Building multilingual vocabularies

As you saw in Step 2.4, word learning is a core aspect of language learning, and the number of words children know in a language – vocabulary size – is a good predictor of later language and literacy development in that language. In turn, vocabulary size has been linked to:

• Input: the amount and richness of the language that children hear from their adult caregivers. Hearing many different word types and many different types of sentences gives children access to a rich and varied language experience that facilitates their own learning. Conversations and book reading, where children take an active role in language interaction, are particularly conducive to word learning and to language learning in general.
• Processing speed: young children’s speed of processing, so the faster children are at recognising words when they hear them (processing them), the more words they will learn over time.

How input and processing speed vary and interact

Input

While language input is crucial regardless of whether children are learning one or many languages, it’s important to remember that input to multilingual children also varies according to which language the child is addressed in:

• whether it contains words from both languages (ie, code-switches)
• whether it comes from speakers who are themselves multilingual, and
• whether it accounts for a relative majority or minority of the total amount of input.

These are all important sources of variation in multilingual children’s linguistic environment that contribute to their word learning.

Processing speed

Another source of variation is children’s own speed of lexical processing. Speech is fast with an average of 2.5 words per second. Quick and accurate recognition of a word can free up resources to attend to other useful aspects of the context and therefore contribute to more detailed and more robust lexical representations. Children who are efficient at lexical processing will quickly recognise a word in the speech stream and will need fewer exposures to the phonological form of the word, speeding up the process of vocabulary acquisition.

As well as varying among children for the reasons outlined above, speed and input also interact. The speed of lexical processing is input-dependent and research with bilingual children¹ has shown that it is language-specific. Relative lexical processing efficiency in children’s two languages between the ages of 2 and 3 is positively associated with their relative vocabulary knowledge in their two languages. So, for example, two-year-olds who have more input in Spanish compared to English are faster to access words in Spanish than in English and will learn more words in Spanish than in English. The language-specific relationship between word knowledge and processing suggests that a child’s skills at recognising and learning words are not just a product of their own individual characteristics, but are closely related to the input they receive.

The importance of cognates and the role of crosslinguistic alignment

Children can become bilingual in many different ways and may hear and use their two languages in different contexts, at different times, and with different speakers. When it comes to vocabulary, their knowledge is distributed across two languages, and it may only partially overlap. Bilingual children may know some words in only one of their languages if their opportunities are limited to that language. For example, they may know mathematical terminology only in the language of schooling and not in their home language if they never use the home language for school homework. Or they could know the names of certain foods or religious practices only in their home language if they never encounter them outside of their family or home language community; these are known as ‘singlets’. This is completely natural and it also applies to adult bilinguals. At the same time there will be words that children know in both languages because they refer to concepts that they routinely encounter across different contexts in both languages; these are known as ‘translation equivalents’ or ‘doublets’.

A subset of these doublets holds a special place in bilingual vocabularies; they are words that sound similar, have similar meanings and can also have similar spelling; for example, the English word ‘banknote’ and the Ukrainian word ‘banknota’. These words are called ‘cognates’ and bilingual children – like adults – typically show a cognate effect, ie, they are more accurate in correctly naming or identifying cognates and they are also faster to do so because of the crosslinguistic overlap of sound and meaning.

There are, however, factors that constrain the cognate effect, both due to characteristics of the words themselves, and because of children’s individual differences in terms of their bilingual background. Words that are more frequent and that are acquired early are more likely to benefit from a cognate effect. Correctly identifying cognates can be easier in the non-dominant language because the child already has a good representation in their dominant language and is more likely to guess what the unfamiliar word might mean.

Cognates can therefore facilitate vocabulary acquisition and children learning pairs of languages with more cognates (eg, Catalan and Spanish) make faster progress in increasing their overall vocabulary size than children learning two languages that are less related to each other and that will have fewer cognates (eg, English and Mandarin).

Another important thing to remember is the cultural dimension of word meaning. Even when two words are cognates it does not necessarily mean that the meaning is the same across two languages². For example, while the words ‘coffee’ and ‘caffé’ are cognates across English and Italian, the cultural connotation of these two drinks is quite different across these two cultures. Being bilingual often means being bicultural too, and learning words in two languages relies on a cultural and social understanding of what words mean not only in different languages, but in different cultures.

Discussion

Although English is spoken in both the UK and USA, some words that sound the same have different meanings in the two countries. For example, ‘mean’ is the opposite to ‘generous’ in British English but ‘nasty’ or ‘bad-humored’ in American English. Can you think of words that sound similar in different languages but have different cultural connotations?

References

1. Hurtado, N., Grüter, T., Marchman, V. A., & Fernald, A. (2014). Relative language exposure, processing efficiency and vocabulary in Spanish–English bilingual toddlers. Bilingualism: Language and Cognition, 17(1), 189-202.
2. Cultural influences on word meanings revealed through large-scale semantic alignment. Bill Thompson, Seán G. Roberts & Gary Lupyan. Nature Human Behaviour, 4(10), 1029-1038.