, scientific studies of language acquisition focused mainly on monolingual children learning their native tongue, be it English, Chinese, French or one of the many other languages of the world. Such a focus was surprising, since there are fewer monolingual people across the globe than there are bilinguals. Studies show that the vast majority of the world's children actually grow up learning more than one language simultaneously. And, with the huge changes in population distribution in recent decades, even here in the UK, bilingualism is becoming the norm rather than the exception.
An increasing number of pupils at all levels of schooling, from nurseries to colleges, hear English at school but go home to households where the native language is different. In response to this, scientists have broadened their perspectives and are now examining language acquisition in new ways to discover whether there are significant differences in the development of infants who grow up monolingually and those simultaneously mastering two or more languages. These comparisons are yielding interesting results that have prompted further questions. One of the most fundamental questions is, does bilingualism confer a cognitive advantage to young children or does it instead slow their learning process?
SPEECH VS LANGUAGE
Before we begin to tackle this question, we must point out an important distinction between 'speech' and 'language'. Speech is about sound patterns, whereas language refers to structure and meaning.
The languages of the world can be divided into three families, according to the rhythmic structures of their sound patterns:
1. Stress-based languages such as English, Dutch and German, where stress is placed differently within words (for example, in English, /dUctor/ versus /guit.r/)
2. Syllable-based languages like French, Spanish and Catalan, where stress in most words is equal across all syllables
3. Mora-based languages like Japanese, where the rhythmic structure is measured in terms of whether a syllable is long or short (for example, /Honda/ in Japanese has three units, a long 2-unit /hon/ and a short 1-unit /da/).
At birth, infants know nothing about language per se, but they do know something about speech. As we saw in 'Life before birth' (Developing Brain series, Part 1, Nursery World 19 May 2010), from the seventh month in the womb a foetus can hear its mother's voice. So, several weeks before birth, babies are already becoming accustomed to the sound pattern of their mother's speech.
Foetuses and babies who are exposed to more than one language must, therefore, learn both to distinguish between and develop an ear for different, and potentially conflicting, sound patterns. But for the young language-learner, does this additional demand translate into a developmental advantage (an additional level of sensitivity, facilitating language acquisition), or a developmental disadvantage (an additional load hindering progress)?
Research shows that just hours after birth, infants can already distinguish their mother's voice from other female voices, and respond differentially to recordings of speech in the mother's tongue versus speech in languages with contrasting sound patterns (for example, English vs Chinese). In fact, newborns show a clear preference for their mother-tongue and will suck harder on a dummy to hear it. But what if a baby has heard two different languages in utero and is born into a bilingual household?
Janet Werker, an infant speech processing expert from Canada, set out to answer this question. Her studies show that newborns who heard their mothers regularly speak two languages during pregnancy showed no preferential sucking for either language. This suggests that right from birth, monolingual children display speech sensitivities that are different from those of multi-lingual children.
BILINGUAL VS MONOLINGUAL INFANTS
The results of studies of bilingual language acquisition are often confounded by the fact that they tend to focus on children from immigrant families with low socio-economic status and the potentially negative effects that learning two languages at once might have had on them.
With this in mind, it is difficult to ascertain whether any language delays found in these populations are due to bilingualism itself, to factors linked with low socio-economic status, or are instead the result of a combination of the two.
This type of research also takes place late in development, in that most bilingual studies examine the language - vocabulary and grammar - of school-age children. So, little is known about the infant roots of language outcome.
In order to surmount these problems, the Canadian research team led by Werker, as well as Nuria Sebastian-Galles from Spain, have looked specifically at bilingualism in educated families of middle socio-economic status in order to remove immigration and poverty as a confound. Moreover, instead of looking exclusively at later language acquisition, they have focused on early speech perception - the early roots of sensitivity to language-related input.
