Post written by François Grosjean.

As I wrote in a first post on this topic (see here), when I put the finishing touches to some of my posts, I often tell myself that the phenomenon that I have just described remains enigmatic. In this post, I will come back to topics relating to language choice, code-switching and language interferences (also known as transfers). We know a lot about each topic at the linguistic level–each phenomenon has been described extensively–but much less at the cognitive and neurolinguistic levels.

When speaking to monolinguals, or to bilinguals who only know one of their languages, bilinguals are incredibly adept at adopting the right language quickly and efficiently. They also excel at keeping to just that language and closing out the other(s) (see here). If they speak the language fluently and have no accent in it, they may "pass" as monolinguals.

Thanks to research in the area, it would seem that it is the speaker's executive control which manages cognitive processes such as attention, working memory, planning, inhibition, mental flexibility, and so on, which also manages language control in bilinguals. The brain structures involved are most probably the caudate nucleus, the prefontal cortex, the anterior cingulate cortex (ACC) and the supramarginal gyrus, according to neuropsycholinguist Jubin Abutalebi.

Although much advance has been made in our understanding of language control, and the brain structures that underlie it, when it comes to specific bilingual behavior, things become a bit more opaque. For example, we know that when bilinguals are in a monolingual mode, the other language(s) may sometimes seep through in the form of interferences, that is deviations from the language being spoken stemming from the influence of the other, deactivated language. They are of two sorts–static and dynamic–and the mystery revolves around the latter. These are fleeting intrusions of the other language such as the accidental pronunciation of a sequence based on the rules of the other language, or the momentary use of a word or grammatical structure from the wrong language (see here).

Bilinguals are often not aware of interferences such as these since they are accidental. It is only when their interlocutor asks what they meant by word X, or corrects their utterance, or looks at them in a strange way, that they realize, after the fact, that the other language has slipped in. We know a lot about the linguistics of interferences–whole books have been written about them–and yet they remain enigmatic in terms of the underlying cognitive and neurolinguistic operations that are involved in producing them. A whole new area begs to be researched at these levels.

Bilinguals also spend time in a bilingual mode, speaking to other bilinguals who share their languages and with whom they feel comfortable intermingling their languages. They choose a language based on who they are speaking to, what the context is, what it is that they will be talking about, and what it is they want to accomplish during the interaction. They then bring in the other language when the need arises. One common way of doing so is to code-switch, that is to shift completely to the other language for a word, a phrase or a sentence before reverting back to the base language (see here). They can also borrow, that is bring in a word or short expression from the other language and adapt it morphologically, and often phonologically, into the base language. There is also the possibility of taking a word from the language being spoken (the base language) and adding a meaning to it based on a word in the other, guest, language (see here).

All these operations that take place in a bilingual mode are, once again, well described linguistically but still in need of extensive research at the cognitive and neurolinguistic levels. For example, what happens when a person, such as a teacher or an interpreter, is in a bilingual mode perpceptually, i.e. both languages are being perceived and hence are active, but only one language is being produced? As we saw in an earlier post on language teachers (see here), even though many usually only use one language in class, i.e. the language being learned by their students, they have the other language ready to intervene in case someone asks a question in it or produces a code-switch. But they rarely allow themselves to intermingle their languages, although this is changing in some school contexts.

As for interpreters (see here), they too are in a bilingual language mode when listening: they have to hear the input (source) language but also the output (target) language, not only because they have to monitor what they are saying but also in case the speaker uses the target language in the form of code-switches. However, they must deactivate or inhibit the production mechanism of the source language so that they do not simply repeat what they are hearing, as sometimes they do when they get very tired.

In sum, how does the bilingual manage to be in a bilingual mode during perception but, at the same time, be in a monolingual mode during production, as in the case of language teachers and interpreters at work, among others? This is another mystery which hopefully will find a clear explanation, both cognitively and neurolinguistically, in the years to come.

Photo of the Sherlock Holmes silhouette from Shutterstock.


Abutalebi, Jubin (2008). Neural aspects of second language representation and language control. Acta Psychologica, 128(3), 466-78.

Grosjean, Francois (2013). Speech production. Chapter 3 in Grosjean, François & Li, Ping. The Psycholinguistics of Bilingualism (pp. 50-69). Malden, MA & Oxford: Wiley-Blackwell.

"Life as a bilingual" posts by content area.

François Grosjean's website.

You are reading

Life as a Bilingual

Misunderstanding in the Multilingual Workplace

Picking the wrong words and meanings in multilingual communication

Bilingualism in the Sky

Pilot and air traffic controller communication in a foreign language

A Multivocal Self

Growing up in Korean, English, Chinese and Japanese