In one set of studies, Sebastian-Galles compared infants raised in bilingual Spanish-Catalan families with those raised in monolingual environments. She further distinguished between infants raised with two languages from the same language family (Spanish/Catalan) and infants raised with two languages from different language families (Spanish/English). Her aim was to discover whether the bilingual babies showed any advantages over their monolingual counterparts. Or would their speech perception abilities turn out to be worse? She also wanted to know if there were significant differences between infants brought up with, for instance, English and Spanish, compared with infants brought up with more closely related languages like Catalan and Spanish.
The Spanish studies revealed that bilingual upbringing confers both advantages and disadvantages. All infants, whether monolingual or bilingual, are born with the ability to discriminate between sounds from all the world's languages. Even if they only hear, for example, English in their immediate environment, they can still hear the differences between sounds from very different unfamiliar languages such as Hindi, Japanese, Swahili or French.
However, by around nine months, monolingual infants exposed to only one language undergo what is known as 'perceptual narrowing' - that is, they specialise in the fine discrimination of the sounds from their native tongue and lose the ability to discriminate language sounds that do not belong to their own language. By that age, they also become proficient at detecting subtle mispronunciations in their mother tongue (for example, they will react if a speaker says /gog/ instead of /dog/). But what about infants raised in families where two languages are spoken?
Interestingly, Sebastian-Galles found that infants brought up bilingually were delayed in both perceptual narrowing and the ability to detect mispronunciations. It is as if the brain of the bilingual infant takes longer to commit to the sounds of the different speech patterns that it processes. In other words, it has a more flexible approach to speech sounds because of hearing two ranges of different sounds and different words referring to the same object (for example, /dog/ /perro/). This flexibility results in delayed specialisation.
Similar research in Canada suggests that bilingual babies actually pay less attention to detail; they initially process speech sounds more globally than their monolingual counterparts. But this is a just a minor delay. Within a few weeks or months, bilingual infants also develop perceptual narrowing, but for two languages rather than one. This is true of infants hearing speech from two different language families, such as English and French.
Interestingly, infants exposed to two languages with similar intonation patterns, like Spanish and Catalan, show an advance over monolingual infants. The Spanish-Catalan bilinguals succeeded earlier in discriminating languages from the same language family.
To summarise, monolingual infants are earlier at discriminating their mother-tongue from languages from different language families but take longer to discriminate between languages within the same language family. By contrast, bilingual infants are able to make these fine, within-language-family discriminations earlier than monolinguals.
LEARNING VOCABULARY: BILINGUAL TODDLERS
Another difference between monolingual and bilingual children emerges during toddlerhood. It had been known for some time that monolingual children abide by what is called the Mutual Exclusivity Constraint. This is tested in the following way.
The toddler is shown two toys (say a car, for which he already knows the name, together with a strange-shaped object that he's never seen before and doesn't have a name for).
The experimenter then asks: 'Give me the blinket' (a nonsense word). Children abiding by the Mutual Exclusivity Constraint will pass the object for which they don't yet have a name, presuming that the car would not have two different names. However, children raised bilingually will pass either object. In other words, they don't abide by the Mutual Exclusivity Constraint when learning vocabulary, because they are accustomed to objects having more than one name.
DELAYED OR ADVANCED?
We now know that, despite the findings from earlier research on families from low socio-economic backgrounds, bilingual infants and toddlers are not in fact delayed in language acquisition. Nor are they simply in advance.
The pattern of acquisition differs between monolinguals and bilinguals as a function of different aspects of speech and language. The best way to capture the differences between monolingual and bilingual children is to recognise that each group follows a somewhat different, yet highly adaptive developmental trajectory, according to the environments in which they are raised. In other words, there is more than one way to acquire one's native tongue.
NON-LINGUISTIC ADVANTAGES
Another study carried out on babies in continental Europe by Agnes Kovacs and Jacques Mehler found that infants who are raised in bilingual households have what psychologists call better 'executive function'.
Executive function involves the ability to plan, to solve problems, to rapidly shift attention between goals, and to monitor and control motor functions. Their experiment ran as follows: seven-month-olds who had been exposed to two languages from birth were compared with seven-month-olds who had experienced just one language.
The babies saw two white squares on a computer screen and heard a short, made-up word. There were nine words in total, and each time a word was said a puppet appeared immediately afterwards, always in the same square.
As the experiment proceeded, all the babies - bilingual or monolingual - started anticipating where the puppet would appear and looked at that square just before the puppet actually appeared. Clearly, this simple task doesn't require much in the way of executive function. However, the next nine trials had a puppet appear in the other square.
The infants' task was then to learn that the link between words and where the puppet appeared had changed. Interestingly, only the bilingual seven-month-olds succeeded. Unlike their monolingual peers, the bilinguals revealed an early capacity for attention switching.
Moreover, the advanced executive function of the bilinguals wasn't restricted to this task. It also generalised to their visual abilities. When the puppet's appearance was prompted by three shapes instead of nonsense words, bilingual infants were still superior to monolingual infants at shifting attention and changing their predictive gazes.
Thus, bilingual infants are somewhat better than their monolingual counterparts at inhibiting automatic reactions and in using mental control. This is likely to arise from their constant need to focus on one language, while simultaneously inhibiting interference from the other language, and interestingly it generalises beyond language-relevant tasks to general cognition. So, even before they can speak, a bilingual advantage emerges in the ability to suppress something already learned and to update predictions using new information - a very useful cognitive skill.
GOOD FOR ADULT BRAINS
Although infants can acquire two languages simultaneously with amazing ease, learning a new language as an adult can be a painfully embarrassing task. Yet researchers have recently found that being multi-lingual is even a protective factor against ageing! So, although being brought up bilingually gives rise to a few temporary delays in child development, it seems to confer lifelong benefits.
REFERENCES
- - Bialystok, E and Craik, FIM (2009). 'Cognitive and linguistic processing in the bilingual mind', in Current Directions in Psychological Science 19(1), 19-23
- - Byers-Heinlein, K, Burns, TC, Werker, JF (2010). 'The roots of bilingualism in newborns' in Psychological Science 21(3), 343-348
- - Byers-Heinlein, K, and Werker, JF (2009). 'Monolingual, bilingual, trilingual: Infants' language experience influences development of a word learning heuristic', in Developmental Science 14(4), 488-499
- - Karmiloff, K and Karmiloff-Smith, A (2000). Pathways to Language: From foetus to adolescent. Developing Child Series, Harvard University Press
- - Karmiloff-Smith, A (2010). 'Multiple trajectories to human language acqusition: Domain-specific or domain-general?' in Human Development 53, 239-244
- - Sebastian-Galles, N (2010). 'Bilingual language acquisition: Where does the difference lie?' in Human Development 53, 245-255
- - Werker, JF, and Byers-Heinlein, K (2008). 'Bilingualism in infancy: First steps in perception and comprehension of language' in Trends in Cognitive Sciences 12(4), 144-151
- - Werker, JF, Byers-Heinlein, K, Fennell, C (2009). 'Bilingual beginnings to learning words'. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1536), 3649-3663.
MOTHER AND DAUGHTER
Annette Karmiloff-Smith studied in Geneva with Jean Piaget, where she completed her doctorate, and is now a 
Professorial Research Fellow at the Centre for Brain and Cognitive Development, Birkbeck, University of London. She was awarded a CBE for her work on cognitive neuroscience, and is the author of eight books and more than 200 scientific articles and chapters. Her research covers infancy to adolescence across numerous cognitive domains, with a particular speciality in infants with genetic syndromes.
Kyra Karmiloff has a BSc in Anthropology and an MSc in Psychological Research Methods from University College London. Before giving birth to her three sons, she was a research assistant at the Centre for Studies in Language at Cambridge University.
As well as being a novelist, she is co-author with her mother of three books on child development and on language acquisition.
